5V Rectifier Intermittent Due to Heat?

[clismi]

I'm repairing a Tektronix 468 oscilloscope and believe I have an intermittent 5V bridge rectifier.

After the scope is off for days or hours and then powered up, I sometimes have the voltage at the output of the rectifier and sometimes I don't, but I always have the correct rectifier input. Whenever the 5V power supply is working, it always fails within several minutes. If I spray the rectifier with freeze mist or put a screwdriver blade on the rectifier (like a heat sink, I guess) the power supply will start working again (less than a minute later). When I try to wiggle the rectifier or tap it or the PCB, it doesn't affect the power supply.

The power supply uses a linear regulator with an op-amp controlling a series output transistor.

Has anyone ever seen a rectifier fail like this or should I keep looking? My experience with rectifiers has been they either work or don't. Am I missing anything? This circuit board isn't easy to get out so I can only see the component side now.

Thanks!

 

[Militoy]

There seems to be a zillion ways a semiconductor can fail - after more than 30 years in the business - I'm surprised by something I see at least once a week. Your use of cold spray indicates that you have some experience troubleshooting - and even though I don't recollect seeing a bridge fail in the way you describe - conductive materials do move around with heat. I would not hesitate to change the bridge, if I observed what you describe. You should, of course, try to make sure that the cold spray isn't affecting the downstream regulator at the same time you cool the bridge.

 

[NSklavos]

You may also try reflowing the solder on the rectifier. Constant heating & cooling on equipment that old can produce a cold (dry) solder connection.

 

[davenn]

then again there may be nothing wrong with the bridge rectifier !

its just failing with overheating due to excess current drain by some other faulty component elsewhere along that supply rail.
The tek 46x series scopes are well known for those sorts of faults. commonly electrolytic or tantalum caps cooking.
Have you actually measured the 5V rail is it indeed 5V plus/minus ~ 0.3V ?
a substantial drop in voltage would indicate a faulty component. see if you can isolate sections of the scope that use that rail and see what effect that has on the bridge temp.
this is what I had to do with my Tek 465B scope

it would be interesting to know the current drain on that rail ?

Dave

 

[clismi]

5V Power Supply Measurements

Those are some interesting ideas. Thanks for the help.

An interesting development--the more this scope is on, the better it works. That sounds like an electrolytic capacitor to me.

As for the 5 V supply, when it is working, I measure 4.93 V. The manual says it should read between 4.92 and 5.09 volts. After a few minutes, the output drops to about 4.80 V. So it is a little on the low side. The other voltages are within range.

I haven't measured the ripple on the power supplies yet, but I think I'll do that and measure the current as suggested. I don't see a spec for current draw though. That would be really helpful.

I like the idea of looking for bad electrolytic capacitors, and I have an Sencore LC53 that I can test them with.

Again, thank you so much for the help and suggestions. I have all day tomorrow, Labor Day, to work on the scope, and am looking forward to some fun. I'll check for bad solder joints on the rectifier first.

 

[clismi]

More info

The filter capacitor on the 5V power supply had high leakage current. I reformed it on a Sencore LC53 and the leakage test looked good after that. Now the oscilloscope's 5V power supply stays on for a long time.

The output of the rectifier has 60 cycle ripple, which means one side of it is open, right? I'll order a rectifier and a capacitor since it will cost less than $4 for both.

I hope that solves my other problem. With the scope's coupling on GND, the displayed waveform looks good when you first turn it on, and the 5V supply is within tolerance at 4.99 V. After about 10 or 20 minutes, I get a 60 cycle ripple on the scopes displayed GND reference, it eventually increases in amplitude, the 5V supply goes out of tolerance (4.0 to 4.2 V), and eventually the rectifier has no output. I suppose it is overheating. After a couple minutes the 5V output comes back and the process repeats.

What do you think?

 

[davenn]

The filter capacitor on the 5V power supply had high leakage current. I reformed it on a Sencore LC53 and the leakage test looked good after that. Now the oscilloscope's 5V power supply stays on for a long time.

The output of the rectifier has 60 cycle ripple, which means one side of it is open, right? I'll order a rectifier and a capacitor since it will cost less than $4 for both.

I hope that solves my other problem. With the scope's coupling on GND, the displayed waveform looks good when you first turn it on, and the 5V supply is within tolerance at 4.99 V. After about 10 or 20 minutes, I get a 60 cycle ripple on the scopes displayed GND reference, it eventually increases in amplitude, the 5V supply goes out of tolerance (4.0 to 4.2 V), and eventually the rectifier has no output. I suppose it is overheating. After a couple minutes the 5V output comes back and the process repeats.

What do you think?

the bridge could possibly be faulty or it could still be dried out electros no longer doing their smoothing job and allowing the ripple to show up. If its an easy bridge to replace, go for it, it cant hurt

remember the scope is prob 30 or so yrs old a large portion of the electro caps are probably drying out. Personally I dont shag around with them I wouldnt bother trying to reform them either, it it does work, prob only a very short term fix.
Swap them....Swap them....Swap them.... hahahaha
did I say it enuf ? Swap them.... it will save a lot of heartache down the track

cheers
Dave