Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

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 28th August 2013, 11:43 PM   #131
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
I was able to scare up a model for the BSS159, so I did some simulations. I'm moderately pleased with the results for such a simple circuit, especially as the high order distortion procts seem to be down in the mud compared to
2nd and 3rd. I need to take my bench prototype and shift the feedback point to the top of the resistor as in the schematic, instead of at the top of the IR LED - resistor stack. That may clean up the sound a bit.
Attached Images
File Type: gif quarnel.GIF (28.4 KB, 418 views)
  Reply With Quote
Old 31st August 2013, 06:34 PM   #132
gman76 is offline gman76  United States
diyAudio Member
 
Join Date: Apr 2007
You have a *basement* in sili valley? Nice work, I might adapt some of this for a headphone amp. Oh aren't light bulbs microphonic? I know vacuum tubes are.
  Reply With Quote
Old 31st August 2013, 06:53 PM   #133
gman76 is offline gman76  United States
diyAudio Member
 
Join Date: Apr 2007
Forget that comment (microphonic), you're running so much current that it doesn't matter. But I do know this-- lamps have a positive TC. I'm still trying to explain the behavior that nelson speaks of (midrange hump), maybe the thermal time constant at work?

I remember building a weinbridge oscillator back in the 80s based a simple 3w lamp. Worked great, 0.1% THD, from such a simple circuit, lamp used as a feedback element.
  Reply With Quote
Old 31st August 2013, 11:01 PM   #134
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
I had to buy a ~100 year old house to get that basement.
  Reply With Quote
Old 6th September 2013, 05:31 AM   #135
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Update - I've been looking more closely at gain-phase plots and am concerned that the current edition of L'Fake bottoms out at 0 dB instead of continuing on to a gain less than one. I wonder whether this is an artifact of the gain-phase analyzer (or my setup), or a real problem, as this sort of thing indicates that there is not enough gain margin, such that hooking up the amp to something like a cable and a pair of speakers (reactive load!!!) could cause oscillation problems to rear their ugly head.

Having said that, the time domain response (square wave test) looks quite benign. However, the square wave test was done with a resistive load with short leads.

At any rate, Ill be probably spending some time this weekend at work with the gain-phase analyzer and an amp or two, trying to suss out whether the analyzer or my simulations (which predict all is well) are telling the truth. I have an HP 3575 gain-phase meter at home that I could press into service for point measurements near gain crossover - it might be able to supply an independent point of view. Plugging my ears and doing the square wave test with a speaker and reasonably long cables might also be instructive. I have run into situations where a resistive load is hunky-dory, but the amp in question takes off when hooked to a pair of real speakers ( my "Kingfisher" tube amp is a case in point). That one took some fixing.

Edit - the time domain and frequency domain plots I'm citing are at posts 108 and 109 in this thread.

Last edited by wrenchone; 6th September 2013 at 05:35 AM.
  Reply With Quote
Old 10th September 2013, 08:54 PM   #136
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
This afternoon, I ran a gain-phase plot on the most recent version of the "Fake SIT" amp. The results are appended. On both plots, gain is 8 dB/div, and phase is 45 degrees/div. Zero dB and zero degrees are both in the middle of the page (there are 10 vertical divisions, so the zero point is 5 boxes up from the bottom of the plot).

The bottom plot of the two shows gain and phase for the original amp. The gain (top trace) never reaches 0dB (it even peaks up a bit) before the phase (bottom trace) hits rock bottom, which is disturbing. This is in spite of the square wave response looking reasonably decent. This is an indication of an oscillation problem waiting to happen. Behavior like this explains why some amps might look benign on the bench, but take off and sing like a tree full of birds once they're hooked up to long cables and speakers. Another characteristic of a marginally stable amp appears to be excruciating sibilants, especially on voices.

The top plot shows the same amp with a 33nF + 10 ohm Zobel network hooked across the output terminals. The gain now passes cleanly through 0dB, and stays below it. There is only 39 degrees of phase margin, but that can be (and will be) amended with a change in compensation, now that the amp is behaving itself around the zero crossing. I intend to examine all my amps this way in the future.

This investigation also points out why an output Zobel network is a good idea.
Attached Images
File Type: gif QNel_res.GIF (43.5 KB, 255 views)

Last edited by wrenchone; 10th September 2013 at 08:58 PM.
  Reply With Quote
Old 11th September 2013, 06:01 PM   #137
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
BTW, none of this sort of squirreley behavior showed up in the simulations, which shows that you can only trust them so far...
  Reply With Quote
Old 19th September 2013, 06:03 AM   #138
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
I just picked up some 1000W linear quartz-iodine bulbs on fleabay that should prove useful for biasing up a L'Fake Lite at a supply voltage that should offer more efficiency/less wasted dissipation. We'll see how they look when they arrive. I won't spend much time on this for right now, as there are several other projects I want to finish for this year's Burning Amp. I will at least throw a bulb on a bench supply to determine the 2A bias point.
  Reply With Quote
Old 25th November 2013, 01:36 AM   #139
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Just to stir the pot a bit, here's a zen-sh fake with a modulated current source load. Gain is about 5X, and THD appears too be a fairly linear function of output voltage, dominated by 2nd and 3rd. The results shown are a simulation with 1V input drive. A 0.2V input signal gives me 1/5 the THD, with a similar harmonic distribution. In practice, I would use the IRFP260 for the big output mosfets rather than the 360s shown in the schematic - those were the models I had available in PSpice.
Attached Images
File Type: gif fake_zen.gif (29.0 KB, 145 views)
  Reply With Quote
Old 25th November 2013, 01:52 AM   #140
The one and only
 
Nelson Pass's Avatar
 
Join Date: Mar 2001
Quote:
Originally Posted by wrenchone View Post
hooking up the amp to something like a cable and a pair of speakers (reactive load!!!) could cause oscillation problems to rear their ugly head.

...BTW, none of this sort of squirreley behavior showed up in the simulations
Those peaky loudspeakers and cables again!

  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
If You Don't Have a SIT, Fake It... wrenchone Analog Line Level 93 30th December 2012 12:05 AM
Behind SIT 2 pdyn Pass Labs 0 13th September 2012 08:54 AM
SIT-1 and SIT-2 owner manuals Nelson Pass Pass Labs 25 12th September 2012 07:16 PM
WTB: 6.5" woofer to replace broken woofer in cheap band pass subwoofer Jimmy154 Swap Meet 2 5th October 2010 02:00 PM
The elusive Horn...can I have it all? Fast1one Full Range 13 5th October 2007 05:50 AM


New To Site? Need Help?

All times are GMT. The time now is 06:39 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