I really appreciate these detailed, analog-focused descriptions. I know soundbars and Atmos setups do similar things via DSP, but it’s really helpful to see this illustrated from basic fundamentals.
When i was 18 i worked add a recording studio here in Van Nuys California called Sound City. They had two electrodyne mixing boards and a very limited number of outboard devices. I think each Studio had two tektronix limiters and a coil which could be patched in to flip the phase of an input if needed. They also had two live Echo Chambers which were really cool. Even when they upgraded to Neve consoles they really didn't add any outboard equipment. And even with that limited equipment they had tons of gold records on the wall that were recorded in those Studios. Simple can work 🙂
Matthew Polk did this EXACT thing in the 80s with the Polk Audio SDA line of home audio loudspeakers. There was even an interconnect between the left and right speakers, and special instructions regarding what type of receivers or power amps could and could not be used (because of common grounds). The top of the line model from that series had eight 6 1/2" woofers and four dome tweeters on each side, placed into special groups just as you explained. He did it all at power amp level, and the crossovers in each speaker were four rack spaces tall. Those speakers are still revered by home audio folks today.
Very cool stuff and there are many interesting and adventurous implementations of creating spacious audio. I believe the Polk audio stuff was designed for use as stereo speakers placed in conventional locations and to enhance the stereo further using lots of speakers and processing and such. The reason I dived into this MS speaker adventure was that it requires no processing and uses four speakers and a very simple wiring technique and can achieve very cool and interesting sound from a single enclosure and some fairly inexpensive speakers. I made a second one and put in rechargeable batteries so it could operate off the wall or off of the batteries. Fun stuf
The SDA speakers had the tweeters side by side. The reason for that was to cancel out the left speaker in the right ear and visaversa. I found that design somewhere back in the 70s and built one. Pretty cool.
This is an interesting idea! I'm not an audio pro, but I keep coming back because I always learn something. Doing something new with speakers is always more interesting than the struggle to perfect the same ordinary speaker designs.
This was actually done by a bluetooth speaker manufacturer named Riva some years ago! Look up Riva Turbo X and Riva S - they used a single front firing fullrange driver and two side firing fullrange drivers in combination with mid-side processing to create a sound field!
Very cool and yes it can be done with processing and this concept is kind of the basis of all those sound bars out there. My goal was to build it with no processing so that it could work with any stereo signal and not have all that weird phasey sound that sound bars have and most processed versions seem to embrace
@@DaveRat take a look at line driver/receiver chips DRV134 but from THATS, they can be operated in a creative way that would be very convenient in a controllable stereo widening mid-side setup. I've always liked this idea for environmental background music in a space you can walk around in.
This was interesting to watch, I grew up using the old KF730's. it was always interesting to see how they had side firing woofers and how they sumed up in the center to create a single mono sound. It's pretty cool to see how you can keep the stereo sound moving from one extreme to the other with it suming together nicely.
Very cool and side firing speakers producing low frequencies pretty much act almost identically to a forward firing speaker except they have the little advantage of slightly increased energy forward and rearward and slightly decreased energy off to the sides in the low frequency.
The mid-side speaker already exists. It works very well. It is produced by Aspen Pittman Designs (currently run by his daughter after his passing) and is available on Sweetwater. The technology is called Centerpoint Stereo. The speaker is called the Spacestation v3 and the bigger unit the Spacestation XL. Keyboard players LOVE it, and it makes a great PA in small to mid-sized venues. And I have done great things with electric guitar.
Hmmm, that seems like a glorified sound bar. They say they have some patented 3d audio controllable front to back image control Not sure if that the same thing as the speaker in the vid that passively recreates Mid-Side with no processing, not 3d, and does not use any patented concepts, just simple logical wiring that works with just about any stereo amp enhances the stereo image
@@DaveRat No, totally NOT a glorified sound bar. It takes in stereo and puts out mid/side for a stereo effect everywhere (not just the center between speakers). It's hard to believe without actually hearing it but it's extraordinary. There is a matrix and a speaker special made by Eminence set at 90 degrees from the front main speaker and tweeter. Many are skeptical until they actually hear it.
@@DaveRat It has a front facing closed back speaker that gets fed L+R, and an open back speaker below it rotated 90 degrees that gets fed L-R. So very much a mid-side amp. They do sound quite stereo. Fender made a line of amps called SFX that used the design. It's been a while since I looked at the patent, but I think they also crossed-over the low frequencies to only be reproduced by the L+R speaker.
Cool and interesting! How does it address the out of polarity rear sound radiation of the side facing speaker to not be asymmetrically destructive with the forward facing speaker? With mid side recording, the symmetry is achieved by having 2 versions of the figure 8 mic signal of opposing polarity But with an open back speaker, you only have one version of that signal So while the fig 8 pattern can be created, unlike fig 8 where the forward mic is in polarity with the left signal on the left and in polarity with the right signal on the right, with a single version, you are face with a choice of the forward speaker being in polarity with left or the right. But perhaps that is where the patented processing comes in. Messing with freq dependant polarity ( which is actually phase and time shifts) could address some of those issues. That is why I used the sound bar reference, as sound bars tend to implement freq dependant processing ( delay and phase) to force the sound to sound wider. I see that as a vastly different approach than this speaker and mid side recording that use no processing, no delay, no phase shift. Just put polarity reversal. Mid side recording can be done with a y cable, a polarity rev cable and 3 inputs. And actually can be achieve totally passively and requires no processing. And interesting. Would be cool to hear the speaker and or know what they are actually doing to the signal.
@@DaveRat I'm not an expert on the processing system. There is a single pot called "Width" and when you dial it all the way down the sound collapses to mono sounding from the front speakers. Dial it up it gets wide. Dial it too far up and it collapses some and doesn't sound so great. Some specific sources benefit from different settings, like single sources vs, a stereo track. I was able to get a huge guitar sound by running two mono signals into two mixer tracks the EQ'ing them slightly differently and panning them 50% + - L R. Many keyboard players say it's the best Leslie simulator speaker there is. There are some rare instances where a stereo mix (older recorded music) just hits the matrix wrong and sounds weak. However, the wide sound experience carries from room to room. It's wild. And kind of makes it the perfect party PA for a house. The music is ever present everywhere while guests can talk over it (it somehow doesn't interfere with talking like regular PAs do. ) Anyway, I see you are very interested in the concept and thought you'd be interested in what work has been done before. IT isn't very widely known except that Fender licensed the technology for their acoustic guitar amplifier.
Dave you are such a huge inspiration for me and probably about 90% of the people who watch your Channel! I just want to thank you for continuing to teach us and share your knowledge
I remember doing this sometime in the early 80s after reading an article in an amateur radio magazine describing this technique. In addition, each of the side speakers included a 47 ohm or so wire wound pot connected as rheostat in series for adjusting volume and something like a 10 or so microfarad cap, also in series, for high-pass filtering and some additional phase shift.
I made a speaker that is exactly what you did here exept using a dsp and using a center channel instead of a left and right , so thats 3 channels out of stereo , Differential stereo out of the sides and mono from the front , so far it is the best speaker i have ever heard , it fills a room like nothing else does while still having a center channel to handle voices or foreground noises .
WOW! I've recorded mid-side for years, but you're the only one I know of that could reverse engineer it to actually work in a speaker configuration! Always love your videos. Can't wait to see the next one!
I'm already using this with a matrix, through 4 AUX busses and it works great. I could use 3 busses, but am using two stereo amps. So nice you've built this speaker box!
@@anttikinnunen7904 I've tried a lot and ended up putting the mid speakers on top of the L and R (sides), this works well to widen or narrow the stereo image. I could use a true center speaker, but I'm not looking for surround effects, just enhancing the stereo image. Other things I've tried is rotating and letting the L and R speakers bounce off the walls with a slight delay to make it less direct, and placing the L and R speakers in line with my ears, like earphones and a true representation of how M+S is recorded. That was not very useful, because the direction of the sound changes in that case when putting in more or less side signals. Experimenting with vinyl records this way is amazing, by changing polarities and levels magic occurs and you can hear things isolated that would be impossible otherwise.
I did this same thing, but only using one amp, and four speakers, which I can adjust with a single potentiometer and also can switch the added two speakers in or out with a simple switch. Works great!
👍🤙👍 Yes this only uses one stereo amplifier and I did do a switch that converts it to stereo or stereo Plus side speakers. And then I built another cabinet just like it so I could compare and created some other fun wiring that adds EQ to the side speakers as well as put in a amp with a subwoofer out and added a fifth speaker as a subwoofer on a modular removable rear plate.
It would be really cool to take some multi-track stems and create a mixed song where you have one instrument panned full left, another full right, then one 20% left, another 20% right, and then something centered (mono). Then you could use that mixed track for your demonstrations.
This was a fun experiment to watch. I can definitely see myself in my older years when my boys are older just tinkering around trying different things. Love the video, can't wait to show some friends this.
This seems to be a variation on the poor man's surround scheme that I used back in the '80. The only difference is I would run the "surround/ out of polarity " speaker pair in series with anapprprat L pad then connect across left + and right + . This scheme worked well for my purposes but did necessitate using an amplifier with a common speaker ground. As a side note my mom had a Buick Regal where the factory " miss wired" the front speakers in the same manner as you did. When we got the fader set just right we ,in fact, did get the best stereo effect in the neighborhood.
I have used this exact wiring but placing each speaker in each corner of my listening space. For some recordings it made a phenomenal effect. At some point we were watching family videos that were filmed with a stereo capable camera while someone was called out in the video. The stereo enhancing was such that it fooled us into thinking the call occurred for real. The called person even got out of the room to search for the caller... until she also responded in the video, haha. I never thought of placing the four drivers in one single enclosure, now I have an urge to replicate your build. How do you think sound would change if the enclosure had separate airtight chambers for each driver?
Very cool and yes I've seen the surround sound type versions of this. As far as separate chambers, The main result would be a reduction in low frequency output as it would reduce the enclosure size. Since the surround sound speakers typically don't have much low end the forward-facing stereo speakers can use the entire box to get added low frequency. But if you have something that is low-end heavy to one side, The side firing speakers now have the entire enclosure to reproduce that low frequency. By segmenting it into smaller enclosures you either need to make the whole box bigger or sacrifice low frequency. I made another one of these and you can't really see it in the video but they're modular. The rear panel is removable and replaceable with a panel that has a fifth speaker that acts as a subwoofer. Also that rear module can be swapped out to make the speaker have a battery powered option. Those are super fun and they sound really cool
@@DaveRat I'm not sure, you may have inadvertently edited out the blinks. Haha I was semi joking. I found myself wide eyed and unblinking also, and I was very immersed in the content, so no complaints!
@@DaveRat I am exploring this information and doing some experimenting of my own with stereo guitar. I'd like a small but effective stereo guitar set up for my home practice space. So far I've relied on speaker separation and had no prior experience with this type of strategic phase cancelling
Brian Eno had a description on is ambient records about hooking up a third speaker between the plusses of your amp. As a kid I did notice this speaker had one of the normal speakers as an in phase friend, so ended up having two of these in reversed polarity, similar to your setup, but spaced. It did bring up extra layers from Eno's On Land LP, for instance. Then I noticed my amp did not like it, so I ended up with two rear speakers, that both had left and right speakers in them, and a small slit between the drivers was the air output. So I added the signals acoustically, rather than electronically. I do not think I was ever satisfied all the time, but I have forever liked sound being dislocated from the speakers. 40 years later, In theater, I often add Omniwave speakers, to pull the sound out of the main speakers. From what I learned as a kid, I just need some energy from the mid range to pull the voices to a virtual location between the speakers. I do not need them to be phase aligned so much, but time alignement is crucial.
Was just going to comment about the Brian Eno method. You could probably recreate the third speaker with a phase-inverting matrix (rather than using the amp method, which halves the resistance). Would be an interesting comparison to hear it. The "Triphonic" Eno method might be easier to implement in a live environment as it only requires one additional speaker stack?
There are a whole bunch of cool ways to wire the speakers for various versions of this. And ways to add capacitors and chokes to EQ, and all can be done passively. Actively opens up a whole infinite slew of other options. The stuff that I've been working on I wanted to make a speaker that sounds cool and runs off of standard stereo Bluetooth and creates a wide stereo image from a single speaker
Hi Dave, In the future video, would we be able to take a look at how to optimize your four speakers setup? I'd love to dive deeper into what's actually happening in the amp when you "cross-load" the channels like this. Maybe even investigate with some oscilloscopes or impedance meters. When playing back mono audio, my assumption is that the amp is only loaded by the main L and R speakers since the voltage between the positive terminals of the two channels is zero. Would that mean for any regular stereo track, any mid information in the waveform wouldn't contribute to the load and we would only have to calculate what happens in the side scenario? My main concern is with impulses, timing, and frequency response. How do these properties differ from the conventional LR to when the amp is cross-loaded with the L-R/R-L speakers?
Yes I will try and do another video on it and show some other cool wiring techniques. As far as the amp is concerned the left speaker loads the left channel with 8 ohms and the right speaker loads the right channel with 8 ohms. And the side speakers which are two 8 ohm speakers in series do the exact same thing as running an amp bridge mono into 16 ohms. And if you look up bridge mono or look at any of my videos I did on bridge mono, running bridge mono into 16 ohms is the same as running stereo into 8 ohms. So with an 8 ohm load on each channel and bridged mono into 16 ohms the amp season equivalent of 4 ohms per channel.
This looks and sounds awesome! Could this be made active easily? I can imagine this paired with a smaller sub to make a more immersive take on the column PA systems that could also be very portable and easy to set-up/tear down.
Really neat kit! Good explanation of the concept too. I knew what mid-side meant in terms of mixing (e.g. mid-side eq in a DAW), but this gave me a whole new perspective on it. Good stuff!
This general idea has been a around a long time, historically wiring a single speaker in the center of the back deck of your car to the plus signals of both Left and Right speaker (the difference). Produced a pleasing "Side" only sound that made it almost sound like a third, sometimes ambient channel depending on source material.
Yes is numerous implementations of various wiring to create surround sound. Brian Eno and hafler and pretty much every soundbar that's ever been made. This particular version is based on a mid-side design radiating a wide stereo image from a single point or a single enclosure. The concept here was to explain how mid side works and also show people that using a simple stereo amplifier and some inexpensive off-the-shelf speakers they can make a super cool ultra wide stereo enclosure with some simple wiring.
@@DaveRat And great job, BTW. I enjoyed this episode a lot. As an aside, most people don't realize that stereo LP records are encoded Mid-Side (actually L+R and L-R) for mono compatibility....but mostly because it's just very cool!😎.
Really nice video, thank you ! I don't know if it's been mentioned yet but Bang & Olufsen have a few mid/side bluetooth speakers in their range, like the Beolit 17, 20 and the Beosound Level and with the smartphone app you can indeed make them "project" sound closer or further away from the speaker and i think it's neat and better than the more typical stereo design on smaller portable speakers like these
And yes these B&O speakers rely heavily on DSP to work but i think it's sexy that they can use a single (mono) amp for the two tweeters and still get some sort of stereo image out of it. They don't use any really fancy patented technology either in fact B&O doesn't brag about this tech, they don't pretend it's magical and will create a 3D like soundstage or whatever, they just say you can dial in the width of the sound in their app with a simple setting.
Interesting and those things look cool and would be fun to hear. They say they have 360° sound radiation and the width or closer and further adjustment is interesting. Often that is done with ambience or adding ambience or some sort of reverberation type thing. I don't see any information indicating the design is related to mid-side though. In fact I think the fact that it is 360° kind of prevents it from being a mid-side design. But yes very cool stuff I have a B&o turntable and they make beautiful and interesting gear
@@DaveRat i learned about this mid / side thing by an engineer on the B&O facebook group, i own both of these speakers and it's quite funny how you can indeed make the sound larger or narrower even if the three tweeters of the Beolit 17 are wired to the same amplifier. The Beosound Level has two tweeter at it's extremities that plays the side channel while a center larger tweeter plays the mid. On the Beosound Level the effect is more pronounced, if you sit in front of it and play with the settings it truely make the sound goes really wide to just mono. And here again, the Beosound Level is only using a single amp for both tweeters which are wired out of phase. On both of these speakers, putting them to the widest setting does make the music a little weird cause it often deeps the vocals and make all the instruments that are panned left right or have been recorded in stereo, really loud. And if you put the setting all the way to the mid only it makes the sound very flat and hard panned instruments will almost disappear. It never fully isolate the mid or the side channel of course but i think it goes as far as 50%.
This has really got me thinking! If we were to wire up a PA system following your diagram, a speaker received the same signal from both + terminals, I agree cancels out but would that cause any damage? The driver to over exert, amps not discharging properly etc? Could we archive the same thing by wiring up a PA as we normally would, then adjusting the polarity in a system controller/amp’s DSP?
Connecting of speaker between the two hot terminals of an amplifier is also known as bridge mono. You can do a RUclips search on that or look at other videos that I did on the subject. If the amp is sent the same signal to both channels nothing comes out of the speaker. If you send something different to both channels like an equal and opposite signal then you get twice the voltage out of the amplifier. As far as loading down the amplifier each channel sees one half of the load impedance. So if you wire an 8 ohm speaker across the two hots of an amplifier then each channel sees four ohms. Some amplifiers are actually built with each channel being two channels in bridge mode. This is done because bridge mode more efficiently uses the power supply of the amplifier and allows creating higher voltages therefore reducing the amount of loss in speaker cables. The downside of bridge mode is that it lowers the impedance of the amp sees and running bridge mono into four ohms loads each channel down to 2 ohms which may be thermally or electronically less than ideal for some amplifiers. On the other hand some amplifiers can run impedances well below a single ohm.
Great video, what if I want a mono speaker instead of left and right. Can I just sum left and right in this passive matrix or will that destroy the side feed? Also.. can I make this matrix in a dlr connector box and feed amplifiers instead of speakers? Is it possible to make a passive xlr MS matrix box. Left and right in and mid plus side out?
Yeah that could be done, And I thought about exploring that angle, you could build a passive resistive voltage divider network to derive the mono center speaker signal and run that into a mono amp into the mono center speaker And then run the stereo signal into a stereo amp and wire it with the two side speakers in series as I show in the video, across the two hots of the stereo amp.
Great explanation. Without a mixing console this could be done with 2-channel processing purely analog or with modern DSP. I believe the "IMX Expander" on radio shack EQ's did this back in the 80s/90s.
Arrgh, just lost a longish post due to internet dropping off. About adding a L+ and R+ speaker to the rear at a party in a wood. Sounded great whichever way I was pointed. Blew one person's mind... but it *was* a party. At my home studio setup, right in the middle, I have two small monitors wired to the L and R (the L to the right-most small box and vice versa) with the levels much lower than the main L&R...Lovely tight centre, and still sounds wide. Tried out of polarity on one of these two centre boxes which gives some odd perspectives! Maybe a delay on both of these... hmmm........ (I like rows of dots)
very good demo of a slightly complex thing like your vid on line arrays with the mini speakers. this has applications for club systems (for dance music) that want a sprinkle of a bigger stereo image across a big space at high SPL. Appriciate your vids.
Interesting unit. Seems to be a quite different adventure as it uses processing to emulate spaciousness using 2 speakers, but interesting none the less. Here is what I found on it: The Hughes AK-100 Sound Retrieval System (SRS) was first developed in the early 80’s by Arnold Klayman to create an immersive 3D sound field using only two speakers by using the process known as ‘head-related transfer functions’. This process creates a more spacious ambience within the stereo field, producing strong special cues for different instruments in an audio mix.
This reminds me of the Carver Sonic Holography technique, where he sold a box that would do cross-mixing of left/right channels where the cross-mixed signal would also be inverted. This type of psychoacoustics has intrigued me for a long time; I once saw a Bose soundbar about the size of a normal center channel speaker that created a credible left/right wide stereo effect and I have been interested in the idea of doing something similar in a bluetooth speaker to try to produce a more credible stereo effect from physically small speakers.
Super cool and yes I've seen many implementations and versions of this done for surround or with various forms of processing. I thought this simple straightforward approach that's based on MS recording would be a fun project and it worked well beyond what I expected it to do
Just made this setup in Ableton using groups and the stock Utility plugin. Effect Group -> (Normal), (Wide group) -> (normal phase), (utility with phase inverse and swapped channels). Sounds to me like it's exactly what the width control of what the utility plugin and many others actually do, but it's interesting to be able to have the extra control over the channels and be able to process the wide vs normal signal and also to know what it's doing.
I built this speaker so that a normal stereo Bluetooth signal or any stereo signal could be used and recreated by a relatively small speaker and create a wide stereo image. I believe there's advantages to doing it acoustically using multiple speakers versus trying to synthesize it by altering the signals and running them into a standard stereo speaker setup. This setup adds clarity and dimension without creating that phasey sound that happens when you try and create it with signals and run it through a standard speaker pair. Another example that sounds phasey and undesirable to me are all of those TV sound bars that are trying to do a similar thing.
@@DaveRat Yeah, I'm keen to try and make a physical speaker as well. Very cool concept and a great use of the mid-side concept. I was just having fun messing around with it as a mixing tool with that Ableton setup, definitely does add a lot of phaseyness but I am keen to experiment more with it. In my test before I filtered out the lows starting at about 500hz in the wide group to add only a little bit of the effect for extra space and it sounded alright on a first listen. Love your work Dave :)
Hi dave, I hope all is well. Dont know if u remember last year u sold me the outter shell panels for a sennheiser md441-U because one side of the mics leather panels was ripped off . You sent me both of the panels, but I only needed one side, so i sent back the other side to you . I hope u received it. BUT ANY WAYS... I refurbished a pair of yamaha ns10s, and the woofers r very thin, and moisture is starting to affect the woofer, making them soft. Unfortunately, one side of the woofer ripped, and i had to repaired it.. My question is .. is there any kind of waterproof spray that u can recommend that i can use for that woofer since the woofer is made outta sum kind of paper? Im using a dehumidifier in my studio now. But i would like to add sum protection on the woofers since they are very old
It may slightly change the sound but any coating willl do that anyway. We use lacquer spray paint as it's quite liquid and absorbs into the paper rather than coating it and it will add waterproofing and some rigidity back into the cone
Q Sound... Some kind of virtual stereo to me is like MS equation but with a HPF from lows up to the mids at center summing and after that polarity reversal of the one side.
Cool stuff. I feel like you could make this simpler, and with three speakers. You have your single mid speaker, then Ldiff, and Rdiff. Wiring: L and R amp + and - go to the + / - of the mid speaker to sum the two parts; then, Rdiff and Ldiff are wired in series (- to +) between L+ and R+ on the amp, producing the difference between L and R to the mid. Do you have a CNC machine to cut these enclosures?
Yes, have a CNC. I was aiming to do this using a standard stereo input and or stereo amp without any processing such that it can be plugged into any normal stereo output. To do the three speakers, the challenge is getting the mono signal for the center speaker without processing.
@@DaveRat yes, but wouldn't you get that by running the + of both L and R amp channels to the center speaker? Hmm, I guess that doesn't quite remove the side signals, though... not unless they're quite out of polarity.
"...work on ways to implement this live. I believe it can be done with a matrix..." This sounds like a good job for an Outline Newton! There are ways that a stereo signal can be routed, polarity reversed, eq'd and delayed that are a bit beyond what can be done in a typical console matrix that may benefit your experiments. I've read that Rat has some in the inventory.
shouldn't there be physical separation inside the box as well, so the L and R speakers do not move the L-R and R-L speakers? and even L/R baffle as well?
I thought about that and actually no physical separation improves The low frequencies. This is because the side speakers don't have very much low end with a stereo signal that has fairly mono low frequencies. This means the entire enclosure can be used for the low end of the stereo speakers. But if low end is panned hard left or right then All the low end comes out of one of the side speakers and one of the forward speakers and reverse polarity out of the other side speaker. This in effect gives all of the enclosure to one speaker creating a deep low end. If you had four separate enclosures then all of the speakers would have their own tiny enclosure so you would have to either make the speaker bigger or suffer less low frequency overall. This speaker is actually modular and the back panel comes off and I made a second back panel that has a little subwoofer in it that also uses the same enclosure. And that sounds really cool with even more low end
I made a dj plugin that does from Stereo to L-R | R-L on a pot, the results are surprising, some cases when the producer used an exciter it sounds like a low res mp3 mess, some cases it pulls out samples they couldn't be heard in normal playback and for most tracks it's the strongest bass cut I've ever known.
It's a good idea and an interesting project. But I would probably try it out differently: Each normal speaker chassis has some kind of an almost fig 8 pattern, with some non-balance, caused by the magnet, cage and cone shape. So if you'd mount two identical speakers with their cones facing towards each other and wire them so, that they will be "out of pohase" (actually: IN phase, cause they are pointing in two different directions) - you will get a "perfectish" speaker for your S-signal... If you combine this "S-speaker" (mounted sideways) with a third speaker, that is centered and sends out the M-signal, you'll get a proper MS-speaker arrangement. Now you can use a 2Ch poweramp, and send L-R to the side chassis and L+R (mono signal) to the center OR you use a MS-recording, and directly send the S signal to the side chassis and the M to the center... No idea, how this would sound, but it would be a proper MS-speaker setup then, analog to a MS-microphon arrangement. The problem might be the big difference between microphones and speaker: Speaker need to transmit POWER, so they need to move air... So for getting a proper result with low frequencies, you'll need large systems that can move a lot of air... But anyway, this would be another very interesting project...
Hmmm, I believe there may be a flaw to that concept. A mid side recording is 2 mics recorded but to creat mid- side you actually use 3 channels. Forward, fig 8 in and fig 8 out. To playback mid side you also need 3 channels One side speaker need signal with the fig 8 left in polarity and fig 8 right out of polarity. The other side speaker needs the fig 8 right in polarity and fig left out of polarity. The design you described is missing a channel that is needed .
@@DaveRat Nope, you're wrong. A MS recording contains 2 channels, mid and side signal. If you want to transfer it into a standard L-R stereosignal, of course, you need to add and subtract the side signal to/from the mid. This is what's usually happening within these MS-mixing matrixes. But if you'd have a proper MS speaker system, with one speaker having a proper fig8 pattern and being oriented 90°, this speaker could transmit the pure S-signal. In addition with a center speaker, carrying the M signal, this all should add up (acoustically) to a proper stereo information. This setup would only require 2 channels and a two channel amplifier; one for the M and one for the S signal. Cheers, Salossi
What are the problems, in your view, of the approach of using a single mid speaker with a planar speaker turned 90 degrees sent the differential signal?? Thank you. Very interesting.
With a planer speaker, it will be in polarity on one side and out of polarity on the other. The forward faxing speaker will either be in polarity with one side or the other of the planar speaker With mid side, the fig 8 is split into 2 inputs, one in reverse polarity to the other But that can't directly be done physically. Unless you use two planal speakers with reverse polarity and then block the rear radiation of both. Which is exactly what I am doing but not using a planalr as a normal speaker is easier to block the rear radiation of.
After some pondering, actually a planar speaker turned sideways would work. Mono to the center speaker, left - right to the planar speaker. Such that the left side of the planar reproduces the left signal in polarity reproduces the right signal out of polarity. And the right side of the planar speaker reproduces the out of polarity right signal, out of polarity because it's the back of the speaker and the in polarity left signal out of polarity. So the right hand side reproduces the right signal in polarity and the left signal out of polarity. The net outcome is correct using two speakers as you describe
The mid-speaker wants to have either a ported or sealed enclosure to capture the low frequencies. Putting them all in the same enclosure allows them to share the space and makes the overall enclosure size smaller for a given amount of low end
So is this what they do in sound bars for home theater setups (with maybe some additional techniques as well I assume)? And what was that first song btw?
There may be some soundbars out there that do something similar but from what I've heard most of them have numerous speakers in them and they try and do beam steering. Things like send the signal with a slightly different delay to various speakers such that it shoots the sound off to the side or off-center or center. This has the advantage of putting different sounds in different places but it has a disadvantage of sounding kind of weird and phasey as well.
Yes similar concept. Michael for this was the package The setup into a single speaker to create a wide stereo image. Similar to what TV sound bars do but sounds better and not all phase weird sounding like those sound bars are
Dave, I guess you need to record the figure 8 with two seperate mics, cause the switched figure 8 of a single mic only has on output with both signals coherent in one cable, so no way to split this signal. With two mics and seperate signals it will work. Or am I wrong?!
To do mid-side the figure eight is just a single mic. Figure 8 Mic is just split to two channels. One channel has the figure 8 mic in polarity panned hard one way and the other channel has the figure 8 mic out of polarity panned hard the other way. And the directional mic is panned center.
@@DaveRat OK, thanks. Now I get it, you use 2 channels panned and use the polarity differences then. 1 Mic saved but some 2 speakers more needed. right?
Thanks for making this video. I'd love to try making something like this myself! 12:51 wouldn't wiring both positive terminals to the speakers make it so similar sounds between channels would become direct current at the speaker? They wouldn't just cancel each other out or am I wrong? If that is the case, you would have to sum the signals first then high pass to get rid of any DC (maybe in processing). What would that potentially do to the amp if there is no processing? For a small setup like this I'm guessing it's fine.
No it won't make DC. All identical signals would raise the voltage on both terminals at the same time so the votage difference between the terminals would be 0 volts for identical signals For signals that are different, it would just play the signals that are different as a musical waveform
@@DaveRatThanks for responding! I'm enjoying thinking about this. I probably need to do some tests to confirm this. You are correct there wouldn't be a voltage differential between the terminals, but the voltage on the voice coil would rise to match that voltage which would induce a magnetic field. Testing voltage between the terminals is only part of the circuit. The voltage would have to also be tested to ground. A voltage difference at the terminals would indicate the speaker is moving, right? If you start at 0v and apply a positive voltage from both channels on the amp there would have to be some amount of voltage rise on the coil. I may need to test that with my amp and a voltmeter.
@Zatore_ voltage to ground is not relevant unless somehow the speakers are referenced to ground. Think of Phantom power where a dynamic mic voice coil has 48 volts DC on both sides of it and has no impact on the mic whatsoever. A coil of wire or a speaker or a resistor or just even a piece of wire will be impacted by the difference in voltage applied to it on each side. You could put a hundred volts on both sides of a piece of wire and there would be no current flow. You can put a thousand volts on each side of a resistor and there would be no current flow And you can put any voltage you want on one side of a speaker and put the identical voltage on the other side of the speaker and there will be no current flow And without current flow through the coil you won't get sound or motion or heat or anything at all. It is the difference in voltages applied that makes things happen
My idea is very basic, just having the matrix/routing electronically, and having 3 speakers. Stereo pair for side, centre for mid. If you use balanced mode radiators you'll get super wide dispersion, and you can point the speakers away from one another, really filling a space.
Hmmm, pointing a speaker sideways does not alter its polarity or phase. If the speaker is not mounted in an enclosure then pointing it sideways will give you a null where you have cancellation between the front and the back. If the speaker is mounted in an enclosure and you only hear the front of the speaker, then it doesn't matter which way it points, The speaker is in polarity in all directions. Exceptions to this rule are open back speakers and flat panel speakers that the back is not sealed. So for this enclosure since all the backs are together inside the box and all the fronts point outside of the box, All the speakers are in polarity no matter which way they point Unless if they are wired out of polarity in which case they would be out of polarity no matter which way they point
If you take a speaker that's in a box and in polarity and point it forward, and measure the sound it will be in polarity in front behind above and below. If you take a speaker in a box and point it backwards that is in polarity, it will also be in polarity in front behind above and below and off to the sides. Polarity is a 180° or 0° polarity has no 90° shift. It's all or nothing Yes or no in or out. Phase is a frequency dependent shift. Here's a video on monitors versus main PA polarity and I believe I cover the direction that speakers are pointed in it. ruclips.net/video/d_3NOvt-gNg/видео.htmlsi=xDXCqNqAWVzSv0rm
Dave - I do MS recording and think this is neat. But I just play mine back in surround sound and I get near 300 degree imaging. But how might we play back MS with crosstalk cancellation for 3D imaging in the room rather than just from the front pair?
hey dave, lets say you were doing a little show, a couple dual 18 subs and some loudspeakers stacked on top of them. would it be insane to use three 60 degree speakers on top of the sub stack, left center right, with the center in mono? i have two srx828sp bass boxes and, now i have 3 of the srx835p tops, because one broke and i had to get another while it's in service. what do you think?
Not the subject of this video, but I have a question: Besides L-Acoustics (which I've seen in a few of your videos), what other PA manufacturers are your favorites and why?
@@DaveRat I’ll have to check them out! What are your thoughts on Meyer and high end JBL (a couple other of the most common ones I see being used at big concert venues)?
JBL's okay, they're more geared towards upper mid-level gear. Meyer is geared towards higher end stuff and the new panther stuff sounds pretty good. The challenge with Meyer is it's all self-powered and very heavy. And Meyer tends to over process things in my opinion. L-acoustics is known for being super high-fi and clear high volume stuff. D&B is known for getting really loud and punchy and also sounding very clear
@@DaveRat I went to a concert with L acoustic arrays. I think the subs were EV but the mix sounded extremely clean and clear even indoors. I know JBL makes a wider range of quality than the other brands here (which seem to only make top end gear). I’ve heard some cruddy sounding JBL (though I’m sure part of it was the system not being set up right) and I’ve heard some great sounding JBL…seems like a huge difference in quality between, say, their EON series and their VTX series. Thanks for the replies, btw! If you haven’t done this already I think it’d be cool if you did an in depth comparison video of these commonly used brands we see at big concerts. Always looking to get more perspectives on this stuff!
Hey Dave! Anyway we can get some stereo reference tracks with that speaker statem in both configurations? It'd be really cool to get down and dirty on analyzing what the summation does!
@@jimpemberton I'm afraid I would have to find room for one. All of the audio from my computer and the TV runs through my sound mixer so I could stop running everything through my PA speakers with a cabinet like that.
It sounds really cool, works really well for a center TV speaker with or without left right speakers. Also, pointing it into a corner or a wall sounds super cool as the reflected sound fils the room and gives almost a surround sound vibe
So this is a mid-side speaker for a stereo signal pair. If you have a mid-side signal pair, then the equivalent speaker setup as the mid-side microphone setup *would* be valid, right?
The signal setup for mid side is 3 channels. Forward mic which has some left and right info, both in polarity Left firing speaker which has left info in polarity and right info out of polarity And right firing mic which has right info in polarity and left info out of polarity. If you use just 2 signals, one for the forward mic and one for the fig 8 mic and use a single side firing speaker like a flat panel speaker that has a figure 8 pattern, that speaker can either be sent the in polarity left/out of polarity right or the in polarity right / out of polarity left signal. But to reproduce mid-side recordings properly you need both of those signals not just one or the other.
Correction, after readinf some comments and pondering, looks like yes, a single flat panel speaker or open back speaker would work for the side firing speakers. And just needs a left-right or right minus left signal.
Could you do the L(-R) and R(-L) as electronic summing opposed to mechanical summing ? Doing this with introduction of timing delay/offset of the out of phase signal could be used for a small amount of steering ? Just a brain fart Edit: to clarify my comment I meant doing the phase inversion and delay prior to the amp. Could you also had a LPF to remove the high frequencies in the summation of out of phase driver on each side ? second edit: could you use fir filters to adjust the phase on the side firing woofers to acoustically sum but provide offset signals with the minus odd the opposite stereo field added ? If that makes any sense 🤷♂️
Yes much can be done electronically and TV sound bars are pretty good example of the exploration of that. That said, I personally don't like many if any TV sound bars as they tend to sound phasey and are often difficult to find intelligibility with. Increase the stereo image and ruin sound quality and vocal clarity. Conversely this simple non-electronic totally passive version, sounds clear as a bell and increases stereo imaging but doesn't sacrifice intelligibility and a simple to implement
The Widest 3D Sound-field Ive ever experienced, came from my EPI 100v speakers (using the model with the upgraded plastic-bezel tweeter version). It seems the Tweeters Inverted cone (concave vs convex), causes the sounds to extend much further to the opposing sides, than a standard dome tweeter. That, and likely because of its high-energy nature, due to its more powerful magnets + more powerful voice coils. I had them about 8 feet apart, and you could walk almost within 1foot of the extreme left, and still get a perfect stereo image from both channels.. without Any notable drop-off in the stereo sound. The sound also extended back into the room at least 15 feet deep... before any drop off in the sound field was noted. They sound best, on the floor, and closest to a wall... despite being a little bigger than Bookshelf speakers. You can be standing, sitting, or laying down... and you still wont have any notable vertical drop-off issues. Thats how absolutely amazing their spread is. In addition, their accuracy is Audiophile grade. They will produce almost zero distortion... even at high volume levels... allowing you to hear NEW things in the music, that you have spend over 2 decades enjoying. For example... finally understanding lyrics that the singer was singing... that you never could make out on any other speakers / headphones (due to the vocals being semi-distorted with the musical instruments... and inferior drivers that are unable to maintain a flawless sonic reproduction, leading to Muddy Blending from micro-distortions). You can find them fairly cheap, used. Or.. you can buy from one of the former employees, that sells them hand-made.. at his website: "humanspeakers" dot com This attempt at a wide field, might be workable for concert PA speakers... but on a small scale, I hear virtually no differences. Furthermore... you dont typically need a +30 foot sound field in a standard apartment / living room. A lot of people are listening in much smaller room sizes. BOSE used a trick to get an extended sound field.. using dirt cheap low-fidelity drivers, by using multiple drivers facing different directions. I personally despise Bose, as their drivers are Inferior Garbage, that have no business representing themselves as "Audiophile" grade speakers. Interestingly enough... another very wide stereo soundstage speaker that I experienced when I was like 15 yrs old (Im 50 now), was at a friends house. I asked him where the sound was coming from... and he pointed to these Octagon shaped coffee tables. I believe the top-most area of each octagonal side... had the vents for sound to exit through them... and the sounds filled the entire room, in a very 3D way. I dont know what the make / model was... and or what drivers they used in them. All I can say, is that they really dont make speakers anywhere near as good as the late 70s era speakers. These low power ECO drivers.. dont have enough cone control, and exhibit a lot of micro-distortions, as well as have a very poor stereo imaging spread. And while many praise expensive Planer drivers for their sonic accuracy... they seem to have some of the most Directional of sounds, with the most NARROW of sound-fields. You have to sit in the perfect position, else the stereo effect drops off Dramatically. I could never deal with that. Id rather have the wide spread of my EPIs, with "slightly" less fidelity... to get a much better overall sonic 3d experience.
I'm not sure I know they have some cool tech for below frequency and getting lots of low and out of a small enclosure. A lot of people don't like bose but I personally very much appreciate their tech and the fact they're owned by MIT gifted to them. For a while they were very kind of overpriced cheapish consumer but they've come around and they're technology is always been very innovative
Hey Dave, long time member (had to create a new account).. kind of off topic question but the subject of polarity reminded me.. I recently watched a video of someone reversing the polarity of their overheads because the sound that hits the mic after the drum is hit is a negative pressure wave in his eyes.. the positive pressure wave goes to the floor (under a drum kit). What are your thoughts on this?
A lot to unpack there. Many people, (I believe incorrectly) believe absolute polarity matters. I have never seen or heard anything that proves it matters whether the initial pressure wave is positive or negative That said, what does matter is the polarity of combinations of signals, the more similar the signals, the more polarity matters. Two identical signals summed together out of polarity will cancel completely I do believe polarity of overhead mics can matter. Not so much for the cymbals, but for the other sounds that the mics pick up If your floor Tom is the same distance from your ride symbol mic as you're overhead is from the floor tom, then the floor Tom is going to bleed into both the overhead and the ride symbol mic. When the overhead mic and ride symbol mic are mixed together, those two versions of the floor tom will either sum together or cancel out to some degree. If the overhead mic is in polarity with the right symbol mic, then the low end from the floor Tom will be additive in both of those yet delayed in time compared to the sound picked up from the floor tom mic. By reversing polarity of the ride symbol mic, some of that floor tom low frequency from the combined ride cymbal and overhead mic can be canceled out, possibly cleaning up the four tom sound. I did some videos on stage mic polarity if you haven't seen them you should check them out because they cover this subject
This is a fairly simple wiring and there have been many dives into this over the years. Hafler, Eno and others have done versions of various wiring to create surround or augmented stereo from stereo signals. Often putting speakers in the rear of the room Just about every TV sound bar does some sort of electronic version that usually sounds like phasey crap. I wanted to specifically focus on a simple mid-side implementation such that a single speaker can would offer clear wide stereo without that phasey sound of a sound bar. I have some other cool wiring methods and added a little subwoofer speaker to the box as well.
For veterans like Rat, it must be a little annoying that it's called immersive sound. It's just as much a marketing issue as it is a processing issue, a lot!. I think that before taking the step towards such systems, sound engineers should be more concerned with acoustics. If you've mastered acoustics, it's the next level.
Ha, yeah, The adventure of surrounding people with sound is a challenging one because any energy spent on pulling focus away from me band or artist strictly has less than ideal results. Money and time and technology spent on increasing sound system complexity so that various sounds can be put in various places for a period of radiate from various locations is often better spent on improving the quality of a simple system to cover the venue better with a higher quality sound. That is not to say that systems like L-Isa Don't have value and offer assets, but they can easily be more distraction and detrimental to the sound. My biggest reservation about L-isa, is been having multiple small clusters may offer flexibility and where the sound radiates from but it lacks the clarity and power of 2 giant long line arrays. In order to get an L-isa system to match the power and clarity of a stereo rig and also offer the advantages radiating sounds from various locations, pretty much need the entire stereo rig you would have had anyways and they need to add a bunch of smaller clusters as well which is an expensive and considerable undertaking.
anyone have any experience with the space station center point stereo? I really want a setup like that to get the leslie sound without need the leslie.
@@DaveRat have you ever had any interest in leslie speakers? they sound unlike any other system I’ve heard and I’ve to re-create the magic of one in person
Yes they are very cool and I can't remember which band I was mixing but there was a song that needed a Leslie sound and I did a bunch of research and at the time ended up with a Dynacord CLS222 Leslie simulator as the best solution for the situation. Not the real thing but it was close enough for the live mix. If I was on the musician side of things, having an actual real world Leslie would be very cool. I've mic'ed a bunch of them in various bands, One of the aspects that's interesting is that The physical diameter of the rotation of the horns or woofer output is important. You can't get a lot of Leslie sound if the physical size of the Leslie is not big enough. Bigger is better and it can't really be shrunk if you want to get that full Doppler effect
I have for decades espoused a mono+ambient approach to 'stereo' listening, based on breaking the sweet-spot paradigm. in the mid 80s I worked on a tv soap that switched to stereo, & we tried quite hard to get the broadcaster- the whole national network- to adopt M-S instead of left-right stereo. imagine you are shooting a two-shot at a table, one actor either side of the centre line. you pan them left & right. then you cut to the reverse angle, & pan them the opposite way. distracting, isn't it? which is why we don't do it in the cinema. in fact, we recorded everything M-S anyway, with a small schoeps figure-8 fastened to the usual 416 or 816. [on location, individual radio mics went on a mono track, & we'd have to fake the boom into stereo because the VTRs didn't have enough tracks.] the idea was that you'd "put the viewer in the room with the actors" using the ambience, but keep all the main dialogue in the centre, where the screen is. pretty much what happened in most cinemas too, at least until surround caught on. so we lost that battle. but there's no reason why an m-s matrix couldn't be installed on most hifi systems, to drive a centre speaker & something like a small array of surround speakers dotted about the place & with some wired out-of-polarity. kind of like hafler in reverse..... remember that brian eno diagram on one of his ambient albums? the same trick that a lot of hifi companies used for faking quad, back when that was the thing.
Super cool and yes you get it! And this has been implemented in many different ways over the years by different people and companies. And I thought this cool simple speaker that allows anyone to build it themselves would be a fun project and a chance to help explain MS recording
Typically 5.1 or at least true 5.1 is actually mixed as separate content so there are six channels of audio with the 0.1 being the subwoofer and five separate audio mixes for the other five channels. This setup extracts all the stereo information and send it to separate speakers that only reproduce the stereo information and none of the mono information. Then that stereo information is added back in with the left stereo information added back in in polarity to the left speaker and the right stereo information added back in in polarity to the right speaker and the left stereo information added back in out of polarity to the right speaker and the right stereo information added back in out of polarity to the left And all this is done with very simple wiring and a stereo amp. No artificial processing or crap added to the sound
![Megan Thee Stallion - Roc Steady (feat. Flo Milli) [Official Video]](http://i.ytimg.com/vi/HozPPyqW0dY/mqdefault.jpg)
![Twenty One Pilots - “The Line” (from Arcane Season 2) [Official Music Video]](http://i.ytimg.com/vi/E2Rj2gQAyPA/mqdefault.jpg)
I really appreciate these detailed, analog-focused descriptions. I know soundbars and Atmos setups do similar things via DSP, but it’s really helpful to see this illustrated from basic fundamentals.
Fun and yes, I also like the non processed path of purity and simplicity.
When i was 18 i worked add a recording studio here in Van Nuys California called Sound City. They had two electrodyne mixing boards and a very limited number of outboard devices. I think each Studio had two tektronix limiters and a coil which could be patched in to flip the phase of an input if needed. They also had two live Echo Chambers which were really cool. Even when they upgraded to Neve consoles they really didn't add any outboard equipment. And even with that limited equipment they had tons of gold records on the wall that were recorded in those Studios. Simple can work 🙂
@dreammix9430 super cool and what an honor to work there!
@@DaveRat it really was. I'm very grateful for that experience
🤙👍🤙
Matthew Polk did this EXACT thing in the 80s with the Polk Audio SDA line of home audio loudspeakers. There was even an interconnect between the left and right speakers, and special instructions regarding what type of receivers or power amps could and could not be used (because of common grounds). The top of the line model from that series had eight 6 1/2" woofers and four dome tweeters on each side, placed into special groups just as you explained. He did it all at power amp level, and the crossovers in each speaker were four rack spaces tall. Those speakers are still revered by home audio folks today.
Very cool stuff and there are many interesting and adventurous implementations of creating spacious audio.
I believe the Polk audio stuff was designed for use as stereo speakers placed in conventional locations and to enhance the stereo further using lots of speakers and processing and such.
The reason I dived into this MS speaker adventure was that it requires no processing and uses four speakers and a very simple wiring technique and can achieve very cool and interesting sound from a single enclosure and some fairly inexpensive speakers.
I made a second one and put in rechargeable batteries so it could operate off the wall or off of the batteries.
Fun stuf
The SDA speakers had the tweeters side by side. The reason for that was to cancel out the left speaker in the right ear and visaversa. I found that design somewhere back in the 70s and built one. Pretty cool.
Very cool!
This is an interesting idea! I'm not an audio pro, but I keep coming back because I always learn something. Doing something new with speakers is always more interesting than the struggle to perfect the same ordinary speaker designs.
Awesome and thank you for hanging out and for commenting!
This was actually done by a bluetooth speaker manufacturer named Riva some years ago! Look up Riva Turbo X and Riva S - they used a single front firing fullrange driver and two side firing fullrange drivers in combination with mid-side processing to create a sound field!
Very cool and yes it can be done with processing and this concept is kind of the basis of all those sound bars out there.
My goal was to build it with no processing so that it could work with any stereo signal and not have all that weird phasey sound that sound bars have and most processed versions seem to embrace
@@DaveRat take a look at line driver/receiver chips DRV134 but from THATS, they can be operated in a creative way that would be very convenient in a controllable stereo widening mid-side setup. I've always liked this idea for environmental background music in a space you can walk around in.
This was interesting to watch, I grew up using the old KF730's. it was always interesting to see how they had side firing woofers and how they sumed up in the center to create a single mono sound. It's pretty cool to see how you can keep the stereo sound moving from one extreme to the other with it suming together nicely.
Very cool and side firing speakers producing low frequencies pretty much act almost identically to a forward firing speaker except they have the little advantage of slightly increased energy forward and rearward and slightly decreased energy off to the sides in the low frequency.
The mid-side speaker already exists. It works very well. It is produced by Aspen Pittman Designs (currently run by his daughter after his passing) and is available on Sweetwater. The technology is called Centerpoint Stereo. The speaker is called the Spacestation v3 and the bigger unit the Spacestation XL. Keyboard players LOVE it, and it makes a great PA in small to mid-sized venues. And I have done great things with electric guitar.
Hmmm, that seems like a glorified sound bar. They say they have some patented 3d audio controllable front to back image control
Not sure if that the same thing as the speaker in the vid that passively recreates Mid-Side with no processing, not 3d, and does not use any patented concepts, just simple logical wiring that works with just about any stereo amp enhances the stereo image
@@DaveRat No, totally NOT a glorified sound bar. It takes in stereo and puts out mid/side for a stereo effect everywhere (not just the center between speakers). It's hard to believe without actually hearing it but it's extraordinary. There is a matrix and a speaker special made by Eminence set at 90 degrees from the front main speaker and tweeter.
Many are skeptical until they actually hear it.
@@DaveRat It has a front facing closed back speaker that gets fed L+R, and an open back speaker below it rotated 90 degrees that gets fed L-R. So very much a mid-side amp. They do sound quite stereo. Fender made a line of amps called SFX that used the design. It's been a while since I looked at the patent, but I think they also crossed-over the low frequencies to only be reproduced by the L+R speaker.
Cool and interesting! How does it address the out of polarity rear sound radiation of the side facing speaker to not be asymmetrically destructive with the forward facing speaker?
With mid side recording, the symmetry is achieved by having 2 versions of the figure 8 mic signal of opposing polarity
But with an open back speaker, you only have one version of that signal
So while the fig 8 pattern can be created, unlike fig 8 where the forward mic is in polarity with the left signal on the left and in polarity with the right signal on the right, with a single version, you are face with a choice of the forward speaker being in polarity with left or the right.
But perhaps that is where the patented processing comes in. Messing with freq dependant polarity ( which is actually phase and time shifts) could address some of those issues.
That is why I used the sound bar reference, as sound bars tend to implement freq dependant processing ( delay and phase) to force the sound to sound wider.
I see that as a vastly different approach than this speaker and mid side recording that use no processing, no delay, no phase shift. Just put polarity reversal.
Mid side recording can be done with a y cable, a polarity rev cable and 3 inputs. And actually can be achieve totally passively and requires no processing.
And interesting. Would be cool to hear the speaker and or know what they are actually doing to the signal.
@@DaveRat I'm not an expert on the processing system. There is a single pot called "Width" and when you dial it all the way down the sound collapses to mono sounding from the front speakers. Dial it up it gets wide. Dial it too far up and it collapses some and doesn't sound so great. Some specific sources benefit from different settings, like single sources vs, a stereo track.
I was able to get a huge guitar sound by running two mono signals into two mixer tracks the EQ'ing them slightly differently and panning them 50% + - L R.
Many keyboard players say it's the best Leslie simulator speaker there is.
There are some rare instances where a stereo mix (older recorded music) just hits the matrix wrong and sounds weak. However, the wide sound experience carries from room to room. It's wild. And kind of makes it the perfect party PA for a house. The music is ever present everywhere while guests can talk over it (it somehow doesn't interfere with talking like regular PAs do. )
Anyway, I see you are very interested in the concept and thought you'd be interested in what work has been done before. IT isn't very widely known except that Fender licensed the technology for their acoustic guitar amplifier.
Dave you are such a huge inspiration for me and probably about 90% of the people who watch your Channel! I just want to thank you for continuing to teach us and share your knowledge
So cool and thank you! I really appreciate the positive input and it makes me really happy to know the info I share is appreciated!
I remember doing this sometime in the early 80s after reading an article in an amateur radio magazine describing this technique. In addition, each of the side speakers included a 47 ohm or so wire wound pot connected as rheostat in series for adjusting volume and something like a 10 or so microfarad cap, also in series, for high-pass filtering and some additional phase shift.
Super cool and will try and do a vid with alternate wiring adventures
You’re my spirit animal, Dave this is rad. Love watching you come up with stuff.
Awesome and thank you!!
for real he always has such an interesting take and then backs it up with his experiments
Honored!
Mr. Rat, Dave, you've inspired me with other video's to build my own line array with 2x6 cabinets. And it's going good!
I love that keep me updated!
I made a speaker that is exactly what you did here exept using a dsp and using a center channel instead of a left and right , so thats 3 channels out of stereo ,
Differential stereo out of the sides and mono from the front , so far it is the best speaker i have ever heard , it fills a room like nothing else does while still having a center channel to handle voices or foreground noises .
Very cool
WOW! I've recorded mid-side for years, but you're the only one I know of that could reverse engineer it to actually work in a speaker configuration! Always love your videos. Can't wait to see the next one!
🤙👍🤙
I'm already using this with a matrix, through 4 AUX busses and it works great. I could use 3 busses, but am using two stereo amps. So nice you've built this speaker box!
Have you experimented with speaker placements?
@@anttikinnunen7904 I've tried a lot and ended up putting the mid speakers on top of the L and R (sides), this works well to widen or narrow the stereo image. I could use a true center speaker, but I'm not looking for surround effects, just enhancing the stereo image. Other things I've tried is rotating and letting the L and R speakers bounce off the walls with a slight delay to make it less direct, and placing the L and R speakers in line with my ears, like earphones and a true representation of how M+S is recorded. That was not very useful, because the direction of the sound changes in that case when putting in more or less side signals. Experimenting with vinyl records this way is amazing, by changing polarities and levels magic occurs and you can hear things isolated that would be impossible otherwise.
👍🤙🤙
I did this same thing, but only using one amp, and four speakers, which I can adjust with a single potentiometer and also can switch the added two speakers in or out with a simple switch. Works great!
👍🤙👍 Yes this only uses one stereo amplifier and I did do a switch that converts it to stereo or stereo Plus side speakers. And then I built another cabinet just like it so I could compare and created some other fun wiring that adds EQ to the side speakers as well as put in a amp with a subwoofer out and added a fifth speaker as a subwoofer on a modular removable rear plate.
What are the speakers and amplifier you are using for these kind of things? I would love a go-to speaker and amp for DIY stuff like this.
It would be really cool to take some multi-track stems and create a mixed song where you have one instrument panned full left, another full right, then one 20% left, another 20% right, and then something centered (mono). Then you could use that mixed track for your demonstrations.
Good idea I have some live recordings I could mess around with and do something fun like that with
reminds me of the old quadraphonic setups.
🤙👍🤙
This was a fun experiment to watch. I can definitely see myself in my older years when my boys are older just tinkering around trying different things. Love the video, can't wait to show some friends this.
Cool cool Dey Dey!
This seems to be a variation on the poor man's surround scheme that I used back in the '80. The only difference is I would run the "surround/ out of polarity " speaker pair in series with anapprprat L pad then connect across left + and right + . This scheme worked well for my purposes but did necessitate using an amplifier with a common speaker ground. As a side note my mom had a Buick Regal where the factory " miss wired" the front speakers in the same manner as you did. When we got the fader set just right we ,in fact, did get the best stereo effect in the neighborhood.
Very cool and interesting about the Buick!
I have used this exact wiring but placing each speaker in each corner of my listening space. For some recordings it made a phenomenal effect. At some point we were watching family videos that were filmed with a stereo capable camera while someone was called out in the video. The stereo enhancing was such that it fooled us into thinking the call occurred for real. The called person even got out of the room to search for the caller... until she also responded in the video, haha.
I never thought of placing the four drivers in one single enclosure, now I have an urge to replicate your build.
How do you think sound would change if the enclosure had separate airtight chambers for each driver?
Very cool and yes I've seen the surround sound type versions of this.
As far as separate chambers, The main result would be a reduction in low frequency output as it would reduce the enclosure size. Since the surround sound speakers typically don't have much low end the forward-facing stereo speakers can use the entire box to get added low frequency. But if you have something that is low-end heavy to one side, The side firing speakers now have the entire enclosure to reproduce that low frequency.
By segmenting it into smaller enclosures you either need to make the whole box bigger or sacrifice low frequency.
I made another one of these and you can't really see it in the video but they're modular. The rear panel is removable and replaceable with a panel that has a fifth speaker that acts as a subwoofer.
Also that rear module can be swapped out to make the speaker have a battery powered option.
Those are super fun and they sound really cool
Superb content, thanks. Also the lack of blinking really helps me focus on the information
Do I not blink?
@@DaveRat I'm not sure, you may have inadvertently edited out the blinks. Haha I was semi joking. I found myself wide eyed and unblinking also, and I was very immersed in the content, so no complaints!
@@DaveRat I am exploring this information and doing some experimenting of my own with stereo guitar. I'd like a small but effective stereo guitar set up for my home practice space. So far I've relied on speaker separation and had no prior experience with this type of strategic phase cancelling
Brian Eno had a description on is ambient records about hooking up a third speaker between the plusses of your amp. As a kid I did notice this speaker had one of the normal speakers as an in phase friend, so ended up having two of these in reversed polarity, similar to your setup, but spaced. It did bring up extra layers from Eno's On Land LP, for instance.
Then I noticed my amp did not like it, so I ended up with two rear speakers, that both had left and right speakers in them, and a small slit between the drivers was the air output. So I added the signals acoustically, rather than electronically. I do not think I was ever satisfied all the time, but I have forever liked sound being dislocated from the speakers.
40 years later, In theater, I often add Omniwave speakers, to pull the sound out of the main speakers. From what I learned as a kid, I just need some energy from the mid range to pull the voices to a virtual location between the speakers. I do not need them to be phase aligned so much, but time alignement is crucial.
Was just going to comment about the Brian Eno method. You could probably recreate the third speaker with a phase-inverting matrix (rather than using the amp method, which halves the resistance). Would be an interesting comparison to hear it. The "Triphonic" Eno method might be easier to implement in a live environment as it only requires one additional speaker stack?
Very cool and interesting thank you
There are a whole bunch of cool ways to wire the speakers for various versions of this.
And ways to add capacitors and chokes to EQ, and all can be done passively.
Actively opens up a whole infinite slew of other options.
The stuff that I've been working on I wanted to make a speaker that sounds cool and runs off of standard stereo Bluetooth and creates a wide stereo image from a single speaker
This passive method was best illustrated by David Hafler’s pseudo surround plan from days gone by.
Looking forward to hearing more from this project
🤙👍🤙
Hi Dave,
In the future video, would we be able to take a look at how to optimize your four speakers setup? I'd love to dive deeper into what's actually happening in the amp when you "cross-load" the channels like this. Maybe even investigate with some oscilloscopes or impedance meters.
When playing back mono audio, my assumption is that the amp is only loaded by the main L and R speakers since the voltage between the positive terminals of the two channels is zero. Would that mean for any regular stereo track, any mid information in the waveform wouldn't contribute to the load and we would only have to calculate what happens in the side scenario?
My main concern is with impulses, timing, and frequency response. How do these properties differ from the conventional LR to when the amp is cross-loaded with the L-R/R-L speakers?
Yes I will try and do another video on it and show some other cool wiring techniques.
As far as the amp is concerned the left speaker loads the left channel with 8 ohms and the right speaker loads the right channel with 8 ohms.
And the side speakers which are two 8 ohm speakers in series do the exact same thing as running an amp bridge mono into 16 ohms.
And if you look up bridge mono or look at any of my videos I did on bridge mono, running bridge mono into 16 ohms is the same as running stereo into 8 ohms.
So with an 8 ohm load on each channel and bridged mono into 16 ohms the amp season equivalent of 4 ohms per channel.
This looks and sounds awesome! Could this be made active easily? I can imagine this paired with a smaller sub to make a more immersive take on the column PA systems that could also be very portable and easy to set-up/tear down.
Really neat kit! Good explanation of the concept too. I knew what mid-side meant in terms of mixing (e.g. mid-side eq in a DAW), but this gave me a whole new perspective on it. Good stuff!
This general idea has been a around a long time, historically wiring a single speaker in the center of the back deck of your car to the plus signals of both Left and Right speaker (the difference). Produced a pleasing "Side" only sound that made it almost sound like a third, sometimes ambient channel depending on source material.
Yes is numerous implementations of various wiring to create surround sound. Brian Eno and hafler and pretty much every soundbar that's ever been made.
This particular version is based on a mid-side design radiating a wide stereo image from a single point or a single enclosure.
The concept here was to explain how mid side works and also show people that using a simple stereo amplifier and some inexpensive off-the-shelf speakers they can make a super cool ultra wide stereo enclosure with some simple wiring.
@@DaveRat And great job, BTW. I enjoyed this episode a lot. As an aside, most people don't realize that stereo LP records are encoded Mid-Side (actually L+R and L-R) for mono compatibility....but mostly because it's just very cool!😎.
👍🤙👍
Really nice video, thank you !
I don't know if it's been mentioned yet but Bang & Olufsen have a few mid/side bluetooth speakers in their range, like the Beolit 17, 20 and the Beosound Level and with the smartphone app you can indeed make them "project" sound closer or further away from the speaker and i think it's neat and better than the more typical stereo design on smaller portable speakers like these
And yes these B&O speakers rely heavily on DSP to work but i think it's sexy that they can use a single (mono) amp for the two tweeters and still get some sort of stereo image out of it. They don't use any really fancy patented technology either in fact B&O doesn't brag about this tech, they don't pretend it's magical and will create a 3D like soundstage or whatever, they just say you can dial in the width of the sound in their app with a simple setting.
Interesting and those things look cool and would be fun to hear.
They say they have 360° sound radiation and the width or closer and further adjustment is interesting.
Often that is done with ambience or adding ambience or some sort of reverberation type thing.
I don't see any information indicating the design is related to mid-side though.
In fact I think the fact that it is 360° kind of prevents it from being a mid-side design.
But yes very cool stuff I have a B&o turntable and they make beautiful and interesting gear
@@DaveRat i learned about this mid / side thing by an engineer on the B&O facebook group, i own both of these speakers and it's quite funny how you can indeed make the sound larger or narrower even if the three tweeters of the Beolit 17 are wired to the same amplifier. The Beosound Level has two tweeter at it's extremities that plays the side channel while a center larger tweeter plays the mid. On the Beosound Level the effect is more pronounced, if you sit in front of it and play with the settings it truely make the sound goes really wide to just mono. And here again, the Beosound Level is only using a single amp for both tweeters which are wired out of phase.
On both of these speakers, putting them to the widest setting does make the music a little weird cause it often deeps the vocals and make all the instruments that are panned left right or have been recorded in stereo, really loud. And if you put the setting all the way to the mid only it makes the sound very flat and hard panned instruments will almost disappear. It never fully isolate the mid or the side channel of course but i think it goes as far as 50%.
Very cool and interesting and thank you! Yes that sounds like what can be done with polarity and multiple speakers.
Very cool
Dolby Pro Logic was a sort of mid side speaker system
🤙👍🤙
This has really got me thinking!
If we were to wire up a PA system following your diagram, a speaker received the same signal from both + terminals, I agree cancels out but would that cause any damage? The driver to over exert, amps not discharging properly etc?
Could we archive the same thing by wiring up a PA as we normally would, then adjusting the polarity in a system controller/amp’s DSP?
Connecting of speaker between the two hot terminals of an amplifier is also known as bridge mono. You can do a RUclips search on that or look at other videos that I did on the subject.
If the amp is sent the same signal to both channels nothing comes out of the speaker.
If you send something different to both channels like an equal and opposite signal then you get twice the voltage out of the amplifier.
As far as loading down the amplifier each channel sees one half of the load impedance.
So if you wire an 8 ohm speaker across the two hots of an amplifier then each channel sees four ohms.
Some amplifiers are actually built with each channel being two channels in bridge mode. This is done because bridge mode more efficiently uses the power supply of the amplifier and allows creating higher voltages therefore reducing the amount of loss in speaker cables.
The downside of bridge mode is that it lowers the impedance of the amp sees and running bridge mono into four ohms loads each channel down to 2 ohms which may be thermally or electronically less than ideal for some amplifiers.
On the other hand some amplifiers can run impedances well below a single ohm.
Great video, what if I want a mono speaker instead of left and right. Can I just sum left and right in this passive matrix or will that destroy the side feed? Also.. can I make this matrix in a dlr connector box and feed amplifiers instead of speakers? Is it possible to make a passive xlr MS matrix box. Left and right in and mid plus side out?
Yeah that could be done, And I thought about exploring that angle, you could build a passive resistive voltage divider network to derive the mono center speaker signal and run that into a mono amp into the mono center speaker
And then run the stereo signal into a stereo amp and wire it with the two side speakers in series as I show in the video, across the two hots of the stereo amp.
Excellent video not sure that I fully understand. But very interesting! One question what amp and speakers are you using?
Really interesting to appreciate this clinics, it motivates me to apply in live bands world. Thanks one again Dave.
👍🔧👍
Great explanation. Without a mixing console this could be done with 2-channel processing purely analog or with modern DSP. I believe the "IMX Expander" on radio shack EQ's did this back in the 80s/90s.
🤙👍🤙
Arrgh, just lost a longish post due to internet dropping off. About adding a L+ and R+ speaker to the rear at a party in a wood. Sounded great whichever way I was pointed. Blew one person's mind... but it *was* a party.
At my home studio setup, right in the middle, I have two small monitors wired to the L and R (the L to the right-most small box and vice versa) with the levels much lower than the main L&R...Lovely tight centre, and still sounds wide.
Tried out of polarity on one of these two centre boxes which gives some odd perspectives! Maybe a delay on both of these... hmmm........ (I like rows of dots)
very good demo of a slightly complex thing like your vid on line arrays with the mini speakers. this has applications for club systems (for dance music) that want a sprinkle of a bigger stereo image across a big space at high SPL. Appriciate your vids.
👍🤙👍
I love recording in Mid side, acoustic guitars and overhead drums in particular.
👍🤙👍
Huughes Research made a processor to do the same thing. Hughes AK-100 SRS Sound Retrieval System
Interesting unit. Seems to be a quite different adventure as it uses processing to emulate spaciousness using 2 speakers, but interesting none the less. Here is what I found on it:
The Hughes AK-100 Sound Retrieval System (SRS) was first developed in the early 80’s by Arnold Klayman to create an immersive 3D sound field using only two speakers by using the process known as ‘head-related transfer functions’. This process creates a more spacious ambience within the stereo field, producing strong special cues for different instruments in an audio mix.
This reminds me of the Carver Sonic Holography technique, where he sold a box that would do cross-mixing of left/right channels where the cross-mixed signal would also be inverted. This type of psychoacoustics has intrigued me for a long time; I once saw a Bose soundbar about the size of a normal center channel speaker that created a credible left/right wide stereo effect and I have been interested in the idea of doing something similar in a bluetooth speaker to try to produce a more credible stereo effect from physically small speakers.
Super cool and yes I've seen many implementations and versions of this done for surround or with various forms of processing.
I thought this simple straightforward approach that's based on MS recording would be a fun project and it worked well beyond what I expected it to do
Just made this setup in Ableton using groups and the stock Utility plugin. Effect Group -> (Normal), (Wide group) -> (normal phase), (utility with phase inverse and swapped channels).
Sounds to me like it's exactly what the width control of what the utility plugin and many others actually do, but it's interesting to be able to have the extra control over the channels and be able to process the wide vs normal signal and also to know what it's doing.
I built this speaker so that a normal stereo Bluetooth signal or any stereo signal could be used and recreated by a relatively small speaker and create a wide stereo image.
I believe there's advantages to doing it acoustically using multiple speakers versus trying to synthesize it by altering the signals and running them into a standard stereo speaker setup.
This setup adds clarity and dimension without creating that phasey sound that happens when you try and create it with signals and run it through a standard speaker pair.
Another example that sounds phasey and undesirable to me are all of those TV sound bars that are trying to do a similar thing.
@@DaveRat Yeah, I'm keen to try and make a physical speaker as well. Very cool concept and a great use of the mid-side concept. I was just having fun messing around with it as a mixing tool with that Ableton setup, definitely does add a lot of phaseyness but I am keen to experiment more with it. In my test before I filtered out the lows starting at about 500hz in the wide group to add only a little bit of the effect for extra space and it sounded alright on a first listen. Love your work Dave :)
You can switch Abletons Utility from Width, to Mid/Side Mode by right clicking on the width knob and choosing it✌️
Hi dave, I hope all is well. Dont know if u remember last year u sold me the outter shell panels for a sennheiser md441-U because one side of the mics leather panels was ripped off . You sent me both of the panels, but I only needed one side, so i sent back the other side to you . I hope u received it.
BUT ANY WAYS...
I refurbished a pair of yamaha ns10s, and the woofers r very thin, and moisture is starting to affect the woofer, making them soft. Unfortunately, one side of the woofer ripped, and i had to repaired it..
My question is .. is there any kind of waterproof spray that u can recommend that i can use for that woofer since the woofer is made outta sum kind of paper?
Im using a dehumidifier in my studio now. But i would like to add sum protection on the woofers since they are very old
It may slightly change the sound but any coating willl do that anyway. We use lacquer spray paint as it's quite liquid and absorbs into the paper rather than coating it and it will add waterproofing and some rigidity back into the cone
Thank u so much
such a freaking cool idea! Also the best description on M/S recording I've heard!😆 Definitley need to see more!
Super cool and thank you
Q Sound... Some kind of virtual stereo to me is like MS equation but with a HPF from lows up to the mids at center summing and after that polarity reversal of the one side.
👍🤙👍
Cool stuff. I feel like you could make this simpler, and with three speakers. You have your single mid speaker, then Ldiff, and Rdiff. Wiring: L and R amp + and - go to the + / - of the mid speaker to sum the two parts; then, Rdiff and Ldiff are wired in series (- to +) between L+ and R+ on the amp, producing the difference between L and R to the mid.
Do you have a CNC machine to cut these enclosures?
Yes, have a CNC.
I was aiming to do this using a standard stereo input and or stereo amp without any processing such that it can be plugged into any normal stereo output.
To do the three speakers, the challenge is getting the mono signal for the center speaker without processing.
@@DaveRat yes, but wouldn't you get that by running the + of both L and R amp channels to the center speaker? Hmm, I guess that doesn't quite remove the side signals, though... not unless they're quite out of polarity.
"...work on ways to implement this live. I believe it can be done with a matrix..." This sounds like a good job for an Outline Newton! There are ways that a stereo signal can be routed, polarity reversed, eq'd and delayed that are a bit beyond what can be done in a typical console matrix that may benefit your experiments. I've read that Rat has some in the inventory.
Very cool and yeah hopefully I'll get a chance to test it out
shouldn't there be physical separation inside the box as well, so the L and R speakers do not move the L-R and R-L speakers? and even L/R baffle as well?
I thought about that and actually no physical separation improves The low frequencies.
This is because the side speakers don't have very much low end with a stereo signal that has fairly mono low frequencies.
This means the entire enclosure can be used for the low end of the stereo speakers.
But if low end is panned hard left or right then All the low end comes out of one of the side speakers and one of the forward speakers and reverse polarity out of the other side speaker. This in effect gives all of the enclosure to one speaker creating a deep low end.
If you had four separate enclosures then all of the speakers would have their own tiny enclosure so you would have to either make the speaker bigger or suffer less low frequency overall.
This speaker is actually modular and the back panel comes off and I made a second back panel that has a little subwoofer in it that also uses the same enclosure. And that sounds really cool with even more low end
I made a dj plugin that does from Stereo to L-R | R-L on a pot, the results are surprising, some cases when the producer used an exciter it sounds like a low res mp3 mess, some cases it pulls out samples they couldn't be heard in normal playback and for most tracks it's the strongest bass cut I've ever known.
Super cool and interesting
So clever. It's obvious once you see it, but I've never seen it. Great!
🤙👍🤙
It's a good idea and an interesting project. But I would probably try it out differently:
Each normal speaker chassis has some kind of an almost fig 8 pattern, with some non-balance, caused by the magnet, cage and cone shape. So if you'd mount two identical speakers with their cones facing towards each other and wire them so, that they will be "out of pohase" (actually: IN phase, cause they are pointing in two different directions) - you will get a "perfectish" speaker for your S-signal... If you combine this "S-speaker" (mounted sideways) with a third speaker, that is centered and sends out the M-signal, you'll get a proper MS-speaker arrangement. Now you can use a 2Ch poweramp, and send L-R to the side chassis and L+R (mono signal) to the center OR you use a MS-recording, and directly send the S signal to the side chassis and the M to the center... No idea, how this would sound, but it would be a proper MS-speaker setup then, analog to a MS-microphon arrangement.
The problem might be the big difference between microphones and speaker: Speaker need to transmit POWER, so they need to move air... So for getting a proper result with low frequencies, you'll need large systems that can move a lot of air... But anyway, this would be another very interesting project...
Hmmm, I believe there may be a flaw to that concept. A mid side recording is 2 mics recorded but to creat mid- side you actually use 3 channels.
Forward, fig 8 in and fig 8 out.
To playback mid side you also need 3 channels
One side speaker need signal with the fig 8 left in polarity and fig 8 right out of polarity. The other side speaker needs the fig 8 right in polarity and fig left out of polarity.
The design you described is missing a channel that is needed .
@@DaveRat Nope, you're wrong.
A MS recording contains 2 channels, mid and side signal. If you want to transfer it into a standard L-R stereosignal, of course, you need to add and subtract the side signal to/from the mid. This is what's usually happening within these MS-mixing matrixes.
But if you'd have a proper MS speaker system, with one speaker having a proper fig8 pattern and being oriented 90°, this speaker could transmit the pure S-signal. In addition with a center speaker, carrying the M signal, this all should add up (acoustically) to a proper stereo information.
This setup would only require 2 channels and a two channel amplifier; one for the M and one for the S signal.
Cheers, Salossi
Interesting, ok, yes, that makes sense.
Interesting experiment as usual, you are genius Dave. Looking forward into what comes next.
👍🤙👍
Do the amplifier output channel grounds need to be at the same potential? Some amps float the negative terminals and cant be tied together.
It should work with most stereo amps. Except if the amp has transformer outputs and they are fully isolated, that may not work
I am reminded of the Groove Tubes SPX SpaceStation with two drivers face to face delivering the Side information.
👍🤙👍
What are the problems, in your view, of the approach of using a single mid speaker with a planar speaker turned 90 degrees sent the differential signal??
Thank you. Very interesting.
With a planer speaker, it will be in polarity on one side and out of polarity on the other.
The forward faxing speaker will either be in polarity with one side or the other of the planar speaker
With mid side, the fig 8 is split into 2 inputs, one in reverse polarity to the other
But that can't directly be done physically. Unless you use two planal speakers with reverse polarity and then block the rear radiation of both.
Which is exactly what I am doing but not using a planalr as a normal speaker is easier to block the rear radiation of.
After some pondering, actually a planar speaker turned sideways would work.
Mono to the center speaker, left - right to the planar speaker. Such that the left side of the planar reproduces the left signal in polarity reproduces the right signal out of polarity.
And the right side of the planar speaker reproduces the out of polarity right signal, out of polarity because it's the back of the speaker and the in polarity left signal out of polarity. So the right hand side reproduces the right signal in polarity and the left signal out of polarity.
The net outcome is correct using two speakers as you describe
Would it make more sense for the mid speaker to be a sealed closed back or an open back design?
Thank you for your responses.
The mid-speaker wants to have either a ported or sealed enclosure to capture the low frequencies. Putting them all in the same enclosure allows them to share the space and makes the overall enclosure size smaller for a given amount of low end
So is this what they do in sound bars for home theater setups (with maybe some additional techniques as well I assume)?
And what was that first song btw?
There may be some soundbars out there that do something similar but from what I've heard most of them have numerous speakers in them and they try and do beam steering. Things like send the signal with a slightly different delay to various speakers such that it shoots the sound off to the side or off-center or center. This has the advantage of putting different sounds in different places but it has a disadvantage of sounding kind of weird and phasey as well.
@@DaveRat oh ok. Cool, thanks! What about that song? Sounded good
great! what are the amp constraints for this special speakers connections? Does a class D works for ex, without risk of damage?
I'm using class D in the speaker in the video
@@DaveRat thanks for the precision!
👍🤙👍
reminds me of brian eno's ambient speaker set up he recommended for ambient 1! mono speaker behind listener with only side channel
Yes similar concept. Michael for this was the package The setup into a single speaker to create a wide stereo image. Similar to what TV sound bars do but sounds better and not all phase weird sounding like those sound bars are
Dave, I guess you need to record the figure 8 with two seperate mics, cause the switched figure 8 of a single mic only has on output with both signals coherent in one cable, so no way to split this signal. With two mics and seperate signals it will work. Or am I wrong?!
To do mid-side the figure eight is just a single mic.
Figure 8 Mic is just split to two channels.
One channel has the figure 8 mic in polarity panned hard one way and the other channel has the figure 8 mic out of polarity panned hard the other way.
And the directional mic is panned center.
@@DaveRat OK, thanks. Now I get it, you use 2 channels panned and use the polarity differences then. 1 Mic saved but some 2 speakers more needed. right?
Read some of the other comments there are setups where people are doing mid side playback with just two speakers
Thanks for making this video. I'd love to try making something like this myself!
12:51 wouldn't wiring both positive terminals to the speakers make it so similar sounds between channels would become direct current at the speaker? They wouldn't just cancel each other out or am I wrong? If that is the case, you would have to sum the signals first then high pass to get rid of any DC (maybe in processing). What would that potentially do to the amp if there is no processing? For a small setup like this I'm guessing it's fine.
No it won't make DC. All identical signals would raise the voltage on both terminals at the same time so the votage difference between the terminals would be 0 volts for identical signals
For signals that are different, it would just play the signals that are different as a musical waveform
@@DaveRatThanks for responding! I'm enjoying thinking about this. I probably need to do some tests to confirm this.
You are correct there wouldn't be a voltage differential between the terminals, but the voltage on the voice coil would rise to match that voltage which would induce a magnetic field. Testing voltage between the terminals is only part of the circuit. The voltage would have to also be tested to ground. A voltage difference at the terminals would indicate the speaker is moving, right? If you start at 0v and apply a positive voltage from both channels on the amp there would have to be some amount of voltage rise on the coil. I may need to test that with my amp and a voltmeter.
@Zatore_ voltage to ground is not relevant unless somehow the speakers are referenced to ground. Think of Phantom power where a dynamic mic voice coil has 48 volts DC on both sides of it and has no impact on the mic whatsoever.
A coil of wire or a speaker or a resistor or just even a piece of wire will be impacted by the difference in voltage applied to it on each side. You could put a hundred volts on both sides of a piece of wire and there would be no current flow. You can put a thousand volts on each side of a resistor and there would be no current flow
And you can put any voltage you want on one side of a speaker and put the identical voltage on the other side of the speaker and there will be no current flow
And without current flow through the coil you won't get sound or motion or heat or anything at all.
It is the difference in voltages applied that makes things happen
always appreciate your excellent nerdery Dave
🤙👍🤙
YESSSS. I have been thinking of this idea for a while, so glad you're trying it out!
My idea is very basic, just having the matrix/routing electronically, and having 3 speakers. Stereo pair for side, centre for mid.
If you use balanced mode radiators you'll get super wide dispersion, and you can point the speakers away from one another, really filling a space.
🤙👍🤙
how does it affect it because you put the speakers on the side, surely that's 90° out of phase to the front speaker?
Hmmm, pointing a speaker sideways does not alter its polarity or phase.
If the speaker is not mounted in an enclosure then pointing it sideways will give you a null where you have cancellation between the front and the back.
If the speaker is mounted in an enclosure and you only hear the front of the speaker, then it doesn't matter which way it points, The speaker is in polarity in all directions.
Exceptions to this rule are open back speakers and flat panel speakers that the back is not sealed.
So for this enclosure since all the backs are together inside the box and all the fronts point outside of the box, All the speakers are in polarity no matter which way they point
Unless if they are wired out of polarity in which case they would be out of polarity no matter which way they point
@@DaveRat yes but having a speaker behind at 180°, does I thought 90° does as well
If you take a speaker that's in a box and in polarity and point it forward, and measure the sound it will be in polarity in front behind above and below.
If you take a speaker in a box and point it backwards that is in polarity, it will also be in polarity in front behind above and below and off to the sides.
Polarity is a 180° or 0° polarity has no 90° shift. It's all or nothing Yes or no in or out.
Phase is a frequency dependent shift.
Here's a video on monitors versus main PA polarity and I believe I cover the direction that speakers are pointed in it.
ruclips.net/video/d_3NOvt-gNg/видео.htmlsi=xDXCqNqAWVzSv0rm
Dave - I do MS recording and think this is neat. But I just play mine back in surround sound and I get near 300 degree imaging. But how might we play back MS with crosstalk cancellation for 3D imaging in the room rather than just from the front pair?
Isn't there a fender acoustic amp that is M/S? One forward facing driver with a closed back and another just below and perpendicular for the sides.
Yeah that's what people are posting and sounds interesting
The processing for this would be fairly straightforward to build with an ARM chip (daisy seed for example) and a few codec chips.
They're intriguing little drivers Dave... Are they off your stock shelf, or did you buy some little units to play with?
hey dave, lets say you were doing a little show, a couple dual 18 subs and some loudspeakers stacked on top of them. would it be insane to use three 60 degree speakers on top of the sub stack, left center right, with the center in mono? i have two srx828sp bass boxes and, now i have 3 of the srx835p tops, because one broke and i had to get another while it's in service. what do you think?
That's kind of a good idea and that should work and give more of the room a stereo sound
@@DaveRat Awesome, thanks. now i dont feel so dumb for having an odd number of speakers.
Cool project,, 👍😎👍
Thanks Dave,, 💖🙏💖
👍🤙👍
Not the subject of this video, but I have a question: Besides L-Acoustics (which I've seen in a few of your videos), what other PA manufacturers are your favorites and why?
I like the D&B stuff. They make some cool products that are useful and sound good.
@@DaveRat I’ll have to check them out! What are your thoughts on Meyer and high end JBL (a couple other of the most common ones I see being used at big concert venues)?
JBL's okay, they're more geared towards upper mid-level gear. Meyer is geared towards higher end stuff and the new panther stuff sounds pretty good. The challenge with Meyer is it's all self-powered and very heavy. And Meyer tends to over process things in my opinion.
L-acoustics is known for being super high-fi and clear high volume stuff. D&B is known for getting really loud and punchy and also sounding very clear
@@DaveRat I went to a concert with L acoustic arrays. I think the subs were EV but the mix sounded extremely clean and clear even indoors. I know JBL makes a wider range of quality than the other brands here (which seem to only make top end gear). I’ve heard some cruddy sounding JBL (though I’m sure part of it was the system not being set up right) and I’ve heard some great sounding JBL…seems like a huge difference in quality between, say, their EON series and their VTX series.
Thanks for the replies, btw! If you haven’t done this already I think it’d be cool if you did an in depth comparison video of these commonly used brands we see at big concerts. Always looking to get more perspectives on this stuff!
🤙👍🤙
Hey Dave!
Anyway we can get some stereo reference tracks with that speaker statem in both configurations? It'd be really cool to get down and dirty on analyzing what the summation does!
Amazing concept, Dave! If you come up with a kit, I'm in.
Wound this be a one source speaker for a home sound system ?
@@78tag I think it could be, and one that might sound great.
@@jimpemberton I'm afraid I would have to find room for one. All of the audio from my computer and the TV runs through my sound mixer so I could stop running everything through my PA speakers with a cabinet like that.
It sounds really cool, works really well for a center TV speaker with or without left right speakers.
Also, pointing it into a corner or a wall sounds super cool as the reflected sound fils the room and gives almost a surround sound vibe
Again a very interesting video! Thanks, Dave! Love your work! 🙂
🤙👍🤙
So this is a mid-side speaker for a stereo signal pair. If you have a mid-side signal pair, then the equivalent speaker setup as the mid-side microphone setup *would* be valid, right?
The signal setup for mid side is 3 channels.
Forward mic which has some left and right info, both in polarity
Left firing speaker which has left info in polarity and right info out of polarity
And right firing mic which has right info in polarity and left info out of polarity.
If you use just 2 signals, one for the forward mic and one for the fig 8 mic and use a single side firing speaker like a flat panel speaker that has a figure 8 pattern, that speaker can either be sent the in polarity left/out of polarity right or the in polarity right / out of polarity left signal.
But to reproduce mid-side recordings properly you need both of those signals not just one or the other.
Correction, after readinf some comments and pondering, looks like yes, a single flat panel speaker or open back speaker would work for the side firing speakers. And just needs a left-right or right minus left signal.
2:56 could introduce channel 3 a little bit? I got the figure eight. Pretty exciting!
Dave it would be great to make that speaker into a kit!!
This is super cool! Thank you for the fascinating video!
What do you think about dipole loudspeaker? Can it add more enhance stereo image in room? What do you think about linkwitz loudspeakers?
Nice and great enclosure design.
🤙👍🤙
Could you do the L(-R) and R(-L) as electronic summing opposed to mechanical summing ?
Doing this with introduction of timing delay/offset of the out of phase signal could be used for a small amount of steering ?
Just a brain fart
Edit: to clarify my comment I meant doing the phase inversion and delay prior to the amp. Could you also had a LPF to remove the high frequencies in the summation of out of phase driver on each side ?
second edit: could you use fir filters to adjust the phase on the side firing woofers to acoustically sum but provide offset signals with the minus odd the opposite stereo field added ? If that makes any sense 🤷♂️
Yes much can be done electronically and TV sound bars are pretty good example of the exploration of that.
That said, I personally don't like many if any TV sound bars as they tend to sound phasey and are often difficult to find intelligibility with. Increase the stereo image and ruin sound quality and vocal clarity.
Conversely this simple non-electronic totally passive version, sounds clear as a bell and increases stereo imaging but doesn't sacrifice intelligibility and a simple to implement
The Widest 3D Sound-field Ive ever experienced, came from my EPI 100v speakers (using the model with the upgraded plastic-bezel tweeter version). It seems the Tweeters Inverted cone (concave vs convex), causes the sounds to extend much further to the opposing sides, than a standard dome tweeter. That, and likely because of its high-energy nature, due to its more powerful magnets + more powerful voice coils. I had them about 8 feet apart, and you could walk almost within 1foot of the extreme left, and still get a perfect stereo image from both channels.. without Any notable drop-off in the stereo sound. The sound also extended back into the room at least 15 feet deep... before any drop off in the sound field was noted.
They sound best, on the floor, and closest to a wall... despite being a little bigger than Bookshelf speakers. You can be standing, sitting, or laying down... and you still wont have any notable vertical drop-off issues. Thats how absolutely amazing their spread is. In addition, their accuracy is Audiophile grade. They will produce almost zero distortion... even at high volume levels... allowing you to hear NEW things in the music, that you have spend over 2 decades enjoying. For example... finally understanding lyrics that the singer was singing... that you never could make out on any other speakers / headphones (due to the vocals being semi-distorted with the musical instruments... and inferior drivers that are unable to maintain a flawless sonic reproduction, leading to Muddy Blending from micro-distortions).
You can find them fairly cheap, used. Or.. you can buy from one of the former employees, that sells them hand-made.. at his website: "humanspeakers" dot com
This attempt at a wide field, might be workable for concert PA speakers... but on a small scale, I hear virtually no differences. Furthermore... you dont typically need a +30 foot sound field in a standard apartment / living room. A lot of people are listening in much smaller room sizes.
BOSE used a trick to get an extended sound field.. using dirt cheap low-fidelity drivers, by using multiple drivers facing different directions. I personally despise Bose, as their drivers are Inferior Garbage, that have no business representing themselves as "Audiophile" grade speakers.
Interestingly enough... another very wide stereo soundstage speaker that I experienced when I was like 15 yrs old (Im 50 now), was at a friends house. I asked him where the sound was coming from... and he pointed to these Octagon shaped coffee tables. I believe the top-most area of each octagonal side... had the vents for sound to exit through them... and the sounds filled the entire room, in a very 3D way. I dont know what the make / model was... and or what drivers they used in them. All I can say, is that they really dont make speakers anywhere near as good as the late 70s era speakers. These low power ECO drivers.. dont have enough cone control, and exhibit a lot of micro-distortions, as well as have a very poor stereo imaging spread.
And while many praise expensive Planer drivers for their sonic accuracy... they seem to have some of the most Directional of sounds, with the most NARROW of sound-fields. You have to sit in the perfect position, else the stereo effect drops off Dramatically. I could never deal with that. Id rather have the wide spread of my EPIs, with "slightly" less fidelity... to get a much better overall sonic 3d experience.
Super interesting thank you
You gotta hear polk's SDA speakers if theyre still set up at the Mossimo HQ in California. Crosstalk cancellation gives a freakishly wide stereo image
Always innovating. Love it.
🤙🤙 Anthony!
Did you ever get around to making a kit? Love your experiences mr rat
Working on it. Got vust but it's in my list of adventures
Does the Bose Wave Radio use a similar tech?
I'm not sure I know they have some cool tech for below frequency and getting lots of low and out of a small enclosure.
A lot of people don't like bose but I personally very much appreciate their tech and the fact they're owned by MIT gifted to them.
For a while they were very kind of overpriced cheapish consumer but they've come around and they're technology is always been very innovative
Hey Dave, long time member (had to create a new account).. kind of off topic question but the subject of polarity reminded me.. I recently watched a video of someone reversing the polarity of their overheads because the sound that hits the mic after the drum is hit is a negative pressure wave in his eyes.. the positive pressure wave goes to the floor (under a drum kit).
What are your thoughts on this?
Based on his logic, you would think that every mic on a drumkit should be reverse polarity except the snare side mic
A lot to unpack there. Many people, (I believe incorrectly) believe absolute polarity matters.
I have never seen or heard anything that proves it matters whether the initial pressure wave is positive or negative
That said, what does matter is the polarity of combinations of signals, the more similar the signals, the more polarity matters.
Two identical signals summed together out of polarity will cancel completely
I do believe polarity of overhead mics can matter.
Not so much for the cymbals, but for the other sounds that the mics pick up
If your floor Tom is the same distance from your ride symbol mic as you're overhead is from the floor tom, then the floor Tom is going to bleed into both the overhead and the ride symbol mic.
When the overhead mic and ride symbol mic are mixed together, those two versions of the floor tom will either sum together or cancel out to some degree.
If the overhead mic is in polarity with the right symbol mic, then the low end from the floor Tom will be additive in both of those yet delayed in time compared to the sound picked up from the floor tom mic.
By reversing polarity of the ride symbol mic, some of that floor tom low frequency from the combined ride cymbal and overhead mic can be canceled out, possibly cleaning up the four tom sound.
I did some videos on stage mic polarity if you haven't seen them you should check them out because they cover this subject
Very neat stuff indeed!
👍🤙👍
Didn’t Polk Audio try to do something like this back in the 80s?
This is a fairly simple wiring and there have been many dives into this over the years. Hafler, Eno and others have done versions of various wiring to create surround or augmented stereo from stereo signals. Often putting speakers in the rear of the room
Just about every TV sound bar does some sort of electronic version that usually sounds like phasey crap.
I wanted to specifically focus on a simple mid-side implementation such that a single speaker can would offer clear wide stereo without that phasey sound of a sound bar.
I have some other cool wiring methods and added a little subwoofer speaker to the box as well.
Are you going to be at Infocomm? thanks for all your interesting videos
Yes headed to infocom Thurs
@@DaveRat I was there as well. Would have liked to have met you. Thanks for your videos and response
Yes I'll try and be better about posting when I go to trade shows and I like to meet fellow sound humans!
This is amazing. Thanks for sharing
👍🤙👍
Fhis is absolutely awesome!!!!
I love mid side tricks, my favorite way to mess with audio. This had me click real fast
Awesome!!
this man is such a damn legend..
🙈🙈
This video has inspired me to ask you to do a deep dive into the spatial/immersive PA system world
I will ponder that.
For veterans like Rat, it must be a little annoying that it's called immersive sound. It's just as much a marketing issue as it is a processing issue, a lot!. I think that before taking the step towards such systems, sound engineers should be more concerned with acoustics. If you've mastered acoustics, it's the next level.
Ha, yeah, The adventure of surrounding people with sound is a challenging one because any energy spent on pulling focus away from me band or artist strictly has less than ideal results.
Money and time and technology spent on increasing sound system complexity so that various sounds can be put in various places for a period of radiate from various locations is often better spent on improving the quality of a simple system to cover the venue better with a higher quality sound.
That is not to say that systems like L-Isa Don't have value and offer assets, but they can easily be more distraction and detrimental to the sound.
My biggest reservation about L-isa, is been having multiple small clusters may offer flexibility and where the sound radiates from but it lacks the clarity and power of 2 giant long line arrays. In order to get an L-isa system to match the power and clarity of a stereo rig and also offer the advantages radiating sounds from various locations, pretty much need the entire stereo rig you would have had anyways and they need to add a bunch of smaller clusters as well which is an expensive and considerable undertaking.
This in some parts reminds me of the Hafler-matrix fake quad stuff.
🤙👍🤙
anyone have any experience with the space station center point stereo? I really want a setup like that to get the leslie sound without need the leslie.
I saw several comments to this video where people mention the space station that may be able to help
@@DaveRat have you ever had any interest in leslie speakers? they sound unlike any other system I’ve heard and I’ve to re-create the magic of one in person
Yes they are very cool and I can't remember which band I was mixing but there was a song that needed a Leslie sound and I did a bunch of research and at the time ended up with a Dynacord CLS222 Leslie simulator as the best solution for the situation.
Not the real thing but it was close enough for the live mix. If I was on the musician side of things, having an actual real world Leslie would be very cool.
I've mic'ed a bunch of them in various bands, One of the aspects that's interesting is that The physical diameter of the rotation of the horns or woofer output is important. You can't get a lot of Leslie sound if the physical size of the Leslie is not big enough.
Bigger is better and it can't really be shrunk if you want to get that full Doppler effect
could this be used for like a computer surround sound bar?
Absolutely and it would work great
Lovely options. 👍
👍🤙👍
Cool Dave, very nice idee.
👍🤙👍
that is really interesting! thank you for sharing this!
Super cool and thank you
I have for decades espoused a mono+ambient approach to 'stereo' listening, based on breaking the sweet-spot paradigm. in the mid 80s I worked on a tv soap that switched to stereo, & we tried quite hard to get the broadcaster- the whole national network- to adopt M-S instead of left-right stereo.
imagine you are shooting a two-shot at a table, one actor either side of the centre line. you pan them left & right. then you cut to the reverse angle, & pan them the opposite way.
distracting, isn't it? which is why we don't do it in the cinema.
in fact, we recorded everything M-S anyway, with a small schoeps figure-8 fastened to the usual 416 or 816. [on location, individual radio mics went on a mono track, & we'd have to fake the boom into stereo because the VTRs didn't have enough tracks.]
the idea was that you'd "put the viewer in the room with the actors" using the ambience, but keep all the main dialogue in the centre, where the screen is. pretty much what happened in most cinemas too, at least until surround caught on.
so we lost that battle. but there's no reason why an m-s matrix couldn't be installed on most hifi systems, to drive a centre speaker & something like a small array of surround speakers dotted about the place & with some wired out-of-polarity.
kind of like hafler in reverse..... remember that brian eno diagram on one of his ambient albums? the same trick that a lot of hifi companies used for faking quad, back when that was the thing.
Super cool and yes you get it! And this has been implemented in many different ways over the years by different people and companies. And I thought this cool simple speaker that allows anyone to build it themselves would be a fun project and a chance to help explain MS recording
Isnt this just the rear channel in a 5 channel surround setup?
Typically 5.1 or at least true 5.1 is actually mixed as separate content so there are six channels of audio with the 0.1 being the subwoofer and five separate audio mixes for the other five channels.
This setup extracts all the stereo information and send it to separate speakers that only reproduce the stereo information and none of the mono information.
Then that stereo information is added back in with the left stereo information added back in in polarity to the left speaker and the right stereo information added back in in polarity to the right speaker and the left stereo information added back in out of polarity to the right speaker and the right stereo information added back in out of polarity to the left
And all this is done with very simple wiring and a stereo amp. No artificial processing or crap added to the sound