[f] AMB on Facebook

Daniele's Beta24 thread

beta24

Re: Daniele's Beta24 thread

Postby amb » April 7th, 2018, 12:47 am

What unit are those "Idss" numbers in? They sure don't look like mA. Instead of hooking up a DMM in series with parts by lifting a pin, it might be more useful to measure all voltages in the circuit and compare against an operating point PDF. I'll send you a copy. All voltages marked are relative to ground.
User avatar
amb
Site Admin
 
Posts: 10379
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: Daniele's Beta24 thread

Postby daniele86 » April 7th, 2018, 12:58 am

Sorry

Moreover punctual JFET Idss values are:
Q41: 7.70mA
Q42: 8.04mA
Q43: 7.60mA
Q44: 7.80mA
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

Re: Daniele's Beta24 thread

Postby amb » April 7th, 2018, 1:12 am

Those are pretty good Idss matches.
User avatar
amb
Site Admin
 
Posts: 10379
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: Daniele's Beta24 thread

Postby Deej49 » April 8th, 2018, 11:47 am

Good luck... I am about to start putting my Beta24 together. A fabulous design. I had MBL 8011AM monoblocks and I am optimistic for Beta24 to be a strong competitor.
Deej49
 
Posts: 2
Likes: 0 post
Liked in: 0 post
Joined: December 14th, 2017, 1:58 pm
Location: Denver
Country: Australia (au)

Re: Daniele's Beta24 thread

Postby daniele86 » April 27th, 2018, 3:27 pm

Hi all,
I've worked on my boards in the last few days, that's what happened:

    After checking JFET and re-soldering to board-1, it starts working, even after a lot of checks it was an unbelievable cold joint. :oops:
    So I completed what I hope it's a good Power-up initial setup (but the 22R10W are on).
    The second board didn't warm up to much (except for the 4 diodes D7 D8 D9 D10 as mentioned in Ti assembly instruction, so no panic)
  • configuring Voltages of 4V across R7 (which gives VR9=4.1V VR8=3.99V VR10=4.15V) and
  • configuring 133mV across R11 and 132mV (not so easy to achieve because it needs time to stabilize and seems also that changes during the board warm up)
  • MOSFET bias setup:
    I have now that with VR2 to full clockwise, VR45 reach 27Vdc max. The voltages across R45 R46 don't go up more than those 27mV.
    And if I continue with the setup and rotate clockwise also VR3, VR45 and VR46 fall down to 1mV
    VR2 measured off-board is 209 Ohm at his maximum rotation,
    I've got VO+O- = 350mV,
    Vreg stable at 34.5Vdc
    but Vunreg+ falls from 41Vdc to 32Vdc! What? :o
  • If I turn back fully counterclockwise VR2 (and VR3) I see that
    voltages around R17 R18 R21 R22 are more or less correct, Q9 to Q10 are very hot
Could it be a problem similar to Pietreh? https://www.amb.org/forum/initial-set-up-issue-beta24-board-t3689.html
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

Re: Daniele's Beta24 thread

Postby amb » April 28th, 2018, 4:12 am

You may need to also adjust VR4 at the same time as VR2 and VR3 to make it possible to achieve proper bias and get near zero DC offset. Tweak VR2 and VR3 until there is some voltage increase across the output Caddock resistors, and note the DC offsets on each side of the amp. Tweak VR4 a bit to balance the bias currents and DC offsets if necessary. Then repeat the above several times until you could reach the target output stage bias and low DC offset. You may or may not be able to achieve all of the above, depending on the specific issue, but it's a good start.
User avatar
amb
Site Admin
 
Posts: 10379
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: Daniele's Beta24 thread

Postby zhanqian » May 1st, 2018, 9:15 am

Hi Daniele,

I also encountered the MOSFET bias setup problem, I had try many ways, in the end I solved it by measure and match the Idss of Q1-Q4, Q41-Q43. You can see my post here https://www.amb.org/forum/ye-beta24-build-t3739-20.html, I think you can try pick the more matching ones for Q1-Q4 from those JFETs. Hope it may help with your setup problem.
zhanqian
 
Posts: 27
Likes: 14 posts
Liked in: 10 posts
Joined: October 24th, 2017, 8:02 pm
Location: Shenzhen
Country: China (cn)

Re: Daniele's Beta24 thread

Postby daniele86 » May 4th, 2018, 2:54 pm

amb wrote:You may need to also adjust VR4 at the same time as VR2 and VR3 to make it possible to achieve proper bias and get near zero DC offset. Tweak VR2 and VR3 until there is some voltage increase across the output Caddock resistors, and note the DC offsets on each side of the amp. Tweak VR4 a bit to balance the bias currents and DC offsets if necessary. Then repeat the above several times until you could reach the target output stage bias and low DC offset. You may or may not be able to achieve all of the above, depending on the specific issue, but it's a good start.


I've tried with small steps of 5mV at a time but before reaching the target, maintaining low offset, again bias top at 26-27 mV (and Vunreg falls from 40.5V to 31V)
Is it normal having such under voltage between G and Vunreg (both +&-) under this conditions? All BJTs and diodes are tested ok, even Res. are of the same value of the other board.

zhanqian wrote:Hi Daniele,

I also encountered the MOSFET bias setup problem, I had try many ways, in the end I solved it by measure and match the Idss of Q1-Q4, Q41-Q43. You can see my post here https://www.amb.org/forum/ye-beta24-build-t3739-20.html, I think you can try pick the more matching ones for Q1-Q4 from those JFETs. Hope it may help with your setup problem.


I've read carefully your posts and get out that values from my system [mA]:

Q1 = 7.89
Q3 = 7.99
Q2 = 7.60
Q4 = 7.71

Q41 = 8.02
Q43 = 8.03
Q42 = 7.74
Q43 = 7.95

I'll look deeper and maybe swapping Q1-Q3 or Q2-Q4 as mentioned around here.
thanks
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

Re: Daniele's Beta24 thread

Postby amb » May 4th, 2018, 3:57 pm

If you still have the 22Ω current limiting resistors on the Vunreg rails, then it's normal for the Vunreg rail voltage (as seen by the amp board) to drop as you turn up the bias.
User avatar
amb
Site Admin
 
Posts: 10379
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: Daniele's Beta24 thread

Postby daniele86 » May 13th, 2018, 2:10 pm

Switched Q2-Q4.
Another quadcopter flying... 8-)

2 people like this post.
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

Previous

Return to β24 fully-differential power amplifier

Who is online

Users browsing this forum: No registered users and 0 guests

Quick Links

AMB audio DIY resources
AMB audio shop