The Tale Of 5 Tweeters - SDA Tweeter Replacement Guide

DarqueKnight

mhardy6647 wrote: »

Do you have any reference tweeters that are different than this cadre to test?

I have tested several copies of each tweeter, including brand new copies of the RD0194 and RD0198, and the 5 kHz dip still appears.

mhardy6647 wrote: »

I wonder if 5-ish kHz dip is endemic to these tweeters or an artifact of the test.

I wondered about that. Moving the microphone closer or farther away did not get rid of the 5 kHz dip. The FR plot below is of my SDA SRS 1.2TLs in the same room. There is a small dip in the response at 5 kHz, but nothing close to what is shown in the tweeter response plots.


Figure 19. Room response of SDA SRS 1.2TLs in tweeter testing room.

The SDA CRS+s in my office at work do show a dip at 3.5 kHz and at 5 kHz.


Figure 20. Room response of SDA CRS+s in office at work.

I did a search on "tweeter dip at 5 kHz" and found that some speakers have a dip at 5 kHz to improve the smoothness and midrange "presence" of the speaker.

@KennethSwauger would you ask Stu if these tweeters are supposed to have a dip at 5 kHz?

DarqueKnight

Near Field Measurements Redone - 5 kHz Dip Gone!

After further study and experimentation, the previous near-field frequency response measurements were redone. The previous measurements were taken at a distance of 9 inches. The new measurements were taken at a distance of 2 inches, with the microphone pointing at the exact center of the tweeter dome.

The following quote is taken from a white paper by Jeff Bagby, "How to Achieve Accurate In-Room Quasi-Anechoic Free-Field Frequency Response Measurements Down to 10 Hz":

"To begin with we need to understand what a far-field response really is. To be in the far-field is not as far as you may think. You are effectively in the far-field when your microphone is at a distance that is 3-5 times the radiating diameter of the driver."

Therefore, for a near field measurement for a 1" tweeter, the microphone should have been no further than 3 to 5 inches away. At the prior 9 inch microphone distance, room effects were coming into play.

Also from Bagby, "...the usable upper limit of the near field response is defined as
Fmax = 4311/ radiating diameter (in inches). Below this frequency the near-field response will be accurate and free of room and cabinet diffraction effects, above this frequency the near-field data is less reliable and less accurate."

Another rule for near-field measurement is that the microphone needs to be spaced from the center of the driver approximately 0.10 times the effective radiating radius of the driver. In the case of a 1" tweeter, this distance would have been 1/20th (0.05"). This was impractical in my case due to reflections from the tripod holding the microphone. A distance of 2 inches from the center of the tweeter domes resulted in smooth response curves without any dips due to room effects.

SDA Tweeter Frequency Response Measurements - Two Inch Distance


Figure 21. SL1000 frequency response.


Figure 22. SL2000 frequency response.


Figure 23. SL2500 frequency response.


Figure 24. SL3000 frequency response.


Figure 25. RD0194 frequency response.


Figure 26. RD0198 frequency response.


Figure 27. Composite plot of all six tweeters. Black line: RD0194. Red line: RD0198. Light blue line: SL1000. Green line: SL2000. Purple line: SL2500. Dark blue line: SL3000.


Figure 28. Comparison of RD0194 and SL2000 frequency response.


Figure 29. Comparison of RD0194 and SL1000 frequency response.


Figure 30. Comparison of RD0198 and SL2500 frequency response.


Figure 31. Comparison of RD0198 and SL3000 frequency response.

SDA Tweeter Harmonic Distortion Analysis - Two Inch Distance

In order of least harmonic distortion to most, the tweeters are: RD0198, RD0194, SL3000, SL1000, SL2500, SL2000.

The legend for the harmonic distortion plots is as follows:

Black line at top - frequency response.
Dark blue line second from top - summation of 2nd - 5th harmonics.
Light blue line - 5th harmonic.
Green line - 4th harmonic.
Purple line - 3rd harmonic.
Red line - 2nd harmonic.


Figure 32. SL1000 harmonic distortion analysis.


Figure 33. SL2000 harmonic distortion analysis.


Figure 34. SL2500 harmonic distortion analysis.


Figure 35. SL3000 harmonic distortion analysis.


Figure 36. RD0194 harmonic distortion analysis.


Figure 37. RD0198 harmonic distortion analysis.

References

1. Jeff Bagby, "How to Achieve Accurate In-Room Quasi-Anechoic Free-Field Frequency Response Measurements Down to 10 Hz", white paper, January 3, 2014, presented at Loudspeaker Design Workshop”, Kokomo, IN, December 14, 2013.

2. Christopher J. Struck, Steve F. Temme, " Simulated Free Field Measurements", Journal of the Audio Engineering Society, Vol. 42, No.6, June 1994.

3. Michael LaLena, "Speaker Testing and Analysis", https://www.diyaudioandvideo.com/Tutorial/SpeakerResponseTesting/

westmassguy

Nice work Ray.

DarqueKnight

SDA Tweeter Rankings

Throughout the years it has been said that the SL2000 was the worst performing/worst sounding of the SDA tweeters. The quantitative measurements and physical inspections done in this study seem to support that. The SL2000 rated at the bottom in terms of build quality, balanced frequency response, and harmonic distortion among the six tweeters tested.

Build Quality

1. RD0198
2. RD0194
3. SL3000
4. SL2500
5. SL1000
6. SL2000

Frequency Response

1. RD0198
2. RD0194
3. SL3000
4. SL2500
5. SL1000
6. SL2000

Harmonic Distortion

1. RD0198
2. RD0194
3. SL3000
4. SL1000
5. SL2500
6. SL2000

K_M

DarqueKnight wrote: »

SDA Tweeter Rankings

Throughout the years it has been said that the SL2000 was the worst performing/worst sounding of the SDA tweeters. The quantitative measurements and physical inspections done in this study seem to support that. The SL2000 rated at the bottom in terms of build quality, balanced frequency response, and harmonic distortion among the six tweeters tested.

Build Quality

1. RD0198
2. RD0194
3. SL3000
4. SL2500
5. SL1000
6. SL2000

Frequency Response

1. RD0198
2. RD0194
3. SL3000
4. SL2500
5. SL1000
6. SL2000

Harmonic Distortion

1. RD0198
2. RD0194
3. SL3000
4. SL1000
5. SL2500
6. SL2000

Sounds about right.
By the way, did you ever get around to doing the Frequency Response of the MW mentioned earlier??

DarqueKnight

No.

K_M

DarqueKnight wrote: »

Tony M wrote: »

I said I've heard over the years there was a rise in amplitude around the 5khz. region.
Is my memory wrong about the all the discussions about why the SL2000 sounds so bad most of the time?

The 5Khz region is where I was concentrating my curiosity at in the graphs. As you can see and.

Prior to your comment, I had never heard any mention of an amplitude rise in the 5 kHz region. All the discussions I know about were concerning the 5 dB rise in the 13 kHz region.

With the tweeters mounted in "Free Space" with no speaker baffle, you are seeing the effects of Baffle edge diffraction taking place at a higher frequency, than when mounted in the speaker.

http://www.linkwitzlab.com/diffraction.htm
http://ampslab.com/blog/2017/09/06/dome-tweeters/

The small dimensions of the face plate create a ripple effect when frequencies are a wavelength of the dimensions radiating from the tweeter dome.

In this case, it is approximately 4.5" wide, and there for a null around approximetly 5000hz and a resultant rise at about 10khz and about 2.5 khz.

Since the height dimension is smaller, diffraction will occur at a different frequency and help mask some of the ripple of up and down levels with rising frequency.

sevendogsbsd

Thank you for this article. After 31 years, I updated my SDA2A's SL2000 tweeters to the RDO194-1 from Midwest Speaker Repair. I did not do any before and after critical listening but I swear they sound like a new pair of speakers. I had thought about buying new speakers and that perhaps mine were just old but this simple and inexpensive update brought them back to life. Thanks!

geppy1

Great stuff I never really had much of a problem with the SL 1000

gp4jesus

Do my eyes deceive me or does the RD0 194 frame separate cleanly from the tweeter allowing mounting in a RD0 690(RTi A series) frame?

Thnx Tony

DarqueKnight

The RD0194 tweeter can be taken out of its frame. I do not know it can be mounted in an RD0690 frame, or if doing so would compromise the 194's performance.

gp4jesus

1000 thanks for your thorough research.

DarqueKnight wrote: »

The RD0194 tweeter can be taken out of its frame.

Great! Thnx for the quick reply.

DarqueKnight wrote: »

I do not know it can be mounted in an RD0690 frame...

likely. Several years ago someone on this forum successfully replaced a ‘690* w/an aftermarket ring tweeter** similar to a Polk 700 series speaker - LOVED the results!
* RTi A9s
** XO upgrades didn’t “do it” for him

DarqueKnight wrote: »

...or if doing so would compromise the 194's performance.

possibly - thanks for the cautionary advice.

Tony

gp4jesus

194s in any RTi A series speaker is a nice idea and will remain just that - 194s are anything BUT a drop in.

A comparison of your 194 photos to a 690’s physical etc, measurements demands MAJOR surgery to the closure primarily due to ~twice the magnet diameter* among other things - *sigh*
* 690 series ~1”

Thanks DK

Tony

K_M

Posted the SL-2000 response (as measured by member D-K)
But after closer examination, realized this is not really showing any peak at all, but simply a slightly rising response of about 2 db at around 12-14 khz.

What "smoothing" was used to generate the plots?

F1nut

From the November 1987 issue of Stereo Review,

"We noted with interest that the 5-dB tweeter resonance peak at 13,000 Hz that we found in the earlier version was again present in the SDA-1c."

K_M

F1nut wrote: »

From the November 1987 issue of Stereo Review,

"We noted with interest that the 5-dB tweeter resonance peak at 13,000 Hz that we found in the earlier version was again present in the SDA-1c."

I get there is a peak, (from the rta11t, stereophile) but D-K's response curves appear to have a lot of smoothing applied was all.

westmassguy

deleted

wolfie62

2 years ago I bought a mint pair of 1989 RTA 8Ts. I wanted, and still do, a pair of RTA 12(B), but don’t have the room.

The 8Ts use the same 6501 MW as the 12s. So, I get similar sound as the 12s.

After listening to my stock 8Ts, the original tweeters, SL2000, have a coloration that sounds like crumpled Mylar potato chip bags when reproducing scratches on records. I bought a pair of 0194 tweeters and replaced the SL2000s. Very smooth and detailed, no coloration.

2 weeks ago I bought a pair of 1984 SL1000 Peerless silk dome tweeters; price of $30 was too good to pass up! Near mint condition, taken from the RTA 12Bs.

I put those in, in place of the 0194 tweets.

Just my preference, but I like the Peerless more than the 0194 units!

But none of this comparison would have been possible without DK’s work and others who contributed! I would not have known what to substitute, nor what to expect!

This work is unparalleled in its excellence any where else!

Choices are a very good thing! I love Polk speakers, have since first hearing the RTA 12s in 1980!

Kudos for those in here keeping it real and factual!

I bow to the greatness.......

F1nut

The Peerless are entirely the wrong tweeter for your speakers.

xschop

Question for DarqueKnight, Is there an aftermarket 1" silk dome tweeter that you believe would be a good replacement or substitute for the RD0198 if it were mated properly to the RD0194/198 frame plate?
Thank you again for all your research.

DarqueKnight

If there is, I don't know of it. Some insight is to be gained by Polk having to re-engineer replacements for the SL2000 and SL3000 tweeters. I don't think it's that easy to find off-the-shelf tweeters that are close in performance and electrical characteristics to the originals.

DOTK

New guy here with another ?? for DarqueKnight

I have a pair of 1984 CRS with the SL1000. Bought a pair of the RDO194 from Midwest to replace the two I have running (sda tweeter pair fuses pulled). Seems past installs have modified the base of the RDO194s or enlarged the mounting hole/template on the speaker face. Looking at the 4 mounting screws for each to their mounting plates the spacing looks similar.

Could I just remove the old drivers from the SL1000 mounting plates and screw on the drivers from the RDO194 onto the SL1000 plates? Would keep from having to modify plates or speaker face?

Great work on your part...many thanks!

pitdogg2

DOTK wrote: »

Could I just remove the old drivers from the SL1000 mounting plates and screw on the drivers from the RDO194 onto the SL1000 plates? Would keep from having to modify plates or speaker face?

Great work on your part...many thanks!

If I'm understanding correctly NO you cannot switch the front tweeter face plates. The SL1000 tweeters have the dome and glued on the face plate. The connection points are also different for the wires. The new RD-0194 tweeters will need to have the old hole modified a little to keep the speaker front bezel from forcing the connection leads into the magnet assembly on the RD-0194. The screw holes are also off just a little.

DarqueKnight

pitdogg2 wrote: »

DOTK wrote: »

Could I just remove the old drivers from the SL1000 mounting plates and screw on the drivers from the RDO194 onto the SL1000 plates? Would keep from having to modify plates or speaker face?

Great work on your part...many thanks!

If I'm understanding correctly NO you cannot switch the front tweeter face plates.

You are understanding correctly.

Dr_Wu

I was able to trim the flange from the RD0194 face plate (it's plastic), using a Dremel cutting wheel; taking care to protect the screw holes, then smoothing the cut edges. The face plate then fit within the old cut-outs and mounted flush to the baffle. I didn't experience a difference in screw hole location.

xschop

Dr_Wu wrote: »

I was able to trim the flange from the RD0194 face plate (it's plastic), using a Dremel cutting wheel; taking care to protect the screw holes, then smoothing the cut edges. The face plate then fit within the old cut-outs and mounted flush to the baffle. I didn't experience a difference in screw hole location.

Diddo. I've done the same just to see if they fit the factory 5jr baffle. Afterwards you can simply fill in the old SL1000 baffle mount holes with wood glue, let dry, then redrill for RD0194 holes.

DOTK

Thanks all for the replies! Think i'll get out the Dremel and follow Dr Wu's lead.

Next question is where can I get the info on recapping/rebuilding the Gen 2 original CRS "2 board" crossovers? Have searched but cant find it yet?

xschop

A 6" table belt sander makes very quick work of the mod.

Dr_Wu

That would have been nice to have!

DarqueKnight

DOTK wrote: »

Next question is where can I get the info on recapping/rebuilding the Gen 2 original CRS "2 board" crossovers? Have searched but cant find it yet?

Here is the schematic:

https://us.v-cdn.net/5021930/uploads/attachments/1/8/5/1/1/26679.pdf