[f] AMB on Facebook

Daniele's Beta24 thread

beta24

Re: Daniele's Beta24 thread

Postby daniele86 » March 10th, 2018, 5:17 am

I've rechecked the BJT in the Common-Mode Feedback, removing Q49 and Q50 that have the C-B joined.
All looks good (0.6 -0.7Vdc)

I've taken some measurements of this part of the board.
The huge difference with the other board are "-1.5", "-6.0" and "-17" (in the board that's working: "0.2", "-0.26" and "-12" respectively)

I'll check the other BJTs
You do not have the required permissions to view the files attached to this post.
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 » March 10th, 2018, 5:44 am

I wonder if something is wrong with CR6. The voltages you measured actually tells us that D11 and D12 are good (each have about 11V across them, close enough to 12V). But somehow a high current is being pulled through Q47 down D11, R49, D12, and most likely CR6.
User avatar
amb
Site Admin
 
Posts: 10377
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 » March 10th, 2018, 6:17 am

amb wrote:Perhaps go beyond just the CMFB section and test the BC550C and BC560C transistors everywhere else?

Did you measure that "scratch" to see if it breaks a trace or causes a short? If there is no measured problem then it wouldn't be the cause. I seem to recall it was merely a blemish on an internal plane and it's not near anything else, if I remembered correctly.


Yes, all the BC550C and BC560 of CMFB section were checked. I also changed Q47-49-51. I have some other BJTs, but not JFET. I found it on ebay, how difficult is the matching of JFET? How many JFET is necessary for a good matching (for 1 board)?

The scratch is on the botton layer, near V+ pin and seems only superficial. There is no short between V+ and V-.

I'm going to change CR6 diode.


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 » March 10th, 2018, 6:28 am

I don't think the JFETs are bad, but even if they were, you should not be buying any of them from eBay, because chances are good you will be getting counterfeits. You do need a fairly large number of devices to achieve a good quad-match, because the Idss variation between devices could be a very wide (6mA-12mA), and we want to match to 0.5mA or better...

If you really need to get matched JFET quads for replacement, PM me.
User avatar
amb
Site Admin
 
Posts: 10377
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 » March 10th, 2018, 7:31 am

CR6 changed. Same sad voltage.
Later I'll check the BJT other than Common Mode Feedback
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 daniele86 » March 10th, 2018, 10:01 am

How can I check JFETs if I desolder them? is it possible?

I think the problem is always on CMFB section..
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 » March 10th, 2018, 1:25 pm

Read this article JFETs: The New Frontier, Part 1 by Erno Borbely. In there there is a section about testing JFETs which should be useful to you, if you remove the JFET to test offboard. The description and diagram is for an N-channel device. For P-channel simply reverse the power supply polarities. You can use ±12V instead of ±10V to power the circuit if it's more convenient, or use two 9V batteries.
User avatar
amb
Site Admin
 
Posts: 10377
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 » March 13th, 2018, 5:42 am

amb wrote:I don't think the JFETs are bad..


Hi Ti!
So what do you think about this problem? I changed CR6 and also Q47-49-51. Do you think the problem is always here in the CMFB section?

Can I test the CMFB section without V+ and V-? Only with V + reg and V-reg from S22.

Another question, suppose that I made an error during installation and I mixed the JFET so now they are no longer matched, could cause this problem?
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

Daniele's Beta24 thread

Postby amb » March 13th, 2018, 8:42 am

Without the V+ and V- supplies the global feedback loops will be inoperative. It’s useless to test in this manner.

Grossly mismatched JFETs can indeed cause problems. You may have read my description of the β24 topology akin to a hovering quadcopter, where the four quadrants of the amp circuit hang in balance. If the JFETs are badly mismatched, then the circuit may not be able to be re-balanced by adjusting the trimpots and “fall over”.
User avatar
amb
Site Admin
 
Posts: 10377
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, 12:34 am

I've tested some more components with my tiny amount of free-time:
Desoldered BJTs (Q50 Q52) hFE are inbound of datasheet specifics and nearly close each other

Moreover punctual JFET Idss values are:
Q41: 770
Q42: 804
Q43: 760
Q44: 780

I don't think their wrong... so next days I'll check hFE of remaining Q45-Q48, and change Capacitors.

Could it be useful to check current flowing thought CR6 CR5 desoldering a pin and connecting MMD in series?

Thanks a lot.
daniele86
 
Posts: 164
Likes: 13 posts
Liked in: 11 posts
Joined: June 28th, 2012, 11:53 pm
Country: Italy (it)

PreviousNext

Return to β24 fully-differential power amplifier

Who is online

Users browsing this forum: No registered users and 20 guests

Quick Links

AMB audio DIY resources
AMB audio shop

cron