2A3 PP Can my OPT do the job? - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 5th February 2012, 08:34 PM   #1
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Default 2A3 PP Can my OPT do the job?

I am working on a 2A3 PP amp which I currently have breadboarded with an Edcor CXPP25-8-5K. At this time I am running it open loop. The frequency response curve that I measured is attached. The other two pictures show a 10khz square wave at the grid of one of the 2A3's, and at the OPT secondary into an 8ohm non-inductive resistor. The frequency of the ringing on the photographs coincides with the dip and peak on the freq. response curve, with the peak at about 50khz. I was planning to try adding a resistor and capacitor in series, paralleled from the primary center-tap to the plate of each 2A3 to try and neutralize the ringing. Is this a good place to try and squash the ringing; if not, where should I consider? Is this the right approach, or is there a better compensation network I should look at? With the peak at only 50khz, is it possible to achieve a good response at 20khz with this OPT, or am I doomed? I would eventually like to add some feedback, but I would like to see a better open loop behavior before doing it. I'm hoping that one of you has fought this fight before and can point me in the right direction. Thanks!
Joel
Attached Images
File Type: jpg Response curve--open loop.jpg (98.7 KB, 333 views)
File Type: jpg 10khz at 2a3 grid.JPG (209.7 KB, 325 views)
File Type: jpg 10khz at output.JPG (205.4 KB, 317 views)
  Reply With Quote
Old 5th February 2012, 09:04 PM   #2
diyAudio Member
 
HollowState's Avatar
 
Join Date: Aug 2006
Location: Taxland, New Jersey
Quote:
Originally Posted by jgray View Post
With the peak at only 50khz, is it possible to achieve a good response at 20khz with this OPT, or am I doomed?
No. you're not doomed. For a realistic square wave response out to 20KHz, you should use a 2KHz fundamental. With all the harmonics that make up a square wave, 2KHz will equate by a factor of ten up to 20KHz. By using 10KHz, you're asking the transformer to pass 100KHz and beyond.
__________________
"It is hard to imagine a more stupid or more dangerous way of making decisions than by putting those decisions in the hands of people who pay no price for being wrong." ~Thomas Sowell
  Reply With Quote
Old 5th February 2012, 09:21 PM   #3
diyAudio Member
 
Join Date: Dec 2011
Sometimes it is interesting to look at the schematics of the audio section (modulator) of A.M. broadcast transmitters. They're generally push pull driving push pull. Often the input transformer has split secondaries (two wires coming out instead of a center tap) and the low sides of the drive coils each have a resistor to ground and a resistor to the opposite output plate. So it's balanced feedback. The feedback resistors are actually a number of smaller resistors in series (the plates are at 3000 Volts), with small caps across each for some lead compensation. As far as transformers ringing go, I know some put a series RC across the primary. I was thinking the R would lower the Q quite a bit, but that it wasn't that good to have more C lowering the resonant frequency. I've been wanting to try putting a series RLC across the primary to bring the impedance right back down in the same place it peaks. I've never seen anyone else do it, but I think it should work well. Hopefully there wouldn't be much energy at that frequency as I'd like to get away with a pretty tiny axial choke. If you're running an OPT with 4 and 16 Ohm taps (4 is effectively a CT), grounding the 4 Ohm and using the common and 16 for feedback to the output cathodes is a really strong start at cleaning things things up, but doing that would require separate filament windings. Still keep the speaker connected wherever it was. With that tightening up the output section response, there should be less difficulty adding any global feedback. For things like this it never hurts to dig up an old HP 4800A vector impedance meter for measuring transformer characteristics and a HP 3575A gain and phase meter to get a better idea what compensation/feedback networks might work well to close the loop. Good output transformers always seem to be an issue. I've wondered if the David Hafler patent has run out, or can be licensed reasonable. The patent number is in the old Dynanco manuals. It not only interleaved windings for tighter coupling (better balance and less leakage inductance) but it minimizes capacitive currents by keeping coils and ends of coils with the lowest voltage differences near each other. To do that best come the trick probably not in most other segmented transformers, have a coil (or several??) wound backwards. That allows the lower voltage end be where it is wanted while still having the desired phase at the other end. Some claim to make as good of transformers, but I serious doubt it. If any do use reverse would coils I'd like to know about it. There have been some older articles testing/comparing transformers including running square waves and the differences were huge. In one amp article transformers had such poor coupling/leakage it actually helped to have different RC networks across each half of the primary.

Last edited by riccoryder; 5th February 2012 at 09:22 PM. Reason: typo
  Reply With Quote
Old 5th February 2012, 09:55 PM   #4
diyAudio Member
 
the_manta's Avatar
 
Join Date: Nov 2006
Location: Munich, Bavaria
Tried it and it worked OK.
I used the screen tabs for that. Just like here: Roehrentechnik. That guy makes really good OPT's, so he definitely knows what he's doing.

Generally this doesn't look too bad. This Edcor OPT performs much better than I thought. My Hammond 1627SE is much, much worse.


Quote:
Originally Posted by riccoryder View Post
I've been wanting to try putting a series RLC across the primary to bring the impedance right back down in the same place it peaks. I've never seen anyone else do it, but I think it should work well. Hopefully there wouldn't be much energy at that frequency as I'd like to get away with a pretty tiny axial choke.
What you need the L for ? Leakage inductance of the OPT is your L
__________________
Терпенье и труд все перетрут
  Reply With Quote
Old 6th February 2012, 05:29 PM   #5
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Thanks for the comments. The scope trace for 2khz looks pretty square, but with some overshoot and ringing. I will try and post a picture of that tonight.

The OPT only has a single 8ohm secondary so I can't do the balanced feedback idea.

The link to Roehrentechnik was interesting. I hadn't thought of using the UL taps as a possibility for a compensation network connection. I will probably try both the anode to center tap and the anode to UL tap and see what works better.
  Reply With Quote
Old 6th February 2012, 05:53 PM   #6
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Riccoryder, I just spotted this link (in the tube amp photos area) to a company that makes transformers with reverse windings like you were wondering about. I don't know them, or their performance, but the comments in the forum seemed positive.
Sound Carrier
  Reply With Quote
Old 7th February 2012, 12:18 AM   #7
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Here's the 2 kHz square wave at the secondary of the OPT into an 8 ohm non-inductive load. The 50 kHz ringing is still visible.
Attached Images
File Type: jpg 2khz at output.JPG (204.2 KB, 238 views)
  Reply With Quote
Old 19th February 2012, 11:09 PM   #8
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Default Can anyone explain this?

Here's an update. I tried a compensation network in several places without much luck including: across the OPT secondary, from the 2A3 anodes to the primary center tap, from the 2A3 anodes to the screen taps, and between the 2 anodes. What finally seemed to work was to put an R and C in series from one screen tap to the other. I'm not sure why this seemed to work better than between the two anodes. Can anyone explain this? Attached are a few pictures showing how the ringing changed with a 2kHz square wave. The four pictures are for: no network, a .033uF and R, a .047uF and R, and a .01uF and R. I think the .01uF result looks the best. Which one would you go with?
Attached Images
File Type: jpg With no cap.JPG (351.1 KB, 194 views)
File Type: jpg With .033uf cap.JPG (362.0 KB, 33 views)
File Type: jpg With .047uf cap.JPG (309.9 KB, 33 views)
File Type: jpg With .01uf cap.JPG (382.9 KB, 39 views)
  Reply With Quote
Old 20th February 2012, 07:59 AM   #9
45 is offline 45  Italy
diyAudio Member
 
45's Avatar
 
Join Date: Dec 2008
Location: UK
Quote:
Originally Posted by jgray View Post
Which one would you go with?
The one which sounds better. Compensated square waves might look prettier but this doesn't mean a better sound.....
You might also see if you can get rid of ringing with a small cap in parallel to fb resistor instead of adding stuff between the plates. This may work just fine as your 2 KHz wave without compensation also tells the ringing is damped.
Cheers,
45
  Reply With Quote
Old 21st February 2012, 01:20 AM   #10
jgray is offline jgray  United States
diyAudio Member
 
Join Date: Sep 2010
Location: Chicago NW suburbs
Hi 45, thanks for the comments. I will try and do a careful listening test and see if my ears are "golden" enough to tell the difference.

I am still running without feedback yet, so can't do the idea of putting a cap in parallel with the feedback resistor. When I do add the feedback, I will for sure experiment with that. On another amp I built, that cap made a big difference.
Regards,
Joel
  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
What to do with this PP OPT? pwan Tubes / Valves 1 7th November 2008 02:36 AM
Use for 5K:8 PP OPT? dsavitsk Tubes / Valves 25 26th February 2008 07:56 AM
Sizing PP OPT alexg Tubes / Valves 14 11th June 2007 07:19 AM
James SE OPT for 300B & 2A3 on eBay argo Tubes / Valves 0 20th January 2005 11:21 AM
2a3 Se Amp With 5k Opt? zxx123 Tubes / Valves 4 1st September 2003 12:01 PM


New To Site? Need Help?

All times are GMT. The time now is 05:37 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