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 26th November 2010, 03:16 PM   #31
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Eureka, CA
Quote:
Originally Posted by richwalters View Post
I tried this sometime ago using high transconductance unscreened tubes, the 6CL6,12BY7 gang, and with the minimal damping and excellent ft, self oscillation can be a problem. Beware of the layout using such gyrators with a near infinite AC load Z.
The feedback resistor connected to the output tube plate is an AC load and sometimes also a DC load for the driver pentode. It's very well controlled and I've not had any oscillation problems using 6CL6 or other high gm pentode.

I agree the thin sound and rounded 10 KHz square wave are not likely to be related except perhaps caused by a high output damping factor. Thin sound could simply be flat LF response. The square wave rounding could simply be your OPT response and gNFB would be needed to help out.
But what is the sine wave frequency response? Maybe the high end is good enough.

The driver voltage and current changes are not a bad move though. Gyrators wouldn't hurt either but none of that is likely to change the thin-ness or the rounding of the square waves at this point.

Cheers,

Michael
  Reply With Quote
Old 26th November 2010, 03:53 PM   #32
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
Here is the open loop response of the amplifier (1watt) into a dummy load. (blue trace)
The black trace shows the result when I used an RC-network gNFB.
Transformers resonance is at 44,5 kHz

These where all measured without plate to plate NFB
Attached Images
File Type: png 807 combined.png (3.9 KB, 193 views)
__________________
Real tubes have top-caps
  Reply With Quote
Old 26th November 2010, 04:18 PM   #33
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
Wow, it is starting to make sense..!
Red trace is with shunt feedback added and the gNFB through RC network.
The green trace is with shunt feedback and no gNFB.
If you compare it with the blue one, you'll notice the similarities and a little bell inside your head should begint to ring.

Now, how do i fix it ?
Attached Images
File Type: png 807 combined2.png (5.7 KB, 195 views)
__________________
Real tubes have top-caps
  Reply With Quote
Old 26th November 2010, 04:47 PM   #34
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Eureka, CA
Quote:
Originally Posted by hidnplayr View Post
Wow, it is starting to make sense..!
Red trace is with shunt feedback added and the gNFB through RC network.
The green trace is with shunt feedback and no gNFB.
If you compare it with the blue one, you'll notice the similarities and a little bell inside your head should begint to ring.

Now, how do i fix it ?
Dingdingding

OK, the OPT peaking is increasing the NFB at high frequency, causing less gain.

I'm guessing (without detailed observations) that the peaking is much greater on the primary side as the P-S coupling decreases, resulting in greater effect on the feedback vs. the output, hence the greater loss with local feedback.

This could be the issue Sheldon is noticing as well.

The feedback or the OPT or both could be compensated. I'd be inclined to try a zobel across the OPT primary as an experiment.

Great work!

Michael
  Reply With Quote
Old 26th November 2010, 04:51 PM   #35
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
So.. I added a load of gNFB no RC network this time (altough turning it upside-down might have worked..)

Black = open loop
Blue = with plate to plate feedback
Red = same as blue, but with gNFB

The other pics are square wave (1khz and 10khz) at 1 watt into the dummy load.

For the listening test.. I will need an additional gain stage!
Attached Images
File Type: png 807 shunt3.png (4.3 KB, 194 views)
File Type: jpg 26112010365.jpg (311.1 KB, 195 views)
File Type: jpg 26112010366.jpg (290.5 KB, 184 views)
__________________
Real tubes have top-caps
  Reply With Quote
Old 26th November 2010, 05:24 PM   #36
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
Now look what happens when I return on my steps and remove the gNFB and use 220k for the plate to plate resistors.. (green trace)
Remember, the red trace is the one with the load of gNFB.
Attached Images
File Type: png 807 shunt4.png (5.0 KB, 75 views)
__________________
Real tubes have top-caps
  Reply With Quote
Old 26th November 2010, 06:30 PM   #37
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
..So i replaced the 0.91µF capacitors I was using to couple the feedback resistors, with 47nF ones..
This creates LF roll-off in the feedback circuit, compensating the LF roll-off of the output Xformer..
(evil/genius plan, I leave it up to you to decide..)

Et voila!
No gNFB and pretty decent square wave response, even with a real speaker as load. Just like I wanted..

The green trace is that of the circuit I just described.
Attached Images
File Type: png 807 shunt5.png (4.9 KB, 83 views)
__________________
Real tubes have top-caps
  Reply With Quote
Old 27th November 2010, 05:17 PM   #38
diyAudio Member
 
hidnplayr's Avatar
 
Join Date: Sep 2008
The previous idea obviously was a stupid one, as it made the DF higher where you want it to be small, on the lower frequency's..

Another way, wich I'm testing now, is to make an arrangement with resistors, in baby huey style. And accross the resistor that connects the two phases together, I placed a zobel.
Now there is less damping in the highest frequency's, but the High frequency roll-off of the amplifier is much better then without the zobel..
__________________
Real tubes have top-caps
  Reply With Quote
Old 28th November 2010, 09:57 AM   #39
diyAudio Member
 
Join Date: Mar 2003
Location: Zürich
Quote:
Originally Posted by revintage View Post
The driver should be run at as high current where it has its sweet spot, check curves. Go for a D3a or E280F at 150V/150V/20mA. Also add gyrators instead of anode resistors. MJK has shown elsewhere how it should be done.
Hi Revintage

I have followed quite a lot of the recent discussions on gyrators, but can't remember about them being combined with pentodes. Can you please provide me with some further details, a link, or some keywords to search?

Thanks, Erik
__________________
my surname is indeed 'de Best': neither misspelling nor snobbism! Ask SY!
  Reply With Quote
Old 29th November 2010, 08:52 AM   #40
diyAudio Member
 
Join Date: May 2007
Hey Erik,
This one made by MJK in an earlier thread says it all:

Click the image to open in full size.
  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
EL84 PP using 5K plate to plate primate OPT jmillerdoc Tubes / Valves 5 9th March 2010 09:14 PM
Plate Current VS Plate Voltage Graphs Captn Dave Tubes / Valves 2 13th February 2009 08:57 PM
ECC83 - Long Plate vs Short Plate poynton Tubes / Valves 16 17th February 2008 07:42 AM
Subwoofer Plate Amplifier Base Plate supplier? jwells777 Parts 1 11th May 2003 07:40 AM


New To Site? Need Help?

All times are GMT. The time now is 05:15 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright İ1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2