Balanced F5 in a Small Footprint - Page 15 - diyAudio
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 20th May 2013, 03:37 AM   #141
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
Quote:
Originally Posted by kasey197 View Post
Absolutely rigorous analysis as always from you lhquam

The real world measurements may well be only half of what the sims suggest -
but the interesting thing in your simulations are the VERY significant difference in HF impact of the caps in the cascode vs non casc implementations.

What do you think thats attributable to, given the small difference (bw casc and no casc) for the no caps versions ?

Also I have some caps handy that might suit - if you pm me an address, i can drop them in the mail.
My guess is that the significant decrease of THD at high frequencies with cascoding is due to eliminating most of the Miller effect due to the gate-drain capacitance of the drive MOSFET.

What caps did you use? The ones I am planning to order are Nichicon UHW0J153MHD .015 Farad Z=0.010 ohm.
  Reply With Quote
Old 20th May 2013, 03:39 AM   #142
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
Quote:
Originally Posted by kasey197 View Post
also fwiw, i found that paralleling two low-z caps had better results than one larger cap of equivalent value (due to reduction in esr from leads etc).
Yes, I was guessing that many parallel caps would be better.
  Reply With Quote
Old 20th May 2013, 04:13 AM   #143
The Dastardly Dad of Three
diyAudio Member
 
Join Date: Sep 2009
I used the rubycon MCZ but it seems these may be EOL now. Looks like ones you're suggesting may suit well even after de-rating for low frequencies. Let us know what you find ....
  Reply With Quote
Old 20th May 2013, 02:02 PM   #144
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
[QUOTE=lhquam;3497373]My guess is that the significant decrease of THD at high frequencies with cascoding is due to eliminating most of the Miller effect due to the gate-drain capacitance of the drive MOSFET.

I am trying to better understand why the 3rd (odd) harmonics are so greatly reduced at high frequencies by the combination of the caps and cascoding.

I did a simulation of an unbalanced cascoded configuration with a 4 ohm load. The left plot shown below is a plot of the 3rd harmonic with 1pF and .1F caps. It is nearly identical to the 3rd order plot for the balanced configuration. Of course, with the unbalanced configuration, there is much less even order harmonic cancellation, therefore the THD (not shown) is much higher. This result suggests that the caps with cascoding would be helpful for the unbalanced F5, particularly if the N and P channel MOSFETs are better matched in transconductance and internal Rs.

In the right plot shown below, is a similar simulation for a cascoded F5 circlotron with caps across the source resistors which DOES NOT show a reduction in the 3rd harmonic. I believe that this is due to to topology of the global feedback loops. In the balanced F5, the 2nd harmonics are mostly cancelled by the push-pull outputs before the feedback loop. In the F5 circlotron, each feedback loop receives the output from a single-ended output stage with is highly polluted with 2nd harmonic.
Attached Images
File Type: jpg F5unbal-casc-10W-Cscaps-mix3.jpg (65.7 KB, 156 views)
File Type: jpg F5-circ-casc-10W-Cscaps-mix2b.jpg (73.2 KB, 141 views)
  Reply With Quote
Old 20th May 2013, 02:44 PM   #145
The Dastardly Dad of Three
diyAudio Member
 
Join Date: Sep 2009
Quote:
Originally Posted by lhquam View Post

In the right plot shown below, is a similar simulation for a cascoded F5 circlotron with caps across the source resistors which DOES NOT show a reduction in the 3rd harmonic.
The lines in the right chart shows some reductions ? Those are h3 lines right ?
  Reply With Quote
Old 20th May 2013, 03:31 PM   #146
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
Quote:
Originally Posted by kasey197 View Post
The lines in the right chart shows some reductions ? Those are h3 lines right ?
What I meant to say is that the cascoded circlotron does not exhibit the same behavior at high frequencies as that of the balanced and unbalanced push-pull output stages. Another plot (not shown) shows H3 vs. frequency for an UNCASCODED circlotron with the same values for Cs. That plot is nearly identical to that for the cascoded circlotron. I do not fully understand this result. Clearly something more than the Miller capacitance is involved.
  Reply With Quote
Old 21st May 2013, 02:17 AM   #147
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
After more simulations and analysis of the circuits, I do not understand why the balanced F5 with output FET cascoding and big caps across the output FET source resistors has such low third (odd) harmonics at high frequencies. Is this a simulation artifact? Can I reproduce this behavior in an actual circuit?
  Reply With Quote
Old 21st May 2013, 06:24 AM   #148
The Dastardly Dad of Three
diyAudio Member
 
Join Date: Sep 2009
i read somewhere that caps can screw up THD sim results and the solution is to put a voltage source across each capacitor.

Measure the voltage across each cap in the sim, add correctly polarized voltages across each cap and just set it to the voltages you measured.
  Reply With Quote
Old 21st May 2013, 01:29 PM   #149
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
Quote:
Originally Posted by kasey197 View Post
i read somewhere that caps can screw up THD sim results and the solution is to put a voltage source across each capacitor.

Measure the voltage across each cap in the sim, add correctly polarized voltages across each cap and just set it to the voltages you measured.
Thanks for the suggestion, I will give it a try. I have been very careful in setting up the simulations and making sure that the FFT results are not corrupted, but I am at a loss to explain the high frequency results. They are just too good to believe.
  Reply With Quote
Old 21st May 2013, 04:08 PM   #150
lhquam is offline lhquam  United States
diyAudio Member
 
Join Date: Mar 2010
Location: Olalla, Oregon: Land of the 100 Valleys
OK, I have clue to what is occurring to cause the 3rd harmonic to be so low with cascoding and caps across the source resistors. Here is a simulation of an F5-style push-pull output stage only, and an F5-style circlotron output stage with and without cascoding. The switches S1-S4 are there to allow the simulation commands to enable or disable the cascode FETs. From left to right the images are the schematic, AC analysis with cascoding, and AC analysis without cascoding. In both plots 0.047F caps were across the source resistors.

As you can see, with cascoding there is a significant peak in the output stage gain at high frequencies. When global feedback is applied in the full amplifier, this means that there will be a higher feedback factor at higher frequencies, thereby decreasing the harmonic distortion components.
Attached Images
File Type: jpg VDMOS-cascode-asc.jpg (289.5 KB, 113 views)
File Type: jpg VDMOS-cascode-ac.jpg (203.7 KB, 112 views)
File Type: jpg VDMOS-nocascode-ac.jpg (206.5 KB, 98 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
small footprint speaker for kick bass BigBoned Subwoofers 12 24th October 2011 01:30 AM
Small footprint SUB using multiple small drivers FredrikC Subwoofers 5 29th August 2010 05:01 AM
Small footprint SUB using multiple small drivers FredrikC Subwoofers 0 25th August 2010 08:12 PM
First time horn help.. small footprint Matt@Nearfield Multi-Way 2 17th May 2005 02:55 PM


New To Site? Need Help?

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