The term waveguide was introduced by Earl Geddes to describe horns that he developed based on his equations to define the directivity of a horn. As he said in the interview (and on his website) all waveguides are horns, but only horns that are based on his waveguide theory are waveguides. Of course the term waveguide has now been absorbed by the audio world and the original definition is fading into obscurity. But the main thing to remember is that if it looks like a horn and quacks like a horn, it's a horn. In the video I misspoke and said "constant" directivity instead of "controlled" directivity. Sorry, I'm not working from a carefully prepared script written by my team of support staff. I'm only human and make mistakes. You probably make a few yourself and I forgive you. Now onto the fun stuff! The wild and crazy phase plug experiments were interesting. I think I'll make a pair of the 50's tail lights for my horns (not waveguides...) from maple just because they look nifty. Like I said, I'm never going to hear the difference either way, but I'll enjoy looking at them. Also a side note: Guys tend to take all of this audio stuff way too seriously, and zoom in on the smallest details, blowing their importance out of proportion. There is this idea that if you do all of the little things "right", then you'll reach audio nirvana. And if you ignore those little things you obviously can't be taken seriously and any speakers you make will be subpar. This is simply not true. To get great sounding speakers you need to do the big things right. That will take them 99% of the way. That last 1% is in those 100 little details that can be tweaked. If tweaking details is your thing, then go for it. But stop fooling yourself on their significance. You don't get a "night and day difference" when you round the corners on the cabinet. Smoothing out a 2db ripple at 5Khz won't "add 3 feet to the soundstage depth" or some other miraculous improvement. Understand that we don't listen with our ears - we listen with our brains and our brains do all kinds of fun stuff with what our ears hear. The interview with Earl Geddes: ruclips.net/video/nhe8VfuTg08/видео.html
If you have the capability to equalise then the best by far approach is to make the on and off axis response the same giving a good power response that can now be manipulated however you want. Pro audio went down this path decades ago and now just laughs at the HiFi worlds half arsed attempts to get good room response. So learn some maths, design your horn to be constant directivity and accept the topend rolloff on axis and then equalise. All your little manufacturing fun can then be directed at reducing throat distortion and getting the directivity to match throughout the complete response for all drivers in the system.
I like your down to earth and no BS explanation of things! A lot of times frequency response charts say nothing, most oldtimer audiophiles can't even hear beyond 15khz, and just trying stuff out and seeing its impact makes things way more clear and easy to understand for laymen like me!
Nicely done John. You did a good job on this video. Welcome to the world of phase plugs. They give with one hand and take away with the other. Two things I would add. Nothing major in the least. The term waveguide is borrowed from radio antennas. Earl used waveguide as much for marketing as he did for setting himself and his work apart from other horn design work. It has been said that a waveguide is there to control directivity and a horn to both control directivity and add gain to the output device because of load matching the diaphragm to the air in a way that better matches the pressure generated by the diaphragm to the surrounding air. A horn allows more molecules of air to get a kick in the butt than just running the dome into the room. Excited molecules of air are what gets us sound. I'm happy that you did some EQ measured on and off axis. Again EQ is not magic. It causes as many problems as it solves. Nice to see you keeping it real. One last observation. The region from 2 to 5 kilohertz can shift your ambience a lot more than anywhere else. I know you understand this. Even a db and change can be noticed under the right circumstances. That I have played with many times. Again a lot depends on your recordings. And how they are made. If it is a two mic recording and all the cues to convince our ear/brain system are due to inter-channel cross talk and recording room reflections then you have a auditory magnifying glass to listen with. If it is a recording done with mono micing and mixed by a skilled engineer, then you have less of what can trick our ear/brain into thinking what is being reproduced in front of us is real. Thoroughly enjoyed watching. Keep en coming.
EQ can definitely do that if you overuse it (I have in the past) and don't look at everything it affects (I've done that too...). But it's a tool like any other that can fine tune to get the results you want. Much better if you can do that on an individual driver level, rather than globally, and limit how aggressively you use it. Having a digital crossover with PEQ filters for each driver is a true luxury and makes the passive crossover option seem stone age in comparison.
What I find interesting about Earl's interview is the multiple subwoofers which I knew about but didn't know he was the one to arguably do the research first, and also the passive vs electronic crossover design, which I've also heard that passive can be "as good" as an electronic if designed correctly in synergy with the mechanical aspect of the loudspeaker.
Great video! I made a pair of spears using full range drivers and removed the dust cover for a phase plug. I did exactly what you did- quick prototypes to see if I was headed in the right direction. I used aluminum, copper, brass and wood. The final result was bullet shaped teak with a flared brass sleeve down the center. I find it amazing how much little changes can shape sound.
Yes! There are just as many "right" ways of building speakers as there are manufacturers. If one way, be it horn/waveguide, dipole, bipole, Omni or cardioid (my current personal designs over the last few years) then everyone would be clamouring for those. Also besides the brain/ear don't forget how we are simply swayed by the way things look. Keep up the awesome work, always looking forward to your next video, and can't wait to see your final design!
Why do you use foam as a phaseplug? Because it;s easy to shape? A wooden dowel or a screw or a small coin works too. Because all you need to accomplish here is to slow down or lengthen the path for the sound from the tip of the dome relative to the sound from the surround. If you can get them in phase by redirecting them around an object, you get the same effect as with forcing them to go through foam. Except foam additionally attenuates it.
The term waveguide is a marketing/advertising term since all HiFi people "know for sure" that horns are colouring the sound. The industry now calls them waveguides and voila all the colouration is gone. Magic!
For that upper roll off you can compensate by adding a bypass capacitor to the series resistor you'll probably use to bring the sensitivity more in line with your mid-driver. The trick, of course, is picking the cap value (it won't be large) but it essentially turns that resistor into a frequency controlled resistor. As the frequency rises, the impedance of the capacitor lowers which will shunt the signal around the resistor. Of course, if you're using DSP this is moot.
i love these videos, this series, this channel. and the comments, which are also mostly great, have just enough audiophile nervosa insecuritas to keep it spicy!!!
i notice it's just 1/24 smoothing, that's an impressive response and not just for diy speakers... the off axis is also somewhat in line with on axis, a little exception around 10kHz w/ -5dB, no big deal at all. that speaker will produce a very pleasant and accurate listening experience. (of course the off axis also refers to vertical) the "new idea" of speaker fq response puts greater importance on the off-axis compared to the older approach where the goal was to achieve a flat on-axis response and don't mind the off-axis too much once the listening position was chosen. ------------------- this new idea is in fact a better understanding of how we perceive speakers in a room and it turns out the reflections play a larger role than was initially assumed. cumulatively more reflected acoustic energy reaches our ears than the direct sound, up to the point where the high fqs are so directional that there's v little wall reflection. a speaker that maintains it's response off-axis relative to the on-axis (be it flat or not) will produce wall reflections that are similar in tone w/ the direct sound coming from the speaker. otherwise we hear two different tonalities... when a speaker is measured in a room, the mic also pics up the reflections and without windowing the response (cut the recording before the reflection arrives) there is no way to tell what part of the sound is from the speaker and what comes from a reflection. so applying aggressive EQ can't really solve the problem since the speaker itself may be flat on axis and it's the off-axis that creates the non-linearity. hope this makes a little sense :) (they have these days setups and software that can eliminate the room reflections when measuring speakers in rooms, demonstrably identical to anechoic measurement of same speakers. the best performing speakers based on listening tests are the ones w/ a consistent on and off axis response, even if not particularly flat.)
All true for untreated rooms, but off axis response (and reflections) are less important in an acoustically treated room. The room these are in is treated and that's why the 1/24th response is so clean without gating.
This is interesting! I have been thinking about directivity, and I've noticed coaxial designs covering a large portion of the low frequency drivers with a wave guide, or drivers covered completely in the case of Genelec. (Or 2 small drivers spaced apart, in the case of an MTM....which is a coaxial speaker in my view.) I think the phase plug creates a tighter wave guide and forces all the energy to propogate from the perimeter. I suspect this creates a more defined directivity pattern. But just a theory!
Loving the videos John! Absolutely recommend checking off-axis. I recently spent days playing with phase plugs and got the on-axis greatly improved. Then when I checked off-axis it was horrific 😢 Days of work for nothing! I was doing solid phase plugs. Maybe less chance to go wrong with a bit of foam. IMO it is probably slowing the propagation speed in that area and changes the wavefront shape. KEF say a spherical wave front is best for waveguides but pistonic drivers don't quite do that even if domed. A 3d printer is brilliant for making phase plugs of all shapes.
@@Audio_Simon I was going to try a disk after the bullet, and then got sidetracked on the foam. And before I knew it the day was consumed. I think that while doing these experiments was interesting, the net gain isn't worth the effort. Most people won't hear the reduction in output at 17Khz, so it's just for the "flat to 20k" bragging rights that it has any value.
There's not a lot of content beyond 15khz in most music, so that makes the gains even less. But with that said, I will try a disk and see how it performs. Easy enough to try it while I'm doing the final setup on the speakers.
@@IBuildIt I'll look forward to hearing how it works out. I'm someone who found significant benefit adding supertweeters with a 16khz 24LR acoustic filter to a speaker with Neo3 planar tweeter so we obviously hear quite different.
Well, often the difference (not necessarily improvement) that guys hear when they add a super tweeter is the comb filtering that happens when two drivers are cancelling each other out at certain frequencies. Unless the super tweeter is extremely close to the tweeter physically or there is no overlap at at all with the tweeter in the frequencies it plays, comb filtering is a given. If you take a piece of music you like and open it in a program like Audacity, you can see how much high frequency content there is and at what level. Also normal hearing is most sensitive in the 2khz to 5khz range, and less so at low and high frequency.
@@IBuildIt My thinking is that the main audible effect was from changing the power response of the system. Much like a rear or upward firing tweeter, helping add back some of the HF reflections that provide cues for 'spaciousness' that get easily absorbed in room or simply don't exist from a narrow directivity tweeter. It certainly made the in-room response at listening position more linear, because without it the in-room response dropped off rapidly above 15khz or so. Given that you said you can not hear the top frequencies it would be a really interesting experiment to see if you hear any effect adding a dedicated 'super' tweeter that you could easily switch on and off, perhaps with a bit of dsp for easy tweaking. Just an idea for an interesting video!
Rounding corners with radii that small makes no difference. Mathematically the radius should be closer to half the baffle step frequency length but one thing you CAN do is introduce a smooth 45 degree angle part way across the baffle which pushes the virtual source point at impedance change from cab to air neatly at the edges firing out at 45 degrees not reflecting directly towards the listener. Waveguide as a term has been in use for decades by the way. Not so sure your friend there was the first. Perhaps in the USA maybe?
My guess is that in the real world, where most of us adults can't hear above 15k hz, , the refined hunk would be the most useful version. Experimenting with a few versions of that shape could get rid of the 10k dip.
Are you testing them in the listening room or do you have something specially built for testing? I’d be curious to know more about your testing process.
30 inches for everything shown in this video. When I properly set up the speakers I'll move out to a full meter away, but it won't look much different. In case you missed it, there are several videos here showing what I did in the room, along with measurements.
Btw - I believe BOSE calls its long snake-like tubes in its speakers, "Acoustic Wave Guides". You know what Audiophiles call them? Artificial sounding, fake-bass.
Excellent sounding speakers must have well built physical boxes of whatever shape they take. Well built finely crafted speakers have to be constructed to be as solid and free from vibrations and resonances. The construction techniques may be more important than the actual physical materials they are made from. Some boxes/cabinets have exotic construction and all those internal braces and “waveguides” should further dampen unwanted and undesirable vibrations further strengthening and create nonresonant enclosures. Next quality drivers need to be used in a manner consistent with the design objectives. Putting expensive drivers in poorly built enclosures are bad design decisions. Quality components in the crossovers are important. Cheap stuff won’t help expensive drivers create what they were designed to do. Then there is the speaker placement and the room itself also parameters that will affect the quality of the speaker in ways more important than rounding off corners, etc. OB designs are another concept entirely. I think some people may have developed very acute hearing sensitivity just like some people have outstanding visual acuity and they can perceive sonic nuisances that 90% of the average listeners are unable to perceive especially those of us who do have limitations in out hearing response. (Above 12kHz I’m seriously deficient especially in my right ear) so I have to be practical in my speaker choices. Excellent presentation John!! Thanks. Btw John shopping at Walmart yesterday I encountered a gentleman who was your doppelgänger. I almost asked him how was the OB speakers coming along 😂
How a speaker sounds from under 1ft away from it... is very different from how a speaker sounds at 7 to 10 ft from it. As such, many of these measurements and recordings, are likely not optimal to how you would be hearing / experiencing them. (even with the perfect room treatment) The question is... what happens to the tweeters soundwaves, altered by these horns/waveguides... by the time they reach your ears, at about 8ft away? Its very possible that what seems good on the readings from a few inches away from the speakers... ends up sounding poorly, once those altered soundwaves.. reach your ears at 8ft away. Also, the kind of changes may ramp up in scale, of extreme levels of differences... depending on how much power they are being fed. Anyone whom has seen experiments with aerodynamics and fluid dynamics... knows how complex these things actually are. How a very small change, can cause very unexpected / big changes.
@@KipdoesStuff I know exactly why he measures from close up. You obviously have not been reading the replies, in depth. The point here... is that a Horn (or "Wave Guide"), alters the way that the soundwaves are projected. This could theoretically cause the waves shape, to become inverted, as it hits the guide, and or as it travels through the air. Thus, if you recorded two instruments in a room, using a Bineural microphone... and played this recording back on this wildly altered tweeter... it could end up causing the 3D audio positions of each instrument... into different places in the room. It might flatten the very depth of the image. It might partially or fully invert the image, depending on how far away you are from the driver. But what it wont be... is Accurate to the Recorded Image. You could then put on a pair of high end Audiophile grade headphones, and easily tell the difference in the 3d soundstage / image... from the poor quality speakers / drivers. Here is another thing that many Audio-Measurement freaks do not seem to understand... From my own personal experience, when listening to different kinds of tweeters: 1) While Ribbon Tweeters may have incredible accuracy... they have a very NARROW soundstage / sweet spot. Most of these high end speakers will sound so much worse, once you move a mere foot out from the center of them. Why is this? Its likely due to the fact that the drivers tape, is actually facing 90 degrees to the listener. This causes image distortions right from the very start. 2) Many Dome tweeters, are often too soft, and get ate up when there is heavy bass playing in the music track. Most especially at higher volume levels. 3) The most 3d sounding tweeter I had heard, was a dual pole ribbon tweeter, that had two vertical 1" tall ribbons... that were full 360 degree circles, mounted next to each other. These sat directly on top of small bookshelf speakers, allowing full 360 degree projection. The problem? Because the sound was being output in every direction at the same time... the sounds would end up bouncing off the back walls, and other areas... and arriving at your ears at different timings. It was making ECHO like effects... and completely ruined the intended Image of the recorded music and movies that were playing on them. 4) The best Tweeters Ive every heard... are from the EPI 100v speakers. They have the same concave shape that larger drivers have. They project an incredibly wide 3d image / sweet spot. You can move almost 90 degrees to one of the speakers, and still be getting an incredible 3d stereo image. Not only that.. but they vanish.. as if they are not coming from a speaker at all. Its like the instruments are actually producing the sounds in the room... exactly where the musicians were recorded. If I put a blind-fold on you... you likely could not find the speakers, even with a solid 10 min of searching. Now, Im not saying that Ive heard it all. Not even close. What I am saying... is that there is a LOT more to these "Measurements" than people realize. I can put two speakers with virtually identical readings / specs in front of you... and each will sound completely different, when you do an A to B comparison. Its not that one could not manage to figure out the best and most accurate ways to measure speakers. Its that the people whom think they are Experts in this field... are far more clueless than they even realize. And worse yet... they are so EGO bound, that they are not willing to even take the time to understand what the issues are.
@@KipdoesStuff Part 2: Im going to add one more point. While these dudes that make Measurements are so gung ho about "The SCIENCE"... They completely IGNORE the science behind what makes high end DRIVERS so much different from low spec drivers. For example... If you have a weak magnet, and weak driver coils... as well as a heavier cone material... you will get distortions. Why? Its pure Physics. If I tell you to hop in a Pickup Truck... accelerate to 60 mph in two seconds... then STOP in less than 2 seconds... can your truck do it? Nope. The "Driver" (Engine) is not powerful enough to be able to accelerate that fast. And the Brakes are not powerful enough to STOP the truck that fast. Now, imagine a speakers cone. The Musical wave, expects a lightning fast acceleration in one direction... and then within a fraction of a second later.. it expects to both stop and accelerate in the Opposite direction... at the same blinding fast acceleration rate. If your magnet and coils are not powerful enough... they cant get the cone to accelerate as fast as the music intended. And likewise... the driver can not stop and accelerate in the opposite direction, without some delays from the over-shoot... because its not powerful enough to stop it instantly. That is Compounded by a heavier cone mass... rather than a lightweight composite cone material. That said... if the cone material is too light and flexible.. it will distort due to the very air pressure, as it moves. These types of things are not brought up by many amateur speaker builders. They simply grab some fairly cheap Chinese drivers off of parts express, and call it a day. Many of these drivers might not even be made without flaws to begin with. Some may not have been measured correctly from the factory supplied Data-Sheets. Some may not match each other as a pair (in spec values, and in sound production). And just tossing a Poor, under-powered, highly distorting Driver into a fancy box, with an AWESOME EQ / Crossover... isnt going to make them sound spectacular. Is not going to make them "Audiophile" grade in experience / output / performance. Mariah Carey, I believe has an 8 Octave vocal range. You cant take a Paula Abdul vocal box (driver), and expect to get Mariah Careys level of sounds out of it. Not even with a lot of digital EQ and Effects. Just like you cant expect to race your Toyota Camry, against an F1 race car. There are huge reasons why these things are so vastly different in performance.
@@IBuildIt Part 1 - I have built one set, and modified others. My first one, was to take a PC speaker that had a horribly designed plastic shell + a poor pizero tweeter and flaky amp, apart. I cut some MDF strips, and made a general box shape... moving the loose panels around, to form different sizes, while the woofer was playing. I was surprised to discover, that the driver sounded Much better than the original enclosure. And my first lesson was learning that they sounded best with an open baffle... rather than when I tried placing a backing block on them. Finding the shape of the box that sounded best... I then marked, cut, and glued it in place. Then made the front panel to hold the driver and assembled it. Next up, I took a tweeter from a larger home speaker box, and placed it on top of the speaker, facing the ceiling. My idea was to make the tweeter fire up into a cone, to spread the sound in 360 degrees. Temporarily, since I didnt have the Cone that I needed... I cut a few wooden dowels "Legs", about 3.5" in length... and placed them on top of the speaker box.. at the four corners. I then set a piece of MDF on top of them. I was pleasantly surprised at how good they sounded. A huge upgrade from originals tweeter... and much better bass sound, than the originals farty fake-bass (tuned port) design. I initially didnt thing the tweeters were going to be powerful enough to spread out without the cone... but it worked so well, that I just decided to use them as they were. These were then wired to a box that I put two 12" woofers inside of a larger box that was under my bed. The reason, was due to me not having the space for the two larger house speakers on either side of my PC desk. And the fact that I really didnt care that much about that spare house speakers sound. The entire setups was actually pretty good. I placed the smaller boxes closer to the ceiling, on the top shelf of the desks hutch. They blended in decently with the 12" woofer box, and I was getting much better imaging, because the tweeters were no longer blocked from being under a desk. From this build, and a few other speaker mods...I quickly realized, how little I knew about the subject.
This all is very interesting, but "two days of my work" is enough to purchase the most expensive and flat home audio tweeter. This is obvious, that your tweeter membrane, Sir, is too heavy, and no trick will help. You have to pay more attention on 10-12kHz range. This is a "clearance". It's very important imho.
@@IBuildIt Better drivers = Better sound. You cant soup up a Gremlin / Yugo , and expect Ferrari / Lamborghini / F1 results. A race car, has so many special attributes, that give it massive performance advantages, such that no standard vehicle can come close to matching them. Stiffer Frame - to prevent twisting under cornering stresses Superior and Stiffer suspension High power, specially tuned, and super-charged motor High power fuel pump, to get far more gas to the engine, much quicker Programmed Computer, balancing fuel mixture, to gear/RPM for best possible performance Special tires. Wider, and Tackier grip, to stick to the road at high speeds Lower center of gravity precision balance in weight, spread equally across all tires lighter weight parts, and reduced overall weight, for better acceleration and performance special aerodynamics, special cooling channels, and Downforce spoilers special engine cooling enhancements extremely powerful, heavy duty, brake mechanisms While people can have fun running a standard or tweaked car on a race-track... you can not really compare this experience, to driving a specially engineered "Super-Car". Faster acceleration, Faster top speed, superior stopping / braking, superior cornering at high speed, superior traction and responsive control... even in the worst case scenarios. Obviously, not everyone can afford a high end racing car. Many cant even afford to Hop up a used mid-tier sports car. But there are high performance speaker drivers, that are still very affordable, to even the modest of budgets. (much cheaper, of course, than buying an over-priced high end speaker that has been built by a company)
@@IBuildIt And? This is a textile membrane tweeter. It's heavy by default. It was born with significant slope (with an artificial bump at 12kHz to extend a range to 20). This is cheaty tweeter with a loud label. When you increased an acoustic resistance with the horn, textile gave up.
I wonder what would happen if someone tied both of you together by your tails and draped you over a clothesline... Now that would be wild and crazy, I think!
The term waveguide was introduced by Earl Geddes to describe horns that he developed based on his equations to define the directivity of a horn. As he said in the interview (and on his website) all waveguides are horns, but only horns that are based on his waveguide theory are waveguides.
Of course the term waveguide has now been absorbed by the audio world and the original definition is fading into obscurity. But the main thing to remember is that if it looks like a horn and quacks like a horn, it's a horn.
In the video I misspoke and said "constant" directivity instead of "controlled" directivity. Sorry, I'm not working from a carefully prepared script written by my team of support staff. I'm only human and make mistakes. You probably make a few yourself and I forgive you.
Now onto the fun stuff! The wild and crazy phase plug experiments were interesting. I think I'll make a pair of the 50's tail lights for my horns (not waveguides...) from maple just because they look nifty. Like I said, I'm never going to hear the difference either way, but I'll enjoy looking at them.
Also a side note:
Guys tend to take all of this audio stuff way too seriously, and zoom in on the smallest details, blowing their importance out of proportion. There is this idea that if you do all of the little things "right", then you'll reach audio nirvana. And if you ignore those little things you obviously can't be taken seriously and any speakers you make will be subpar.
This is simply not true.
To get great sounding speakers you need to do the big things right. That will take them 99% of the way. That last 1% is in those 100 little details that can be tweaked.
If tweaking details is your thing, then go for it. But stop fooling yourself on their significance. You don't get a "night and day difference" when you round the corners on the cabinet. Smoothing out a 2db ripple at 5Khz won't "add 3 feet to the soundstage depth" or some other miraculous improvement.
Understand that we don't listen with our ears - we listen with our brains and our brains do all kinds of fun stuff with what our ears hear.
The interview with Earl Geddes: ruclips.net/video/nhe8VfuTg08/видео.html
If you have the capability to equalise then the best by far approach is to make the on and off axis response the same giving a good power response that can now be manipulated however you want.
Pro audio went down this path decades ago and now just laughs at the HiFi worlds half arsed attempts to get good room response.
So learn some maths, design your horn to be constant directivity and accept the topend rolloff on axis and then equalise.
All your little manufacturing fun can then be directed at reducing throat distortion and getting the directivity to match throughout the complete response for all drivers in the system.
You need to fix the shape of the horn, not add a phase plug...
Not only educational, but just plain fun!
Best place I know to hang out, have fun while learning - and not be sold a time share.
I like your down to earth and no BS explanation of things! A lot of times frequency response charts say nothing, most oldtimer audiophiles can't even hear beyond 15khz, and just trying stuff out and seeing its impact makes things way more clear and easy to understand for laymen like me!
Hi John, all of the technical info is wonderful but I can’t wait to see the final product. Great job
Nicely done John. You did a good job on this video. Welcome to the world of phase plugs. They give with one hand and take away with the other. Two things I would add. Nothing major in the least. The term waveguide is borrowed from radio antennas. Earl used waveguide as much for marketing as he did for setting himself and his work apart from other horn design work. It has been said that a waveguide is there to control directivity and a horn to both control directivity and add gain to the output device because of load matching the diaphragm to the air in a way that better matches the pressure generated by the diaphragm to the surrounding air. A horn allows more molecules of air to get a kick in the butt than just running the dome into the room. Excited molecules of air are what gets us sound. I'm happy that you did some EQ measured on and off axis. Again EQ is not magic. It causes as many problems as it solves. Nice to see you keeping it real. One last observation. The region from 2 to 5 kilohertz can shift your ambience a lot more than anywhere else. I know you understand this. Even a db and change can be noticed under the right circumstances. That I have played with many times. Again a lot depends on your recordings. And how they are made. If it is a two mic recording and all the cues to convince our ear/brain system are due to inter-channel cross talk and recording room reflections then you have a auditory magnifying glass to listen with. If it is a recording done with mono micing and mixed by a skilled engineer, then you have less of what can trick our ear/brain into thinking what is being reproduced in front of us is real. Thoroughly enjoyed watching. Keep en coming.
EQ can definitely do that if you overuse it (I have in the past) and don't look at everything it affects (I've done that too...). But it's a tool like any other that can fine tune to get the results you want. Much better if you can do that on an individual driver level, rather than globally, and limit how aggressively you use it.
Having a digital crossover with PEQ filters for each driver is a true luxury and makes the passive crossover option seem stone age in comparison.
What I find interesting about Earl's interview is the multiple subwoofers which I knew about but didn't know he was the one to arguably do the research first, and also the passive vs electronic crossover design, which I've also heard that passive can be "as good" as an electronic if designed correctly in synergy with the mechanical aspect of the loudspeaker.
Great video! I made a pair of spears using full range drivers and removed the dust cover for a phase plug.
I did exactly what you did- quick prototypes to see if I was headed in the right direction. I used aluminum, copper, brass and wood. The final result was bullet shaped teak with a flared brass sleeve down the center.
I find it amazing how much little changes can shape sound.
To show “what can be done.” Check. Mission accomplished. Very cool to see the difference in measurements from simple and sophisticated changes.
Yes! There are just as many "right" ways of building speakers as there are manufacturers. If one way, be it horn/waveguide, dipole, bipole, Omni or cardioid (my current personal designs over the last few years) then everyone would be clamouring for those.
Also besides the brain/ear don't forget how we are simply swayed by the way things look.
Keep up the awesome work, always looking forward to your next video, and can't wait to see your final design!
Why do you use foam as a phaseplug? Because it;s easy to shape? A wooden dowel or a screw or a small coin works too. Because all you need to accomplish here is to slow down or lengthen the path for the sound from the tip of the dome relative to the sound from the surround. If you can get them in phase by redirecting them around an object, you get the same effect as with forcing them to go through foam. Except foam additionally attenuates it.
The term waveguide is a marketing/advertising term since all HiFi people "know for sure" that horns are colouring the sound. The industry now calls them waveguides and voila all the colouration is gone. Magic!
That's what it's become, for sure.
I'm not saying you're wrong...but how you say it is entertaining! I like good sound, but I have a deep distrust in "hifi".
For that upper roll off you can compensate by adding a bypass capacitor to the series resistor you'll probably use to bring the sensitivity more in line with your mid-driver. The trick, of course, is picking the cap value (it won't be large) but it essentially turns that resistor into a frequency controlled resistor. As the frequency rises, the impedance of the capacitor lowers which will shunt the signal around the resistor. Of course, if you're using DSP this is moot.
honest video on a very interesting topic, I hope to see more like this
i love these videos, this series, this channel. and the comments, which are also mostly great, have just enough audiophile nervosa insecuritas to keep it spicy!!!
Boy, that's sure a rabbit hole you went down! Speakers are so complex!
i notice it's just 1/24 smoothing, that's an impressive response and not just for diy speakers...
the off axis is also somewhat in line with on axis, a little exception around 10kHz w/ -5dB, no big deal at all. that speaker will produce a very pleasant and accurate listening experience. (of course the off axis also refers to vertical)
the "new idea" of speaker fq response puts greater importance on the off-axis compared to the older approach where the goal was to achieve a flat on-axis response and don't mind the off-axis too much once the listening position was chosen.
-------------------
this new idea is in fact a better understanding of how we perceive speakers in a room and it turns out the reflections play a larger role than was initially assumed. cumulatively more reflected acoustic energy reaches our ears than the direct sound, up to the point where the high fqs are so directional that there's v little wall reflection.
a speaker that maintains it's response off-axis relative to the on-axis (be it flat or not) will produce wall reflections that are similar in tone w/ the direct sound coming from the speaker.
otherwise we hear two different tonalities...
when a speaker is measured in a room, the mic also pics up the reflections and without windowing the response (cut the recording before the reflection arrives) there is no way to tell what part of the sound is from the speaker and what comes from a reflection. so applying aggressive EQ can't really solve the problem since the speaker itself may be flat on axis and it's the off-axis that creates the non-linearity.
hope this makes a little sense :)
(they have these days setups and software that can eliminate the room reflections when measuring speakers in rooms, demonstrably identical to anechoic measurement of same speakers. the best performing speakers based on listening tests are the ones w/ a consistent on and off axis response, even if not particularly flat.)
All true for untreated rooms, but off axis response (and reflections) are less important in an acoustically treated room. The room these are in is treated and that's why the 1/24th response is so clean without gating.
This is interesting! I have been thinking about directivity, and I've noticed coaxial designs covering a large portion of the low frequency drivers with a wave guide, or drivers covered completely in the case of Genelec. (Or 2 small drivers spaced apart, in the case of an MTM....which is a coaxial speaker in my view.) I think the phase plug creates a tighter wave guide and forces all the energy to propogate from the perimeter. I suspect this creates a more defined directivity pattern. But just a theory!
So wild. So crazy. So here for it!
Have you looked into the foam acoustic lenses electrovoice used on their Superdome tweeters
Good stuff Sir. Enjoying your videos.
Having only played with wave guides and horns at microwave frequencies, acoustic is a whole different world.
Unless your a bat and can hear the microwaves !
@@kfl611 Not one bat in the ntire world hears above ultrasonic.
@@JosephLorentzen Of course ! Just saying. Animals are amazing in how wonderfully they perceive the world and their environment.
Horns and wave guides = new name so not to scare people of the horn sound. Waveguide is basically a shallow horn.
Anyone experimenting along the same lines, I suggest using non-hardening modeling clay to quickly form different horn and waveguide shapes.
Do you build waveguides? Would you consider building some and sell them? Thanks!
Loving the videos John! Absolutely recommend checking off-axis. I recently spent days playing with phase plugs and got the on-axis greatly improved. Then when I checked off-axis it was horrific 😢 Days of work for nothing!
I was doing solid phase plugs. Maybe less chance to go wrong with a bit of foam. IMO it is probably slowing the propagation speed in that area and changes the wavefront shape. KEF say a spherical wave front is best for waveguides but pistonic drivers don't quite do that even if domed.
A 3d printer is brilliant for making phase plugs of all shapes.
@@Audio_Simon I was going to try a disk after the bullet, and then got sidetracked on the foam. And before I knew it the day was consumed.
I think that while doing these experiments was interesting, the net gain isn't worth the effort. Most people won't hear the reduction in output at 17Khz, so it's just for the "flat to 20k" bragging rights that it has any value.
There's not a lot of content beyond 15khz in most music, so that makes the gains even less. But with that said, I will try a disk and see how it performs. Easy enough to try it while I'm doing the final setup on the speakers.
@@IBuildIt I'll look forward to hearing how it works out. I'm someone who found significant benefit adding supertweeters with a 16khz 24LR acoustic filter to a speaker with Neo3 planar tweeter so we obviously hear quite different.
Well, often the difference (not necessarily improvement) that guys hear when they add a super tweeter is the comb filtering that happens when two drivers are cancelling each other out at certain frequencies. Unless the super tweeter is extremely close to the tweeter physically or there is no overlap at at all with the tweeter in the frequencies it plays, comb filtering is a given.
If you take a piece of music you like and open it in a program like Audacity, you can see how much high frequency content there is and at what level. Also normal hearing is most sensitive in the 2khz to 5khz range, and less so at low and high frequency.
@@IBuildIt My thinking is that the main audible effect was from changing the power response of the system. Much like a rear or upward firing tweeter, helping add back some of the HF reflections that provide cues for 'spaciousness' that get easily absorbed in room or simply don't exist from a narrow directivity tweeter. It certainly made the in-room response at listening position more linear, because without it the in-room response dropped off rapidly above 15khz or so.
Given that you said you can not hear the top frequencies it would be a really interesting experiment to see if you hear any effect adding a dedicated 'super' tweeter that you could easily switch on and off, perhaps with a bit of dsp for easy tweaking. Just an idea for an interesting video!
Rounding corners with radii that small makes no difference. Mathematically the radius should be closer to half the baffle step frequency length but one thing you CAN do is introduce a smooth 45 degree angle part way across the baffle which pushes the virtual source point at impedance change from cab to air neatly at the edges firing out at 45 degrees not reflecting directly towards the listener.
Waveguide as a term has been in use for decades by the way. Not so sure your friend there was the first. Perhaps in the USA maybe?
Are you talking about putting a 45 on the outside edges of the baffle?
Great job John, practice makes perfect, as they say… 🤓
Keep it going sir….. Thank You.
My guess is that in the real world, where most of us adults can't hear above 15k hz, , the refined hunk would be the most useful version. Experimenting with a few versions of that shape could get rid of the 10k dip.
Are you testing them in the listening room or do you have something specially built for testing? I’d be curious to know more about your testing process.
In my listening room. It's the best place I have it, other than trying to set it up outdoors (and I'm not doing that).
@@IBuildIt by those measurements it seems like a hell of a listening room! How far away was the mic?
30 inches for everything shown in this video. When I properly set up the speakers I'll move out to a full meter away, but it won't look much different.
In case you missed it, there are several videos here showing what I did in the room, along with measurements.
@@IBuildIt awesome, good to know! Thanks!
Horns and wave guides = new name so not to scare people of the horn sound.
Btw - I believe BOSE calls its long snake-like tubes in its speakers, "Acoustic Wave Guides". You know what Audiophiles call them? Artificial sounding, fake-bass.
Good video, good testing, and good looking speakers! Scotch-Brite is spelled with a 'c', btw... ;)
Interesting as always. Can't wait to see the finished product.
I wouldn’t go anywhere near a horny guide’s phase plug!
horns and phase plugs - so many ways to make a dome tweeter sound worse )
It's like all thumbs are fingers but not all fingers are thumbs.
good things!
Excellent sounding speakers must have well built physical boxes of whatever shape they take. Well built finely crafted speakers have to be constructed to be as solid and free from vibrations and resonances. The construction techniques may be more important than the actual physical materials they are made from. Some boxes/cabinets have exotic construction and all those internal braces and “waveguides” should further dampen unwanted and undesirable vibrations further strengthening and create nonresonant enclosures. Next quality drivers need to be used in a manner consistent with the design objectives. Putting expensive drivers in poorly built enclosures are bad design decisions. Quality components in the crossovers are important. Cheap stuff won’t help expensive drivers create what they were designed to do. Then there is the speaker placement and the room itself also parameters that will affect the quality of the speaker in ways more important than rounding off corners, etc. OB designs are another concept entirely. I think some people may have developed very acute hearing sensitivity just like some people have outstanding visual acuity and they can perceive sonic nuisances that 90% of the average listeners are unable to perceive especially those of us who do have limitations in out hearing response. (Above 12kHz I’m seriously deficient especially in my right ear) so I have to be practical in my speaker choices.
Excellent presentation John!! Thanks.
Btw John shopping at Walmart yesterday I encountered a gentleman who was your doppelgänger. I almost asked him how was the OB speakers coming along 😂
How a speaker sounds from under 1ft away from it... is very different from how a speaker sounds at 7 to 10 ft from it. As such, many of these measurements and recordings, are likely not optimal to how you would be hearing / experiencing them. (even with the perfect room treatment)
The question is... what happens to the tweeters soundwaves, altered by these horns/waveguides... by the time they reach your ears, at about 8ft away? Its very possible that what seems good on the readings from a few inches away from the speakers... ends up sounding poorly, once those altered soundwaves.. reach your ears at 8ft away.
Also, the kind of changes may ramp up in scale, of extreme levels of differences... depending on how much power they are being fed.
Anyone whom has seen experiments with aerodynamics and fluid dynamics... knows how complex these things actually are. How a very small change, can cause very unexpected / big changes.
He measures from that distance to reduce the variables from vibration in the room. Its not rocket science.
@@KipdoesStuff I know exactly why he measures from close up.
You obviously have not been reading the replies, in depth.
The point here... is that a Horn (or "Wave Guide"), alters the way that the soundwaves are projected.
This could theoretically cause the waves shape, to become inverted, as it hits the guide, and or as it travels through the air.
Thus, if you recorded two instruments in a room, using a Bineural microphone... and played this recording back on this wildly altered tweeter... it could end up causing the 3D audio positions of each instrument... into different places in the room. It might flatten the very depth of the image. It might partially or fully invert the image, depending on how far away you are from the driver.
But what it wont be... is Accurate to the Recorded Image.
You could then put on a pair of high end Audiophile grade headphones, and easily tell the difference in the 3d soundstage / image... from the poor quality speakers / drivers.
Here is another thing that many Audio-Measurement freaks do not seem to understand...
From my own personal experience, when listening to different kinds of tweeters:
1) While Ribbon Tweeters may have incredible accuracy... they have a very NARROW soundstage / sweet spot. Most of these high end speakers will sound so much worse, once you move a mere foot out from the center of them.
Why is this? Its likely due to the fact that the drivers tape, is actually facing 90 degrees to the listener. This causes image distortions right from the very start.
2) Many Dome tweeters, are often too soft, and get ate up when there is heavy bass playing in the music track. Most especially at higher volume levels.
3) The most 3d sounding tweeter I had heard, was a dual pole ribbon tweeter, that had two vertical 1" tall ribbons... that were full 360 degree circles, mounted next to each other. These sat directly on top of small bookshelf speakers, allowing full 360 degree projection.
The problem? Because the sound was being output in every direction at the same time... the sounds would end up bouncing off the back walls, and other areas... and arriving at your ears at different timings. It was making ECHO like effects... and completely ruined the intended Image of the recorded music and movies that were playing on them.
4) The best Tweeters Ive every heard... are from the EPI 100v speakers. They have the same concave shape that larger drivers have. They project an incredibly wide 3d image / sweet spot. You can move almost 90 degrees to one of the speakers, and still be getting an incredible 3d stereo image.
Not only that.. but they vanish.. as if they are not coming from a speaker at all. Its like the instruments are actually producing the sounds in the room... exactly where the musicians were recorded. If I put a blind-fold on you... you likely could not find the speakers, even with a solid 10 min of searching.
Now, Im not saying that Ive heard it all. Not even close. What I am saying... is that there is a LOT more to these "Measurements" than people realize.
I can put two speakers with virtually identical readings / specs in front of you... and each will sound completely different, when you do an A to B comparison.
Its not that one could not manage to figure out the best and most accurate ways to measure speakers. Its that the people whom think they are Experts in this field... are far more clueless than they even realize. And worse yet... they are so EGO bound, that they are not willing to even take the time to understand what the issues are.
@@KipdoesStuff Part 2: Im going to add one more point. While these dudes that make Measurements are so gung ho about "The SCIENCE"...
They completely IGNORE the science behind what makes high end DRIVERS so much different from low spec drivers.
For example... If you have a weak magnet, and weak driver coils... as well as a heavier cone material... you will get distortions.
Why? Its pure Physics.
If I tell you to hop in a Pickup Truck... accelerate to 60 mph in two seconds... then STOP in less than 2 seconds... can your truck do it?
Nope. The "Driver" (Engine) is not powerful enough to be able to accelerate that fast. And the Brakes are not powerful enough to STOP the truck that fast.
Now, imagine a speakers cone. The Musical wave, expects a lightning fast acceleration in one direction... and then within a fraction of a second later.. it expects to both stop and accelerate in the Opposite direction... at the same blinding fast acceleration rate.
If your magnet and coils are not powerful enough... they cant get the cone to accelerate as fast as the music intended. And likewise... the driver can not stop and accelerate in the opposite direction, without some delays from the over-shoot... because its not powerful enough to stop it instantly.
That is Compounded by a heavier cone mass... rather than a lightweight composite cone material.
That said... if the cone material is too light and flexible.. it will distort due to the very air pressure, as it moves.
These types of things are not brought up by many amateur speaker builders.
They simply grab some fairly cheap Chinese drivers off of parts express, and call it a day.
Many of these drivers might not even be made without flaws to begin with. Some may not have been measured correctly from the factory supplied Data-Sheets. Some may not match each other as a pair (in spec values, and in sound production).
And just tossing a Poor, under-powered, highly distorting Driver into a fancy box, with an AWESOME EQ / Crossover... isnt going to make them sound spectacular. Is not going to make them "Audiophile" grade in experience / output / performance.
Mariah Carey, I believe has an 8 Octave vocal range. You cant take a Paula Abdul vocal box (driver), and expect to get Mariah Careys level of sounds out of it. Not even with a lot of digital EQ and Effects.
Just like you cant expect to race your Toyota Camry, against an F1 race car.
There are huge reasons why these things are so vastly different in performance.
@@johndough8115 Have you designed, built and measured speakers before?
@@IBuildIt Part 1 - I have built one set, and modified others.
My first one, was to take a PC speaker that had a horribly designed plastic shell + a poor pizero tweeter and flaky amp, apart.
I cut some MDF strips, and made a general box shape... moving the loose panels around, to form different sizes, while the woofer was playing.
I was surprised to discover, that the driver sounded Much better than the original enclosure. And my first lesson was learning that they sounded best with an open baffle... rather than when I tried placing a backing block on them.
Finding the shape of the box that sounded best... I then marked, cut, and glued it in place. Then made the front panel to hold the driver and assembled it.
Next up, I took a tweeter from a larger home speaker box, and placed it on top of the speaker, facing the ceiling. My idea was to make the tweeter fire up into a cone, to spread the sound in 360 degrees.
Temporarily, since I didnt have the Cone that I needed... I cut a few wooden dowels "Legs", about 3.5" in length... and placed them on top of the speaker box.. at the four corners. I then set a piece of MDF on top of them.
I was pleasantly surprised at how good they sounded. A huge upgrade from originals tweeter... and much better bass sound, than the originals farty fake-bass (tuned port) design. I initially didnt thing the tweeters were going to be powerful enough to spread out without the cone... but it worked so well, that I just decided to use them as they were.
These were then wired to a box that I put two 12" woofers inside of a larger box that was under my bed.
The reason, was due to me not having the space for the two larger house speakers on either side of my PC desk. And the fact that I really didnt care that much about that spare house speakers sound.
The entire setups was actually pretty good. I placed the smaller boxes closer to the ceiling, on the top shelf of the desks hutch.
They blended in decently with the 12" woofer box, and I was getting much better imaging, because the tweeters were no longer blocked from being under a desk.
From this build, and a few other speaker mods...I quickly realized, how little I knew about the subject.
Horns without compression chamber are called waveguides, that's all.
Vertical installation would be better
Need to taper the hole in your headlight. You are messing with the expansion formula for the horn. blocking vs. filtering.
The formula I used for the horn was "this curve looks okay" :)
But yeah, interesting idea! a horn within a horn!
@@IBuildIt within a horn, within a horn, within a horn...
And I thought I knew about speakers. HA! 😊
I guess I could say John is an audiophile. horn-y about better sound reproduction ;)
All freeways are highways but not all highways are freeways.
All IPA's are beers but not all beers are IPA's. I always loved those comparisons.
Lol. If you are not going to hear a difference between them, why the hack are you making them?
This all is very interesting, but "two days of my work" is enough to purchase the most expensive and flat home audio tweeter. This is obvious, that your tweeter membrane, Sir, is too heavy, and no trick will help.
You have to pay more attention on 10-12kHz range. This is a "clearance". It's very important imho.
You have to loosen your necktie!
@@IBuildIt Better drivers = Better sound. You cant soup up a Gremlin / Yugo , and expect Ferrari / Lamborghini / F1 results. A race car, has so many special attributes, that give it massive performance advantages, such that no standard vehicle can come close to matching them.
Stiffer Frame - to prevent twisting under cornering stresses
Superior and Stiffer suspension
High power, specially tuned, and super-charged motor
High power fuel pump, to get far more gas to the engine, much quicker
Programmed Computer, balancing fuel mixture, to gear/RPM for best possible performance
Special tires. Wider, and Tackier grip, to stick to the road at high speeds
Lower center of gravity
precision balance in weight, spread equally across all tires
lighter weight parts, and reduced overall weight, for better acceleration and performance
special aerodynamics, special cooling channels, and Downforce spoilers
special engine cooling enhancements
extremely powerful, heavy duty, brake mechanisms
While people can have fun running a standard or tweaked car on a race-track... you can not really compare this experience, to driving a specially engineered "Super-Car". Faster acceleration, Faster top speed, superior stopping / braking, superior cornering at high speed, superior traction and responsive control... even in the worst case scenarios.
Obviously, not everyone can afford a high end racing car. Many cant even afford to Hop up a used mid-tier sports car. But there are high performance speaker drivers, that are still very affordable, to even the modest of budgets. (much cheaper, of course, than buying an over-priced high end speaker that has been built by a company)
Scanspeak D2905
@@IBuildIt And? This is a textile membrane tweeter. It's heavy by default. It was born with significant slope (with an artificial bump at 12kHz to extend a range to 20). This is cheaty tweeter with a loud label. When you increased an acoustic resistance with the horn, textile gave up.
I wonder what would happen if someone tied both of you together by your tails and draped you over a clothesline...
Now that would be wild and crazy, I think!