Q: UL PP Instability - Anode Grid2 Snubber - Page 2 - 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 4th July 2012, 11:03 AM   #11
diyAudio Member
 
Join Date: Aug 2008
Quote:
Originally Posted by Merlinb View Post
At what frequency does it start rolling off?
With snubbers installed, roll off starts at 10 - 12 KHz, depending upon value of C (even with 45 pF).
Without snubbers, there is no roll off at audio frequency range.

Book published by G.E.C. "An approach to audio frequency amplifier design" recommends anode-grid2 snubbers for UL 1000 - 2000 pF + 470 - 1500 Ohm.

Last edited by LinuksGuru; 4th July 2012 at 11:08 AM.
  Reply With Quote
Old 4th July 2012, 04:53 PM   #12
diyAudio Member
 
richwalters's Avatar
 
Join Date: Sep 2003
Location: Alps:Tube amp designs over 150W, SMPS guru.
Circuit diagram SVP ?
Most are correct; high gm tubes like the 88 do need an anode to g2 snubber regardless of transformer and unsuprisingly wide values can be effective and just relevant to the output stage. There is far more to the this; I would start by also examining the first stage anode stage if that zobel is fitted because that determine the Bode plot loop gain at the upper end that no response peak exists after gloal NFB is connected.


richy
  Reply With Quote
Old 4th July 2012, 05:27 PM   #13
diyAudio Member
 
Join Date: Jan 2009
Location: planet earth
Hi Linuks,

which otp are you running ?

Mike
  Reply With Quote
Old 4th July 2012, 06:00 PM   #14
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
Quote:
Originally Posted by LinuksGuru View Post
Dumping ringing is not something I'm looking for. Square wave response at 2 KHz is near perfect at several V.

Amps works fine when load is connected to 8 ohm tap, but unstable if to 4 ohm tap.
With another output transformer it oscillates after certain power level reached no matter of load connection (I have 4 different transformers).

Snubbers between anode and grid 2 terminated instability completely, but introduced high frequency roll off. Quite strange, since in another amp I mentioned there was no such side effect.

Any idea how to avoid HF roll off with anode-grid2 snubbers?
If you're seeing a significant rolloff with only 45pF in the snubbers, something is *not* as we think it is.

As DF96 has tried to emphasize, we need to distinguish between measures designed to damp output stage parasitics and measures designed for loop stability. The former are important even in an amp without a global loop.

And as I have tried to emphasize, the output transformer primary snubbers are not a good place to put measures for global loop stability. (However, a Zobel across the secondary is both easy and uniformly positive WRT global stability - maybe 10 or 15 Ohms and 0.1uF.)

All good fortune,
Chris
  Reply With Quote
Old 4th July 2012, 07:23 PM   #15
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Quote:
Originally Posted by LinuksGuru
Quite possible stability affected by the very long interconnection wires, I have separate power supply sitting on another side of the table.
You do have some local supply rail decoupling on the amp side of the table, don't you? Otherwise the long supply wires could act as Lecher lines.

45pF could have an effect - remember we are talking about 10's or 100's of H. You don't need much capacitance to resonate within the audio band, which is partly why decent OPTs are expensive.

Last edited by DF96; 4th July 2012 at 07:28 PM. Reason: comment on 45pF
  Reply With Quote
Old 4th July 2012, 09:49 PM   #16
diyAudio Member
 
Join Date: Aug 2008
Hi, folks !

Thanks a LOT to all who replied with helpful ideas, tips and suggestions!

Schematic is slightly modified Williamson (available in a bunch of sources), switchable between penthode, UL, CFB, UL+CFB, and triode mode, adopted to KT88 / KT120 / 6550, fixed BIAS and 460-490V B+.
I tested it with 4 different custom built 60W / 3.3K - 3.9K transformers.

Since it is draft / test assembly, everything laying across the table, wires are long as road to Siberia (B+/GND - 80 cm, filament - 60 cm, transformer leads - 28 cm, input signal wire - 1 m).
DC supply rails are decoupled with electrolytes and 1000 pF ceramic caps.
Filament common for all 4 tubes, false center tap (made with 2 x 100 Ohm resistors) lifted to 40V with resistor divider from B+.

One transformer worked fine out of the box, another caused oscillation of UL even with GNFB disconnected, third only at high signal level (with GNFB on).
Williamson is very well known to be very picky when it comes to output transformers. However, I managed to get ALL of them to work, with up to 60W in UL mode. Since leakage inductance is in range 8 - 18 mH (depending upon particular transformer), it appears stray capacitance (all iron is quite big !) played major role in stability issues.

The cause of high-frequency roll off (described in my 1st post) was the conjunction of high-frequency RC filter (10K + 500pF) in pre-driver which was installed to cut off everything above audio band, and anode-grid#2 snubbers (seprately they don't have such side effect). Removing RC filter from pre-driver solved this problem.
Anode / grid#2 snubbers were suggested in a book published by G.E.C. "An approach to audio frequency amplifier design". In my case they are 4.7K + 1000 pF.

I didn't tried Zobel at secondary, it was not necessary at this time, additionally, it could "mask" high frequency oscillation if its level is small (once, with another amp, I had 8mV - 20mV at 8 Ohm output terminals).

THD/IMD analysis was not made yet at this moment, I just put away soldering iron, but picture looks promising. There is no stability issues or oscillation with any trafo I have (only slight tremble of picture on oscilloscope screen with output signal at very low level, e.g. 100mV), even at heavy overload / clipping.
Just look at present 2 KHz square wave with worst transformer, which at the beginning didn't worked at all (oscillated in UL mode with GFNB off).
Attached Images
File Type: jpg WL P2P SQ2KHz 2012-07-04.jpg (35.0 KB, 32 views)
  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
520v in PP UL...? JandG Tubes / Valves 4 19th April 2009 03:11 AM
Which do you prefer - EL84 UL PP or EL34 Triode PP - and why? ray_moth Tubes / Valves 4 23rd February 2007 12:22 PM
PP-Triode , PP-UL , PP-Pentode-strapped-as-Triode percy Tubes / Valves 23 22nd January 2007 12:36 AM
Is this EL84M UL PP OK? leadbelly Tubes / Valves 34 27th February 2006 01:53 AM


New To Site? Need Help?

All times are GMT. The time now is 05:52 PM.


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