I own pair of Harbeth’s (M30.2 40th Anniversaries), as well as Spatial Audio, Magnepan, and Focal. They are the sort of speaker that is easy to like, until you compare them with others. Alan Shaw has found the perfect model: build a beautiful-looking box speaker (no curves = cheaper), stuff it with cheap parts, and then engineer it to measure well. He must have the best margins of any domestic manufacturer (i.e. not Chinese-built). People are happy and he is happy, supporting British industry and making a living at it. I don’t fault the guy. If you can use cheap parts and get the sound right, why not?
@@Unpreeeedictable very cool. I have a smaller room, so was considering the new M6 Sapphires. I assume they are far better than the Harbeth 30.2 Anniversary (which is another speaker I was considering as well)?
@@rlowes totally different. The Harbeths do a couple of things very well, especially acoustic and vocal. Think singer-songwriter. The Spatials do EVERYTHING well. My Harbeths are for sale ☺️
I suggest the presenter reads Harwood's BBC paper concerning speaker cabinet construction, which concluded that the best way of combatting cabinet wall radiation was to use a thin-wall approach, allowing damping measures to be made very effective (if I remember correctly). Useful reading for anyone, actually.
Thin wall design is a great approach to speed of decay and using bracing at corners all helps, but that was related to cabinet size and surface area. The LS35a is an example of a small box which has great imaging and a smooth response but for a definate purpose, small outside broadcast trucks and control rooms. They sound really great but, there are limits to what they can do above 20 watts of clean power. These speakers had massive amounts of correction to overcome what was commercially available drivers from KEF at the time. So not tremendously accurate but fit for the intended purpose.
@@adrianfisher9641 Do note that the article refers specifically to the LS3/6, which is a LOT bigger than the LS3/5a - but the principles covered are anyway not exclusive to speakers of any other dimensions.
A lot of measuring but no listening! My Harbeth P3ESR 40th Anniversary sound absolutely incredible and the speakers are designed to be played with the grills on.
I've heard thousands of speakers using cheesy parts. If you think those sound incredible then you have really heard anything that is really incredible.
@@dannyrichie9743you passed English, let’s test that theory. “If you think those sound incredible then you have really heard anything that is really incredible”
The Harbeth P3ESR are reference quality speakers and have an uncanny realness to the sound, especially voices. They are smooth, non-fatiguing and have ultra low distortion. Listen to them and forget about the crossover part count. Many speakers would sound much better with additional crossover network components- such as Zobel network to flatten the impedance and make them more tube friendly; BSC or baffle step compensation that compensates for too much midrange being reflected off of the cabinet baffle, higher order crossovers needing more parts to keep the drivers in their smoothest frequency zone that greatly reduces distortion and lastly notch filtering networks to tame any peaks that the drivers might exhibit. Looks like an abundance of components but they work and work extremely well. Just listen to them, they sound glorious as do all Harbeth models.
They certainly did their work to create a good level of accuracy, but the high parts count using extremely poor quality parts really sucks the life out of the music. These are far from reference quality.
@@dannyrichie9743 Your assertion the speaker has no life is a subjective evaluation. Many experienced people have the opposite reaction. And so it goes.
@@dannyrichie9743 so you do actually set these speakers up and have a listen to music? Do you do that before you open them up and see the mess inside? I fear that if you didn't, just knowing what's inside them would give you a negative bias towards their sound. It's just confusing because they measure so well.
Actually if that’s a P3esr, Stereophile didn’t measure any cabinet vibrations that they thought would cause colourations with these particular Harbeths. Also interesting to note that the new XD version of these apparently has better crossover parts and better wiring and binding posts (and a fair bit more expensive I believe). Would be very interesting to compare the internals. I do think that saying these should be “avoided” is a bit much. Tons and tons of people love these speakers - they must be doing something right. If you want to buy them specifically to upgrade the internal parts, sure yes, avoid. But to recommend avoiding overall, that’s taking things a bit too far I think. Are they overpriced? Like many other production speakers, just looking at parts, sure. But that’s an overall value judgement for the purchaser. Good luck DIYing a cabinet that looks like this.
There are "tons and tons of people" who smoke crack. But I don't believe "they must be doing something right", and I would not recommend it either. An appeal to popularity is not a good basis for forming one's opinion.
@@jdlech I’m talking about hardcore audiophiles and reviewers who consider these to be amazing nearfield monitors for pleasure listening and some who even own a set themselves. Do you think people who listen to tons of gear that decide to purchase these for their own systems are all deluded or lying for some reason? To dismiss a speaker purely by looking at the parts is foolish if you actually care about how these things sound IMO. But everyone is entitled to their own opinions and approach to things. If you value parts, yep these are not going to satisfy. And maybe they’re not detail and resolution monsters because of this, not sure, I’ve never heard them. But it doesn’t mean they’re not really enjoyable to listen to. The problem with this kind of “you must avoid because of this objective quality even if other highly experienced people say they sound great” is getting into Audio Science Review territory where only the objective matters and the subjective doesn’t. Also, the “if everyone jumped off a bridge, would you” comeback is a pretty pointless argument.
@@rlowes There is certainly a balance to be struck. Never underestimate the power of social proof, visual bias, and the ability of our brains to adjust to stimuli over time. Not to say that these sound bad (I've never heard them) but man social proof and visual bias, alone, are strong Jedi mind f's! Over time I've noticed the ASR affinity for Harmon. Bias strikes again.
@@NakeanWickliff absolutely agree. To be clear, I’m not claiming these absolutely must sound amazing, or that I would even like them. I’m am saying however that there’s a pretty high likelihood they would sound great to SOME people who heard them. And to simply say they should never be considered or purchased due to parts quality is BS. If that’s the case, 95% of the production speakers out there should be avoided.
@@rlowes Yup, we're in alignment there for sure. I actually have a pair of Klipsch Heresy speakers that I have gone full circle on. First speakers I owned when starting to actually critically listen. Then I measured and saw the lumpy response, and also compared them to open baffle speakers. After that, I couldn't unhear the Heresy boxy sound and stopped listening to them. Now, I've come back around to loving them in the living room even knowing they're not linear and they have vocal bloom from the box resonance. It's a charming sound to me now and I really like them for what they are. In the end, It's all subjective. Kinda like cruising in an old classic car. You love them because of the beauty of their quirks not despite them.
I once took my GR Research AV-1’s amp shopping. The shop was a Harbeth dealer. I would say that the trounced the P3ESR’s (the original version). The Harbeth’s sounded a bit fuller, but likely due to that cabinet resonance. The dealer was in awe. He spent more time listening to my speakers than I did listening to his amps! I listened to some of the bigger Harbeth models while I was there, and couldn’t wrap my head around to performance to cost ratio.
Hey Danny... I own the sb3's, and I am a musician... When fed a good source they sound like the instuments should. They have real defined soundstage. , and great imaging... Also in the setup manual they show the speakers tweeters facing up. I wouldn't think that would make a difference with your on axis measurements, but you would know better than me. Lastly, one of the main things multiple respected reviewers said about them was that they were very accurate and flat. The audio world is wacky!
The reason the nht is woofer up is actually a psychoactive reason. Psychoacoustically, people tend to locate high frequencies at a point higher than the actual location the sound is coming from. So putting the tweeter down puts the heights psychoacoustically at the level of the woofer. I remember reading an AES paper 50 years ago where they conducted the tests and wrote the paper.
Interesting stuff. I've always liked Totem monitors, and it would be interesting to see how they measure. They use a minimum number of higher quality parts so I always though that's what I was hearing, but I'm sure they don't measure flat, and not sure about spectral decay but it would be interesting to see. They also hand-match their parts in pairs so the left and right speakers match. I've got their Rainmakers, Sky Monitors, and Kin Monitors now around my house. I chose all of them just by comparing them to similarly priced speakers, and these sounded better. It would be cool to see if I could take the Sky's up a notch though. Hmm.
I have a friend of mine who has been a professional loudspeaker engineer for a very long time. Put this way, he was designing professional products 20 years before GR Research jumped into the DIY market. His experience includes designing numerous well reviewed products for a number well known brands across many products lines from lifestyle Bluetooth speakers, car audio, in-wall, on-wall, in-ceiling outdoor, as well as high-end 2-channel, home theater, AND DIY KITS. I asked him for his frank response to the Harbeth section of the video. He has absolutely no connection with Harbeth, and offered this remarks strictly as a critique of GR Research: To start with, what GR Research portrays in its You Tube video is not in any way an objective evaluation of the Harbeth product. GR Research makes its money selling people on the concept that they are capable of improving on other manufacturing design work, mostly by the simplistic concept that all it takes is higher grade crossover components, and by touting the superiority of his own DIY kit products. While this is his prerogative, again it is neither objective, and definitely is self-serving. To this gentlemen’s credit, he recognizes that the Harbeth speaker measures well, noting the well done grill concept that is part of the Harbeth line. He also mentions that the impedance has no issues, but also vaguely mentions the tweeter impedance is “kind of high at that end”, which is a ludicrous statement. The reality is that all voice coil type tweeters have a frequency dependent reactive (AC inductive and capacitive elements as opposed to just being a DC resistance) rise in the impedance and it is normal for it to be relatively high at 20 kHz. This is deliberately misleading. Next, the host points out that the box has obviously high resonance and is somehow deliberately “lossy”. His claim is that this makes the low end bass “boomy” and colored in a way “that people like”, which is also not true, at least according to Dr. Floyd Toole’s extensive work at the National Research Council (Canada) and Harman International on explaining the correlation between objective measurement and listener preference. He then proceeded to validate his criticism by taping the box with his fingers. Besides the fact that I really didn’t hear any “ringing” when he tapped the box, if you look closely at the back of the box, there is a layer of damping material stapled to the rear baffle, something traditional in British speaker engineering, and I suspect, though the host didn’t bother showing you a view of the inside of the enclosure, this same material is also used to damp all four walls of the enclosure. I might have been more receptive to this off-hand and unsupported conclusion had he done something like offer accelerometer plots on the Harbeth enclosure verses something he considered more appropriate, but that was not offered. While I find his comment on the impedance and the enclosure annoying, his comments on the crossover, which he draws, a really inaccurate conclusion from, are also not at all accurate. First, parts count. He implies that this is a ridiculous number of parts for a two-way network, and that the network has over “30” components (actual count is 20). Without the schematic, it’s really hard to judge, but here is what I see. Five inductors are likely used as three in the woofer circuit, 2 as part of a 3rd order low-pass filter topography, with the third inductor likely used to control some top-end peaking in the woofer. I suspect the other two smaller inductors are used in the tweeter circuit, one as part of a 3rd order high-pass filter topography, and the other as either a LCR conjugate circuit on the tweeter resonance, or as a response contouring network. He makes a big point of describing these crossover parts as “cheap” bad sounding network components, which both an exaggeration and misleading. The capacitors are all polypropylene film caps, but obviously in a package format this amateur engineer doesn’t recognize. They are not Mylar caps! (I had a picture of the cap but RUclips will not allow me to post it here) Polypropylene Film Capacitor Temperature Range - 55 C to + 85 C Dissipation factor 0.1% MAX. at 1KHz, 25 Degree Capacitance 104 F Voltage Rating 400 v The 4 larger capacitors on the top left of the Harbeth circuit board are obviously paralleled together for the woofer circuit. So, instead of using a single non-polar electrolytic, which frankly would have worked adequately in this application, Harbeth spent a lot of money paralleling four much more expensive poly caps in this circuit where one cheap cap would have done the job! All of the capacitors in the tweeter circuit on the left side of the circuit board are in pairs, one large cap and one small. This is an obvious detail the host has overlooked, as not only are both caps in the pair polypropylene, but the smaller poly caps are bypass caps, and well known technique in high-end speaker design which is done to overcome parasitic resistor and inductance at the upper end of the caps operating frequencies…it tends to provide more “air” in the tweeter’s subjective perception. There are 11 capacitors total, but in effect, the four in the woofer circuit are acting as 1 cap, and the other 8 caps in the tweeter circuit are actually just 4 appropriately bypassed capacitors. His gasping at the parts count is really ingenuous and misleading to consumers who are not loudspeaker engineers. In terms of parts count, if we count the paralleled caps and bypass caps as just 1 single cap, the parts count is now just 16 parts, half of the horrible gasping “30” OMG parts the host is making such an issue over. He also states that this many parts can lead to a smeared image, loss of high-end detail and soundstage. Sorry, but while that may be his opinion, it’s just not true. The host seems to have come to the conclusion that the number of parts used is because of poor drivers. That is simply not true in this case either. The number of parts used in a crossover varies for a number of reasons, but generally it’s either cost or design philosophy. For example, a typical high-end tweeter high-pass network by itself might contain as many as 12 parts, 3 in a 3rd order topography, 2 attenuation resistors to balance the circuit, 2 more resistors if there is a level switch for the tweeter, 2 in a CR conjugate to contour (voice) the upper harmonics, and 3 parts in an LCR conjugate circuit on the tweeter resonance. This is not excessive and none of these parts are related to SPL correction. The host at GR Research seems to think that changing crossover parts for more expensive parts is his hill to die on. While I may not employ ferrite core inductors in my work, they are not as horrendous as is being pointed out and replacing them with air core inductors may not really save the day as he implies. The Rogers LS35a employed ferrite core inductors, and I don’t believe that the LS35a sounded awful or even mediocre to most listeners. Like all design decisions in loudspeaker engineering, most are often some kind of tradeoff. Cored inductors have substantially lower loss due to have less wire and less dc resistance, and if well designed, do not always present hearable distortion when they saturate at high current levels. In summary, GR Research’s conclusion that this Harbeth two-way monitor is a lost cause for his company to improve is horribly misleading, and simply not the case. The number of crossover parts is not a negative, and his not even being knowledgeable enough to know that the capacitors actually are polypropylene really puts in question his portraying himself as an “expert”. His failure to explain that the number of caps is due to a sound engineering practice of by passing and paralleling capacitors in a crossover network in order to deal with parasitic capacitor inductance and resistance, does not impress me, and is decidedly a disservice to his listeners.
Yeah, there is always someone out there that thinks they know better than industry professionals. He might want to stick with the lifestyle Bluetooth, car audio, in-wall, in-ceiling and outdoor market and leave the high end market to those that specialize in that area. And he thinks that this is not an objective evaluation? Measurements are nothing but objective. So is reporting what I see. If he really thinks that mentioning the tweeter impedance is high is a ludicrous statement, then he really doesn't understand the importance of balancing the impedance or how different types of amps react differently to different impedance loads. What I described regarding the box of the Harbeth's was just restating what they are known for. If he knew anything about that company or their products then he would know that. Paralleling multiple caps to make a larger value is not a plus, and when the values are different it causes some smearing from the dissimilar dissipation rates of the caps. I suspected they were paralleling some value but didn't feel it was worth beating them up further. Caps are also likely polyester and not polypropylene. They are known for using polyester film caps. See this link: www.widescreenaudio.be/wp-content/uploads/2017/02/shl5_crossover_filter.jpg The ones used in this model aren't even that good. And absolutely the use of multiple parts of really low quality (small gauge iron core inductors and sand cast resistors) will have a cumulative effect. The more or them you use the more it degrades the signal. If your buddy doesn't get that then he really has no business responding to the video in a critical manor.
Great stuff. On the Harbeths, though: It's a misconception to say they were designed with a resonance to give extra bloom. That's not it at all. It's a common belief, though, and it really gives British engineering short shrift. This kind of cabinet design harks back to the BBC, which decided, in their in-house monitors, to tackle cabinet resonances by damping thin walls with lossy materials--bitumen, not No Res, but same principle. They did this rather than make the cabinets either heroically thick, or from inert materials, which were unavailable at the time, outside of the aerospace industry. Instead, their engineers chose to permit resonances, so long as they were low in frequency and fairly low Q, so they'd be nearly inaudible. It was, at the time, the most cost-effective way to get low coloration enclosures. In other words, this wasn't designing IN coloration, it was just crafty engineering to design OUT coloration.These speakers are second-cousins to the BBC LS3/5As and, as such, they share the design philosophy.
Translation: there are better, more advanced cabinet materials available today, but British companies haven't kept up with the tech advances. Probably because they "like" or have gotten used to the traditional colored resonant sound from the old-school thin wall cabinets with bitumen.
It sounds like the Harbeth was going after a “sound” pardon the pun. That they achieved at all cost expense of finding the cheapest parts worked out beautifully for them. That’s probably a speaker put into a category like people who like the harsh piercing sound of a class D amp. Compared to the nice smooth sound have a good high quality tube amp. The Harbeth speaker crossover kit would probably be approaching $600 in parts to identically duplicate every component on their crossover and way as much is the speaker hanging off of the back. Not to mention having to add mechanical speaker supports gluing them in to the inside of the cabinet for extra support and adding all the no-Rez. To the inside of the cabinet would take away that mechanical residence of the speaker singing that the speaker manufacture was relying on to give them part of their sound signature. Totally agree on this little English speaker if you’re not happy with the sound don’t try to fix it give it to somebody who enjoys it and start over with a new different speaker
Adding a so called higher quality resistor, capacitor, or inductor will do nothing to better the sound. The value of the components used, meaning the resistive or capacitive numerical value, is all that matters. The engineer masterfully creates his crossover with common parts , of which the value is carefully selected. You cannot argue with the many that find some of these speakers sound just fine as is. Most of what you recommend as upgrades are basically overkill.
Tony, when I sold hi fi years ago, the Japanese brands would all use parts with only 5% value tolerance. Everyone else would use up to 20% so IMHO it’s the tolerance quality of the parts that fundamentally counts. If you have say 5 cheap parts in a board and they are all graded at 20% tolerance, how can you be sure that they all work at near value all the time?
Tony, if good enough is good for you, that’s great. But we on the outside like to enhance or equipment to the better for the sound and not the measurement. We want better sound and depth. Then you need to change and upgrade. Look at speaker designer Mats Buchardt. He’s been selling the S400 Mk1, for some year. Now he upgraded them to Mk2, with much more spec crossover with great components and the sound improved. Look at all the reviewers verdict. What is the conclusion of this? Not only Danny knows the value of great parts. The way you write makes me wonder if you have designed one of these speaker/crossover that Danny displayed.
Any component in series from the amp to the speaker voice coil is going to affect the sound. That's just basic electronics 101. The question is whether or not it can perform as needed. Reasonable tweeters crossed over with poly caps? One can hear a soundstage. With an electrolytic in the same spot, the soundstage effectively disappears.
@@seejayfrujay I'm afraid that this viewpoint displays a complete lack of understanding about how electronics work. Passing an electronic signal through a more expensive resistor of the same value as a cheap one, will sound identical. The electricity doesnt care how expensive a resistor it passes through, a resistor is a resistor. It's possible more expensive ones could have better tolerances for matching, and may have better longevity, but they won't sound different. that IS electronics 101. The fact you think otherwise, suggests you don't actually know electronics at all.
@@1061shrink1061 Always always the ad hominem when someone is talking through their hat. No, a cheap voltmeter will not tell you what is happening. And yes I do, from vacuum tube to MOSFET. My background is physics. You will observe a marked difference in the electronic performance of a electrolytic capacitor versus a poly or other type when you perform frequency sweeps and pulse testing with the appropriate equipment. It takes a bench in other words.
There are two speakers that are widely-used as near fields in pro audio, the Yamaha NS10 and the BBC LS 3/5A. The NS10's can sound rough, but they give engineers good insight into how their mixes will translate to lower-end consumer audio gear, such as car stereos, boom boxes, cheap hi-fi's, Bluetooth speakers and ear buds. The LS 3/5A was made for BBC's studios with the emphasis on mixing/mastering for speaking voice reproduction over the radio and TV. They are specific tools of the audio engineering trade, but neither speaker would be my choice for home hi-fi music reproduction. Oddly, the Harbeth version of the LS 3/5A (as reviewed) is aimed at wealthy audiophiles. Go figure?
I designed an upgrade for the Yamaha NS10 that is being used in about 10 different studios. I am surprised that they were ever used in mixing or mastering. They are not that accurate and don't sound very good either.
@@dannyrichie9743 Amen to that. I cut my teeth writing and producing electronic music using NS10m's; they always sounded lazy to me. No matter what I programmed to have pizzazz and color, they would drearily and reluctantly dribble out like a terrible waiter. Of all the monitoring set ups available to me at the time, I will have to say they were the most accurate - a friend with some PA kit in their living room, or some luxurious and well-presenting headphones were much more satisfying to playback what I thought I was creating. But at any level the NS10s just sounded like they were being played very quietly and seemed positively gleeful when the room we were in just became overloaded, as did I. An upgrade sounds interesting - would it involved replacing the cabinet and all internal components and drivers ? Thanks for all these videos, it's great to have learned another approach to speaker engineering, and it's one that has been added to my own plans.
I've never heard a Harbeth speaker that I liked, and have always been baffled by the overwhelmingly positive press. Danny's analysis is "consonant" with my experience.
They are expensive so people like to show that off even though 95% of everyone couldn't care less or tell any difference anyway. People on stevehoffman's praise them, I am guessing so they can let others know they spent so much money.
Only Danny can take a pair of speakers that a buzillion people love (and not just these three) and “prove” how lousy they are - because his graphs tell him so. I would love to see how many speakers Danny could listen to without seeing them, and never having heard them previously, and tell us which ones had “poor” caps.
I do think Danny would fix that, if he plays his own list of music he knows, then he would be able to say what’s good and bad. The other stuff are called experience. Many years of experience of speakers and speaker parts.
..and he’s right probably 99% of the time. I have done some crossover upgrades myself on some highly renowned (B&W matrix models, several) loved by a gazillion. Along the lines of Danny’s, and the difference was not subtle. Similar on budget models like KEF’s Q150 (used some of Dannies parts on that one). Again, not subtle. Sounded really good before. Sounded great after. My wife served as the blind tester, and not being an audiophile said things like: “that one sounds like it has a little of a haze. This other one is much clearer. I hear more details.” Her words, not mine. Anyway, I think Danny’s points maybe overstated, but no doubt, take a speaker that sounds great and is loved by many, give it some of his crossover treatments, and I am convinced that it will sound even greater. So much so that you wonder why the designers did not do it that way int he first place.
And that is why I enjoy watching this channel... Because you can look past a pretty box. Some people get so fixated on the look of a speaker that they can't hear the sound that comes out of them.
I have the M.L. lx 16, there is a setting on my Marantz SR 5011 that rolls the highs off a little, it made a huge difference, plus I pulled them out a few more inches, wow I have fallen in love with these little beauties all over again, the sound and clarity is incredible. I have owned quite a few speakers, but these have become my favorites.
Hello fellow Martin Logan owner. The LX16's are a little old now. The new M.L. Motion 35XTi are very very nice for the small bookshelf form factor. You might want to audition them at some time. Not saying make a change, but the updates have made a nice improvement in the sound, and you would likely notice the differences. SKOL
Hi Danny. I traded an email with Alan Shaw of Harbeth. He gave a detailed reply on the Harbeth user group about caps and inductors etc. He mentioned that you had not mentioned in your video, that the Harbeth has poly caps. He told me those poly caps are good quality and will last a 100 years. Harbeth sells speakers faster than they can make them. They must be doing something right. Have you listened to them?
My pair of NHT Super One XU speakers seem to have a similar issue as the SB3, as I've wound up placing them upside down or else they sound out of phase. I know they're entry-level speakers with cheap xovers. I paid $65 used for them, and they sound pretty good, so I can't complain.
Interesting on the Harbeths - perhaps it’s designed so each iteration can have a slightly improved crossover and thus seem better than the last version? I have always used LS3/5as, and bought some external crossovers built for them by a company called Cicable, and designed by someone who probably knows most about BBC type loudspeakers and crossover design than anyone else. These gave a massive improvement in THD reduction and increased resolution and transparency. Now I have Stirling Broadcast’s V3 version of the LS3/5a which also uses a crossover using higher quality components (the pcb is about the same size as the Harbeth and also mounted on the back of the cabinet, but uses huge air cored inductors and also has less components on it), and it also has that great transparency and resolution, and with modern SEAS drivers that match very well it images fantastically and has better power handling. I use them with their bass extenders plus some M&K subwoofer crossed over around 60Hz and they are the best I’ve heard.
That's my NHT!! Sad it didn't qualify, but ready to do a diy build. Top binding post was always an annoyance. I like the sound in general, it extends down but I do use a Parts Express DIY Kit DSP subwoofer. Shopping for a DIY kit right now. Thank you GR for your time and advice!
I have a xls encore kit sitting in a box doing nothing,the xovers are put together and sent Danny a pic and he said it looks good.i would sell for 175.00
I think it’s difficult for you to pass comment on the Harbeth’s without actually listening to them. I am English, and I worked as a producer in BBC radio where the studio monitor’s were either Rogers, Spendor or Harbeth’s, all of which were fairly neutral and accurate. I put my trust in BBC engineers. I also think your comments on the cabinets construction are misleading, they are a “thin walled” construction created by BBC engineers for small monitors like these and the Rogers LS3/5a, which I own an original pair of - there is no boominess with the cabinets, there is no resonance either. Yes, they have a “British” sound which is mainly uncoloured so you can listen for extended periods without getting fatigued. I would suggest you get a pair and hook them up to a decent system then pass judgement. As a lot of people are saying here; it really is what they sound like and in my opinion these little Harbeth’s sound great, as do all the Harbeth range. There are many recording and mastering studios and TV control rooms in the uk that use Harbeth as reference monitors, I’m assuming the guys that select these speakers know their stuff.
I think your not aligned with the intent of the video, and I’m not sure the speakers were not listened too. Don’t know if that was ever confirmed, and I’d find it unnatural if anything evaluated isn’t listened too. The point of the video is to use data to baseline the conversation and assure everyone is aligned to describe how speakers are putting out frequencies from the lows to the highs, and how much the speaker changes the sound as compared to the music in the studio. It seems Your contradicting the video because you’re concluding based upon your preference, and what you think. This is that audiophile subjectivity (“British sound”) I’m sure there’d be no comments about how thick the cabinet is, if the output meets the intent of the data output. Likely, an observation based upon the output performance as it was measured
It's also not fair for you to pass judgment on GR Research without listening to one of their top line 2-way monitors and comparing it to your beloved BBC speakers... It may blow your whole worldview right out of the water...
Well said,i can't believe Danny was slighting the engineers like Dudley hardwood,and his peers,so far all he knows is "buying off the shelf drivers and build crossover with sonic caps and loudspeaker design is over. I had started my audiophile journey with p3esr and ended with superhl5plus, In this process i tried a heck of 60 pairs of speakers before reaching the verdict that harbeth's sound is pretty close to perfection.
Mastering music and listening to it are two very different things. In masteri g you're looking not for accuracy. But, instead a sound that you as the engineer want for a track. If your ear has been attuned to the color of a speaker type, such as the Harbeth sound, you're going to want that color when in the edit. Listening often times means much the same in that you're looking for a sound you as the listener desire. Given that many British speakers were designed to accommodate the BBC of the 60s and 70s that particular group of engineers, that were used to a very different standard of reproduction than many would accept today, many still hold on to that "British" tone. Even if it means reproducing the original sound poorly. As a reference, consider B&W speakers. They've been designed as British speakers to be the opposite of that warmish tone that many associate as being British and instead sound just downright harsh. The Harbeths like many Tannoy products sound like something an engineer from 40 years ago would find pleasing. If for any reason the fuzziness that kind of reproduction presents the listener with helped to mask some of the noise LPs and tape tended to cause. We have better source materials today and don't need to veil music to remove noise beyond equipment interference. As such, speakers needn't be built this way anymore.
I always wondered what Danny would do if someone sent in a pair of Harbeth. “Surely Danny wouldn’t tamper with Alan Shaw’s work” is what I always thought to myself. Turns out I was right, but for a completely different reason. Disappointing, but must be one of the most highly regarded monitors out there for a reason… how did it sound Danny?
I was just sent one of them. With that level of parts quality a lot of the life would be sucked out of the music even if the drivers were the best drivers in the world.
Now what if these Harbeth speakers sound good, to a lot of people they seem to do, even though they contain cheesy bits!? They sell upgraded versions of these speakers, that probably sounds better. The upgrade is of course better components, the drivers are probably the same. Harbeth have made a well aclaimed business out of making speakers people like, hours spent testing and listening during development must mean something. Two listeners rarely agree on everything they listen to, regardless of inside components. I bet that some people listening to Dannys speakers will say they don´t like the sound, again regardless of the interior bits. But i totally agree that it´s a shame and wrong saving money inside when we are talking about expensive products!
There's nothing wrong in using steel cored inductors if applied properly. If everyone would use expensive 10ga inductors and fancy poly caps everywhere, most of loudspeaker cost would be in crossover not drivers. I guess there's is some priority when it comes to designing and it makes sense to use expensive parts only where it counts. Zobel networks or RLC traps don't need expensive parts. Only parts that are in series with transducer benefit from better quality parts. Tons of make up and facial surgery won't make every woman into supermodel, or racing slicks won't turn average sedan into race car. There are same rules which apply in proper engineering. On top of that, Danny doesn't show the crossover schematic and where these cheezee parts are utilized. Electrolitic caps - maybe someone should take a look into service manual of their electronic equipment: amps, DACs all use them in the signal path.
GR I appreciate the effort but I have to respectfully disagree on the Martin Logan LX16. On the 30 degree off-axis response, much can be attributed to Mayer-Norton impedance aggregations, perhaps from bipolar junction inductance, somewhat akin to stator hysteresis and non-complementary diode configuration. The crossover inhabitance ratios, when viewed within impedance truculence vectors, are stationary, yet sinusoidal. The big dip at ~ 3500 is more measured than perceived by the listener, due to cubic ionizaton when viewed withing a Thévenin linear framework (measurement) rather than a inductor co-valence (listener). The vertical off-axis is excellent as is phase subjugation and transient variance deflection points. A structural recapitulation, especially when woofer isotope superpositions are considered (i.e. Horowitz, Ravel) can be quite apparent during on-axis phase subduction when aligned with polarity confluence. As always, I appreciate your reviews and keep up the good work!
Sir Maxwell, thanks for that entertaining analysis. I'll be sure to avoid bipolar junction inductance in the future. When my speakers' junction inductance becomes depressed, the high frequencies fade. I try to keep them as manic as possible!
Great insight into speaker manufacturers I get the impression that this guy has to tread a fine line re his impressions of speaker companies products and quality Seems at first glance most speaker manufacturers cut corners to save a buck or two at the expense of sound quality....
Hi Danny, you can add another to this list: ELAC UB51 bookshelf. You evaluated these speakers for me earlier this year and we agreed they were not good candidates for upgrade due to crossover complexity & size vs cabinet access.
Hi Danny, your videos are inspiring and educational. Do you have a video showing the equipment and exactly how you measure everything in your graphs. For an amateur like myself, it would be invaluable to learn as I'm wondering how I can replicate this. Thx.
Interesting on the Harbeth . As the driver are supposed to be be high quality control drivers . Maybe with the right crossover they might not need that many parts !
LOVE IT - LOVE IT!!! Thanks Danny, as always a great perspective shared. I own [not bragging - not apologizing] Harbeth 30.2, knowing its strengths and (possibly many) weaknesses. I could not go with the 3PSR (sp?) Familiarity with these Harbeth speakers, the family line-up, and the BBC monitor speaker legacy... *I firmly agree with all you said* The Harbeths are "pleasing to listen to" but, I cannot build (amps) and do reference listening through them... I do not trust them, as a way of saying it. I am using (first order) Silverline SR17S for my critical listening at the moment. I would love to hear and will one day experience GR products first hand (hopefully) in my home.
It would be interesting to have those Harbeth drivers measured to see or hear why they needed such an intense crossover. The original LS 3/5 A had 3 autotransformers and 15 more parts. For a 2way, Then there are well regarded 2way designs that just used a cap on the tweeter and nothing else. Strange technology.
The BBC decided on the frequency output matching to within 1/2 db across the frequency range, hence why LS3/5A's and other models are licenced, to get that frenquency shape the crossover needed a lot of components, strangely Kef who originally held the license used cheap components in their crossovers?!?!... still amazing sounding speakers for the radio, which is what the BBC designed them for, as near field vocal studio monitors... shame you need 100w to gett them driving properly at 84db....
The SPL chart from NHT SB3, one of most refined bookshelfs, is completely different from Stereophile, which is in overall quite flat. It has a slight hump around 800 hz but is within 3dB but does not have the serious slump around 3khz (dipped down 7dB) as shown here. As you know around 3 khz is in the midrange frequency range. Dipping down 7dB should be quite audible. I had the speakers but I could attest SB3 is one of sweet sounding speakers in midrange. I definitely have not heard any recess in that frequency range. Something wrong with Danny's measurement.
First, parts as they be, the Harbeth's sound damn good, probably the best small monitor I've ever heard. In the end, that's what matters. Second, supposedly with the new XD versions Harbeth has changed the parts in their crossover significantly.
If that model is the best small monitor you've ever heard then you are in for a treat. Every kit we offer out classes them in every way and can be built for significantly less money.
@@sebd6307 I have to agree. There is no doubt that Danny KNOWs a LOT about MEASURING and he loves big ass parts and crossovers. But, some of the most wonderful, musical sounding speakers don't have those and just don't measure perfect. Sorry. My ears have to be the final judge and if I don't like that flat sound what should I do? Turn in my EARS, lol. Decide that my hearing is IRRELEVANT. Learn to love sound I don't? Sometimes I watch this channel and see those elephant crossovers and I just sigh! Sure you can upgrade everything, but does that really help in ALL cases? I cannot worship at the Temple of Measurement. But, more power to Danny. Because lots of people love HIS sound and do worship him and his designs. It is subjective and that's OK! That's also the PARADOX. That what measures flat does NOT sound good to everyone, but does to some, perhaps even many? But that thought never seems to cross his mind. It is NOT possible?
That third speaker you showed as the final one in your review was staggering! I have never seen so many components used in a crossover. Says a lot about the speaker.
Mentioned cheesy parts in the Harbeth crossover. Suddenly I had the urge to eat cheese. Yumm. Thanks! Also, I'll be looking into your speaker offerings. Thanks!
Those NHT speakers are using that inverted driver pair that i remember from my Mission Cyrus speakers i had back in the late 80s/early 90s. They were recommended to be used at slightly below ear level, like many of the speaker support stands that were available at the time, so that the crossover "beam" pointed slightly upwards to ear level. If I remember rightly!
yeah... they can mean or not mean as much as one wants or needs or doesn’t. i like erin’s method of listening first and then measurements after confirm his listening experience.
Hey Danny,I priced the round trip shipping costs but the Can. Govt. wants almost as much for the import duty (based on insured cost). I am considering one of your cheaper kits (for not much more money) to get an idea of the sound of your speakers. Cheers,Darrin
Speakers are built to a price point, which can always be extended. He's actually praised quite a few speakers. There are, however, many speakers out there do not warrant their price point, and there's nothing wrong with pointing that out.
@@thunderpooch Okay, sure, there are people that enjoy the peaks and valleys that some speakers add to the music being reproduced. However, the goal for the majority of manufacturers is to have a smooth response.
@@thunderpooch Flattening a speaker simply means its response curve will faithfully reproduce the sound of the instrument(s) played back. Anything else is a modification of the original sound. It is the ONLY true approach toward accurate sound.
Hi Danny, I've been looking into the Lii Audio range of speakers. Decware and Cantuck Audio use them in the open baffle 'Betsy' speaker. I've been weighing up whether to get a pimped out version of the XLS Encore or some of the 15" full range drivers and a 15" woofer from Lii Audio. Shipping to Australia is always going to be costly.
The P3ESR is a highly acclaimed and (almost) universally loved mini monitor that’s been made for over a decade. How to square this with Danny’s review? It seems that other factors matter more than having expensive parts in the crossover.
Greatly appreciate your speaker videos. Always interesting. Regarding the slight zooming issue, I think it's your autofocus on your camera. Set it to manual, not auto, and make sure the focus ring witness marks on the barrel of the lens is set approximately to the distance from the lens to your seated position. The lens is searching for focus I think.
As for the Harbeth---If we believe objective measurement indicates quality then we shouldn't worry about the nature of the parts used, only the measured results. Shaw has evidently used parts adequate to achieve the objective results desired. It looks like the maker of the video is trying to play the subjective and objective cards at the same time.
Objective measurements don't indicate quality. They indicate accuracy. That is only part of the picture. Parts quality matter a LOT. They can measure great and sound great, or measure great and sound like a speaker playing through a wet rag. You'll never get great sound from budget level parts that are smearing the signal and cause loss of details and spatial ques.
at 5:00, is the 'ringing' at 800hz, the big blister in the spectral decay, is it caused by the woofer driver itself, the damping inside the speaker, lack of a port, or what ?or just a driver flapping out of control around that frequency? or is it just an issue to be fixed with crossover design?
Stereophile reviewed the SB3 and it got high marks with sound and measurements. So I don't know where this is coming from. I have the SB3 and they sound great. Great bass and soundstage from them.
The one that I just measured had a huge hump in the woofers response with a lot of stored energy there. If it wasn't for that I would have seen what could be done with them. They are loaded with cheesy parts too. If you think those sound great then you haven't really heard great yet.
In the Stereophile review, the issues in the 800Hz range were mentioned several times. However IMO several speakers out there have some type of measured imperfection yet some things some can be heard by some and not others. Take it for what it is and understand that that Danny's goal is and how it differs from reviewers. He's trying to correct the measured imperfections... no speaker or room is perfect, some measure better than others, some sound better than others! ______________________________ From the Stereophile review: "However, I was a bit perturbed by the peak apparent in the octave between 600Hz and 1kHz. This will tend to throw midrange detail forward a little in the soundstage, as well as make the speaker a little intolerant of high-level voices." _______________ "Figs. 5 and 6 show the NHT's lateral dispersion, the latter having the off-axis response curves normalized to the on-axis response. The peak at 800Hz is persistent, which, all things being equal, might be expected to add a slight nasality to the SB-3's balance. As BJR noted no upper-midrange coloration, this peak is perhaps low enough in level that it slips by unobserved, other than its effect on the perception of recorded detail."____________ "The SB-3's step response (fig.8) confirms that both drive-units are connected with positive acoustic polarity, with the tweeter's step smoothly handing over to the woofer's. The undulations apparent in the latter are associated with the 800Hz peak, as can be seen in the cumulative spectral-decay plot (fig.9)."______________ "Overall, this is quite good measured performance, though I was bothered by the existence of the 800Hz peak."
To those who think measuring with inadequate electronic equipment that is simply generation behind the most sensitive instrument we have , our ears. Tubes! They don't measure well but to those who think they don't sound fantastic really needs to get their ears checked and learn how to listen. The same with speakers to an extent. Every good recording or engineer have tubes in the recording chain. That messes with measurement but is better for what matter most, it sounds just better. But preferences do matter even if the result is not realistic compared to the actual recording session. Inches, we improve by inches compared to live sound.
Previously watching so many RUclips reviewers emote enthusiastically about these speakers, especially the fairly expensive Harbeths as a near field monitor, makes me question either their ears, taste, or truthfulness. I guess you can "like" something, even if it doesn't perform well, as long as it crosses off some of your expectations. But wow, when you opened that up and we saw the army of parts needed to make it sound "good" just blew me out of the water. Crazy. Why would they do this? Why not start out with better drivers in the first place and use less, but better, crossover components? I just don't get it. Sort of like your review comparing the $350 Wharfedale to a $4000 Revel showed you don't always get more for your money (but should). There's nothing wrong with the laid back, smooth, "British sound", but there ought to be a better way of achieving that. Thanks. Hoping someone will send you the KLH Model 5 one day.
Curious why you think the drivers are not of good quality, especially when looking at the spectral decay plots? Alan Shaw of Harbeth will tell you that the proprietary Harbeth-developed RADIAL material that the mid/woof is made of is one of the best materials he's used in 40 years of speaker design and manufacture to accurately reproduce the human voice. I agree that the specification of parts utilized in the crossover could be better, but on the other hand, one can counter with the view that it's hard to argue with the results: the speaker measures _very well_ and the spectral decay is really excellent, as well.
I've heard a couple of smart audio engineers make similar statements about Bose's satellite/subwoofer systems: "No highs, now lows, must be a Bose", followed by a diatribe about the crossover design.
@@tombrennan6312 Well, the host of this video points out that the speaker is performing like an instrument and adding it’s own color to the sound, plus the cheap components are smearing the sound and leaving a lot of resolution and clarity on the table. It just doesn’t give you everything from the recording. That said, people may not want “everything” in a recording. My step mother absolutely cannot stand music from any speaker except her absolutely POS iHome speaker that softens and veils everything that comes out of it. Sharp details startle and fatigue her. I wouldn’t say she’s typical (or the kind of person an audio company would profit off of having as a customer), but there are other people like her and thus there is a market for pretty, soft, background speakers.
Harbeth, speakers with too many parts tend to suck the life of the speakers, also it becomes too inefficient typical of Harbeth speakers, you need to use a high powered amp to drive it to a reasonable level, also to get the dynamic range.
(LX16? Are you looking at the right sheet of paper? Also, these require break-in) I beg to differ; I tried eight high rated pairs of speakers over the last two years, and I finally tried the Martin Logan 15i speakers. The 15i speakers are the best sounding bookshelf's I've ever owned. Sometimes, the measurements don`t tell the whole story.
@Danny Richie You didn't listen to a pair of 15i speakers properly placed and dialed in, did you? No, you measured it, right? And mistakenly called it a different model. I am sure your equivalent of these speakers would probably impress me a lot, right? Why not send me a pair to evaluate? If I like them enough, I will buy them. If not, I will pay for the return shipping. On another note, how was the crossover quality on the ML 15i speakers? They are $449 a speaker retail. I paid $269 a speaker through an outlet on ebay.
@@scottbennett3119 95% of all models going out the door are DIY kits. We don't send out samples. Comparing them to something like our X-LS Encore is like asking me how my Smart Car compares to my Corvette (yes I own them both). The capacitors in the tweeter circuit of the ML's were budget level poly caps (fair), but everything else was super cheap and bottom of the barrel quality wise. The woofers response was so bad on them (too much ringing and break up) that there was no fixing them. Oh, and I'll even get my kids names and employees names mixed up on a regular bases. Nothing new there.
@@dannyrichie9743 Thanks for getting back to me. I do like the mid to high end on the MLs, I don`t listen at higher levels much, so I haven't experienced obvious ringing or cone break up due to my wife being with me a lot of the time. They seem very good at the levels I usually listen at when crossed to my sub at 80 Hz. I would bet better parts would be helpful at the levels I listen at. I have limited strength, so I have to stick with small bookshelf models, so I am able to dial them in myself. I have been keeping an eye on your X-LS Encore though. Have a good one...
@@scottbennett3119 So, you want everyone to acknowledge that your Martin Logan is some miracle speaker because you paid for it? No, it is a piece of crap. Now, move on and enjoy your speaker.
Big part of problem with “high end audio” is the lack of a standard of sound quality (and it’s measurables) which is accepted by everyone. Once “everything goes”, companies will spend money where it counts for the profitability and sustainability of their business: they will spend on appearances as opposed to substance. On Marketing as opposed to R&D and build quality. From your valuable Channel, it appears that it takes a serious effort to PROPERLY build a decent speaker: good design, well dampened and braced cabinet, as-complex-as-required crossover with good quality parts to keep the response flat, since no drivers are “perfect”. Even highly regarded companies like ATC could upgrade some components (better braced and dampened cabinets), to quote your channel. So where does this leave the poor consumer when looking for a decent pair of speakers at a price point which doesn’t break the bank? And all these considerations do not even mention good room acoustics (the feasibility of applying room acoustic treatments to anything but a dedicated listening room while staying married), to assume the full potential of these aforementioned speakers…this hobby is not for the faint of heart…
I would be interested to see/hear what an upgraded external crossover for the Harbeth would measure/sound like. You could remove the old junky crossover, install some stuffing or no-res inside. Also, you could install tube connectors to the external crossover and then run high quality speaker wire through the cabinet back directly to the drivers. Or alternatively, you could remove the original crappy crossover and use an active digital crossover instead.
I could not wait to see this video... knew it would be good... This grounded [Danny kind of] perspective flies in the face of truthfulness, Alan (the designer) claiming "perfection" with his Radial driver technology... dismissing so many other noteworthy designs. He speaks so sharply of the LS5/9 speaker / its' drivers, the woofer being poor technology... WELL... The Rogers LS5/9 (in spite of numerous, documented tonal anomalies) was one of my favorite speakers that I have ever owned and deeply regret selling (to please my wife) it was just fun to listen to the 15 years I owned them. Flawed can be fun, just .not good. as a reference tool. I value accuracy, closer to the mark Danny is defining for us.
You are the coolest guy on the planet. Seriously, I really mean that. And you just drip with confidence and expertise. It would be worth driving from where I live in Maryland to Texas just to shake your hand. Thanks so much for your videos.
I would think the average Harbeth buyer doesn't care about what you're talking about. They care about how it sounds, in the moment. And they're not cheap, so the buyer is not pinching pennies, and neither is the manufacturer necessarily. I'm thinking someone bought the wrong kind of speakers for them, if these ended up in your hands.
If anyone really cares about how a speaker sounds then that one should not be in consideration. They are a long way from the expectations of the buyer in that price range.
The Harbeths are clearly trying to be a modern day take on the LS3/5A which in it's day had quite a complex crossover network due to the raw response issues of the KEF drivers. Surely modern drivers should be much better behaved than the T27 and B110 from the seventies you would think. Something obviously wrong if the Harbeths need so much crossover intervention.
I get it! But a lot of these designs were based on 70’s or 80’s thinking. We had no FFT or the ability to measure science back then so the boxes and the behavior of them were big unknowns. Celestion looked at this problem with the SL 700 and later the budget version the SL 600, using science and laser scan plotted drivers and fixed the box problems to produce one of the most perfect examples of a small to medium size speakers. Measurements and science produce predictable and highly tunable results and are critical. Voicing after the event causes the manufacturers to try to fix issues using cheap components where actually designing the product from the outset makes really outstanding products from the beginning. And…… then there is international taste and expectations about what quality actually is!
I googled and they did a 40th anniversary edition and- upgraded internals. Hificorner have a picture of the different insides and blog post about it. Cost for the "Anniversary edition" extra £845. I also agree, with such a good reputation I didn't expect Harbeth to have gone that way. But I am enjoying the reality of someone opening these things up and making it clear what's good and what's not.
@@willjohnson2722 it's easy to dismiss Danny for being such a small player in the audio market as many do. But, when he pulls back the curtain and shows stuff like this you suddenly realize why the little pizza shop across town makes a better product than the Domino's down the street: accountability means more at such a level.
@@akr01364 thank you for your comment. I agree 100% and with the sentiment. I am not a tinkerer. But I do love things to be the best they can be. And so I love channels like Danny's where things are opened up- to show what's inside, speakers, group tests etc. What's it really like? Most of the videos I've seen seem to show products that are engineered to sound the best they can at the lowest possible cost= the maximum profit. And that's understandable when people are running a business, because dealers take a large cut, VAT on top etc etc. High quality hifi is already a niche industry, production volumes are already low- I don't expect people are making a huge amount of money even when they are charging higher prices for lower quality products. But it's still a bit sad to see lower quality components in a product with a high price tag. Catch 22 in Hifi though, in my experience in the UK anyway- say you get a speaker made at the local boutique. It costs a fortune because it has high quality materials and components and that's why it sounds amazing. If you're like me and a lot of audio enthusiasts, let's say- in a year or two, you change system, downgrade, upgrade, whatever. Who's going to buy your speaker? Good luck selling it without a massive loss. And so over the years I've come to understand why people do stick to more mainstream products sometimes- not always because of the sound quality, but often because of the ease of buying, selling and support. And people don't want to buy modded products for the same reason they don't want to buy modded cars. However- a known product, upgraded by a known and respected engineer, like Danny- that's a super interesting proposition. And as always with the internet that's cool- people can find your videos years later and learn- like I did here.
@@willjohnson2722 this reminds me of an interesting back and forth that occurred in one of his comments sections: post modification resale value. He noted that a speaker featuring a well done modification he'd put together typically sells for a higher pricevthan one that wasn't. That discussion made sense to me as one of his more famous kits is for the Sony CS bookshelf speaker. The standard model sells for roughly $120 USD. One I came across on Ebay featuring the GR Research kit was, sans reserve, $550 at the time and likely went for more. Imagine what one of Sonos Faber's models would go for? And if there were to be a problem, he'd be able to help remedy it. A pretty decent middle ground when you think about it.
"a lot of these older speakers they really weren't designed that well to begin with..so to rebuild some of those around an old cabinet sometimes does not make any sense" Go and tell that to Kendrick Sound who rebuild old speakers like JBL for example being the basis of there successful business.
yes, I heard those Harbeth speakers recently and certainly are not worth €3000 here in europe. Danny speaks the truth. Even the cheapest GR Research kits will outperform them due to quality parts and great crossover design. Beware of sexy shiny exotic cabinets they are definitely soaking up your money !
i recently made these X-Statik's they are extraordinary indeed, all the money on the parts/design/Danny genius - not a lot on the sturdy painted cabinets - happy days ruclips.net/video/R389WYYgSLQ/видео.html
Wow, i am absolutely shocked. Harbeth are not cheap speakers, yet you found them to be a total mess. Just goes to show, no matter how much you pay for a loudspeaker. There is rubbish at all price points. I won't ever be buying a pair of Harbeth after seeing this!!!
@@markholder6851 I honestly dont know, but I hope Mr Alan Shaw is asshamed of himself. In fact I'm appalled by so many so called speaker companies using crap parts. Total rip off. Shame...
Since the NHT has such a nice and fairly large cabinet cant one install more suitable drivers that require a simpler crossover after having flipped over the cabinet? It seems that the majority of ready made speakers need to be sent straight to the hospital for treatment after purchase .
There's actually a "kludgy" (cheap) way to upgrade those Harbeth: move the crossover outside from the speaker, and put it into an external box. The box is really small and the woofers air movement stresses the components since the placement is too close and it will crack those poor cement someday. . as for the sound... there's really nothing special about those Harbeth, it does sounds pleasant (coloration, that is) as Nearfield, but it didn't behave really good for 2+ meter listening position even with a better power amps; I've guess people just think them as "decent all-rounder" ONLY if you sit really close to it.
Danny’s analysis is similar to that of a Medical Specialist. Quality testing, analysis and diagnostic assessment of a “speaker’s health”. Dr. Danny reveals the true “health” of a speaker the comes across his “examination” table. Always informational, factual and accurate presentations. How much is it worth to a prospective speaker buyer to pay for a high quality “Piano Grade” finish versus true audiophile high accuracy and high performance audio reproduction? Yes some may want to pay for that high grade visual appearance but personally for me I never sit and stare at my speakers just for their visual appearance and even while seriously listening I pay little attention to how the speaker looks but I want a fine experience of enjoying whatever music I choose to indulge in. Thanks Dr. Danny!! 😉
Cicable LS3/5A crossover exists just for the reason of a lack of space being available to place a crossover too large for the cabinet So I would suggest you tell your client of such a solution for the Harbeth P3esr (se/xd)
A channel like this isn’t going to popular with people that wanna live in ignorance bliss of the high amount of $$$ they paid for whatever speaker it is they have It reminds me of fanboys back in the day that fought to death to defend their phone brand, video game console brand, TV brand and what have you. It’s very hard to accept objective truth if it goes against what you bought I see a lot of those people in this comment section.
Super interesting to see the Harbeth characterization. It seems clear that if you want a fully assembled beautiful cabinet in a two way monitor the Tyler Acoustics monitor is the way to go since it already has a GR Research well designed crossover and parts quality.
fyi - my system takes about an hour to an hour and 1/2 to open up - most of my speakers took at least a year to burn in - keep the faith ok - I run a Benchmark DAC1 as my source, a Creek Destiny 2 Integrated - a Hafler HA75 tube amplifier used as a tube line stage ( running Telefunken E88CC's or is the other way around from Kenny at Upscale Audio ) and Harbeth HL5's but I listen to things like Everything Like the Girl, Alice in Chains Unplugged, and Karen Carpenter singing "This Masquerade' so what the f do I know .......
Lots of posters are missing the entire point regarding parts quality . How long is that crossover going to last before the cheaper parts start to degrade? It might sound great….for now.
lesser quality capacitors limits detail according to gs research this does concern me when looking for speakers in a specific price point I was wanting the linton heritage 85th anniversary but gs research was saying the crossover was using budget components
please do a tear down/components evaluation of the Elac Uni-Fi 2.0’s. i’m EXTREMELY interested in what your thoughts on their crossover would be. i would gladly send you my pair for evaluation!
Great video, I really like your honest, results based feedback. Unfortunately for me, now I am questioning my little Harbeth P3ESR's. I suppose that I need to talk to about which kit and components you would recommend as replacement. I would be willing to build a set of your speakers and make some side by side comparisons with the Harbeth's to come to my own conclusion on what I like better.
Why are questioning your p3esr now .????? Do they still sound good ? If so,…why the questioning !? I love my p3esr ,…yes, they are not perfect ( which speaker is) ? But , I love the sound they give , colorations and all ! Heck…world is coloured !
I have always defended that some speakers you can modify ,…others you just ruin them ! I have had a pair of mission 751 like 30 years ago, fabulous speakers , Pull them apart , modify them with good poly caps , better resistances , better intern cable . Result: a speaker that was better in everything detail , hifi , dryer and faster bass, but……something was gone , the fabulous 3D soundstage these speakers kan do ……..was gone,…..they turned forward and cold , So,….I do believe that cheap parts or not , sometimes you do have engineers that know what they do , and they realize that the voicing of a particular component in the crossover is the one that must be used ! Is like driving an old jaguar , it gives a special kind of pleasure , now modify that same car with top notch motor , suspensions , and in the end you will a car with better specks …….but ….does it give you more pleasure driving ??? Relative to harbeth p3esr ……common !!!! Who whants to improve a speaker that works good as it is , I do have a pair , I personally would not modifying them even if it was for free, I like their voicing , ….. thus modifying = another voicing . Perfection is not always the path to completion . If I like a certain sort of coloration ….why modifying it ! What’s the point ? And…..is not truth to the signal !? Who cares. Going to listen to some music, Bye .
The reason the Harbeth crossover has so many parts is to correct for non-linear drivers. Better drivers would cut the crossover parts count by two thirds.
No, the reason is that the kind of response the designer is aiming at requires this kind of complexity when using a passive crossover. Using different drivers wouldn't help much.
@@matthiasmeyer1124 You mean like flat response? I’d like to see raw response graphs for the individual drivers. I’ll bet they’re all over the map. Linear drivers don’t require anywhere near this level of correction.
@@ChicagoRob2 The P3ESR isn't a particularly flat voiced speaker. Flat isn't difficult to achieve. There are many first-timer DIY projects that measure flatter than the Harbeth. For some of the dips and humps of the P3ESR I know that they are intentional - not for lack of skill or due to budget constraints. I have a P3esr clone project at home, using a well-measuring driver and I need a similar parts count to shape the reponse. I can't vouch for the Harbeth woofer but the tweeter is a SEAS driver and superb in an infinite baffle. All raggedness of the tweeter derives from the cabinet. I can make cabinets that perform better in terms of measurements, but subjectively the Harbeth baffle wins for me.
in the first speaker: would there be an option to put a helmholz resonator inside the box that counter-acts that 800Hz resonance? where is it coming from? is this a standing wave between top and bottom? probably too hard to add pieces of wood inside and glue it down to form a new chamber with a small hole that connects it.
These speaker manufacturers have a lot to answer too … only a handful you have reviewed that you would recommend ,keep up the hard work Danny ✌️💪🏽. Ps is there much you can do with electrostatics “quad” ?
I have a question, what about removing all these passive components and just running a digital crossover at the DAC level? So instead of using a stereo amplifier to individually amplify each speaker we would use two stereo amplifier with each channel amplifying a specific driver instead of the full speaker set. Thi would also allow for way quicker iterative improvements wgere it would be possible to quickly measure and apply a different filter and then measure again to get the FR to look as good as reasonably doable?
I own pair of Harbeth’s (M30.2 40th Anniversaries), as well as Spatial Audio, Magnepan, and Focal. They are the sort of speaker that is easy to like, until you compare them with others. Alan Shaw has found the perfect model: build a beautiful-looking box speaker (no curves = cheaper), stuff it with cheap parts, and then engineer it to measure well. He must have the best margins of any domestic manufacturer (i.e. not Chinese-built). People are happy and he is happy, supporting British industry and making a living at it. I don’t fault the guy. If you can use cheap parts and get the sound right, why not?
Which Spatials do you have David?
@@rlowes M3 Sapphires. Fantastic speakers.
@@Unpreeeedictable I am saving up for a pair of those as I rather doubt any will come up used
@@Unpreeeedictable very cool. I have a smaller room, so was considering the new M6 Sapphires. I assume they are far better than the Harbeth 30.2 Anniversary (which is another speaker I was considering as well)?
@@rlowes totally different. The Harbeths do a couple of things very well, especially acoustic and vocal. Think singer-songwriter. The Spatials do EVERYTHING well. My Harbeths are for sale ☺️
I suggest the presenter reads Harwood's BBC paper concerning speaker cabinet construction, which concluded that the best way of combatting cabinet wall radiation was to use a thin-wall approach, allowing damping measures to be made very effective (if I remember correctly).
Useful reading for anyone, actually.
Thin wall design is a great approach to speed of decay and using bracing at corners all helps, but that was related to cabinet size and surface area. The LS35a is an example of a small box which has great imaging and a smooth response but for a definate purpose, small outside broadcast trucks and control rooms. They sound really great but, there are limits to what they can do above 20 watts of clean power. These speakers had massive amounts of correction to overcome what was commercially available drivers from KEF at the time. So not tremendously accurate but fit for the intended purpose.
@@adrianfisher9641 Do note that the article refers specifically to the LS3/6, which is a LOT bigger than the LS3/5a - but the principles covered are anyway not exclusive to speakers of any other dimensions.
A lot of measuring but no listening! My Harbeth P3ESR 40th Anniversary sound absolutely incredible and the speakers are designed to be played with the grills on.
I've heard thousands of speakers using cheesy parts. If you think those sound incredible then you have really heard anything that is really incredible.
@@DougMen1 I get to hear an incredible system every time I turn mine on. I even passed college English, but I do speak Texan.
@@dannyrichie9743you passed English, let’s test that theory. “If you think those sound incredible then you have really heard anything that is really incredible”
The Harbeth P3ESR are reference quality speakers and have an uncanny realness to the sound, especially voices.
They are smooth, non-fatiguing and have ultra low distortion.
Listen to them and forget about the crossover part count.
Many speakers would sound much better with additional crossover network components- such as Zobel network to flatten the impedance and make them more tube friendly; BSC or baffle step compensation that compensates for too much midrange being reflected off of the cabinet baffle, higher order crossovers needing more parts to keep the drivers in their smoothest frequency zone that greatly reduces distortion and lastly notch filtering networks to tame any peaks that the drivers might exhibit.
Looks like an abundance of components but they work and work extremely well. Just listen to them, they sound glorious as do all Harbeth models.
They certainly did their work to create a good level of accuracy, but the high parts count using extremely poor quality parts really sucks the life out of the music. These are far from reference quality.
@@dannyrichie9743 Your assertion the speaker has no life is a subjective evaluation. Many experienced people have the opposite reaction. And so it goes.
@@tombrennan6312 The only way anyone would have an opposite subjective evaluation would be if their reference was a Bose wave radio.
Well, there’s different kinds of distortion.
@@dannyrichie9743 so you do actually set these speakers up and have a listen to music? Do you do that before you open them up and see the mess inside? I fear that if you didn't, just knowing what's inside them would give you a negative bias towards their sound. It's just confusing because they measure so well.
Actually if that’s a P3esr, Stereophile didn’t measure any cabinet vibrations that they thought would cause colourations with these particular Harbeths.
Also interesting to note that the new XD version of these apparently has better crossover parts and better wiring and binding posts (and a fair bit more expensive I believe). Would be very interesting to compare the internals.
I do think that saying these should be “avoided” is a bit much. Tons and tons of people love these speakers - they must be doing something right. If you want to buy them specifically to upgrade the internal parts, sure yes, avoid. But to recommend avoiding overall, that’s taking things a bit too far I think.
Are they overpriced? Like many other production speakers, just looking at parts, sure. But that’s an overall value judgement for the purchaser. Good luck DIYing a cabinet that looks like this.
There are "tons and tons of people" who smoke crack. But I don't believe "they must be doing something right", and I would not recommend it either. An appeal to popularity is not a good basis for forming one's opinion.
@@jdlech I’m talking about hardcore audiophiles and reviewers who consider these to be amazing nearfield monitors for pleasure listening and some who even own a set themselves. Do you think people who listen to tons of gear that decide to purchase these for their own systems are all deluded or lying for some reason?
To dismiss a speaker purely by looking at the parts is foolish if you actually care about how these things sound IMO. But everyone is entitled to their own opinions and approach to things. If you value parts, yep these are not going to satisfy. And maybe they’re not detail and resolution monsters because of this, not sure, I’ve never heard them. But it doesn’t mean they’re not really enjoyable to listen to.
The problem with this kind of “you must avoid because of this objective quality even if other highly experienced people say they sound great” is getting into Audio Science Review territory where only the objective matters and the subjective doesn’t.
Also, the “if everyone jumped off a bridge, would you” comeback is a pretty pointless argument.
@@rlowes There is certainly a balance to be struck. Never underestimate the power of social proof, visual bias, and the ability of our brains to adjust to stimuli over time. Not to say that these sound bad (I've never heard them) but man social proof and visual bias, alone, are strong Jedi mind f's! Over time I've noticed the ASR affinity for Harmon. Bias strikes again.
@@NakeanWickliff absolutely agree. To be clear, I’m not claiming these absolutely must sound amazing, or that I would even like them. I’m am saying however that there’s a pretty high likelihood they would sound great to SOME people who heard them.
And to simply say they should never be considered or purchased due to parts quality is BS. If that’s the case, 95% of the production speakers out there should be avoided.
@@rlowes Yup, we're in alignment there for sure. I actually have a pair of Klipsch Heresy speakers that I have gone full circle on. First speakers I owned when starting to actually critically listen. Then I measured and saw the lumpy response, and also compared them to open baffle speakers. After that, I couldn't unhear the Heresy boxy sound and stopped listening to them. Now, I've come back around to loving them in the living room even knowing they're not linear and they have vocal bloom from the box resonance. It's a charming sound to me now and I really like them for what they are. In the end, It's all subjective. Kinda like cruising in an old classic car. You love them because of the beauty of their quirks not despite them.
I once took my GR Research AV-1’s amp shopping. The shop was a Harbeth dealer.
I would say that the trounced the P3ESR’s (the original version). The Harbeth’s sounded a bit fuller, but likely due to that cabinet resonance.
The dealer was in awe. He spent more time listening to my speakers than I did listening to his amps!
I listened to some of the bigger Harbeth models while I was there, and couldn’t wrap my head around to performance to cost ratio.
Hey Danny... I own the sb3's, and I am a musician... When fed a good source they sound like the instuments should. They have real defined soundstage. , and great imaging... Also in the setup manual they show the speakers tweeters facing up. I wouldn't think that would make a difference with your on axis measurements, but you would know better than me. Lastly, one of the main things multiple respected reviewers said about them was that they were very accurate and flat. The audio world is wacky!
Have you heard the KEF LS50 Meta? I'd like to know your perspective of them.
@@LarsonChristopher in a word…fatiguing.
@@jlo8775 What amp / preamp and amp / receiver was used? I have a pair and use Yamaha AS701 and they sound awesome.
Your hearing is damaged.
Hey Neil, That is because crossover parts aren’t as important as Dannie the crossover salesman would like you to believe.
The reason the nht is woofer up is actually a psychoactive reason. Psychoacoustically, people tend to locate high frequencies at a point higher than the actual location the sound is coming from. So putting the tweeter down puts the heights psychoacoustically at the level of the woofer. I remember reading an AES paper 50 years ago where they conducted the tests and wrote the paper.
Interesting stuff. I've always liked Totem monitors, and it would be interesting to see how they measure. They use a minimum number of higher quality parts so I always though that's what I was hearing, but I'm sure they don't measure flat, and not sure about spectral decay but it would be interesting to see. They also hand-match their parts in pairs so the left and right speakers match. I've got their Rainmakers, Sky Monitors, and Kin Monitors now around my house. I chose all of them just by comparing them to similarly priced speakers, and these sounded better. It would be cool to see if I could take the Sky's up a notch though. Hmm.
Who cares if you like them. A crappy measurement could change that for you right.
I have a friend of mine who has been a professional loudspeaker engineer for a very long time. Put this way, he was designing professional products 20 years before GR Research jumped into the DIY market. His experience includes designing numerous well reviewed products for a number well known brands across many products lines from lifestyle Bluetooth speakers, car audio, in-wall, on-wall, in-ceiling outdoor, as well as high-end 2-channel, home theater, AND DIY KITS. I asked him for his frank response to the Harbeth section of the video. He has absolutely no connection with Harbeth, and offered this remarks strictly as a critique of GR Research:
To start with, what GR Research portrays in its You Tube video is not in any way an objective evaluation of the Harbeth product. GR Research makes its money selling people on the concept that they are capable of improving on other manufacturing design work, mostly by the simplistic concept that all it takes is higher grade crossover components, and by touting the superiority of his own DIY kit products. While this is his prerogative, again it is neither objective, and definitely is self-serving.
To this gentlemen’s credit, he recognizes that the Harbeth speaker measures well, noting the well done grill concept that is part of the Harbeth line. He also mentions that the impedance has no issues, but also vaguely mentions the tweeter impedance is “kind of high at that end”, which is a ludicrous statement. The reality is that all voice coil type tweeters have a frequency dependent reactive (AC inductive and capacitive elements as opposed to just being a DC resistance) rise in the impedance and it is normal for it to be relatively high at 20 kHz. This is deliberately misleading.
Next, the host points out that the box has obviously high resonance and is somehow deliberately “lossy”. His claim is that this makes the low end bass “boomy” and colored in a way “that people like”, which is also not true, at least according to Dr. Floyd Toole’s extensive work at the National Research Council (Canada) and Harman International on explaining the correlation between objective measurement and listener preference. He then proceeded to validate his criticism by taping the box with his fingers. Besides the fact that I really didn’t hear any “ringing” when he tapped the box, if you look closely at the back of the box, there is a layer of damping material stapled to the rear baffle, something traditional in British speaker engineering, and I suspect, though the host didn’t bother showing you a view of the inside of the enclosure, this same material is also used to damp all four walls of the enclosure. I might have been more receptive to this off-hand and unsupported conclusion had he done something like offer accelerometer plots on the Harbeth enclosure verses something he considered more appropriate, but that was not offered.
While I find his comment on the impedance and the enclosure annoying, his comments on the crossover, which he draws, a really inaccurate conclusion from, are also not at all accurate. First, parts count. He implies that this is a ridiculous number of parts for a two-way network, and that the network has over “30” components (actual count is 20). Without the schematic, it’s really hard to judge, but here is what I see. Five inductors are likely used as three in the woofer circuit, 2 as part of a 3rd order low-pass filter topography, with the third inductor likely used to control some top-end peaking in the woofer. I suspect the other two smaller inductors are used in the tweeter circuit, one as part of a 3rd order high-pass filter topography, and the other as either a LCR conjugate circuit on the tweeter resonance, or as a response contouring network. He makes a big point of describing these crossover parts as “cheap” bad sounding network components, which both an exaggeration and misleading. The capacitors are all polypropylene film caps, but obviously in a package format this amateur engineer doesn’t recognize. They are not Mylar caps!
(I had a picture of the cap but RUclips will not allow me to post it here)
Polypropylene Film Capacitor
Temperature Range - 55 C to + 85 C
Dissipation factor 0.1% MAX. at 1KHz, 25 Degree
Capacitance 104 F
Voltage Rating 400 v
The 4 larger capacitors on the top left of the Harbeth circuit board are obviously paralleled together for the woofer circuit. So, instead of using a single non-polar electrolytic, which frankly would have worked adequately in this application, Harbeth spent a lot of money paralleling four much more expensive poly caps in this circuit where one cheap cap would have done the job!
All of the capacitors in the tweeter circuit on the left side of the circuit board are in pairs, one large cap and one small. This is an obvious detail the host has overlooked, as not only are both caps in the pair polypropylene, but the smaller poly caps are bypass caps, and well known technique in high-end speaker design which is done to overcome parasitic resistor and inductance at the upper end of the caps operating frequencies…it tends to provide more “air” in the tweeter’s subjective perception. There are 11 capacitors total, but in effect, the four in the woofer circuit are acting as 1 cap, and the other 8 caps in the tweeter circuit are actually just 4 appropriately bypassed capacitors. His gasping at the parts count is really ingenuous and misleading to consumers who are not loudspeaker engineers. In terms of parts count, if we count the paralleled caps and bypass caps as just 1 single cap, the parts count is now just 16 parts, half of the horrible gasping “30” OMG parts the host is making such an issue over.
He also states that this many parts can lead to a smeared image, loss of high-end detail and soundstage. Sorry, but while that may be his opinion, it’s just not true. The host seems to have come to the conclusion that the number of parts used is because of poor drivers. That is simply not true in this case either. The number of parts used in a crossover varies for a number of reasons, but generally it’s either cost or design philosophy. For example, a typical high-end tweeter high-pass network by itself might contain as many as 12 parts, 3 in a 3rd order topography, 2 attenuation resistors to balance the circuit, 2 more resistors if there is a level switch for the tweeter, 2 in a CR conjugate to contour (voice) the upper harmonics, and 3 parts in an LCR conjugate circuit on the tweeter resonance. This is not excessive and none of these parts are related to SPL correction.
The host at GR Research seems to think that changing crossover parts for more expensive parts is his hill to die on. While I may not employ ferrite core inductors in my work, they are not as horrendous as is being pointed out and replacing them with air core inductors may not really save the day as he implies. The Rogers LS35a employed ferrite core inductors, and I don’t believe that the LS35a sounded awful or even mediocre to most listeners. Like all design decisions in loudspeaker engineering, most are often some kind of tradeoff. Cored inductors have substantially lower loss due to have less wire and less dc resistance, and if well designed, do not always present hearable distortion when they saturate at high current levels.
In summary, GR Research’s conclusion that this Harbeth two-way monitor is a lost cause for his company to improve is horribly misleading, and simply not the case. The number of crossover parts is not a negative, and his not even being knowledgeable enough to know that the capacitors actually are polypropylene really puts in question his portraying himself as an “expert”. His failure to explain that the number of caps is due to a sound engineering practice of by passing and paralleling capacitors in a crossover network in order to deal with parasitic capacitor inductance and resistance, does not impress me, and is decidedly a disservice to his listeners.
Yeah, there is always someone out there that thinks they know better than industry professionals. He might want to stick with the lifestyle Bluetooth, car audio, in-wall, in-ceiling and outdoor market and leave the high end market to those that specialize in that area.
And he thinks that this is not an objective evaluation? Measurements are nothing but objective. So is reporting what I see.
If he really thinks that mentioning the tweeter impedance is high is a ludicrous statement, then he really doesn't understand the importance of balancing the impedance or how different types of amps react differently to different impedance loads.
What I described regarding the box of the Harbeth's was just restating what they are known for. If he knew anything about that company or their products then he would know that.
Paralleling multiple caps to make a larger value is not a plus, and when the values are different it causes some smearing from the dissimilar dissipation rates of the caps. I suspected they were paralleling some value but didn't feel it was worth beating them up further. Caps are also likely polyester and not polypropylene. They are known for using polyester film caps. See this link: www.widescreenaudio.be/wp-content/uploads/2017/02/shl5_crossover_filter.jpg The ones used in this model aren't even that good.
And absolutely the use of multiple parts of really low quality (small gauge iron core inductors and sand cast resistors) will have a cumulative effect. The more or them you use the more it degrades the signal. If your buddy doesn't get that then he really has no business responding to the video in a critical manor.
@@dannyrichie9743 HE IS AN INDUSTRY PROFESSIONAL SPEAKER DESIGNER WHO HAS DESIGNED MANY HI END AWARD WINNING SPEAKERS. I'LL JUST LEAVE IT AT THAT.
@@nancyweiner509 Yeah, well so am I. I think my designs have won just about every industry award there is with print and online magazines.
@@nancyweiner509 Wow, I hear you, Danny was extremely disrespectful and owes you an apology.
@@dannyrichie9743 but not much since 2009 by the look of it
Great stuff. On the Harbeths, though: It's a misconception to say they were designed with a resonance to give extra bloom. That's not it at all. It's a common belief, though, and it really gives British engineering short shrift.
This kind of cabinet design harks back to the BBC, which decided, in their in-house monitors, to tackle cabinet resonances by damping thin walls with lossy materials--bitumen, not No Res, but same principle. They did this rather than make the cabinets either heroically thick, or from inert materials, which were unavailable at the time, outside of the aerospace industry.
Instead, their engineers chose to permit resonances, so long as they were low in frequency and fairly low Q, so they'd be nearly inaudible. It was, at the time, the most cost-effective way to get low coloration enclosures.
In other words, this wasn't designing IN coloration, it was just crafty engineering to design OUT coloration.These speakers are second-cousins to the BBC LS3/5As and, as such, they share the design philosophy.
Translation: there are better, more advanced cabinet materials available today, but British companies haven't kept up with the tech advances. Probably because they "like" or have gotten used to the traditional colored resonant sound from the old-school thin wall cabinets with bitumen.
It would be very interesting to see the Harbeth raw drivers response. Surely they're not that bad, that they need all those parts?
It sounds like the Harbeth was going after a “sound” pardon the pun. That they achieved at all cost expense of finding the cheapest parts worked out beautifully for them.
That’s probably a speaker put into a category like people who like the harsh piercing sound of a class D amp. Compared to the nice smooth sound have a good high quality tube amp.
The Harbeth speaker crossover kit would probably be approaching $600 in parts to identically duplicate every component on their crossover and way as much is the speaker hanging off of the back.
Not to mention having to add mechanical speaker supports gluing them in to the inside of the cabinet for extra support and adding all the no-Rez. To the inside of the cabinet would take away that mechanical residence of the speaker singing that the speaker manufacture was relying on to give them part of their sound signature.
Totally agree on this little English speaker if you’re not happy with the sound don’t try to fix it give it to somebody who enjoys it and start over with a new different speaker
Adding a so called higher quality resistor, capacitor, or inductor will do nothing to better the sound. The value of the components used, meaning the resistive or capacitive numerical value, is all that matters. The engineer masterfully creates his crossover with common parts , of which the value is carefully selected. You cannot argue with the many that find some of these speakers sound just fine as is. Most of what you recommend as upgrades are basically overkill.
Tony, when I sold hi fi years ago, the Japanese brands would all use parts with only 5% value tolerance. Everyone else would use up to 20% so IMHO it’s the tolerance quality of the parts that fundamentally counts. If you have say 5 cheap parts in a board and they are all graded at 20% tolerance, how can you be sure that they all work at near value all the time?
Tony, if good enough is good for you, that’s great. But we on the outside like to enhance or equipment to the better for the sound and not the measurement. We want better sound and depth. Then you need to change and upgrade. Look at speaker designer Mats Buchardt. He’s been selling the S400 Mk1, for some year. Now he upgraded them to Mk2, with much more spec crossover with great components and the sound improved. Look at all the reviewers verdict. What is the conclusion of this? Not only Danny knows the value of great parts. The way you write makes me wonder if you have designed one of these speaker/crossover that Danny displayed.
Any component in series from the amp to the speaker voice coil is going to affect the sound. That's just basic electronics 101. The question is whether or not it can perform as needed. Reasonable tweeters crossed over with poly caps? One can hear a soundstage. With an electrolytic in the same spot, the soundstage effectively disappears.
@@seejayfrujay I'm afraid that this viewpoint displays a complete lack of understanding about how electronics work. Passing an electronic signal through a more expensive resistor of the same value as a cheap one, will sound identical. The electricity doesnt care how expensive a resistor it passes through, a resistor is a resistor. It's possible more expensive ones could have better tolerances for matching, and may have better longevity, but they won't sound different. that IS electronics 101. The fact you think otherwise, suggests you don't actually know electronics at all.
@@1061shrink1061 Always always the ad hominem when someone is talking through their hat. No, a cheap voltmeter will not tell you what is happening. And yes I do, from vacuum tube to MOSFET. My background is physics. You will observe a marked difference in the electronic performance of a electrolytic capacitor versus a poly or other type when you perform frequency sweeps and pulse testing with the appropriate equipment. It takes a bench in other words.
There are two speakers that are widely-used as near fields in pro audio, the Yamaha NS10 and the BBC LS 3/5A. The NS10's can sound rough, but they give engineers good insight into how their mixes will translate to lower-end consumer audio gear, such as car stereos, boom boxes, cheap hi-fi's, Bluetooth speakers and ear buds. The LS 3/5A was made for BBC's studios with the emphasis on mixing/mastering for speaking voice reproduction over the radio and TV. They are specific tools of the audio engineering trade, but neither speaker would be my choice for home hi-fi music reproduction. Oddly, the Harbeth version of the LS 3/5A (as reviewed) is aimed at wealthy audiophiles. Go figure?
I designed an upgrade for the Yamaha NS10 that is being used in about 10 different studios. I am surprised that they were ever used in mixing or mastering. They are not that accurate and don't sound very good either.
@@dannyrichie9743 Amen to that. I cut my teeth writing and producing electronic music using NS10m's; they always sounded lazy to me. No matter what I programmed to have pizzazz and color, they would drearily and reluctantly dribble out like a terrible waiter. Of all the monitoring set ups available to me at the time, I will have to say they were the most accurate - a friend with some PA kit in their living room, or some luxurious and well-presenting headphones were much more satisfying to playback what I thought I was creating.
But at any level the NS10s just sounded like they were being played very quietly and seemed positively gleeful when the room we were in just became overloaded, as did I.
An upgrade sounds interesting - would it involved replacing the cabinet and all internal components and drivers ?
Thanks for all these videos, it's great to have learned another approach to speaker engineering, and it's one that has been added to my own plans.
The ns10 was selected because of it average crappy mediocre sound to hear hpw it sounds on most audio...
The Harbeth speakers reviewed are not BBC designed LS35A's.
I've never heard a Harbeth speaker that I liked, and have always been baffled by the overwhelmingly positive press. Danny's analysis is "consonant" with my experience.
i get it "baffled" good one
@@michaelmcintyre9179 best not to get all up in John’s grill
They are expensive so people like to show that off even though 95% of everyone couldn't care less or tell any difference anyway. People on stevehoffman's praise them, I am guessing so they can let others know they spent so much money.
Only Danny can take a pair of speakers that a buzillion people love (and not just these three) and “prove” how lousy they are - because his graphs tell him so. I would love to see how many speakers Danny could listen to without seeing them, and never having heard them previously, and tell us which ones had “poor” caps.
A buzillion people love 901s and eat Big Macs. What the masses like means nothing.
I do think Danny would fix that, if he plays his own list of music he knows, then he would be able to say what’s good and bad. The other stuff are called experience. Many years of experience of speakers and speaker parts.
I want to sponsor a cage match between Rick Beato, GR-Research (they can nominate their champion), and Steve Guttenberg
..and he’s right probably 99% of the time. I have done some crossover upgrades myself on some highly renowned (B&W matrix models, several) loved by a gazillion. Along the lines of Danny’s, and the difference was not subtle. Similar on budget models like KEF’s Q150 (used some of Dannies parts on that one). Again, not subtle. Sounded really good before. Sounded great after. My wife served as the blind tester, and not being an audiophile said things like: “that one sounds like it has a little of a haze. This other one is much clearer. I hear more details.” Her words, not mine. Anyway, I think Danny’s points maybe overstated, but no doubt, take a speaker that sounds great and is loved by many, give it some of his crossover treatments, and I am convinced that it will sound even greater. So much so that you wonder why the designers did not do it that way int he first place.
And that is why I enjoy watching this channel... Because you can look past a pretty box. Some people get so fixated on the look of a speaker that they can't hear the sound that comes out of them.
I have the M.L. lx 16, there is a setting on my Marantz SR 5011 that rolls the highs off a little, it made a huge difference, plus I pulled them out a few more inches, wow I have fallen in love with these little beauties all over again, the sound and clarity is incredible. I have owned quite a few speakers, but these have become my favorites.
Hello fellow Martin Logan owner. The LX16's are a little old now. The new M.L. Motion 35XTi are very very nice for the small bookshelf form factor. You might want to audition them at some time. Not saying make a change, but the updates have made a nice improvement in the sound, and you would likely notice the differences.
SKOL
Hi Danny. I traded an email with Alan Shaw of Harbeth. He gave a detailed reply on the Harbeth user group about caps and inductors etc. He mentioned that you had not mentioned in your video, that the Harbeth has poly caps. He told me those poly caps are good quality and will last a 100 years.
Harbeth sells speakers faster than they can make them. They must be doing something right. Have you listened to them?
They may use some poly caps, but those caps, coils, and resistors are among the cheapest made. And yes, I've heard them.
My pair of NHT Super One XU speakers seem to have a similar issue as the SB3, as I've wound up placing them upside down or else they sound out of phase. I know they're entry-level speakers with cheap xovers. I paid $65 used for them, and they sound pretty good, so I can't complain.
Interesting on the Harbeths - perhaps it’s designed so each iteration can have a slightly improved crossover and thus seem better than the last version? I have always used LS3/5as, and bought some external crossovers built for them by a company called Cicable, and designed by someone who probably knows most about BBC type loudspeakers and crossover design than anyone else. These gave a massive improvement in THD reduction and increased resolution and transparency. Now I have Stirling Broadcast’s V3 version of the LS3/5a which also uses a crossover using higher quality components (the pcb is about the same size as the Harbeth and also mounted on the back of the cabinet, but uses huge air cored inductors and also has less components on it), and it also has that great transparency and resolution, and with modern SEAS drivers that match very well it images fantastically and has better power handling. I use them with their bass extenders plus some M&K subwoofer crossed over around 60Hz and they are the best I’ve heard.
That's my NHT!! Sad it didn't qualify, but ready to do a diy build. Top binding post was always an annoyance. I like the sound in general, it extends down but I do use a Parts Express DIY Kit DSP subwoofer.
Shopping for a DIY kit right now.
Thank you GR for your time and advice!
The GR research xls encores are pretty nice. I still love mine. My other loudspeakers are La Scalas. I switch them out all the time.
I have a xls encore kit sitting in a box doing nothing,the xovers are put together and sent Danny a pic and he said it looks good.i would sell for 175.00
I think it’s difficult for you to pass comment on the Harbeth’s without actually listening to them. I am English, and I worked as a producer in BBC radio where the studio monitor’s were either Rogers, Spendor or Harbeth’s, all of which were fairly neutral and accurate. I put my trust in BBC engineers. I also think your comments on the cabinets construction are misleading, they are a “thin walled” construction created by BBC engineers for small monitors like these and the Rogers LS3/5a, which I own an original pair of - there is no boominess with the cabinets, there is no resonance either. Yes, they have a “British” sound which is mainly uncoloured so you can listen for extended periods without getting fatigued. I would suggest you get a pair and hook them up to a decent system then pass judgement. As a lot of people are saying here; it really is what they sound like and in my opinion these little Harbeth’s sound great, as do all the Harbeth range. There are many recording and mastering studios and TV control rooms in the uk that use Harbeth as reference monitors, I’m assuming the guys that select these speakers know their stuff.
The Rogers Kef LS3/5A has a much better crossover.
I think your not aligned with the intent of the video, and I’m not sure the speakers were not listened too. Don’t know if that was ever confirmed, and I’d find it unnatural if anything evaluated isn’t listened too. The point of the video is to use data to baseline the conversation and assure everyone is aligned to describe how speakers are putting out frequencies from the lows to the highs, and how much the speaker changes the sound as compared to the music in the studio. It seems Your contradicting the video because you’re concluding based upon your preference, and what you think. This is that audiophile subjectivity (“British sound”)
I’m sure there’d be no comments about how thick the cabinet is, if the output meets the intent of the data output. Likely, an observation based upon the output performance as it was measured
It's also not fair for you to pass judgment on GR Research without listening to one of their top line 2-way monitors and comparing it to your beloved BBC speakers... It may blow your whole worldview right out of the water...
Well said,i can't believe Danny was slighting the engineers like Dudley hardwood,and his peers,so far all he knows is "buying off the shelf drivers and build crossover with sonic caps and loudspeaker design is over.
I had started my audiophile journey with p3esr and ended with superhl5plus,
In this process i tried a heck of 60 pairs of speakers before reaching the verdict that harbeth's sound is pretty close to perfection.
Mastering music and listening to it are two very different things. In masteri g you're looking not for accuracy. But, instead a sound that you as the engineer want for a track. If your ear has been attuned to the color of a speaker type, such as the Harbeth sound, you're going to want that color when in the edit. Listening often times means much the same in that you're looking for a sound you as the listener desire. Given that many British speakers were designed to accommodate the BBC of the 60s and 70s that particular group of engineers, that were used to a very different standard of reproduction than many would accept today, many still hold on to that "British" tone. Even if it means reproducing the original sound poorly. As a reference, consider B&W speakers. They've been designed as British speakers to be the opposite of that warmish tone that many associate as being British and instead sound just downright harsh. The Harbeths like many Tannoy products sound like something an engineer from 40 years ago would find pleasing. If for any reason the fuzziness that kind of reproduction presents the listener with helped to mask some of the noise LPs and tape tended to cause. We have better source materials today and don't need to veil music to remove noise beyond equipment interference. As such, speakers needn't be built this way anymore.
I always wondered what Danny would do if someone sent in a pair of Harbeth. “Surely Danny wouldn’t tamper with Alan Shaw’s work” is what I always thought to myself. Turns out I was right, but for a completely different reason. Disappointing, but must be one of the most highly regarded monitors out there for a reason… how did it sound Danny?
I was just sent one of them. With that level of parts quality a lot of the life would be sucked out of the music even if the drivers were the best drivers in the world.
Now what if these Harbeth speakers sound good, to a lot of people they seem to do, even though they contain cheesy bits!? They sell upgraded versions of these speakers, that probably sounds better. The upgrade is of course better components, the drivers are probably the same. Harbeth have made a well aclaimed business out of making speakers people like, hours spent testing and listening during development must mean something.
Two listeners rarely agree on everything they listen to, regardless of inside components. I bet that some people listening to Dannys speakers will say they don´t like the sound, again regardless of the interior bits.
But i totally agree that it´s a shame and wrong saving money inside when we are talking about expensive products!
The right part is the right part whatever the price. There can be compromises based on cost so decisions can be complex
There's nothing wrong in using steel cored inductors if applied properly. If everyone would use expensive 10ga inductors and fancy poly caps everywhere, most of loudspeaker cost would be in crossover not drivers. I guess there's is some priority when it comes to designing and it makes sense to use expensive parts only where it counts. Zobel networks or RLC traps don't need expensive parts. Only parts that are in series with transducer benefit from better quality parts. Tons of make up and facial surgery won't make every woman into supermodel, or racing slicks won't turn average sedan into race car. There are same rules which apply in proper engineering. On top of that, Danny doesn't show the crossover schematic and where these cheezee parts are utilized.
Electrolitic caps - maybe someone should take a look into service manual of their electronic equipment: amps, DACs all use them in the signal path.
GR I appreciate the effort but I have to respectfully disagree on the Martin Logan LX16. On the 30 degree off-axis response, much can be attributed to Mayer-Norton impedance aggregations, perhaps from bipolar junction inductance, somewhat akin to stator hysteresis and non-complementary diode configuration. The crossover inhabitance ratios, when viewed within impedance truculence vectors, are stationary, yet sinusoidal. The big dip at ~ 3500 is more measured than perceived by the listener, due to cubic ionizaton when viewed withing a Thévenin linear framework (measurement) rather than a inductor co-valence (listener). The vertical off-axis is excellent as is phase subjugation and transient variance deflection points. A structural recapitulation, especially when woofer isotope superpositions are considered (i.e. Horowitz, Ravel) can be quite apparent during on-axis phase subduction when aligned with polarity confluence.
As always, I appreciate your reviews and keep up the good work!
Sir Maxwell, thanks for that entertaining analysis. I'll be sure to avoid bipolar junction inductance in the future. When my speakers' junction inductance becomes depressed, the high frequencies fade. I try to keep them as manic as possible!
I couldn't have said it better myself.
That's easy for you to say.
I will love his outcome on the speaker that I have just shipped to him.
Great insight into speaker manufacturers
I get the impression that this guy has to tread a fine line re his impressions of speaker companies products and quality
Seems at first glance most speaker manufacturers cut corners to save a buck or two at the expense of sound quality....
Hi Danny, you can add another to this list: ELAC UB51 bookshelf. You evaluated these speakers for me earlier this year and we agreed they were not good candidates for upgrade due to crossover complexity & size vs cabinet access.
The last speaker is a real eye opener. Geeze!!!
Hi Danny, your videos are inspiring and educational. Do you have a video showing the equipment and exactly how you measure everything in your graphs. For an amateur like myself, it would be invaluable to learn as I'm wondering how I can replicate this. Thx.
Interesting on the Harbeth . As the driver are supposed to be be high quality control drivers . Maybe with the right crossover they might not need that many parts !
LOVE IT - LOVE IT!!! Thanks Danny, as always a great perspective shared.
I own [not bragging - not apologizing] Harbeth 30.2, knowing its strengths and (possibly many) weaknesses. I could not go with the 3PSR (sp?) Familiarity with these Harbeth speakers, the family line-up, and the BBC monitor speaker legacy... *I firmly agree with all you said* The Harbeths are "pleasing to listen to" but, I cannot build (amps) and do reference listening through them... I do not trust them, as a way of saying it. I am using (first order) Silverline SR17S for my critical listening at the moment.
I would love to hear and will one day experience GR products
first hand (hopefully) in my home.
It would be interesting to have those Harbeth drivers measured to see or hear why they needed such an intense crossover. The original LS 3/5 A had 3 autotransformers and 15 more parts. For a 2way, Then there are well regarded 2way designs that just used a cap on the tweeter and nothing else. Strange technology.
I own Rogers LS3/5a, Harbeth LS 5/12... they are both "clogged"
with crossover network complexity = the British way...
The BBC decided on the frequency output matching to within 1/2 db across the frequency range, hence why LS3/5A's and other models are licenced, to get that frenquency shape the crossover needed a lot of components, strangely Kef who originally held the license used cheap components in their crossovers?!?!... still amazing sounding speakers for the radio, which is what the BBC designed them for, as near field vocal studio monitors... shame you need 100w to gett them driving properly at 84db....
The SPL chart from NHT SB3, one of most refined bookshelfs, is completely different from Stereophile, which is in overall quite flat. It has a slight hump around 800 hz but is within 3dB but does not have the serious slump around 3khz (dipped down 7dB) as shown here. As you know around 3 khz is in the midrange frequency range. Dipping down 7dB should be quite audible. I had the speakers but I could attest SB3 is one of sweet sounding speakers in midrange. I definitely have not heard any recess in that frequency range. Something wrong with Danny's measurement.
Our measurements are quite accurate and they are what they are.
lol I kind of feel like if the Harbeths sound as good as they do with cheap parts maybe parts quality isn’t as important as some people think it is…
Really? I thought they sounded horrible and now I know why. Parts matter.
First, parts as they be, the Harbeth's sound damn good, probably the best small monitor I've ever heard. In the end, that's what matters.
Second, supposedly with the new XD versions Harbeth has changed the parts in their crossover significantly.
If that model is the best small monitor you've ever heard then you are in for a treat. Every kit we offer out classes them in every way and can be built for significantly less money.
@@dannyrichie9743 Well, to to be in mind of Mandy Rice-Davies, you would say that, wouldn't you?
@@sebd6307 I have to agree. There is no doubt that Danny KNOWs a LOT about MEASURING and he loves big ass parts and crossovers. But, some of the most wonderful, musical sounding speakers don't have those and just don't measure perfect. Sorry. My ears have to be the final judge and if I don't like that flat sound what should I do? Turn in my EARS, lol. Decide that my hearing is IRRELEVANT. Learn to love sound I don't? Sometimes I watch this channel and see those elephant crossovers and I just sigh! Sure you can upgrade everything, but does that really help in ALL cases? I cannot worship at the Temple of Measurement. But, more power to Danny. Because lots of people love HIS sound and do worship him and his designs. It is subjective and that's OK! That's also the PARADOX. That what measures flat does NOT sound good to everyone, but does to some, perhaps even many? But that thought never seems to cross his mind. It is NOT possible?
It's time for you to hear KEF LS50 Metas. You have a homework assignment now.
@@LarsonChristopher I owned the Kef LS50 for 4 years... I sold them, I own Harbeth 30.2 now, they sound a lot better than the Kefs.
Hi there, one advice is to set your lens on manual focus. As you don't move alot in this video. Please consider. Tq
That third speaker you showed as the final one in your review was staggering! I have never seen so many components used in a crossover. Says a lot about the speaker.
Mentioned cheesy parts in the Harbeth crossover. Suddenly I had the urge to eat cheese. Yumm. Thanks! Also, I'll be looking into your speaker offerings. Thanks!
Those NHT speakers are using that inverted driver pair that i remember from my Mission Cyrus speakers i had back in the late 80s/early 90s. They were recommended to be used at slightly below ear level, like many of the speaker support stands that were available at the time, so that the crossover "beam" pointed slightly upwards to ear level. If I remember rightly!
I think all these speakers sound really good,sometimes measurments mean nothing.....
yeah... they can mean or not mean as much as one wants or needs or doesn’t. i like erin’s method of listening first and then measurements after confirm his listening experience.
Hey Danny,I priced the round trip shipping costs but the Can. Govt. wants almost as much for the import duty (based on insured cost). I am considering one of your cheaper kits (for not much more money) to get an idea of the sound of your speakers.
Cheers,Darrin
lol, Canadians and the BS you all deal with.
No offense but don't you all see a trend in these videos? There are no good speakers unless GR research designed them, or sells a kit.
Speakers are built to a price point, which can always be extended. He's actually praised quite a few speakers. There are, however, many speakers out there do not warrant their price point, and there's nothing wrong with pointing that out.
Yeah, and he thinks flattening speakers is desirable in every instance
It's an infantile approach which shows vast ignorance
@@thunderpooch Okay, sure, there are people that enjoy the peaks and valleys that some speakers add to the music being reproduced. However, the goal for the majority of manufacturers is to have a smooth response.
@@thunderpooch
Flattening a speaker simply means its response curve will faithfully reproduce the sound of the instrument(s) played back. Anything else is a modification of the original sound. It is the ONLY true approach toward accurate sound.
Hi Danny, I've been looking into the Lii Audio range of speakers. Decware and Cantuck Audio use them in the open baffle 'Betsy' speaker. I've been weighing up whether to get a pimped out version of the XLS Encore or some of the 15" full range drivers and a 15" woofer from Lii Audio. Shipping to Australia is always going to be costly.
truth from a good man
The P3ESR is a highly acclaimed and (almost) universally loved mini monitor that’s been made for over a decade. How to square this with Danny’s review? It seems that other factors matter more than having expensive parts in the crossover.
It has a very balanced response, but it is being held back with those parts and to a significant level.
It’s also the most resold speaker of all time too. 3k for the cheap parts is more the point here.
It has a certain sound that is probably more intimate or smooth sounding for a lot of people. Plus they look nice, are small, and a nice story
The sb3s are given a very flat result in another review using same type testing you do , hmmm
Greatly appreciate your speaker videos. Always interesting. Regarding the slight zooming issue, I think it's your autofocus on your camera. Set it to manual, not auto, and make sure the focus ring witness marks on the barrel of the lens is set approximately to the distance from the lens to your seated position. The lens is searching for focus I think.
Autofocus hunting for sure. Set to manual.
As for the Harbeth---If we believe objective measurement indicates quality then we shouldn't worry about the nature of the parts used, only the measured results. Shaw has evidently used parts adequate to achieve the objective results desired. It looks like the maker of the video is trying to play the subjective and objective cards at the same time.
Objective measurements don't indicate quality. They indicate accuracy. That is only part of the picture. Parts quality matter a LOT. They can measure great and sound great, or measure great and sound like a speaker playing through a wet rag. You'll never get great sound from budget level parts that are smearing the signal and cause loss of details and spatial ques.
Well, he is constant in calling out half-assing speaker parts and design.
at 5:00, is the 'ringing' at 800hz, the big blister in the spectral decay, is it caused by the woofer driver itself, the damping inside the speaker, lack of a port, or what ?or just a driver flapping out of control around that frequency? or is it just an issue to be fixed with crossover design?
So many astonishing comments here. Each owner sent in their speaker for consultation and advice, not a pat on the back.
Stereophile reviewed the SB3 and it got high marks with sound and measurements. So I don't know where this is coming from. I have the SB3 and they sound great. Great bass and soundstage from them.
The one that I just measured had a huge hump in the woofers response with a lot of stored energy there. If it wasn't for that I would have seen what could be done with them. They are loaded with cheesy parts too. If you think those sound great then you haven't really heard great yet.
@@dannyrichie9743 Audioholics dissembled the SB3 replacement the Classic Three and never made such claims.
@@mikedinno8413 I can't help what they claim. I am not a reviewer. I am assessing them for an upgrade and I report what I find.
In the Stereophile review, the issues in the 800Hz range were mentioned several times. However IMO several speakers out there have some type of measured imperfection yet some things some can be heard by some and not others. Take it for what it is and understand that that Danny's goal is and how it differs from reviewers. He's trying to correct the measured imperfections... no speaker or room is perfect, some measure better than others, some sound better than others! ______________________________ From the Stereophile review: "However, I was a bit perturbed by the peak apparent in the octave between 600Hz and 1kHz. This will tend to throw midrange detail forward a little in the soundstage, as well as make the speaker a little intolerant of high-level voices." _______________ "Figs. 5 and 6 show the NHT's lateral dispersion, the latter having the off-axis response curves normalized to the on-axis response. The peak at 800Hz is persistent, which, all things being equal, might be expected to add a slight nasality to the SB-3's balance. As BJR noted no upper-midrange coloration, this peak is perhaps low enough in level that it slips by unobserved, other than its effect on the perception of recorded detail."____________ "The SB-3's step response (fig.8) confirms that both drive-units are connected with positive acoustic polarity, with the tweeter's step smoothly handing over to the woofer's. The undulations apparent in the latter are associated with the 800Hz peak, as can be seen in the cumulative spectral-decay plot (fig.9)."______________ "Overall, this is quite good measured performance, though I was bothered by the existence of the 800Hz peak."
To those who think measuring with inadequate electronic equipment that is simply generation behind the most sensitive instrument we have , our ears. Tubes! They don't measure well but to those who think they don't sound fantastic really needs to get their ears checked and learn how to listen. The same with speakers to an extent. Every good recording or engineer have tubes in the recording chain. That messes with measurement but is better for what matter most, it sounds just better. But preferences do matter even if the result is not realistic compared to the actual recording session. Inches, we improve by inches compared to live sound.
Shocking harbeth! I have to say harbeth sounds most easy and pleasing to my ears.
So, do you listen to music or scope traces? We buy what we like, what makes us feel the music more.
no audiophile listens at 30 degrees off axis. it just doesn't happen
But what is going on at 30 degrees off axis is exactly what the side walls will reflect. A poor off axis will cause a poor room response.
Always a great, honest review. Some speakers have benefits to improve and others simply don’t. These 3 will be on my “Avoid” list.
Try them before you buy them and avoid those speakers , human ears tell you many times better than sophisticated testing equipment
GR-R thank-you for bringing truth and facts for the consumer.
Previously watching so many RUclips reviewers emote enthusiastically about these speakers, especially the fairly expensive Harbeths as a near field monitor, makes me question either their ears, taste, or truthfulness. I guess you can "like" something, even if it doesn't perform well, as long as it crosses off some of your expectations. But wow, when you opened that up and we saw the army of parts needed to make it sound "good" just blew me out of the water. Crazy. Why would they do this? Why not start out with better drivers in the first place and use less, but better, crossover components? I just don't get it. Sort of like your review comparing the $350 Wharfedale to a $4000 Revel showed you don't always get more for your money (but should). There's nothing wrong with the laid back, smooth, "British sound", but there ought to be a better way of achieving that. Thanks. Hoping someone will send you the KLH Model 5 one day.
I believe those Harbeth were 4 years in development. Was that how long they took to design the crossover?
Curious why you think the drivers are not of good quality, especially when looking at the spectral decay plots? Alan Shaw of Harbeth will tell you that the proprietary Harbeth-developed RADIAL material that the mid/woof is made of is one of the best materials he's used in 40 years of speaker design and manufacture to accurately reproduce the human voice. I agree that the specification of parts utilized in the crossover could be better, but on the other hand, one can counter with the view that it's hard to argue with the results: the speaker measures _very well_ and the spectral decay is really excellent, as well.
I've heard a couple of smart audio engineers make similar statements about Bose's satellite/subwoofer systems: "No highs, now lows, must be a Bose", followed by a diatribe about the crossover design.
But the speaker does perform well, objectively. So what's the beef?
@@tombrennan6312 Well, the host of this video points out that the speaker is performing like an instrument and adding it’s own color to the sound, plus the cheap components are smearing the sound and leaving a lot of resolution and clarity on the table. It just doesn’t give you everything from the recording.
That said, people may not want “everything” in a recording. My step mother absolutely cannot stand music from any speaker except her absolutely POS iHome speaker that softens and veils everything that comes out of it. Sharp details startle and fatigue her. I wouldn’t say she’s typical (or the kind of person an audio company would profit off of having as a customer), but there are other people like her and thus there is a market for pretty, soft, background speakers.
Interesting to see the crossover on the Harbeth, thanks Danny!
Harbeth, speakers with too many parts tend to suck the life of the speakers, also it becomes too inefficient typical of Harbeth speakers, you need to use a high powered amp to drive it to a reasonable level, also to get the dynamic range.
(LX16? Are you looking at the right sheet of paper? Also, these require break-in) I beg to differ; I tried eight high rated pairs of speakers over the last two years, and I finally tried the Martin Logan 15i speakers. The 15i speakers are the best sounding bookshelf's I've ever owned. Sometimes, the measurements don`t tell the whole story.
Maybe it's time for some new speakers.
@Danny Richie You didn't listen to a pair of 15i speakers properly placed and dialed in, did you? No, you measured it, right? And mistakenly called it a different model. I am sure your equivalent of these speakers would probably impress me a lot, right? Why not send me a pair to evaluate? If I like them enough, I will buy them. If not, I will pay for the return shipping. On another note, how was the crossover quality on the ML 15i speakers? They are $449 a speaker retail. I paid $269 a speaker through an outlet on ebay.
@@scottbennett3119 95% of all models going out the door are DIY kits. We don't send out samples. Comparing them to something like our X-LS Encore is like asking me how my Smart Car compares to my Corvette (yes I own them both). The capacitors in the tweeter circuit of the ML's were budget level poly caps (fair), but everything else was super cheap and bottom of the barrel quality wise. The woofers response was so bad on them (too much ringing and break up) that there was no fixing them. Oh, and I'll even get my kids names and employees names mixed up on a regular bases. Nothing new there.
@@dannyrichie9743 Thanks for getting back to me. I do like the mid to high end on the MLs, I don`t listen at higher levels much, so I haven't experienced obvious ringing or cone break up due to my wife being with me a lot of the time. They seem very good at the levels I usually listen at when crossed to my sub at 80 Hz. I would bet better parts would be helpful at the levels I listen at. I have limited strength, so I have to stick with small bookshelf models, so I am able to dial them in myself. I have been keeping an eye on your X-LS Encore though. Have a good one...
@@scottbennett3119 So, you want everyone to acknowledge that your Martin Logan is some miracle speaker because you paid for it? No, it is a piece of crap. Now, move on and enjoy your speaker.
Big part of problem with “high end audio” is the lack of a standard of sound quality (and it’s measurables) which is accepted by everyone. Once “everything goes”, companies will spend money where it counts for the profitability and sustainability of their business: they will spend on appearances as opposed to substance. On Marketing as opposed to R&D and build quality.
From your valuable Channel, it appears that it takes a serious effort to PROPERLY build a decent speaker: good design, well dampened and braced cabinet, as-complex-as-required crossover with good quality parts to keep the response flat, since no drivers are “perfect”. Even highly regarded companies like ATC could upgrade some components (better braced and dampened cabinets), to quote your channel. So where does this leave the poor consumer when looking for a decent pair of speakers at a price point which doesn’t break the bank?
And all these considerations do not even mention good room acoustics (the feasibility of applying room acoustic treatments to anything but a dedicated listening room while staying married), to assume the full potential of these aforementioned speakers…this hobby is not for the faint of heart…
There are some gems out there, and we do offer some good considerations in the DIY side of the hobby.
Stick with JBL Studio series. You cannot go wrong and when on sale you get excellent sound quality at a great price
I forgot about the NHTs!!!!!! Are these still being manufactured?
To be fair nht was and is a brand that always was a budgetary spealer . So you can expect top parts on tje crossover. Or terminals what not
I would be interested to see/hear what an upgraded external crossover for the Harbeth would measure/sound like. You could remove the old junky crossover, install some stuffing or no-res inside. Also, you could install tube connectors to the external crossover and then run high quality speaker wire through the cabinet back directly to the drivers. Or alternatively, you could remove the original crappy crossover and use an active digital crossover instead.
I could not wait to see this video... knew it would be good...
This grounded [Danny kind of] perspective flies in the face of truthfulness, Alan (the designer) claiming "perfection" with his Radial driver technology... dismissing so many other noteworthy designs. He speaks so sharply of the LS5/9 speaker / its' drivers, the woofer being poor technology... WELL...
The Rogers LS5/9 (in spite of numerous, documented tonal anomalies) was one of my favorite speakers that I have ever owned and deeply regret selling (to please my wife) it was just fun to listen to the 15 years I owned them. Flawed can be fun, just .not good. as a reference tool.
I value accuracy, closer to the mark Danny is defining for us.
Now that was an enjoyable honest appraisal with a good commercial as well as technical conclusion
You are the coolest guy on the planet. Seriously, I really mean that. And you just drip with confidence and expertise. It would be worth driving from where I live in Maryland to Texas just to shake your hand. Thanks so much for your videos.
Sorprendente! Tengo los Falcon sl3/5a Gold Badge, que están emparentados. Sería muy interesante explorarlos y compararlos con los Harberh!
I would think the average Harbeth buyer doesn't care about what you're talking about. They care about how it sounds, in the moment. And they're not cheap, so the buyer is not pinching pennies, and neither is the manufacturer necessarily. I'm thinking someone bought the wrong kind of speakers for them, if these ended up in your hands.
If anyone really cares about how a speaker sounds then that one should not be in consideration. They are a long way from the expectations of the buyer in that price range.
The Harbeths are clearly trying to be a modern day take on the LS3/5A which in it's day had quite a complex crossover network due to the raw response issues of the KEF drivers. Surely modern drivers should be much better behaved than the T27 and B110 from the seventies you would think. Something obviously wrong if the Harbeths need so much crossover intervention.
I have compared Rogers ls3/5a 15Ohm with them, Rogers sounds better, both were tested with Accuphase E-650
The Heberth, maybe no crossover in the cabinet at all, but an external Xover in it's own box?
miniDSP it.
I get it! But a lot of these designs were based on 70’s or 80’s thinking. We had no FFT or the ability to measure science back then so the boxes and the behavior of them were big unknowns. Celestion looked at this problem with the SL 700 and later the budget version the SL 600, using science and laser scan plotted drivers and fixed the box problems to produce one of the most perfect examples of a small to medium size speakers. Measurements and science produce predictable and highly tunable results and are critical. Voicing after the event causes the manufacturers to try to fix issues using cheap components where actually designing the product from the outset makes really outstanding products from the beginning. And…… then there is international taste and expectations about what quality actually is!
Oh yeah? Look inside at the crossovers.
I'm honestly shocked by the Harbeth. If anyone would've had better bits and pieces I figured they'd do so. What a cheat.
I googled and they did a 40th anniversary edition and- upgraded internals. Hificorner have a picture of the different insides and blog post about it. Cost for the "Anniversary edition" extra £845. I also agree, with such a good reputation I didn't expect Harbeth to have gone that way. But I am enjoying the reality of someone opening these things up and making it clear what's good and what's not.
@@willjohnson2722 it's easy to dismiss Danny for being such a small player in the audio market as many do. But, when he pulls back the curtain and shows stuff like this you suddenly realize why the little pizza shop across town makes a better product than the Domino's down the street: accountability means more at such a level.
@@akr01364 thank you for your comment. I agree 100% and with the sentiment.
I am not a tinkerer. But I do love things to be the best they can be. And so I love channels like Danny's where things are opened up- to show what's inside, speakers, group tests etc. What's it really like?
Most of the videos I've seen seem to show products that are engineered to sound the best they can at the lowest possible cost= the maximum profit. And that's understandable when people are running a business, because dealers take a large cut, VAT on top etc etc. High quality hifi is already a niche industry, production volumes are already low- I don't expect people are making a huge amount of money even when they are charging higher prices for lower quality products. But it's still a bit sad to see lower quality components in a product with a high price tag.
Catch 22 in Hifi though, in my experience in the UK anyway- say you get a speaker made at the local boutique. It costs a fortune because it has high quality materials and components and that's why it sounds amazing. If you're like me and a lot of audio enthusiasts, let's say- in a year or two, you change system, downgrade, upgrade, whatever. Who's going to buy your speaker? Good luck selling it without a massive loss.
And so over the years I've come to understand why people do stick to more mainstream products sometimes- not always because of the sound quality, but often because of the ease of buying, selling and support. And people don't want to buy modded products for the same reason they don't want to buy modded cars. However- a known product, upgraded by a known and respected engineer, like Danny- that's a super interesting proposition. And as always with the internet that's cool- people can find your videos years later and learn- like I did here.
@@willjohnson2722 this reminds me of an interesting back and forth that occurred in one of his comments sections: post modification resale value. He noted that a speaker featuring a well done modification he'd put together typically sells for a higher pricevthan one that wasn't. That discussion made sense to me as one of his more famous kits is for the Sony CS bookshelf speaker. The standard model sells for roughly $120 USD. One I came across on Ebay featuring the GR Research kit was, sans reserve, $550 at the time and likely went for more. Imagine what one of Sonos Faber's models would go for? And if there were to be a problem, he'd be able to help remedy it. A pretty decent middle ground when you think about it.
"a lot of these older speakers they really weren't designed that well to begin with..so to rebuild some of those around an old cabinet sometimes does not make any sense" Go and tell that to Kendrick Sound who rebuild old speakers like JBL for example being the basis of there successful business.
yes, I heard those Harbeth speakers recently and certainly are not worth €3000 here in europe. Danny speaks the truth. Even the cheapest GR Research kits will outperform them due to quality parts and great crossover design. Beware of sexy shiny exotic cabinets they are definitely soaking up your money !
Agree 100%. I would take an Elac or Wharfedale over these any day.
i recently made these X-Statik's they are extraordinary indeed, all the money on the parts/design/Danny genius - not a lot on the sturdy painted cabinets - happy days
ruclips.net/video/R389WYYgSLQ/видео.html
Wow, i am absolutely shocked. Harbeth are not cheap speakers, yet you found them to be a total mess. Just goes to show, no matter how much you pay for a loudspeaker. There is rubbish at all price points. I won't ever be buying a pair of Harbeth after seeing this!!!
And all these so-called hifi reviewers rave on about the greatness of these speakers!! Just shows, who can you trust?!
@@playbackvintagehifihunter9669 i agree with you. Are they being paid to write a glowing review???
@@markholder6851 I honestly dont know, but I hope Mr Alan Shaw is asshamed of himself. In fact I'm appalled by so many so called speaker companies using crap parts. Total rip off. Shame...
He did not say they were a mess. They are relatively flattish but will not have excellent detail.
@@christophern.9234 He did however say "Super Cheesy Parts". I doubt he would use the same description for a genuine LS3/5A.
Since the NHT has such a nice and fairly large cabinet cant one install more suitable drivers that require a simpler crossover after having flipped over the cabinet? It seems that the majority of ready made speakers need to be sent straight to the hospital for treatment after purchase .
Finding drivers that fit the same holes and need the same air space is an impossibility.
Ahhhh.....I wished he has looked into the harbeth 6 months earlier.
If you have a pair and like them, why worry?
@@julesbarry2081 you are half right. At least I get to know what I am paying for.
There's actually a "kludgy" (cheap) way to upgrade those Harbeth: move the crossover outside from the speaker, and put it into an external box.
The box is really small and the woofers air movement stresses the components since the placement is too close and it will crack those poor cement someday.
.
as for the sound... there's really nothing special about those Harbeth, it does sounds pleasant (coloration, that is) as Nearfield, but it didn't behave really good for 2+ meter listening position even with a better power amps; I've guess people just think them as "decent all-rounder" ONLY if you sit really close to it.
Or just buy better speakers.
Danny’s analysis is similar to that of a Medical Specialist. Quality testing, analysis and diagnostic assessment of a “speaker’s health”.
Dr. Danny reveals the true “health” of a speaker the comes across his “examination” table.
Always informational, factual and accurate presentations.
How much is it worth to a prospective speaker buyer to pay for a high quality “Piano Grade” finish versus true audiophile high accuracy and high performance audio reproduction? Yes some may want to pay for that high grade visual appearance but personally for me I never sit and stare at my speakers just for their visual appearance and even while seriously listening I pay little attention to how the speaker looks but I want a fine experience of enjoying whatever music I choose to indulge in.
Thanks Dr. Danny!! 😉
Had high hopes for Nht, always wanted a pair 😕
That harbeth crossover was epic 😁
Cicable LS3/5A crossover exists just for the reason of a lack of space being available to place a crossover too large for the cabinet So I would suggest you tell your client of such a solution for the Harbeth P3esr (se/xd)
A channel like this isn’t going to popular with people that wanna live in ignorance bliss of the high amount of $$$ they paid for whatever speaker it is they have
It reminds me of fanboys back in the day that fought to death to defend their phone brand, video game console brand, TV brand and what have you.
It’s very hard to accept objective truth if it goes against what you bought
I see a lot of those people in this comment section.
Super interesting to see the Harbeth characterization. It seems clear that if you want a fully assembled beautiful cabinet in a two way monitor the Tyler Acoustics monitor is the way to go since it already has a GR Research well designed crossover and parts quality.
fyi - my system takes about an hour to an hour and 1/2 to open up - most of my speakers took at least a year to burn in - keep the faith ok - I run a Benchmark DAC1 as my source, a Creek Destiny 2 Integrated - a Hafler HA75 tube amplifier used as a tube line stage ( running Telefunken E88CC's or is the other way around from Kenny at Upscale Audio ) and Harbeth HL5's but I listen to things like Everything Like the Girl, Alice in Chains Unplugged, and Karen Carpenter singing "This Masquerade' so what the f do I know .......
There were more parts in the Harbeth crossover than the worlds first computer.
... Or in the Apollo capsule.
Not really, just a lot caps in parallel. And some RLC traps.
They are all full of cheesy parts. If they weren't GR research wouldn't make any money.
Lots of posters are missing the entire point regarding parts quality . How long is that crossover going to last before the cheaper parts start to degrade? It might sound great….for now.
lesser quality capacitors limits detail according to gs research this does concern me when looking for speakers in a specific price point I was wanting the linton heritage 85th anniversary but gs research was saying the crossover was using budget components
please do a tear down/components evaluation of the Elac Uni-Fi 2.0’s. i’m EXTREMELY interested in what your thoughts on their crossover would be.
i would gladly send you my pair for evaluation!
Great video, I really like your honest, results based feedback. Unfortunately for me, now I am questioning my little Harbeth P3ESR's. I suppose that I need to talk to about which kit and components you would recommend as replacement. I would be willing to build a set of your speakers and make some side by side comparisons with the Harbeth's to come to my own conclusion on what I like better.
Why are questioning your p3esr now .????? Do they still sound good ?
If so,…why the questioning !?
I love my p3esr ,…yes, they are not perfect ( which speaker is) ?
But , I love the sound they give , colorations and all ! Heck…world is coloured !
I have always defended that some speakers you can modify ,…others you just ruin them !
I have had a pair of mission 751 like 30 years ago, fabulous speakers ,
Pull them apart , modify them with good poly caps , better resistances , better intern cable .
Result: a speaker that was better in everything detail , hifi , dryer and faster bass, but……something was gone , the fabulous 3D soundstage these speakers kan do ……..was gone,…..they turned forward and cold ,
So,….I do believe that cheap parts or not , sometimes you do have engineers that know what they do , and they realize that the voicing of a particular component in the crossover is the one that must be used !
Is like driving an old jaguar , it gives a special kind of pleasure , now modify that same car with top notch motor , suspensions , and in the end you will a car with better specks …….but ….does it give you more pleasure driving ???
Relative to harbeth p3esr ……common !!!! Who whants to improve a speaker that works good as it is , I do have a pair , I personally would not modifying them even if it was for free,
I like their voicing , ….. thus modifying = another voicing .
Perfection is not always the path to completion .
If I like a certain sort of coloration ….why modifying it ! What’s the point ? And…..is not truth to the signal !? Who cares.
Going to listen to some music,
Bye .
The reason the Harbeth crossover has so many parts is to correct for non-linear drivers. Better drivers would cut the crossover parts count by two thirds.
No, the reason is that the kind of response the designer is aiming at requires this kind of complexity when using a passive crossover. Using different drivers wouldn't help much.
@@matthiasmeyer1124 You mean like flat response? I’d like to see raw response graphs for the individual drivers. I’ll bet they’re all over the map. Linear drivers don’t require anywhere near this level of correction.
@@ChicagoRob2 The P3ESR isn't a particularly flat voiced speaker. Flat isn't difficult to achieve. There are many first-timer DIY projects that measure flatter than the Harbeth. For some of the dips and humps of the P3ESR I know that they are intentional - not for lack of skill or due to budget constraints. I have a P3esr clone project at home, using a well-measuring driver and I need a similar parts count to shape the reponse. I can't vouch for the Harbeth woofer but the tweeter is a SEAS driver and superb in an infinite baffle. All raggedness of the tweeter derives from the cabinet. I can make cabinets that perform better in terms of measurements, but subjectively the Harbeth baffle wins for me.
@@matthiasmeyer1124 The sound they are trying to make? Or reproduce? Your statement, while possible right, is wrong on so many levels.
@@matthiasmeyer1124 So, they are designing the crossover to intentionally color the sound.
in the first speaker: would there be an option to put a helmholz resonator inside the box that counter-acts that 800Hz resonance? where is it coming from? is this a standing wave between top and bottom?
probably too hard to add pieces of wood inside and glue it down to form a new chamber with a small hole that connects it.
These speaker manufacturers have a lot to answer too … only a handful you have reviewed that you would recommend ,keep up the hard work Danny ✌️💪🏽. Ps is there much you can do with electrostatics “quad” ?
Well, keep in mind it’s a time honored tradition to point out your competitor’s failings 😅
@@evshrug Danny does not compete with anyone … cheap nasty parts are still cheap nasty parts end of
it's just flabbergasting to me that these renowned manufactures are designing such poor speakers . honestly let's face it.
I have a question, what about removing all these passive components and just running a digital crossover at the DAC level? So instead of using a stereo amplifier to individually amplify each speaker we would use two stereo amplifier with each channel amplifying a specific driver instead of the full speaker set.
Thi would also allow for way quicker iterative improvements wgere it would be possible to quickly measure and apply a different filter and then measure again to get the FR to look as good as reasonably doable?
Imo analog crossovers are probably gonna be better for most sellers. But for diy, DSP is definitely the way to go