Advanced tweaking of Polk SDA's...

daifanshi

I recently bought a pair of 2.3TL's locally from craigslist for a very reasonable price. It's a very good speaker by today's standards, but not superior to a lot of modern speakers. It still is basically a 20 year old design.

I have a background in DIY speakers and I think the design allows for some interesting experimentation.

I have a few ideas I'd like to throw out there.

1. Active crossover! I've read here that many people are willing to spend 300+ dollars on newer crossover components! The biamp terminals give us the opportunity to eliminate the large-signal crossover and their inherent complexity.

Before we decide to bypass the in-speaker crossover completely, we can apply it to some known problems with certain Polk designs. The often maligned 12kHz resonance of the SL2000 tweeter can be easily tamed with an active filter before the main amplifier input without a complex in-speaker crossover redesign. Of course a real active crossover would eliminate all the perceived problems that some people believe $300 worth of capacitors will fix. This also has the added benefit of allowing interesting substitutions of newer (and superior) drivers.

2. Substitution of passive radiator with a tuned port. Those who have used passive radiators in the past probably have encountered problems with excursion limits of passive radiators in some speakers. It would be a simple exercise to replace the passive radiator with a tuned port to get around this problem. I haven't seen any problems with the Polk designs, but maybe somebody out there has a broken passive radiator on an old Polk SDA and needs help. And it would be interesting to see how well it works. This is probably the one I would try first.

ben62670

First welcome to club Polk.
Second your ****:eek: If you kill the passive XO in the highs it would totally throw of the many phase issues of multiple dome tweeters in an array. Third you can't through the low pass out completely because you would kill the SDA effect. Forth the PR's work great in these speakers.
Ben

daifanshi

ben62670 wrote: »

First welcome to club Polk.
Second your ****:eek: If you kill the passive XO in the highs it would totally throw of the many phase issues of multiple dome tweeters in an array. Third you can't through the low pass out completely because you would kill the SDA effect. Forth the PR's work great in these speakers.
Ben

A hybrid active/passive approach for the tweeter array is certainly possible and a decent compromise. It would still be superior to a pure passive for the tweeter array.

Like I mentioned earlier, I hope my passive radiators never get kicked in. If I ever need a replacement, changing the length of a tuned port is a LOT easier than changing the mass-loading of a different PR to make it work right.

ben62670

daifanshi wrote: »

A hybrid active/passive approach for the tweeter array is certainly possible and a decent compromise.

Been there done that. Worked very nicely too. I did an spl sweep. Figured out the true XO point on the mids, and dialed in the highs. I tried 4 tweeters, 3 tweeters, 2 tweeters, and 1. The two tweeters worked the best.

I built these.
http://www.polkaudio.com/forums/showthread.php?t=53273
I used a fork truck to move them around:D Did many experiments with the XO, and I am completely satisfied. Well after I go back to the two Morel tweeters per cabinet.

Have fun, and stick around. Plenty of good folks here.
Ben

Face

Just upgrade the tweeters to the RD0-198's and install new caps, you'll be set.

As Ben mentioned, I can't see how you could go active without killing the SDA effect.

And where are you keeping these speakers that you have to worry about the PR being kicked in? There are aftermarket PR available, plus, OEM PR's pop up on Ebay time to time.

danger boy

i'm with Ben62670, nearly everything under the sun has been tried on the SDA's without much success in improving on their performance. (0utside of the usual caps/resitor and RD0 tweeter upgrades)

try it out on your own 2.3TL's and report back to us here with your findings on a tuned port for the SDA's.

They are your SDA's, you can monkey with them all you want.. but once you build a tuned port for them.. there is no going back. Keep that in mind.

As for the 13kHz spike in the SL2000 tweeter.. you and Candyliquor35 need to get together and discuss this item at length.

daifanshi

danger boy wrote: »

i'm with Ben62670, nearly everything under the sun has been tried on the SDA's without much success in improving on their performance.

try it out on your own 2.3TL's and report back to us here with your findings on a tuned port for the SDA's.

They are your SDA's, you can monkey with them all you want.. but once you build a tuned port for them.. there is no going back. Keep that in mind.

As for the 13kHz spike in the SL2000 tweeter.. you and Candyliquor35 need to get together and discuss this item at length.

I think the SDA effect can be improved by implementing an adjustable group delay to the SDA signal. Instead of relying on the physical separation of the dimensional and stereo drivers, it can be tweaked and optimized with a tweakable all-pass active network.

Regarding the tuned port. Are you saying the change is in some way permanent? How so? I can remove the PR and substitute an appropiately sized piece of plywood with the tuned port mounted to that.

ben62670

The RDO tweeters have no nasty spike at 12k. Big improvement over the SL2000 ear stabbers.

F1nut

You have no idea what you are talking about. Your 2.3TL speakers don't even use the SL2000's. The SDA's that do use that tweeter do not need a complex in-speaker crossover redesign. Just simply replace the SL2000 with the RD0194-1, problem solved.

Doped paper drivers are still the most natural sounding drivers made. There's not a damn thing wrong with the MWxxxx Polk drivers.

Ports better than PR's, yeah right. If you're having excursion issues with the 2.3TL's, you need a better amp.

Here's an idea. Instead of **** up a perfectly good pair of SDA's, sell them and buy something else to experiment on.

daifanshi

Face wrote: »

Just upgrade the tweeters to the RD0-198's and install new caps, you'll be set.

As Ben mentioned, I can't see how you could go active without killing the SDA effect.

And where are you keeping these speakers that you have to worry about the PR being kicked in? There are aftermarket PR available, plus, OEM PR's pop up on Ebay time to time.

I looked at the schematic for the dimensional drive signal and it looks like just a lowpassed version of the opposite side signal. The separation of the dimensional drivers and stereo drivers look to give the correct phase shift. This is really easy to do with an active crossover. You'd need a third pair of amplifiers, but three way actives aren't unusual at all.

From what I can tell the Polk OEM PR's are not available from Polk. They'll get expensive at some point. Have you ever tried to locate and pay for a replacement for a fried Inifinity EMIM midrange? Polk Audio can't support 25 year old speakers forever...

daifanshi

F1nut wrote: »

You have no idea what you are talking about. Your 2.3TL speakers don't even use the SL2000's. The SDA's that do use that tweeter do not need a complex in-speaker crossover redesign. Just simply replace the SL2000 with the RD0194-1, problem solved.

Doped paper drivers are still the most natural sounding drivers made. There's not a damn thing wrong with the MWxxxx Polk drivers.

Ports better than PR's, yeah right. If you're having excursion issues with the 2.3TL's, you need a better amp.

Here's an idea. Instead of **** up a perfectly good pair of SDA's, sell them and buy something else to experiment on.

I thought I was clear that the 2.3TL didn't have excursion problems. But I know for a fact some designs do. I didn't specify any specific speakers.

Regarding the SL2000's. The 2.3TL's have SL3000's for sure. It was an example of problem that can be easily fixed with an active filter.

I detect some hostility.

ben62670

daifanshi wrote: »

I think the SDA effect can be improved by implementing an adjustable group delay to the SDA signal. Instead of relying on the physical separation of the dimensional and stereo drivers, it can be tweaked and optimized with a tweakable all-pass active network.

You mean Carver Holography. You sound like a young me. I was playing with that idea a few years ago. Not worth it. SDA's have a much wider sweet spot than can be implemented via active components. Not trying to discourage you, but after I have played with many of the same ideas I can safely say that as mentioned above that new caps, resistors, tweeters, and the two large inductors being replaced that gains from active manipulation would be mute.
Ben

daifanshi

ben62670 wrote: »

You mean Carver Holography. You sound like a young me. I was playing with that idea a few years ago. Not worth it. SDA's have a much wider sweet spot than can be implemented via active components. Not trying to discourage you, but after I have played with many of the same ideas I can safely say that as mentioned above that new caps, resistors, tweeters, and the two large inductors being replaced that gains from active manipulation would be mute.
Ben

Ben,

Thanks for the insight. I appreciate it. It's nice not to have to defend what I posted for a change...

But I hope you don't mind if I find out "the hard way" just for curiosities sake? I'm afraid I one of those that tweaks more than listens...

Thanks again.

Face

I'm all for tweaking too, but you're talking heresy.

ben62670

daifanshi wrote: »

Ben,

Thanks for the insight. I appreciate it. It's nice not to have to defend what I posted for a change...

But I hope you don't mind if I find out "the hard way" just for curiosities sake? I'm afraid I one of those that tweaks more than listens...

Thanks again.

No I don't mind. Rock on. I think it is great to have an open mind. I spent so many hours (easily a 1000) studying, and manipulating SDA's its not funny, but when it all comes down to a couple simple mods will transform those speakers into something you would be proud to show to any serious HiFi guy. As mentioned you need a real powerful amp to drive them. I would consider 200watts per channel minimum for those other wise you will never truly appreciate what they can do. If you have SL3000 tweeters I would suggest upgrading the XO first.
Ben

DarqueKnight

daifanshi wrote: »

I recently bought a pair of 2.3TL's locally from craigslist for a very reasonable price. It's a very good speaker by today's standards, but not superior to a lot of modern speakers.

Congrats on your purchase. What is it about the sound of your 2.3TL's that you do not like? In other words, what specific performance parameters are you trying to change or improve?

Messing around with a perfectly good pair of speakers without clear performance improvement goals is like going to a plastic surgeon without a clear idea of the results you want to achieve. People who do so often end up looking like Michael Jackson. The SDA analogy to this is the botched up "FrankenPolks" we see from time to time on eBay, Audiogon, and Craiglist.

daifanshi wrote: »

It still is basically a 20 year old design.

So? Most of today's amps, preamps, and speakers are variations on 20, 30, or 40 year old designs.

daifanshi wrote: »

I have a background in DIY speakers and I think the design allows for some interesting experimentation.

During the SDA glory days in the 1980's, you could call Polk and speak to one of their engineers about proposed modifications. I always found them to be quite enthusiastic about suggesting modifications to improve performance. Such advice was helpful in avoiding unfortunate mistakes and wasted time. You might try sending your proposed tweaks to Polk's customer service department and ask that they be forwarded to engineering.

daifanshi wrote: »

The often maligned 12kHz resonance of the SL2000 tweeter can be easily tamed with an active filter before the main amplifier input without a complex in-speaker crossover redesign.

Have you actually heard a pair of SDA's with SL2000 tweeters?

daifanshi wrote: »

Substitution of passive radiator with a tuned port.

Sealed box and passive radiator designs set the standards for low amounts of bass distortion and high bass accuracy and speed. What advantage(s) do you expect from using a tuned port in SDA's?

daifanshi wrote: »

I think the SDA effect can be improved by implementing an adjustable group delay to the SDA signal. Instead of relying on the physical separation of the dimensional and stereo drivers, it can be tweaked and optimized with a tweakable all-pass active network.

Matthew Polk and others at Polk Audio have done extensive research in this area. A good starting point would be to read what has already been done. Matt has shown a willingness to engage in dialog about SDA speaker design. Again, Polk's customer service can forward your inquiry to him.

Since you have 2.3TL's, why so much interest in the SL2000 tweeter?

daifanshi

DarqueKnight wrote: »

Congrats on your purchase. What is it about the sound of your 2.3TL's that you do not like? In other words, what specific performance parameters are you trying to change or improve?

I found out a long time ago that active filters are closest to the ideal for crossovers in speakers when I started for myself in the 1990's. The big problem with passive crossovers in the eletrical interaction between BP, LP, and HP sections that make tuning difficult and time consuming. Especially if you don't have a lot of test equipment. The odd loads presented by passive crossovers are sometime difficult for amplifiers to handle. Having each speaker (or speaker array) driven by a single amplifier without inductors, caps, and resistors in the way presents a very easy load and lets your amplifiers work were they are more comfortable and with maximum dynamic range. Also crossovers are very lossy and rob amplifier power.

DarqueKnight wrote: »

Messing around with a perfectly good pair of speakers without clear performance improvement goals is like going to a plastic surgeon without a clear idea of the results you want to achieve. People who do so often end up looking like Michael Jackson. The SDA analogy to this is the botched up "FrankenPolks" we see from time to time on eBay, Audiogon, and Craiglist.

Active crossover do make an improvement! Without a doubt. Obviously you won't take my word for it. I believe if you heard a properly executed active speaker system you'd be convinced. The problem is that there are very few of them because of cost reasons and marketability. For the dedicated hobbyist it's not so much a problem. My 2.3TL's are old speakers, but I see a lot of potential in them and for the price I paid it allows me to experiment on them without needing to buy new drivers and build new cabinets. Also, since these are my speakers, I should be allowed to do with them as I wish.

DarqueKnight wrote: »

So? Most of today's amps, preamps, and speakers are variations on 20, 30, or 40 year old designs.

These is a specious argument. Automobiles are based on 100 year old designs. Are you claiming cars from the 1980's are better than those today?
There are always going to be incremental improvements in performance over time in any consumer technology because of better materials and techniques.

DarqueKnight wrote: »

During the SDA glory days in the 1980's, you could call Polk and speak to one of their engineers about proposed modifications. I always found them to be quite enthusiastic about suggesting modifications to improve performance. Such advice was helpful in avoiding unfortunate mistakes and wasted time. You might try sending your proposed tweaks to Polk's customer service department and ask that they be forwarded to engineering.

I may do that. CS people are usually a crapshoot, but perhaps Polk is different.

DarqueKnight wrote: »

Have you actually heard a pair of SDA's with SL2000 tweeters?

Yes. But is was many years ago. I don't recall hearing a problem. This is why I am seeking to obtain additional information on the problem beyond just the folklore.

DarqueKnight wrote: »

Sealed box and passive radiator designs set the standards for low amounts of bass distortion and high bass accuracy and speed. What advantage(s) do you expect from using a tuned port in SDA's?

Passive radiators and tuned ports serve EXACTLY the same purpose in a speaker box and should be able to produce equivalent results. Each has it's advantages and disadvantages. For the hobbyist, tuned ports are easy to adjust and tweak because there are really only two degrees of freedom, port diameter and length. And they're cheap as dirt! Passive radiators present difficulties for the hobbyist, since changing the mass loading is all he/she can really do to adjust them (and that has it's limits). Compliance and physical size can't be changed without more difficulty.

Recently I read that the MW6510 wasn't available as a replacement until recently and the MW6503 with different TS parameters was suggested as a replacement. Obviously this kind of change would require changes in the "tuned" elements for the box for ideal performance. Most people would not go through the trouble. But some such as myself would.

DarqueKnight wrote: »

Matthew Polk and others at Polk Audio have done extensive research in this area. A good starting point would be to read what has already been done. Matt has shown a willingness to engage in dialog about SDA speaker design. Again, Polk's customer service can forward your inquiry to him.

I agree, good idea.

DarqueKnight wrote: »

Since you have 2.3TL's, why so much interest in the SL2000 tweeter?

This has been brought up before. I was seeking a simple application of an active network that could fix response problems as a starting point. The SL2000 problem is something that I believe is easy to fix with a notch filter at line level without changing the actual speaker system. I like to mess with this stuff and if someone else can benefit more directly from work I do, then it's worth it for me. I'm sure most people are like that.

cheers.

geoff727

daifanshi wrote: »

The SL2000 problem is something that I believe is easy to fix with a notch filter at line level without changing the actual speaker system.
cheers.

Just a quick, dumb question here....
Rather than using a notch filter to correct a frequency response 'deviation', why not just eliminate the added components of the notch filter and go with a tweeter that has a flatter response to begin with (i.e., the RD-0194/0198 replacements for the SL2000/3000)?

Thanks.

Face

There's more wrong with that tweeter than the peak.

heiney9

Concetrate on a few areas and the SDA's will reward you in spades.

1.) Replace the older tri-laminate tweeter with the appropriate silk replacement RD0198-1

2.) Upgrade/replace all caps and resistors on the x-over board. Deleting bypass caps and polyswitches.

3.) set them up properly in a proper room for maximum performance

4.) use high quality (not nec. quantity) amplification. Although high current high power can achieve superior results if in a large listening room.

5.) use a great source component and great source material.

Over the years many people have tried to improve upon SDA's by trying to add notch filters to tame the older tweeter or active x-overs for whatever reason. Nothing works as well as following the 5 steps above; especially 3, 4 and 5.

H9

geoff727

Pardon me, I probably should have just referred to the SL-2000, not the -3000, when talking about the frequency response deviation.

heiney9

daifanshi wrote: »

I looked at the schematic for the dimensional drive signal and it looks like just a lowpassed version of the opposite side signal. The separation of the dimensional drivers and stereo drivers look to give the correct phase shift. This is really easy to do with an active crossover. You'd need a third pair of amplifiers, but three way actives aren't unusual at all.

From what I can tell the Polk OEM PR's are not available from Polk. They'll get expensive at some point. Have you ever tried to locate and pay for a replacement for a fried Inifinity EMIM midrange? Polk Audio can't support 25 year old speakers forever...

Have you read the SDA "white" paper published by Matthew Polk? It is published in the June 1984 issue of Audio Magazine. It will give a good understanding of why the crosstalk cancellation was implemented as a passive system. Just looking at a x-over isn't going to give you much about the theory of SDA.

Before you even discuss this you should read the white paper.

Also go to POLKSDA.COM and read the Stereo Review article on the 1C's and SRS's and you you'll see that they measured the lowest bass distortion they ever recorded with the Polk passive radiators.

The PR's are better than any ported system I've heard including the LSi series.

H9

DarqueKnight

My question:

DarqueKnight wrote: »

What is it about the sound of your 2.3TL's that you do not like? In other words, what specific performance parameters are you trying to change or improve?

Your answer:

daifanshi wrote: »

I found out a long time ago that active filters are closest to the ideal for crossovers in speakers when I started for myself in the 1990's. The big problem with passive crossovers in the eletrical interaction between BP, LP, and HP sections that make tuning difficult and time consuming. Especially if you don't have a lot of test equipment. The odd loads presented by passive crossovers are sometime difficult for amplifiers to handle. Having each speaker (or speaker array) driven by a single amplifier without inductors, caps, and resistors in the way presents a very easy load and lets your amplifiers work were they are more comfortable and with maximum dynamic range. Also crossovers are very lossy and rob amplifier power.

I directly asked about what sonic improvements you are trying to achieve. I was expecting a response that had something along the lines of "improved bass response", "better imaging", "smoother and more accurate high frequency response".

My point is that I do not understand undertaking a modification exercise unless it gets you closer to the sound you are after.

Are you more interested in electronics theory rather than music reproduction?

heiney9

Also FWIW, Ben did a lot research and had several indepth talks with Polk's Ken Swauger. Feel free to bounce some idea's off Ken or have him ask specific questions of the engineering guys. Raife (Darqueknight) has talked at length with people at Polk and over the years has put together some very valuable SDA information and analysis.

H9

Face

daifanshi wrote: »

But I hope you don't mind if I find out "the hard way" just for curiosities sake? I'm afraid I one of those that tweaks more than listens...

DarqueKnight wrote: »

Are you more interested in electronics theory rather than music reproduction

It appears so.

DarqueKnight

Oh...OK. Thanks. I missed the "tweaks more than listens" response to Ben. Please disregard my prior responses.

Systems

If you want an active speaker then why not buy some good drivers from Madisound or PE and design your own.With carefully selected drivers and crossover slopes and frequencies etc. you will have a very good chance of achieving excellent results.I believe that active speakers have some real performance advantages and I'm current using a fully active setup myself.However as others have mentioned,SDA's because of their unique design are IMO likely very poor candidates for converting to fully active.

daifanshi

geoff727 wrote: »

Just a quick, dumb question here....
Rather than using a notch filter to correct a frequency response 'deviation', why not just eliminate the added components of the notch filter and go with a tweeter that has a flatter response to begin with (i.e., the RD-0194/0198 replacements for the SL2000/3000)?

Thanks.

50 bucks a tweeter is a lot. If you use moderately price op-amps you can probably do the simple notch filter + PS for less than $30 for both speakers. Something like the OPA2134's are inexpensive and very good. But this isn't right for everybody. Some people can't solder, etc....

There might be other problems with the SL2000 that Ben mentioned earlier. But the offensive resonant peak at 13kHz is what is mentioned most often. I even read about in some old reviews.

Harshness and other qualities are often difficult to quantify. If the SL2000 are as poor as people say they are, I wonder why Polk even used them in the first place...

daifanshi

DarqueKnight wrote: »

My question:




Your answer:




I directly asked about what sonic improvements you are trying to achieve. I was expecting a response that had something along the lines of "improved bass response", "better imaging", "smoother and more accurate high frequency response".

My point is that I do not understand undertaking a modification exercise unless it gets you closer to the sound you are after.

Are you more interested in electronics theory rather than music reproduction?

DarqueKnight,

How about "ALL OF THE ABOVE" for sonic improvements? When you remove the passives from the speaker, each speaker or array is controlled independently by it's own amplifier directly on the terminals. Intermodulation between drivers (eg., distortion in the woofer affecting the tweeters) is non-existent because they are isolated. The load on the amplifiers is minimized to a smaller bandwidth and they are less likely to distort. (lower distortion out of the speaker). We should not ignore how a passive crossover can really screw up an amplifier.

Undesirable effects such time-misalignment can be fixed without resorting to physically moving drivers back and forth. Time aligned speakers like the actives I've heard and complicated passive systems like the Thiels have excellent imaging.

Active systems can actually improve bass response by fixing room dependent bass suck-outs and peaks that can make the bass sound "one-note" or just plain weird.

The smoothness of the highs is what the active notch filter for the SL2000 will accomplish.

All this is real and measureable. At least in the systems that I built.

daifanshi

Face wrote: »

It appears so.

However, I do listen to real music to evaluate the effects of what I do. And I do like what I hear.

NotaSuv

Daifanshi

As Ben stated "Rock On'... great to see those that will go against the flow....against the advice of the forum elders.....its that kind of need to know,gotta try this attitude that is responsible for advancments in many areas. Have fun and keep us informed..