Twisted X Overture

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,

I am not sure how many of you have followed my new chip amp design which is based on ideas proposed in this thread.

http://www.diyaudio.com/forums/attachment.php?s=&postid=421299&stamp=1087758354

The idea is a bridged power amp that takes advantage of the "Super Symmetry" effects of the THS4131 and itself is also Symmetric.

So while maturing of the idea is largely to be credited to Nelson Pass and to Terry Aben and others (I don't have time to list them all) who worked very hard to get a discrete circuit going that modeled Nelson's copyrighted schematic, my design is substantially different, in that the only patented IP is wrapped up in the THS4131 and the LM3886. That said I fully recognize the work that went into Super Symmetry and want to be sure to credit that to Nelson Pass and thank him for sharing his ideas on a chipamp which benefits from it.

I wanted a “Symmetric” design but without the big caps in the signal path which the design that Nelson had proposed included, and without the worry of having to license anything from anyone or infringe on anyones patent or IP. The only IP I used was my own and that which was licensed to TI, or that which is readily usable without restriction.

For my part, I also place no restrictions on the material I present here, I simply request that the design I provide only be used for DIY audio. What is free, should remain free. Free as in speech.

My design utilizes the THS4131(or OPA1632 which will also work) to drive the power opamps (LM3886 in this case). The THS4131 is “Super Symmetric” in operation, and that gives my design some serious benefits when it come to CMMR. It is also a very fast part, and has excellent slew rate and settling time.

My design is really nothing more then a LM3886 nested in the feedback of a fully differential opamp. The THS4131 must be compensated with two small value(200pf) to make the circuit stable since the THS4131 is operating substantially faster than the LM3886 at the gains involved.

The input is balanced, and sounds best that way (lowest noise/distortion) but it can also handle single ended input.

There is an optional zobel network on the outputs for those with difficult to drive speakers. They actually can generally be omitted.

The amplifier has 20db of gain and has been tested with 28V rails.

It should be capable of 80W or so into 8ohms and 130W or so into 4ohms.

My heartfelt thanks to Terry Aben for all his support and encouragement, and the rest of the crew on the Pass Labs forum for their guidance in this endeavor. Terry did a superb job on his design which is based on Nelson's published circuit.

It is my hope this project will bring enjoyment to others and perhaps even spawn ideas for other circuits.

Cheers!
Russ
 

Attachments

  • x-amp1.sch.pdf
    69.8 KB · Views: 1,827
PCB PDFs

Here are PDFs for a single side PCB for this project. the PCB is by no means optimal, but it works just fine. I have had excelent results.

PCB PDFs

The copper side is not mirrored and the silk screen is mirrored for toner transfer.

Note that the THS4131 is mounted on the bottom, with pin #1 corner marked with a "1".

I have a much better optimised two sided design which I will be producing very soon.

Cheers!
Russ
 
Geeze Russ... do you slow down? Can you lend me some of your project/play time?!

Do me a favor and continue using a "basic" circuit around the chips themselves, and I will be able to try your variants very quickly indeed (once I source parts) since I went the modular approach... ;) Yet another reason to have done that!

C
 
Carl,

Yes the input impedance is 2.2K, and no it has not been an issue for me, but I am using a preamp which can easily drive a load down to say 500ohms or so. :) I am using my Twisted Sibling preamp. I also have it DC coupled since I know there is no DC on my source.

Alternately you could simply take a good opamp such as the AD8620 and use it as an instrumentation amp in front of the THS4131. That will get you very high input impedance.

You can read about Twisted Sibling here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=79930&perpage=10&pagenumber=1


Cheers!
Russ
 
Re: Super symmetry?

tiltedhalo said:
Am I missing something?? this looks like a balanced line driver circuit , you know a "A typical bridged amp" not super symmetrical looks to have a common feedback loop to the LM and the THS/OPA? if not please explain how it works???


TI Licenced Nelson's Super Symmetry patent in there design of the THS4131. By nesting the feedback of the output stage with the fully differential input stage I gain all the benefits of the Super Symmetry.

Search the forums for more info on Super Symmetry. It is pretty cool stuff. :cool:

The results so far have been very very pleasing. :D

Cheers!:drink:
Russ
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.