Classe DR-25/25 repair - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

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 24th August 2015, 07:35 PM   #1
ungie is offline ungie  Canada
diyAudio Member
 
Join Date: Dec 2002
Location: Ontario, Canada
Default Classe DR-25/25 repair

Hi all,

My Classe 25 (probably a VERY early one given that the PCB designators match the DR-25) has developed a problem on one channel. Resistor R17 (R19 on my amp) on the PCB found in this package has burned:

http://bwgroupsupport.com/downloads/...el%2025-TM.pdf

Because this is a voltage dropper for the front end of the output stage, I'm hoping that the fault lies before the output transistors as they look like they would be a real chore to get out for individual testing! The other channel works fine. I have limited experience with discrete solid state power amps, so any suggestions on where I should start looking would be most welcome.

Thanks!
  Reply With Quote
Old 3rd March 2017, 02:54 PM   #2
ungie is offline ungie  Canada
diyAudio Member
 
Join Date: Dec 2002
Location: Ontario, Canada
Bumping this thread as I want to get back to fixing this amp! I would appreciate it if someone could point me in a direction to start looking at what may be causing this issue! In-circuit checking of all diodes and transistors looks good.
Thank you for your help!
Attached Files
File Type: pdf Classe25.pdf (76.2 KB, 108 views)
  Reply With Quote
Old 3rd March 2017, 04:28 PM   #3
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
I'd suspect C15. 100 V electrolytic, >80 V rails, seems a bit tight if you ask me (depending on what they used). Do you find any measurable shorts after the dead resistor? Anyway, high-voltage breakdown faults can be hard to find.

BTW, I think Q6 and Q7 on the schematic are reversed.
  Reply With Quote
Old 3rd March 2017, 09:49 PM   #4
diyAudio Member
 
Join Date: Mar 2017
Location: Brighton, UK
That resistor mainly isolates the power devices from the rest of the amp circuitry. If it smoked, something to the left of the resistor is faulty. Resistance testing should turn up something rather quickly, with one probe on C15 (+) and the other at various positions in the amplifier.

If simple resistance checking doesn't turn up anything, I would continue by checking any SMD capacitors, as they are known to fail brilliantly in some of the Classe designs.
  Reply With Quote
Old 20th March 2017, 05:31 PM   #5
ungie is offline ungie  Canada
diyAudio Member
 
Join Date: Dec 2002
Location: Ontario, Canada
Quote:
Originally Posted by sgrossklass View Post
I'd suspect C15. 100 V electrolytic, >80 V rails, seems a bit tight if you ask me (depending on what they used). Do you find any measurable shorts after the dead resistor? Anyway, high-voltage breakdown faults can be hard to find.

BTW, I think Q6 and Q7 on the schematic are reversed.
Hi,

C15 is fine and so are all other capacitors on that channel. Interesting to hear your comment about Q6 and Q7...I doubled checked and absolutely they are connected as per the schematic. My early Model 25 has a number of differences from the published schematic and it looks like the earlier version of the PCB was designed to use TO-92 devices in both the Q6 and Q7 positions. The MJE350 and MJE340 devices have twisted legs to fit in the TO-92 footprint and have small heat sinks attached to them. On the later version of this amp the PCB was revised and these devices are mounted directly to the main heat sink. I wonder if they messed up originally and have MJE350 in the wrong position? So far I see no shorts to suggest why that resistor is burning...any other suggestions on where I might look? Is there any harm in lifting the VCC sides of the 33R2 resistors from both rails and powering up through a current limited bench supply without the drive and output transistors being powered?
  Reply With Quote
Old 20th March 2017, 09:34 PM   #6
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Quote:
Originally Posted by ungie View Post
C15 is fine and so are all other capacitors on that channel.
Hmm. Any signs of shorting? As I said, high-voltage breakdown faults can be tricky to find.

Quote:
Originally Posted by ungie View Post
Interesting to hear your comment about Q6 and Q7...I doubled checked and absolutely they are connected as per the schematic.
So the one that connects to R11 is Q6 and it is an MJE350, and Q7 is the MPSA56?

That makes no *******' sense though. Q6/7 basically form a cascode current source to go with the cascode VAS involving Q9/10. The fun part of a cascode is that the lower device (like Q10) can be a high-beta, low-capacitance job rated at a relatively modest breakdown voltage, while the characteristics of the upper one (like Q9) are a fair bit less critical, so you can use a more robust part there. They're even using a bootstrapped bias connection on Q9 to further reduce its influence. For a pnp, the "lower" device has to become the "upper" one, of course.

The rated breakdown voltage of an MPSA56 is 80 V (vs. the 300 V of an MJE350). Your amp has nominal +/-84.2 V supplies, it might see 140-160 V peak-peak at full power (not to mention some thermal abuse). That it has survived for as long as it did is a small miracle in itself. Even if it does not measure dead short C-E as-is, it's going to be degraded and should be replaced while correcting the transistor mixup goof.

Q7 biasing via R10 also looks extremely dodgy to me, it's relying on transistor beta to set emitter voltage, and that's definitely not how it should be done (beta tends to be quite variable). All assuming the schematic is correct. I'd rather see an LED with ~100 F in parallel (giving anything from 1.8-1.9 V for red to ~3 V below VCC+ for blue/white), biased from ground via ~47k or VCC- via 2x 47k in series or 1x 100k (observe voltage rating!), going to Q7 base via ~220 ohms.
Quote:
Originally Posted by ungie View Post
I wonder if they messed up originally and have MJE350 in the wrong position?
Seems so. Guess the schematics guy goofed and the layout/parts person took their word for it. Now how stupid is that.
Quote:
Originally Posted by ungie View Post
Is there any harm in lifting the VCC sides of the 33R2 resistors from both rails and powering up through a current limited bench supply without the drive and output transistors being powered?
Any fully DC-coupled amp like this is not going to work without its frontend. The output would go to one rail and you may end up blowing up poor little C9 and possibly damaging some of the input transistors as well. I'd just poke around with the resistance test as suggested.

Last edited by sgrossklass; 20th March 2017 at 09:47 PM.
  Reply With Quote
Old 21st March 2017, 02:40 AM   #7
diyAudio Member
 
Join Date: Mar 2017
Location: Brighton, UK
Well, you have already done resistance testing. R17 smoked because there was a low resistance to ground. Does R18 show signs of overheating? Zobel network show signs of overheating? Does R17 smoke instantly after it is replaced?

In this case I would replace R17 and R18 with 100 ohm resistors and see which parts are running warm. No load connected, of course.

What is the meaning of the 4 x 100K resistors? This arrangement is not seen in other designs. Why are the c.s. diodes D1 and D2 connected to V- instead of ground?

If you don't get any clues, you will have to bring up V+ and V- slowly starting at about 30 volts. On this type of untested and unique design, you can't assume everything will power up the way you think it does. Various transistors can saturate while the supply is still increasing - all based on variations in part tolerance. My theory is still small value capacitors short circuit. +/- 85 volts is pushing it a little don't you think?

Another way to approach this problem is with a differential probe connected accross various parts - watch the response during power-up. Sometimes you have to let something fail to direct your attention to the problem.

RA
  Reply With Quote
Old 21st March 2017, 08:20 AM   #8
diyAudio Member
 
analog_sa's Avatar
 
Join Date: Aug 2002
Location: Cascais
Quote:
Originally Posted by RestAssured View Post
What is the meaning of the 4 x 100K resistors?

Set the required ccs current by inserting as many resistors in the board as needed?
  Reply With Quote
Old 21st March 2017, 04:57 PM   #9
ungie is offline ungie  Canada
diyAudio Member
 
Join Date: Dec 2002
Location: Ontario, Canada
Yes, I have checked every capacitor in circuit, each resistor value, and each transistor for shorts. I have NOT pulled all the of the parts to check them at this point, but that might be my next step. I have also checked continuity on every circuit trace to look for broken pads or traces or bad solder joints. Output Zobel is also fine. Frustratingly everything checks out ok. I think at this point I will bring up the working channel on a variac until the rails reach +/-40VDC or so (half full rail voltage) and make detailed voltage measurements at all transistor junctions. I will then do the same with the bad channel, unless I see a huge current draw on the positive rail, or R17 starts smoking. I should note that R18 shows no signs of heat damage, but R16 and the paralleled combo of R20/R9/R7 show some discoloration. They measure fine and the working channel shows the same amount of discoloration on those parts. Once I get this amp fully working again I will look at replacing those parts with higher wattage versions and stand them slightly off the PCB.

Thanks everyone for your suggestions so far!

Quote:
Originally Posted by RestAssured View Post
Well, you have already done resistance testing. R17 smoked because there was a low resistance to ground. Does R18 show signs of overheating? Zobel network show signs of overheating? Does R17 smoke instantly after it is replaced?

In this case I would replace R17 and R18 with 100 ohm resistors and see which parts are running warm. No load connected, of course.

What is the meaning of the 4 x 100K resistors? This arrangement is not seen in other designs. Why are the c.s. diodes D1 and D2 connected to V- instead of ground?

If you don't get any clues, you will have to bring up V+ and V- slowly starting at about 30 volts. On this type of untested and unique design, you can't assume everything will power up the way you think it does. Various transistors can saturate while the supply is still increasing - all based on variations in part tolerance. My theory is still small value capacitors short circuit. +/- 85 volts is pushing it a little don't you think?

Another way to approach this problem is with a differential probe connected accross various parts - watch the response during power-up. Sometimes you have to let something fail to direct your attention to the problem.

RA
  Reply With Quote
Old 21st March 2017, 06:30 PM   #10
diyAudio Member
 
Join Date: Mar 2017
Location: Brighton, UK
Quote:
Originally Posted by analog_sa View Post
Set the required ccs current by inserting as many resistors in the board as needed?
Well of course that is the function of 4 paralleled 100K resistors - which are tied to V- in a very un-traditional way. It just is not apparent what is gained, over tying the resistors (of proper value) to ground at the point. The diodes are still providing 1.2v of bias to the ccs.

2 x 85v = 170v between the rails
4 100K in parallel = 25K total
current through diodes is (170v - 1.2v)/25K = 7mA + base currents from 3 transistors. But then with Classe, you always get innovation first, over reliability. Marketing.

RA
  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
25' DVI Cable, 25' TRS/XLR Cables, LL1574 SPDIF-AES Transformer bmwman91 Swap Meet 0 3rd January 2012 07:04 AM
Classe DR-25 transistors mrprytz Solid State 10 14th August 2011 12:11 AM
Shanling Cd-S100/ Music Hall 25 (25.2) fluorescent display source? limono Parts 2 5th August 2011 01:22 AM
cheap (230v 50hz) 160va 25+25 tranny on ebay guitar_joe Swap Meet 0 24th November 2007 10:53 PM
FS: Avel Toroidal Transformer. 1000VA 25+25 acliao Swap Meet 0 29th November 2004 09:46 PM


New To Site? Need Help?

All times are GMT. The time now is 12:28 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki