SDA Crossover: Theory and Technical Questions

staticGenerator

Hello,

My father purchased a pair of SDA 2 speakers in the mid nineties. I remember listening to many old recordings again and rediscovering them, hearing new intricacies. They were amazing, and it's a shame to have only had them in production for such a short period of time.

I wanted to recreate a similar experience in a smaller package designing my own crossover. So I was glad to stumble across this website. I have rudimentary electronics knowledge and loudspeaker design experience. The design of the SDA crossovers appear to be amazingly simple for what they do. Nevertheless, I'm having some problems understanding the function of the circuitry, and I'm hoping some more technical folks can offer some guidance.

Looking at the schematic for the CRS+ here's what I have determined. For the sake of example, let's decide that this circuit is connected to the right channel output on the amplifier: http://www.polkaudio.com/forums/attachment.php?attachmentid=26709&d=1186358580

Here are the conclusions I have made looking at this circuit:

1. The positive connection for the woofer, tweeter and dimensional driver are all wired (including their filtering components) to the positive connector of the positive input for the right channel.
2. The negative connection for the woofer and tweeter of the circuit only are wired (including filtering components) to the negative connector of the negative input for the right channel.
3. The negative connection for the dimensional driver in the right channel connects to the interconnect cable (IC) connection. The IC then connects to the negative connector of the dimensional driver in the left channel so that the all dimensional drivers in a L/R pair are directly connected.
4. The only path I can see by which the common IC connection brings the L/R dimensional drivers to the negative input on the crossover input is through the 16mH coil on the left side of the schematic.

Questions:

1. How does the dimensional driver in the right crossover circuit act to cancel out the the woofer in the left? Since the positive connection in the right dimensional driver is connected to the positive input in the right channel, it seems there is no path by which the positive input of the left circuit can get to the dimensional driver in the right channel crossover circuit.
2. The dimensional driver in the right circuit appears to be wired in-phase. In order for the right dimensional driver to cancel out the signal from the left woofer arriving at the right ear, my understanding is that the right dimensional driver would need to be wired out of phase. I assume therefore that woofer MW6510 and MW6511 are identical, but what would be the negative connector on one driver is actually labeled as the positive connection to achieve the out of phase cancellation of the left woofer?

Thanks in advance for your help on this. I know there could be additional posts that go into the specifics of the SDA technology, but I wasn't able to locate them. Feel free to pass those along if you have the link.

Schurkey

staticGenerator wrote: »

My father purchased a pair of SDA 2 speakers in the mid nineties. I remember listening to many old recordings again and rediscovering them, hearing new intricacies. They were amazing, and it's a shame to have only had them in production for such a short period of time.

I wanted to recreate a similar experience in a smaller package designing my own crossover.

Consider buying a Carver C9 processor for ~$100 and use "ordinary" high-quality speakers and amplification. Some will say the Carver electronic-processing solution is less pleasing than the Polk solution; but it'd be worth your time to give it a try.

staticGenerator wrote: »

Questions:

1. How does the dimensional driver in the right crossover circuit act to cancel out the the woofer in the left? Since the positive connection in the right dimensional driver is connected to the positive input in the right channel, it seems there is no path by which the positive input of the left circuit can get to the dimensional driver in the right channel crossover circuit.

In THIS GENERATION of SDA crossover, the ground side of the dimensional drivers are tied together through a filter network, and of course the SDA interconnect "pin" terminals. Whichever dimensional driver has more voltage (in the appropriate frequency band) at any given instant supplies voltage/current to the negative side of the dimensional driver in the other speaker cabinet. Since the voltage/current is applied to the negative terminal, it in in effect "reverse polarity", causing cancellation of the sound from the other speaker cabinet. Physical location of the dimensional driver provides the timing of this reverse-polarity signal so that it hits our ears at the proper time.

The dimensional driver ALSO is tied to the + terminal for that channel. The filter network allows the dimensional driver to work IN PHASE with the stereo driver in that same cabinet in the bass region, reinforcing the bass from the stereo driver, so that both can power the passive radiator. However, tying the negative terminals of the dimensional drivers together allows the dimensional driver to act independently of the stereo driver in that same cabinet to produce the out-of-phase crosstalk-cancellation signal above the deep-bass frequency region.

staticGenerator wrote: »

2. The dimensional driver in the right circuit appears to be wired in-phase. In order for the right dimensional driver to cancel out the signal from the left woofer arriving at the right ear, my understanding is that the right dimensional driver would need to be wired out of phase.

See above.

Be aware that earlier generations of SDA speakers used a different crossover system to achieve the SDA dimensional signal.

staticGenerator wrote: »

I assume therefore that woofer MW6510 and MW6511 are identical, but what would be the negative connector on one driver is actually labeled as the positive connection to achieve the out of phase cancellation of the left woofer?

Nope. '10 and '11 drivers are similar looking, but very different electrically/mechanically.

dorokusai wrote: »

These are the only drivers, aside from the RD tweeters, being manufactured to support vintage products, such as, Monitor, RTA and SDA. You should print this as a reference and keep it handy for the values it contains. There are no grill parts, cables, crossovers, plates or anything else available for the older loudspeakers, unless you know the Polk G O D.

1 Driver Type Q BL Compliance DC Resistance Fs

3 MW 6501 1.47 5.96 N 3.25 E - 3N/M 7.75 Ohms 31 Hertz
4 MW 6502 1.97 5.16 N 1.47 E - 3N/M 3.52 Ohms 44 Hertz
5 MW 6503 1.82 5.24 N 3.55 E - 3N/M 6.54 Ohms 29 Hertz
6 MW 6509 2.1 4.57 N 3.54 E - 3N/M 8.97 Ohms 30 Hertz
7 MW 6510 1.39 6.32 N 3.41 E - 3N/M 6.57 Ohms 31 Hertz
8 MW 6511 1.405 4.34 N 3.496 E - 3N/M 3.13 Ohms 29 Hertz
9 MW 6512 1.905 4.99 N 1.588 E - 3N/M 3.466 Ohms 40 Hertz

staticGenerator

Thank you Schurkey, there is a lot more going on here. I will also consider the Carver signal processor as you suggested.

staticGenerator

Ultimately, I believe the goal would be to have the dimensional driver in the right speaker offset the signal in the left speaker WITHOUT canceling out common material being fed to both two drivers. Correct? There will be other ways to achieve this goal, so I'm going to consider other possible ways to make this happen (e.g. dual coil drivers in the dimensional driver) and it will be fun to experiment as well.

Schurkey

The simplest (but least-sophisticated) way to experiment with the SDA process is to find a couple of small "bookshelf" speakers that have poor deep-bass, and then place them to the outside of your main speakers. Maybe even throw a towel over the tweeters of the bookshelf speakers, so they're "all midrange".

Put the bookshelf speakers at approximately ear-height, or tweeter-height if that seems to match best with your main speakers. If the main speakers are toed-in to the listening position, you may have to move the bookshelf speakers slightly to the rear, as well as to the outside of the main speakers.

Wire the left bookshelf speaker to the right channel, in reverse polarity. Wire the right bookshelf speaker to the left channel, also in reverse polarity. Use a separate volume control for the bookshelf speakers, so you can tune both the volume level of the SDA signal, and the timing (by playing with the position of the bookshelf speakers in relation to the main speakers.) I would achieve the separate volume control by using a completely different receiver tied to the same source: use your regular system, then add a second receiver, speaker wire set, and the bookshelf speakers, all tied to the same CD player or music server or whatever with a second set of interconnects.

Sit back and enjoy.

It'd be worth your time to read the "white paper" on SDA process supplied by Polk when the series was introduced. I can't find a link to it right now.

staticGenerator

Schurkey wrote: »

The simplest (but least-sophisticated) way to experiment with the SDA process is to find a couple of small "bookshelf" speakers that have poor deep-bass, and then place them to the outside of your main speakers. Maybe even throw a towel over the tweeters of the bookshelf speakers, so they're "all midrange".

That is a simple and easy way to experiment! I'll give that a try.

I found the link to the SDA white paper from the main site by doing search for "SDA technical"

Thanks again.

PolkieMan

So that sounds easy to do and something that most of us have around.
I may try that with set of monitor 4a and an old carver tfm15, thanks.

Schurkey

A bit of a revision on the "separate volume control" which I ran out of edit time to accomplish earlier.

ASSUMING YOU HAVE APPROPRIATE OUTPUT JACKS (or "wye" interconnect cables) you would connect your "regular" system just as it is. IF you have a set of "preamp output" jacks handy, connect your "SDA" receiver/integrated amplifier to the "preamp out" jacks of your main system. Now both the main system and the SDA system increase and decrease in volume with the single volume control of your main system--but--you can use the volume control of the SDA receiver/integrated amp to adjust the level of the SDA speakers independently. I have heard--but not verified--that the SDA drivers should run approximately 1/2 volume of the main midrange drivers. You don't need treble from the SDA speakers, and you don't want much in the way of bass, either. ~100 hz to ~2000 hz is the target.

Ideally, the small "SDA" speakers would be voiced similarly to the main speakers. You'll have fun with differences in speaker sensitivity, amplifier gain, and a hundred other variations between the "main" system and whatever you're using for the "SDA" system...but this is an experiment, not a permanent listening set-up (Well, it's most likely temporary, anyway.)

I also said to put them at tweeter-height; upon further reflection they should probably be at the same approximate height as the midrange driver of you main system. Experiment with placement, see what happens.

I'm guessing that this is exactly what Matthew Himself started with. We have the benefit of seeing how his design matured by examining the series of generational improvements of his commercial products. The first thing to notice is that by putting all the drivers in one cabinet; and by making all the mid-woofers similar if not identical, and then the ability to drive 'em all with a single stereo amp, voicing problems and amplifier gain problems are all eliminated.







Truthfully...as inexpensive as used SDA speakers are, best bet is to simply buy the best used SDAs you can afford, and skip all the experimenting. The big problem here is finding a set within reasonable transportation distance--or having pockets deep enough to provide proper shipping for speakers out of your traveling comfort-zone.

If you ship speakers, (ANY speakers) have the seller connect the + and - terminals on each cabinet with a piece of wire before boxing them up. Connecting the terminals dramatically reduces driver motion as the speakers are bumped around in shipment.

leftwinger57

I might be off base here but this smacks of the Grateful Dead's Wall Of Sound. In effect the 1st line array system ever made like you see all over the place hanging from come-alongs and angled for best dispersion. About the speakers of SDA being out of phase to cancell out one side effectively the same thing was done w/ the microphone system the Dead used w/ the Wall.It was 2 conderser mics of German make I believe called Bruel&Kjaer and were mounted on a mic stand spaced apart w/ the idea being that they were out of phase were the top mic you sang into and the bottom mic picked up all the other sounds but cancelling them out so there was no need for stage monitors which were always a problem or an on stage mixer. On the mic box had a volume control so Jerry and the other singers could make their voices lower or louder as they pleased. This has the same theory way before Polk's time but he found a way to condense it for home audio use. Smart guy. But just like the Wall was an expense the Dead could not afford so it seems also the general puplic buying SDAs especially the big boys. Maybe if he kept it to a 2B like size the technology would have continued.

staticGenerator

nbrowser wrote: »

Interesting tip there, I'm assuming it helps to cancel out vibration caused movement by dead shorting the voicecoils ?

If you've ever had the chance to try and turn the crank on a hand crank generator when the leads are shorted, you'll know that it's nearly impossible to turn - same principle.

westmassguy

staticGenerator wrote: »

If you've ever had the chance to try and turn the crank on a hand crank generator when the leads are shorted, you'll know that it's nearly impossible to turn - same principle.

I'd be more worried about magnet shift, than cone damage.

PolkieMan

That's vey true. But still that piece of wire would be easy to implement for extra insurance.

westmassguy wrote: »

I'd be more worried about magnet shift, than cone damage.