PS can caps, replace or not?

Raul

Active Member
I am finishing the restoration of my Stromberg-Carlson ASR433 (the one that had voltage problems, subject of a thread two weeks ago), I replaced nearly all resistors, coupling caps, all small electrolitycs and also two of the four electrolytics of the can cap (when I was still looking for the cause of the voltage imbalance).
Now the amp works fine and the only original caps remaining are the 20uf electrolytics of the can cap.
Most people say that it's better to replace them all anyway but I am tempted to leave them alone for the following reasons:
- voltages in the amp now are ok and the two capacitors have voltages very close to shematics 312 (315) and 260 (265)
- I measured them with an ESR tester and they measure .99 and 1.1 while two new 20uf 450V measure around 2.2 so worse than the old ones (also the two 10f that I replaced in the PS had very low ESR and probably were still working well)
- the chassis is very shallow, space of work is very limited and to put two new caps there is a bit of a pain in the @ss
What would you do?
Thanks
 
I have found that some electrolytic caps that test OK for ESR do not pass the (more rigorous) leakage test at max working voltage on my Sprague cap checker, so I use ESR for screening and rely on the Sprague leakage test to replace or not. If you're without a traditional cap checker, I'd replace the remaining ones just based on the fact that all the others were bad and then, as Fuxtor says, rest easy knowing the ps is good.
 
High voltage electrolytics need to be tested by a high voltage cap tester.
They may not leak at a few volts but will at 400vdc .
My experience is that the first power supply cap is the one that fails first and then on down the line . It sees the highest ripple and works the hardest.
 
^^this. If some of the caps in the can have failed, the rest probably are not in excellent shape. I'd change them just for peace of mind.

Actually no cap in the can have failed, I had replaced two of them (10uf) during troublesoothing for wrong voltages but they weren't the cause, voltages were exactly the same after replacing them (and their ESR were very good). The cause of the wrong voltages were two resistors of wrong value (coming directly from the factory as they were also in the schematics glued on the bottom of the amp but different from the Sams schematics and from the amp of my friend Charlie, probably mine was an early version still not bug free. After changing these resistors all volatges became correct.
 
It's not the voltages you have to be worried about with bad caps. When power supply


capacitors get dried out or become electrically leaky, they tend to bog down the power transformer which can lead to over heating and/
or failure.
 
It's not the voltages you have to be worried about with bad caps. When power supply
capacitors get dried out or become electrically leaky, they tend to bog down the power transformer which can lead to over heating and/
or failure.

As I've told Raul, I've sometimes left original caps in under those circumstances. I do however have an old Eico cap checker that I rebuilt so I also check the capacitance and leakage. I know he just got something that will probably check capacitance better than his multimeter but I don't think it will check for leakage.

Is there a way to check leakage without a specialized cap checker? If they're not getting hot and the voltages are as they should be, does that tell you anything about leakage?
 
I do not know of another way to test leakage at the operating voltage .
If your amp is quiet, and the can cap does not show any visible signs of leaking or swelling, the power transformer is not warm , the current is not excessive(fuse does not blow) your power supply caps are probably still functional.
That does not mean it will stay that way, it may or may not .

I always power up new to me old amps on a variac with an in-line ammeter to monitor the current as I ramp up the voltage, most good amps pull about 1/2 to 2/3rds the rated fuse amperage. If a power supply cap is leaking you’ll see excessive current.
 
I always power up new to me old amps on a variac with an in-line ammeter to monitor the current as I ramp up the voltage, most good amps pull about 1/2 to 2/3rds the rated fuse amperage. If a power supply cap is leaking you’ll see excessive current.

If a PS cap is drawing excessive current would you expect the voltage for the section of the circuit that it supplies to be significantly lower than normal?

Here's the manual for the Eico Cap tester I use. Perhaps someone can suggest a way to check for leakage based upon its design.
 

Attachments

  • Eico 950b Cap Tester Manual.pdf
    784.1 KB · Views: 6
The EICO instructions for leakage testing electrolytics are pretty clearly stated under "Leakage Tests." One thing, though: the checker will apply hundreds of volts to the electrolytic cap to determine leakage level (using the 'magic eye' tube readout). You cannot have the cap being tested still in the circuit, as the high voltage will also be applied to the other components in the circuit. You need to un-wire the sections being tested, then re-wire them if they pass. Also, as the instructions say, be SURE you get the polarity right on the hook up or you can blow up the cap reversing the voltage. Be sure you are comfortable working with high voltages before using this tester and be sure it is in good working order before use and be sure to carefully follow the instructions.

Finally, be SURE to discharge the capacitors after testing as the instructions say---they will hold the test voltage unless you do and will deliver a nasty shock if you get across them!
 
I've had old caps that worked fine for a while, then failed without any warning. The radio on my desk at work got daily use for probably 3 years, one day I turned it on and it hummed. Original 1958 filter caps, they died around 2015. I had honestly forgotten that I had not rebuilt that radio otherwise I'd have never run it like that.
 
I think I misunderstood your post.

No, I think your answer was on point - that if you measured current with an ammeter it would give you a pretty good idea if the cap was leaking. My reply was a bit incomplete. I should have worded my question about voltage: If you don't want to have to insert an ammeter into the circuit, could you tell if there is leakage if the voltage for that section is significantly low? Since with excessive current draw you might expect voltage to be low. Of course, the question then is how do you define 'significant'?

The EICO instructions for leakage testing electrolytics are pretty clearly stated under "Leakage Tests." One thing, though: the checker will apply hundreds of volts to the electrolytic cap to determine leakage level (using the 'magic eye' tube readout). You cannot have the cap being tested still in the circuit, as the high voltage will also be applied to the other components in the circuit. You need to un-wire the sections being tested, then re-wire them if they pass. Also, as the instructions say, be SURE you get the polarity right on the hook up or you can blow up the cap reversing the voltage. Be sure you are comfortable working with high voltages before using this tester and be sure it is in good working order before use and be sure to carefully follow the instructions.

Finally, be SURE to discharge the capacitors after testing as the instructions say---they will hold the test voltage unless you do and will deliver a nasty shock if you get across them!

Yeah, the Eico tester has come in handy. It does have its limitations (no in circuit testing, unlike an ESR meter) and you must learn to follow the procedures and observe safety rules. I told Raul that he should look around for one. I posted the manual in the hope that someone here might be able to look at the schematic and figure out a way to check for leakage without using a dedicated tester.
 
Well, you'd need to rig up a source of measurable, variable DC voltage up to 500V and install a meter or other device to measure the DC leakage current (current draw) at the desired working voltage--pretty much what the cap testers do. The testers, of course, also have bridge circuitry to measure capacitance value and other parameters, plus the switching to handle different test ranges. My Sprague TO-5 uses meters instead of a 'magic eye' tube.
 
My cap tester uses a neon lamp and some resistors to indicate leakage. As the leakage increases, the bulb flash rate goes up until its just steady. There is a chart in there equating blink frequency to how serviceable the caps are. I considered refitting it for a meter but its in nice, original condition and I don't want to hack it.
 
If he has an adequate multimeter he could remove the fuse from the amplifier and connect the meter in series at the fuse holder. Set the meter to measure current , probably 10a range. If it’s a tube amp I would remove all the tubes except the rectifier tube . Turn on the amp and measure the current . It should be very low except for the rectifier which you will look up how much current the heater draws. This should give you some idea if you have a severely leaking power supply cap.

You could disconnect the b+ from the rectifier and insert the multimeter set to measure current . Put the meter in series with the B+ with all tubes removed except the rectifier and measure . This will give a better idea of any leakage as you won’t be measuring the filament current . If all looks good install the tubes and measure. Calculate the current draw of all the tubes and calculate the difference. These are just ideas
 
I've done the meter between the rectifier and caps before, but honestly if its old enough that I'm concerned about the condition of the caps I generally prefer to just replace them and not have to bother with it. I do own a cap tester but I use it very rarely.
 
Back
Top Bottom