Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Pass Labs This forum is dedicated to Pass Labs discussion.

Beefcake: a beefed-up BA-2
Beefcake: a beefed-up BA-2
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 November 2019, 12:31 PM   #21
JeffYoung is offline JeffYoung  Ireland
diyAudio Member
 
JeffYoung's Avatar
 
Join Date: Jul 2017
Location: Ireland
Beefcake: a beefed-up BA-2
Interesting. It has a pervasive effect:

1) drops distortion by 1/2 (THD @ 1W from .001 -> .00045, so somewhat academic)
2) drops H2 a bit more than H3 (spread between the two has dropped from 9db to 7db)
3) vastly increases the phase margin (50 -> 80)
4) but introduces a response spike around 5MHz which wipes out the gain margin
5) oddly FPBW is down a bit (to 50kHz) even though SR is up (to 26V/uSec)?

I've added an input filter (10K / 120p) to address the instability. It gives me a 10dB gain margin but with a significant loss in frequency response (-3dB 130kHz -> 75kHz, so again somewhat academic?).

Beefcake: a beefed-up BA-2-beefcakecurrentmirror-jpg

Cheers,
Jeff.
Attached Images
File Type: jpg beefcakecurrentmirror.jpg (49.8 KB, 323 views)
  Reply With Quote
Old 4th November 2019, 06:32 PM   #22
Rory Christ is offline Rory Christ
diyAudio Member
 
Rory Christ's Avatar
 
Join Date: Oct 2002
Location: Milwaukee, WI
Hi, Jeff.

I was wondering if you have a screenshot of the full schematic with the current mirror that you could post. I am curious how the current mirror affects the open loop gain.

My understanding is that switching to an active load will greatly increase the open loop gain, which increases the amount of negative feedback, all other things being kept equal (I'm not an expert on this, so my understanding could be off).

I remember coming across a thread from years ago where someone greatly "improved" an Aleph by adding a current mirror, and the response from NP was along the lines of "yeah, I tried that, but I didn't like the sound as much".

The authors I'm familiar with (Cordell, Self, Sloan) seem to treat the addition of a current mirror as a no-brainer in this situation, so I've always been curious about the topic.
  Reply With Quote
Old 4th November 2019, 07:58 PM   #23
JeffYoung is offline JeffYoung  Ireland
diyAudio Member
 
JeffYoung's Avatar
 
Join Date: Jul 2017
Location: Ireland
Beefcake: a beefed-up BA-2
Hi Rory,

I've attached the updated LTSpice file and another one (NOT YET updated with the current mirror) that I use to calculate the OL gain.

I've also attached a JPG in case you don't use LTSpice.

Cheers,
Jeff.
Attached Images
File Type: jpg beefcake_schematic.JPG (184.1 KB, 122 views)
Attached Files
File Type: asc beefcake_sim_ol.asc (14.6 KB, 1 views)
File Type: asc beefcake.asc (14.6 KB, 3 views)
  Reply With Quote
Old 4th November 2019, 08:55 PM   #24
JeffYoung is offline JeffYoung  Ireland
diyAudio Member
 
JeffYoung's Avatar
 
Join Date: Jul 2017
Location: Ireland
Beefcake: a beefed-up BA-2
Ha! Yeah, open-loop gain shot up to 110dB.

Since the FB divider only controls closed-loop gain the net result was adding a boat-load of NFB. (Which I suppose explains why the effects were so pervasive.)

I tried limiting open-loop gain by increasing the resistance in the current mirrors, but it was very finicky (illustrating at least that I'm at the boundary of my knowledge).
  Reply With Quote
Old 4th November 2019, 10:30 PM   #25
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
> I'm at the boundary of my knowledge

I strongly suggest you buy the Douglas Self and Bob Cordell books on amplifier design and take them to bed for a month.

Both have taught self-balancing diff-inputs for many years, and explain how this choice cascades through the design.

Sure you can plow your own path through the jungle, but sometimes you get farther following a map by folks who have been there before.
  Reply With Quote
Old 4th November 2019, 10:43 PM   #26
JeffYoung is offline JeffYoung  Ireland
diyAudio Member
 
JeffYoung's Avatar
 
Join Date: Jul 2017
Location: Ireland
Beefcake: a beefed-up BA-2
Oh, indeed I've spent many months with Bob Cordell (and Morgan Jones). But even if I understand it when reading, it doesn't seem to get to long-term memory till I try to make it work. And then I have to go back and read again....

Not that I'm complaining. I'm a long way from where I started, and it's still fun.
  Reply With Quote
Old 5th November 2019, 05:39 AM   #27
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
It's your hobby. Do what you want.

While he was not the first to notice input pair unbalance effect on THD, Sir Self wrote it up very extensively.

And it's YOUR design. Even Famous Designers break good form. Here is a very recent design from a VERY experienced designer and, if we believe the notes, the input pair is NOT balanced! Certainly no attempt to self-correct balance. The "approximate balance" is set by Q3+R4 against 1k and Vgs of Q6. Unless all the Q3 and Q6 come from the same lot (and they might be, in the first batch) then balance will be not-as-good as Self teaches us.

We must note that JFETs are "softer" so the off-center rise of THD won't be as sharp as Self's BJTs. And this Unnamed Designer has a history of designing to entertain the user rather than technical perfection, and here he *aimed* for non-zero 2nd harmonic.
Attached Images
File Type: gif NP-ACP-2019.gif (14.1 KB, 96 views)

Last edited by PRR; 5th November 2019 at 05:43 AM.
  Reply With Quote
Old 5th November 2019, 06:16 AM   #28
aspringv is offline aspringv  Australia
diyAudio Member
 
Join Date: Sep 2010
Location: Sydney, NSW
Beefcake: a beefed-up BA-2
Maybe pull the doubling up of j74's?

That'd reduce your gain.
  Reply With Quote
Old 5th November 2019, 09:42 AM   #29
indra1 is offline indra1  Indonesia
diyAudio Member
 
Join Date: Nov 2010
Location: Bogor
Quote:
Originally Posted by PRR View Post
... here he *aimed* for non-zero 2nd harmonic.
Paul, I hope you are aware that negative phase 2nd harmonic will cancel some of dynamic driver distortion producing cleaner acoustic output.

Last edited by indra1; 5th November 2019 at 09:45 AM.
  Reply With Quote
Old 6th November 2019, 10:32 AM   #30
JeffYoung is offline JeffYoung  Ireland
diyAudio Member
 
JeffYoung's Avatar
 
Join Date: Jul 2017
Location: Ireland
Beefcake: a beefed-up BA-2
And of course there's the venerable Naim NAP250, with a 1K:22K imbalance. People are still arguing why Julian Vereker made that choice (or even if he knew himself)....
  Reply With Quote

Reply


Beefcake: a beefed-up BA-2Hide 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

Forum Jump


New To Site? Need Help?

All times are GMT. The time now is 09:02 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki