[f] AMB on Facebook

β24 on the move

beta24

Re: β24 on the move

Postby cobretti » August 30th, 2015, 6:56 am

more
You do not have the required permissions to view the files attached to this post.
User avatar
cobretti
 
Posts: 602
Likes: 18 posts
Liked in: 23 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: β24 on the move

Postby amb » August 30th, 2015, 9:56 am

With 1.5pFG feedback compensation caps, you should still have about 1MHz bandwidth. What's the bandwidth of your isolation transformer? Also how did you float your scope? With a cheater plug?
User avatar
amb
Site Admin
 
Posts: 10381
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: β24 on the move

Postby cobretti » August 30th, 2015, 10:57 am

I think I confused you.
I isolated AC main power. I connected scope through AC 120v/120V hospital grade isolation transformer. I modded this transformer by removing ground wire from the receptacle ground prong. I didn't use any input signal transformer if that's what you meant.
I just can not get faster rise time. If I speed the GEN up, then I have almost triangle wave around 1MHz.
User avatar
cobretti
 
Posts: 602
Likes: 18 posts
Liked in: 23 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: β24 on the move

Postby amb » August 30th, 2015, 2:21 pm

Ok. :)
User avatar
amb
Site Admin
 
Posts: 10381
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: β24 on the move

Postby cobretti » November 23rd, 2018, 4:23 pm

After a few years of trouble free operation, I just had a failure. A weird break down, when switching to RCA input, the amp went to 300kHz oscillation about 10Vp-p. First I thought that something in the feedback has blown. Obviously, I was looking for semiconductor fail that was the first thing that come on my mind but measuring all voltages and comparing symmetry of branches didn't show any obvious problems. Regulated and unregulated voltages checked OK. No DC on the output, no burned smell. No cold parts, no overheated parts either. Then I measured 0.95V on one of the Caddock resistors. The ohmmeter measures 95 Ohm. It was pain in the neck to troubleshoot it and find that bastard(my neck really hurts from looking down the board :) ). I am glad I implemented DC protection circuit to my b24 amplifier that safely disconnected my speakers. I would never expect such a failure. Certainly not from passive power component like that.
You do not have the required permissions to view the files attached to this post.
User avatar
cobretti
 
Posts: 602
Likes: 18 posts
Liked in: 23 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: β24 on the move

Postby amb » November 23rd, 2018, 9:19 pm

The question is what happened to cause that oscillation (which is probably what blew the Caddock resistor). Since there was no DC offset problem, why would the protection circuit make any difference?
User avatar
amb
Site Admin
 
Posts: 10381
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: β24 on the move

Postby cobretti » November 24th, 2018, 7:29 am

I was connecting-disconnecting couple of my components. Swapping from XLR interconnects to RCA cables. Left channel RCA cable was already plugged in my preamp to β24. When connecting my right channel RCA cable to amplifier, the dangling RCA barrel accidentally touched one of the speaker terminal. So right side of the amp output got 0V-GND from +-36V regulated PSU through the left channel RCA cable in through the preamp. It was just a few milliseconds hit. Whether the resistor was partially degraded and then I finished it with the touch, I don't know. 30W power resistor should definitely survive it. Then as I was toggling between XLR and RCA inputs, the amp goes into oscillation when RCA input is selected, negative XLR pin connected to ground. I didn't see any oscillation when XLR input selected. Oscilloscope was isolated from GND when probing around.
User avatar
cobretti
 
Posts: 602
Likes: 18 posts
Liked in: 23 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: β24 on the move

Postby amb » November 24th, 2018, 8:47 am

Ok, so a mishap caused a right channel output terminal to be shorted to ground,. Even if it’s only milliseconds in duration, it probably caused enough current flow to damage the output resistor(s). The resistor probably saved the output MOSFETS.

I don’t know what caused the subsequent oscillation that occurred only when RCA input is selected, but in neither situation was there a DC offset, so its dubious that a DC protection circuitry would have helped.

The moral of the story? Don’t be messing with the cables when the power is on.
User avatar
amb
Site Admin
 
Posts: 10381
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: β24 on the move

Postby cobretti » November 24th, 2018, 11:10 am

:D
I just checked all steps in setup instructions and everything looks good. I temporarily air wired 5 of 1.0ohm 3W resistors in parallel to replace the damaged Caddock. All trim pots and voltages looks good. I didn't bother with 22ohm in series unregulated power resistors. The only thing I noticed is that output stage was biased to 250mA or so, it reads 50+mV on Caddocks. Instructions say adjust to 40mV. I wonder if overall long time heat stress could contribute to the Caddock failure. The heatsink gets really hot after 2-3 hours of idle run. I didn't measure the temperature but I can still hold my hand on it. I also checked with my low THD oscillator (since is isolated from the ground) , 0.5Vrms input gives me 10Vrms output into 4ohm resistive load I was sinking 25W with no problem.
User avatar
cobretti
 
Posts: 602
Likes: 18 posts
Liked in: 23 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: β24 on the move

Postby amb » November 24th, 2018, 1:58 pm

If you can still touch the heatsink without burning yourself, and there isn't a sudden and extreme high current surge, then the resistors are not in danger of blowing out. If you bias the output stages to 200mA as specified, at quiescent the heatsinks shouldn't exceed 45°C or so on the Hifi2000 Dissipante case.
User avatar
amb
Site Admin
 
Posts: 10381
Likes: 175 posts
Liked in: 482 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

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

cron