Intermittent Bridge Diode ?

[Goldie99]

I've been looking recently at a power amp for someone, that developed an intermittent fault when it would randomly switch into protection mode, sometimes for a few seconds, sometimes longer, and with no obvious reason. The same symptoms are apparent with no input or speakers connected.

After quite some time, I've finally found an RC202 bridge rectifier (2A, 100V) supplying the protection circuits that appears to be the issue, although none of the internal diodes seems to be shorted or open, at least not on a DMM. Replacement with an new 2A, 200V Vishay bridge rectifier seems to have cured the problem, although I'm confused as to what issue the RC202 might have.

Are intermittent, transient, faults ever apparent with bridge rectifiers ? I've never read anything like that, only failure with an open or shorted diode.

[Jez1234]

It's a rare fault but I have seen it. I had a ReVox PR99 Mk III palmed off on me which had an intermittent fault whereby it would stop and appear to lose all power as if the IEC lead were intermittent. It could work for 6 hours without problem or it could happen at ten minute intervals... really intermittent! After several attempts at repair I eventually traced it to a W01 1.5A bridge rectifier on the tape transport/reel motors control board. On the bright side though I do wonder if this machine had little use and hence little head wear etc due to this!?? As sods law predicts another unrelated fault appeared about an hour after fixing this one. It no longer auto stops at end of play/fast rewind etc. Probably a simple fault but round tuit supplies are low and I could live with this fault... whereas it suddenly dropping out of record and all the lights going out right in the middle of a set by a local band which I was recording was rather more serious!
As I've not even had reason to power it up in five years it's no hurry...

[Sideband]

Not unknown! I've had this happen, admittedly many years ago when a bridge rectifier would suddenly output a low voltage and the item (a tape recorder in this case) would suddenly stop with the mechanism solenoid just clicking everything into stop. Press the play button and away it would go again. When I caught the fault I also noticed the lights dim dramatically and then revert to normal. At first I thought it was a short but then realised that the input power wasn't increasing, it was dropping. After much head scratching, and monitoring of input and output voltages I eventually found the problem....one of the diodes in the bridge was going open. A new bridge cured the problem.

Some years later, I had a similar problem but this time on a portable TV which would give an intermittent small picture with a massive hum bar......

[factory]

There is an eevblog video on the HP 1740A oscilloscope where a diode bridge went intermittent, it's all on youtube. https://www.youtube.com/watch?v=7ocancoZ02Q

David

[Goldie99]

Thanks guys - it's good to know it does happen, and it's an interesting video link too David.
I was actually monitoring the +ve from that rectifier when I noticed it quickly drop, and recover within a couple of seconds - that coincided exactly with all the protection relays kicking out & back in (that was after several hours down various 'rabbit holes' as Aussie Dave puts it...).
Having watched the video, I unfortunately now have the (OCD ?) question of whether it was the rectifier, or the ca. 25 yr. old soldering...?? the attached pics show the solder joints of the original rectifier from above (the underside was fine). As soon as I get a chance, I'll wire up the old rectifier and see if I can get it fail.
In the mean time, thanks again, I'm now much more comfortable that the original problem is indeed resolved. The amp is also now happily sitting in 'Normal' mode, and working properly, since Saturday morning.



Cheers, Alan

[Sideband]

That looks like double-sided print with plated through connections. Highly unlikely it was bad soldering. If any of the pads had failed, that would have been obvious when the rectifier was removed. My bet is was the rectifier.

[MotorBikeLes]

As just said, double sided print. Don't waste time, and risk damage to board. Sharp snips, cut each leg and throw away the component
Apply soldering irion to the side not shown in your photos, feed in solder from the side we saw, and when it flows well, the leg will probably fall out. Don't be "casual", double sided print is easily damaged if not treated carefully and correctly.
Les.

[Goldie99]

Yes, it's a double sided, thru' plated, epoxy board (late '90's), but with minimal if any solder pad on the top surface. I was well aware of the potential for damaging the board, but it fortunately went fine with just solder sucker, braid, and plenty of flux. Good point on the part leg sniping & removal though, where they're accessible it's a good idea - a lot of the parts on this board have the legs folded onto the pads, making it even more critical (luckily not the rectifier).
I agree, I don't think the soldering was the problem, I went all over that board 'tapping' almost every part, and inspected every joint on the solder side under magnification. I didn't find any issues there.

Alan

[lightning]

l had a PA mixing desk that blew the fuse at switch on. The bridge rectifier had a shorted diode but only under load. It tested fine on the diode test facility on my meter.

[Goldie99]

I've actually got the suspect RC202 sitting next to me on the desk at the moment - I ran it for an hour or so at about 12V, 120mA output, and didn't see anything out of the ordinary (monitoring output voltage & current). It's running now at about 250mA, and is maybe beginning to look somewhat suspect (?); the output voltage will suddenly dip for a few seconds, not massively, up to ca. 200 - 300mV, but then suddenly 'recover'.
I'm only using a small encapsulated transformer, and 250mA is about its limit. I'll see if I can find a bigger alternative. The RC202 in its original location in the power amp was running closer to 350 - 400mA, so could well be 'worse' then.