Does anyone have any FFT plots from real Class-AB tube amps? - 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 10th May 2010, 10:32 PM   #11
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
What kind of output tubes? Also what is your driver stage?
I suppose if it is a 200W amp you may have an issue, however if it is a 15W amp, probably not so bad.

I had a similar moment/ feeling when I measured my amp at higher powers.

What does the FFT of the driver stage show?
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 10th May 2010, 11:15 PM   #12
diyAudio Member
 
Globulator's Avatar
 
Join Date: Aug 2004
Location: Cambridge, England.
Quote:
Originally Posted by bigwill View Post
And here's the plot of around 13W into 8ohms (no global feedback in both images), only just a touch into AB I'm hoping real life isn't as bad as this:
Interesting..

Well I can only guess that you are using a larger section of the tube's 'almost straight line' and it's less straight at the extremes. Have you tried moving the bias point and re-simulating?

The answer it seems (if that's the reason) would be to run the tubes at higher voltage so the actual swing is more linear..

The only other effect is OPT saturation - not sure a simulator would 'know' anything about anything else - all Cs and Rs are perfect.
  Reply With Quote
Old 11th May 2010, 02:43 AM   #13
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
I am on my other computer so won't be able to post the measurements till tomorrow, however, I find that GNFB is only partly to blame for the harmonic spectra. When measuring the FFT at the grid of the Output Tube, I found that quite a bit of the distortion is from the drive section. With or Without NFB.

The level and number of higher order spike really go up when the LTP starts to do as Globultor says and hit the extremes. In my case the THD really starts to climb around 35-40Watts of output power.
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 11th May 2010, 12:51 PM   #14
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Here is the FFT spectra of the drive into the output tube. There is no GNFB. The power levels are recorded at the amplifier power level of 1Watt, 10 Watt and 40 Watt output into a 8R non inductive load.

I didn't make much effort to set the noise floor on the measurements and had to rescale slightly at the higher power levels. As you can see, the higher order harmonics are present even before the output tube. What I find interesting is the predominant 3rd order. That I believe is the particular T7 used in the LTP. I can roll tubes and change that. A T7 in this application is marginal at these drive levels, however, plays extremely well at "normal" power levels.

One other point when looking at your sim graphs is the scale. Your noise floor is way down past -140dB. I would think the error of the sim would have a large impact at this point.
Attached Images
File Type: jpg 1W1kHzNoGNFb.jpg (191.1 KB, 123 views)
File Type: jpg 10W1kHzNoGNFb.jpg (197.0 KB, 120 views)
File Type: jpg 40W1kHzNoGNFb.jpg (203.6 KB, 114 views)
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 11th May 2010, 12:56 PM   #15
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
High 3rd order is the tube's way of telling you to get more voltage on the plate! The other thing to watch for is the grid current draw of the output tubes as you near clipping- that will definitely manifest as increasing odd order distortion at the driver.
__________________
You might be screaming "No, no, no" and all they hear is "Who wants cake?" Let me tell you something: They all do. They all want cake.- Wilford Brimley
  Reply With Quote
Old 11th May 2010, 03:40 PM   #16
diyAudio Member
 
Globulator's Avatar
 
Join Date: Aug 2004
Location: Cambridge, England.
Quote:
Originally Posted by SGregory View Post
That I believe is the particular T7 used in the LTP.
Sorry to be picky but what is a T7, and what is an LTP?
TIA!
  Reply With Quote
Old 11th May 2010, 03:57 PM   #17
diyAudio Member
 
boywonder's Avatar
 
Join Date: Jan 2008
Location: So.Cal.
Quote:
Originally Posted by Globulator View Post
Sorry to be picky but what is a T7, and what is an LTP?
TIA!

T7=12AT7

LTP=long tailed pair
  Reply With Quote
Old 11th May 2010, 06:31 PM   #18
diyAudio Member
 
richwalters's Avatar
 
Join Date: Sep 2003
Location: Alps:Tube amp designs over 150W, SMPS guru.
FFT of fully optimised top flight fixed bias 100W HiFi amp; typ AB o/p; conventional Williamson diff driver; concertina p/s. ..boasting 0.1% thd with 20dB global nfb. Trace 0dB = 100W level. ref 1Khz. (Ignore glitch spike at 15Khz) tube line up 4x 6550; 2x 12BY7;1x6BL8

Everyone should be able to get this performance.

richy
Attached Images
File Type: jpg 100W spectrum.JPG (168.6 KB, 98 views)
  Reply With Quote
Old 11th May 2010, 06:50 PM   #19
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
I suppose if I threw 3x the GNFb into the loop I could get close to that too.
At 100Watts my wifes pans fall out of the kitchen cabinets and I lose my happy home. Plus my little amp would go up in smoke at 100W (also bad for wife acceptance factor). Maybe I will try more feedback to see how low of a THD can be obtained at full power.

In the meantime at normal listening levels try this one. 0dB approx 50 Watts, ref 1KHz 2xKT88, 12at7, 12bh7 7.3dB feedback into 8 Ohms. FFT taken at 1Watt output. 0.0076%THD
Attached Images
File Type: jpg 1W1kHz 8R12bh7Vref2.jpg (202.1 KB, 68 views)
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 11th May 2010, 06:55 PM   #20
Svein_B is offline Svein_B  Norway
diyAudio Member
 
Join Date: Oct 2005
Here is a Williamson type amp with more modest 10dB of nfb.
FFT plot at 1W, and at 2dB below clipping.

Notice on both this and on Richy's FFT the IMD with 100Hz from the PS ripple (the spikes on both sides of the fundamental.
You will usually not see this with Spice.

SveinB
Attached Images
File Type: gif MA-pentode.gif (9.1 KB, 64 views)
File Type: gif Pentode-clipping-2db.gif (7.8 KB, 66 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
EU AMP Ecodesign Working Plan 2009-2011 - No way for new Class A - AB - H & Tube Amps qsa Solid State 0 4th May 2010 07:47 AM
Soundstage with Class A/AB amps Samuel Jayaraj Solid State 23 9th January 2008 12:35 PM
Safe Working voltage of output Transistors in class AB amps with +/- supply rajeev luthra Solid State 3 31st July 2004 06:58 AM
Any Class AB Chip Amps ? LS Chip Amps 1 22nd July 2004 08:07 PM
Thermistors in class AB amps ? LBHajdu Pass Labs 10 24th June 2004 09:38 PM


New To Site? Need Help?

All times are GMT. The time now is 11:56 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