SDA-SRS, SRS 1.2, SRS 1.2TL, SRS 2.3TL Crossover Design and Progressive Point Source

[decato]

Introduction

There has been a lot of discussion about which SDA-SRS crossover is best. There has also been some mystery around Progressive Point Source technology. As I continue to learn more about speaker design, I thought I would share some of the latest knowledge I have acquired. According to Polk’s brochure for the SRS series featuring the trilaminate tweeter,

"… the SRS 1.2TL and 2.3TL utilize Polk's Progressive Point Source technology to maintain Constant Vertical Directivity of mid and high frequencies, which prevents undesirable beaming. As frequencies increase, the tweeter array adjusts its radiation area and eventually becomes an ideal point source at the very highest frequencies, eliminating frequency interactions and reflections between multiple drivers."

For the SRS, SRS 1.2, and SRS 1.2TL, I have provided graphs illustrating the frequency response of each tweeter as well as the composite response. A DC resistance of 7.5 ohms was used for the RD0194-1 (SL2000) tweeters while 5.6 ohms was used for the RD0198-1 (SL3000) tweeters. It is interesting to note how the composite response varies among the three models.

Figure 1: SDA-SRS Schematic
Figure 2: SDA-SRS Frequency Response
Figure 3: SDA-SRS 1.2 Schematic
Figure 4: SDA-SRS 1.2 Frequency Response
Figure 5: SDA-SRS 1.2TL Schematic
Figure 6: SDA-SRS 1.2TL Frequency Response


Project

I have always preferred the sound of the 1.2TL over the 2.3TL, mainly because I thought the 2.3TL sounded slightly harsh in the treble. I can now visualize what I was hearing, as there is a small spike in the response of each of the tweeters around 2 kHz. See Figures 7 and 8.

Figure 7: SDA-SRS 2.3TL Schematic
Figure 8: SDA-SRS 2.3TL Frequency Response

A while back I posted a thread about modifying the 2.3TL into a system more like the 1.2TL. I now wish to revisit that project. It is possible to make the 2.3TL sound almost identical to the 1.2TL. The formula, first discussed in http://www.polkaudio.com/forums/showthread.php?t=35433, is now as follows:


Make the low-pass section of the crossover identical to the 1.2TL crossover.
Use 4 MW6503s for the stereo array.
Use 2 MW6510s for the dimensional array.
Follow the schematic in Figure 9 for the high-pass section of the crossover. It will yield the composite frequency response seen in Figure 10, which is almost identical to that of the 1.2TL. Each tweeter’s output increases successively just like in the 1.2TL. The only difference if that there are three instead of four tweeters. Moreover, the phase of the 1.2TL has been maintained in this design.
Applying Soundcoat and adding additional bracing is still a welcome change.


Figure 9: Modified SDA-SRS 2.3TL Schematic
Figure 10: Modified SDA-SRS 2.3TL Frequency Response

How does it sound? Very much like the 1.2TL. How does the 1.2TL sound? Well, you know the answer to that question!

Here are some pictures of the modified crossover. A few components are on the underside of the board. The only drawback to this project is that it is somewhat difficult, as the layout for the high-pass components changes drastically. If you are up for a rewarding challenge, please give it a try. I guarantee you will like what you hear!

Figure 11: SDA-SRS 2.3TL Crossover - Top
Figure 12: SDA-SRS 2.3TL Crossover - Front
Figure 13: SDA-SRS 2.3TL Crossover - Side
Figure 14: SDA-SRS 2.3TL Crossover - Back

- Brian Borowski

[schwarcw]

Nice writeup Brian! I'm always impressed by your research and pursuit of improving the big SDA's. I enjoy it!:)

After I upgraded my 2.3 crossover with Sonicaps, which are physically larger caps than the originals or Solens, I decided that if I even did this again that I would start with building a new board to avoid the mess of caps hanging for dear life over the edge, up in air, etc. Did you give any consideration to using a new board with point to point wiring?

I'm not trying to derail the thread, I'm curious as I know you have worked a lot on various combinations and modifications.

Thanks!

[DarqueKnight]

Excellent writeup Brian. What effects on bass response do you expect from installing a 1.2TL crossover in a smaller cabinet?

[decato]

Did you give any consideration to using a new board with point to point wiring?

Hi Carl,
No, actually. This board seems to be working well. One thing that made it easier for me was that the older Solen capacitors had lengthy flexible leads. I made some modifications recently, after Solen had switched to standard tinned copper leads, but the bulk of that huge stack was already in place.

The board is a bit heavy with the 16 awg inductors on it. I wish Polk had used regular screws like they did in the 1.2TL instead of plastic clips. However, so far everything is holding up just fine.

-Brian

[decato]

Excellent writeup Brian. What effects on bass response do you expect from installing a 1.2TL crossover in a smaller cabinet?

Thanks, Raife.

To answer your question, there was no discernible change in the bass response of the 2.3TL. The low-pass section of the 2.3TL crossover is quite similar to that of the 1.2TL. The only change came when I switched the inductors over to 16 awg (from the default 20/18 awg). I did not bother upgrading the 16 mH inductor as I did in the 1.2TL because I observed that at very low frequencies, the excursions of the two dimensional drivers exceeded that of the four stereo drivers. An inductor with even less DCR might cause the dimensional drivers to bottom out early.

[F1nut]

You've got my wheels spinning.

[Toka78]

Great work, Brian! Who said the SDA was dead? :)

[RichCanDo]

Recently obtained the SDA 2nd Ed. Compendium. Great work! Noticed the schematic for the 1.2TL is not as shown above. Was going to PM DarqueKnight about this, but this appears a good place to post. Factory schematic in Compendium shows for the tweeters, the Positive speaker binding post ending at the Negative speaker terminal. - Negative binding post to Positive tweeter terminals. The reverse of what is shown above.

Above appears correct to me. If so, the factory schematic is wrong. Someone once told me that the binding posts on some speakers leaving the factory had the polarity reversed. If true, it would have been because they were wired at the factory per an incorrect schematic.

Hoping DarqueKnight reads this and can let those with Compendiums know whether the 1.2TL schematic in Compendium is correct or incorrect. Polarity for the 2.3TL is also reversed between binding posts and tweeters in the Compendium schematic.

Edit: spelling!

[DarqueKnight]

In the far recesses of my mind I seem to recall discussing with Ken Swauger the out-of-phase tweeter situation. I think it had to do with some phase shift between the LF and HF sections of the X-over and wiring the tweeters backwards set everything right again.

I will ask Ken to verify. I would expect that a typo of that nature would not get past multiple levels of approval.

[Polk65]

Excellent post Brian.

Here's some nude cousin's. ;)

Steve

[george daniel]

Great post,,got me thinking about my brothers 2.3 tl's :cool:

Another cousin from the deep south

[decato]

I have updated the diagrams for the TL series. The tweeter polarity is now reversed, just as it is in the schematics.

[Larry Chanin]

For the SRS, SRS 1.2, and SRS 1.2TL, I have provided graphs illustrating the frequency response of each tweeter as well as the composite response. A DC resistance of 7.5 ohms was used for the RD0194-1 (SL2000) tweeters while 5.6 ohms was used for the RD0198-1 (SL3000) tweeters. It is interesting to note how the composite response varies among the three models.

- Brian Borowski

Hi Brian,

Very impressive research.

Would you mind describing how you measured the frequency responses?

Thanks.

Larry

[decato]

Would you mind describing how you measured the frequency responses?

Hi Larry,
The frequency response was not measured. The curves represent the simulated responses of the tweeters as dictated by the given crossover circuitry.