What is typical failure of smoothing caps in power supplies

[peterlonz]

Hi,
I am attemting the repair of a JVC AX-211 amplifier circa mid80's.

I have identified & removed from the Circuit board two 6800uF, 50V caps which appeared to have leaked "brown stuff" onto the board.
They appear not swollen or obviously defective otherwise.
My DMM will measure only to 2000uF so I can't simply test.

I have just read the "dried brown stuff" which I thought was leaking & dried electrolyte, may well be discoloured glue & so the caps possibly are OK.

My questions are:
1) How can I test these caps properly.
2) I gather sudden failure is not typical, but can someone advise what usually happens in the gradual failure process & at what stage do you notice something wrong with the appliance.
3) What are the downstream implications for other components - typically.
I ask becasue apart from one transistor that has become "unsoldered" I can find nothing wrong.

BTW no manual is available, I have checked.
Thanks

[vtech]

While there is data for life expectancy of any device, there is no specific time line for capacitor failure. In your case, as you have mentioned, it is very likely the glue that was used to keep the caps in place. Over the years it tends to turn in to a crusty brown substance and usually is not a problem as long as it is not touching a copper trace. I found a picture which is probably identical to yours?
As far as testing a high value electrolytic, short of having a capable tester, there is no clear cut way.
Beside obvious physical signs (bulging/leaking etc), you can always measure the DC rails or better yet use a scope to look for ripple on the line.
In case of a gradual failure, as the cap dries out, the amount of ripple begins to increase which in audio circuitry translates in to a typical "hum" out of the speakers and possible output failure which would instantly blow the fuse. Also, it has a direct effect on DC readings where it begins to decline.
In general, since the main reason for the filter caps are to provide similar-value plus&minus rails, if you are able to measure the DC rails, it is a safe bet that your caps are OK.
Exactly what is the problem with the unit? Is it a power issue?

[peterlonz]

I certainly did not expect someone would post a pic of my unit, yes it looks 100% identical except for the colour of the ribbon cables.
I suspected my caps were OK but only after I read about the glue. The fuse is intact BTW.

Why are smoothing caps glued down, surely they don't generate mechanical vibration & is it likely they will be damaged by re-insertion, if so I might mount them remotely on flying leads so that I can better control the heat of soldering.

I don't have an osciloscope but if I knew exactly where to measure I could get the rail voltage & compare with the tranni output, I think that's what you are suggesting?
The reason I began investigating is that the unit failed to produce any sound when provided with RF signal inputs.

I confess I'm not finding it easy to follow the circuit so I'm unsure where to attempt DMM measurements. Also I'm a bit unsure about the claimed "Two amp block configuration", if there are two amps I can't make them out.
Is it fair to describe this configuration when the tranni is merely cetre tapped?
Can you tell me anything else about this unit, maybe you can point out wher I need to make diagnostic measurements.
Again many thanks, hope you have the time to comment a little more.

[vtech]

The main reason in gluing the bigger caps is to prevent movement while the PC board travels through the wave solder machine in production.
There is no need for you to have them mounted remotely. Isn't there a removable bottom cover on that unit?
You can measure from the ground to each side of the caps which will give you the rail voltages and more likely all right. As far as RF signal, do you mean trying to hook it up to a tuner of some sort? When you say RF signal, there is no way this amplifier can amplify an RF signal since it is designed for Audio not RF(radio frequency).

May want to do a search on amplifier circuits to get familiar with
...I can give you more info but my job keeps me away from here so keep checking back

[peterlonz]

Vtech,
I really appreciate your input, thanks so much.
No there is no bottom access panel on this unit so I had to do quite a bit of dismantling to access the bottom of the PCB & even now the rather limited length of the ribbon cables is restrictive - welcome to the world of electronics repair I suppose!!!
Getting the caps off the PCB was no mean feat, the solder was slow to move in to the copper braid & breaking the glue joint took a fair bit of force.
If there is no reason not to, I will remount on flying leads until I know whether I can get the unit operational.
Thanks for the correction re RF, don't know what I was thinking, & yes I tried a tuner input & also a tape machine, but you always have that doubt is the signal really there or have I ballsed up the connection somehow.
I have beeen busting to ask you how you come to have such a good pic of the guts of this unit & I'm still interested about the claimed Two block configuration.
Don't let work get in the way too much, we only do it to pay the bills right?
Pete

[rip1980]

Wasjust looking over vtech's shoulder andfound a generic schematic to a power supply that should be similar to your unit's.

http://www.errock.co.uk/unreg%20psu%20high%20res.jpg

I don't want to be rude and hotlink. I'd start following the chain, input side of the transformer, output, diode bridge,rails....then (guessing) get a generic schematic if an AB audio amplifier...

If there is any doubt on the caps, just replace them, they're pretty inexpensive. Failure mode can be open or short, or some where in between. Maybe it's just me, but shorting seems more common.

The ramifications are usually UPSTREAM, damaging transformers or rectifiers by excessive current draw.