designing TGM3 - output Triples - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd May 2010, 09:06 PM   #1
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Default designing TGM3 - output Triples

Following on from the last design [TGM2 amplifier I want to see if the design can evolve to achieve better sound. I have some single rail transformers, so this will be a Single Rail design with an output capacitor(s).

For inspiration I have started by looking at the Quad 303. It's an amplifier which came before the famous current dumping designs but people who have listened to both say they prefer the 303 (partly because the 405 was so poorly implemented).

The Quad 303 uses the same bootstrap VAS as my prior designs. With a Single Rail supply it uses a single input device instead of an LTP. It also uses output triples with quasi-complimentary power devices. For TGM3 I'm going to adopt the triples but use a complimentary output. No doubt my version will sound completely different - so be it, I'm not trying to clone the Quad.

Attached is my current approach.

Note that I'm not looking for gobs of power considering that I've only got 36V to play with. So I've used small signal devices everywhere. Simulations say they won't catch fire

Whadya think ??
Attached Images
File Type: jpg TGM3c.jpg (202.5 KB, 1311 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.

Last edited by Bigun; 23rd May 2010 at 09:09 PM.
  Reply With Quote
Old 23rd May 2010, 09:34 PM   #2
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
See where your OLG meter source is? It looks to me like the bootstrap cap is drawing current through it, which contaminates any feedback measurement. You should probably move the bootstrap to the right of the source.

If you plan on running this without a center-tapped trafo, I'm interested in what you think of the minimalist virtual ground I posted about in the TMG2 thread shortly ago.

- keantoken

Last edited by keantoken; 23rd May 2010 at 09:38 PM.
  Reply With Quote
Old 23rd May 2010, 09:52 PM   #3
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
I may toy with the virtual earth, but it seems like it just moves the output cap from one side of the speaker to the other side - and requires more caps in total. I've built one on a small piece of perf board. I also tried it with active devices to hold the centre voltage more firmly than the resistors alone but I haven't done listening tests with it.

Thanks for the tip on the OLG meter, I'll try moving it and see what changes. With the OLG meter I'm really looking at voltages, the meter has perfect zero resistance - so perhaps current flow into the bootstrap cap won't matter ?
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 23rd May 2010, 10:05 PM   #4
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
I am going to move my discussion to this thread now...

Basically what I'm saying is that the cap is no longer directly in series with the speaker - after this now the amplifier has a chance to correct for any distortions caused by it (them). The amp does not care where, with respect to rails, ground is. The amp simply uses ground as a reference; it makes a voltage difference between ground and speaker input. Any distortion caused by the caps only moves the ground reference up or down with respect to rails, and the amp is designed to ignore this as long as power ground and input ground remain the same. When you disconnect input and power ground, the caps essentially become in series with the speaker again.

I realized my explanation in the other thread was lacking... I hope this makes more sense.

While at higher frequencies the cap distortions may slip past the feedback of the amp, this goes for all ambient distortions (I call them ambient because they are induced by rail modulation and peripheral things that the amp is intended to be able to filter out).

If you're not measuring current through the meter, it should be fine. Again, maybe neural pruning...

- keantoken

Last edited by keantoken; 23rd May 2010 at 10:15 PM.
  Reply With Quote
Old 24th May 2010, 01:03 AM   #5
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Do you mean that the power rails are supposed to float wrt virtual ground? You'll want to remove those resistors from the virtual earth or ensure they are of high resistance so that they can float.

This capacitor only virtual earth scenario was recommended by Jean Hirage in his Le Monstre as an option. However, I am concerned that without separate transformer windings for L and R channels there could be a fair bit of cross talk because the two channels will 'talk' via the power rails.

My transformer has only one secondary and needs to feed both L and R channels (in the current set up I can't fit two transformers into the Chasis - I have 4 transformers available and I suppose I could use a large chasis if I had to)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.

Last edited by Bigun; 24th May 2010 at 01:17 AM.
  Reply With Quote
Old 24th May 2010, 01:25 AM   #6
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
Why remove the resistors? This would cause total DC blocking, which would disable the amp. It needs an external bias to keep it near the middle (by relieving rogue DC necessary for input bias current, etc.), and the resistors are for this purpose. Also consider output offset; if the offset is not adjusted and the resistors are too high, the ground will bump up against the rails and may be too sensitive to DC offset to get ground in the right place. I would make the quiescent current of the resistors at least 10 times the total DC current that runs from the rails into ground, hopefully more. For this amp, the only DC running through ground will be Q1's Ib, so you could easily get away with higher value resistors, though I would have at least several mA quiescent current so there are no complications for offset adjustment (when offset is adjusted to zero, ground will also be in the middle of the rails since there is minimal DC to disrupt the balance between resistors).

You could use a different virtual ground for each amp and that would pretty well separate the paths, save for rail modulation which will happen regardless of grounding scheme.

- keantoken

Last edited by keantoken; 24th May 2010 at 01:41 AM.
  Reply With Quote
Old 24th May 2010, 01:53 AM   #7
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
I've edited my previous post so much I'm making a new post just so everyone knows.

I didn't mean for ground to float against the rails. At DC, it would almost be floating, but the resistors are there to help keep it in place.

- keantoken
  Reply With Quote
Old 24th May 2010, 12:56 PM   #8
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Whilst I think about output capacitors perhaps I should simply use the TGM chasis I already have. Now that I've given up on TGM1 I have additional space in the box because it's designed to allow up to 5 channels and so far only 2 are in use.

Now we can use dual rails and the output capacitor can be removed.

I may even be able to modify the existing TGM pcb layout without having to start over
Attached Images
File Type: jpg TGM3f.jpg (268.4 KB, 1190 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 24th May 2010, 09:47 PM   #9
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
I read on one of the krill threads that there is an optimal value for C2. Have you experimented with it? IIRC, if it's too high or too low the benefit is lost. I wonder if the CFP's affect this.

- keantoken
  Reply With Quote
Old 24th May 2010, 10:33 PM   #10
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
I haven't experimented with C2. It's main function is for charge suck-out and I remember settling on 100nF as a value that has been popular. If I remember correctly I used 150nF because the larger cap made pcb layout easier

Looking at the home made pcb from TGM2 it doesn't look too hard to shuffle things around to make up something for TGM3. Heres' the first attempt...based on single pole compensation.
Attached Images
File Type: jpg TGM3pcb.jpg (137.4 KB, 949 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Designing amp to withstand shorted output mr.duck Solid State 27 27th December 2006 04:02 AM
Designing an Output stage for an LM402 dfdye Solid State 7 22nd May 2006 08:13 AM
help designing my 2.5-way edjosh23 Multi-Way 8 26th August 2005 05:40 PM
Designing for guys who like designing Sch3mat1c Tubes / Valves 14 31st August 2003 11:46 AM
Help in designing of current output amp Edo Solid State 24 14th October 2002 09:26 AM


New To Site? Need Help?

All times are GMT. The time now is 03:01 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2