Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

power supply blinking... caps?

Status
Not open for further replies.

parkerbeder

New Member
I have a power supply on a sound board that will blink on and off when it is turned on, about once a second, I saw a couple caps that looked popped, and replaced them. No dice. Any ideas what might make a 110ac to 15 vdc power supply oscilate? even if i knew generally what might be wrong, i could replace a couple of different things... Thanks for any insight!

-Parker
 
I'm wondering if i'm testing the rectifier the right way, while it's in the board, if I test continuity between one of the ~ leads and the + or - lead, it will read a number for a breif instant, then go to 1 (no continuity) no matter which direction i'm going (positive or negative lead from the meter to either ~ or +/- lead. Does this possibly mean that the rectifier is out, or am I not testing it the right way? The rectifier is a gbu8j, datasheets are readily available, but I can't find one with any useful information...

thanks again for any ideas!!!

-Parker
 
Are you using the continuity or diode tester?

The continuity function on some meters uses too lower voltage for diodes to conduct so will show open circuit, even with a good diode.
 
Sorry but that's not enough information, there are lots of different makes and models of Ohmmeters, each will have different test voltages and specifications.
 
I see, it's a cen-tech model 95683, a cheapo. Is there a good way to check it with a 12 volt power supply, and an inexpensive meter such as this one?

Thanks for the replies!

-Parker
 
Why not just hook it up to the mains and very carefully measure the voltage across the large filtering capacitor?

It should read 140V to 160VDC.

It's a good idea to power it from an isolation transformer when doing this, for safety reasons.
 
I'll bet its a cap thats dead, but not bulged...

Regarding capacitors I follow two rules,

When half the total caps in bulged I test them all with a capmeter even if not bulged.

If about 70% of the caps are bulged I don't even bother to test I just replace them all.
Most times after replacing all caps, I test the ones that are not bulged and find many of them are bad but not bulged.
 
what's the best way to test a cap, if you don't have a 'capmeter'? I always used to think you could just test for continuity, if there is any, it' stoast. is that a fair assumption? There were two caps that were bulged in the supply, out of 15-20
 
I think a continuity check would work in some failures. But usually, when they fail, the two leads are still not electrically connected inside the cap.

I suppose you could try to charge them up, then measure the amount of power they dump out. Then you could find their capacitance. That's really too much work, and inaccurate.

Maybe some other guys have a good test without a capacitance meter?
 
Last edited:
You really need an ESR meter to test capacitors but it's not something you can easily build, although you could may be improvise with a signal generator and a 'scope? I've never tried it.
 
You really need an ESR meter to test capacitors but it's not something you can easily build, although you could may be improvise with a signal generator and a 'scope? I've never tried it.

ESR meter's are quite easy to build, and there are many examples on the net, the now defunkt magazine 'Television' even published two designs.

However, it IS inportant, it's high ESR that causes the problems, a 'cap meter' (whatever that might be?) is a useless test unless it's measuring ESR.

You can also test in-circuit while running by using a scope, and simply measuring the ripple on the electrolytics, high ESR means high ripple on the caps - it's really quite obvious the caps are duff.

Obviously any bulging caps are faulty, and I would recommend chaning ALL electrlytics in the PSU - been careful to replace them with high quality, 105 degree, low-ESR versions. The reason for the failure in the first place is poor quality capacitors, not specified correctly for the job.

A famous example in the UK is Sky HD satellite boxes made by Thomson, these have a HUGE failure rate, and a friend of mine checked the spec on the fitted capacitors - at the temperature measured inside the box the rated life of the capacitors is only about 12 months!!!.
 
I hate it when things are designed to fail, or fail by design. :(

The 'cap-meter' I was referring to simply measured the capacitance or the capacitor. After reading some I now realize how important an ESR meter is!

This seems to be a possible candidate? Apparently, you build the circuit, and measure the current that it puts through the meter.

ESR METER
 
Are we all in agreement that the caps are the problem, if the circuit is oscilating? If so, then I'll probably just replace all of them, I'm supposed to use this board for work this friday... :S
 
Is it possible that it's just a load requirement? I'm not sure if multimeters exhibit a relatively high impedance on power supplies, but try putting a resistor across the positive and negative sources and measure it then. Note that you'll want a higher power resistor than 1/4W ones; you can calculate the wattage you need by Ohms Law.

The same problem occured with my PSU, and I thought something was wrong. All it needed was a load to steady the voltage.

Austin
 
Last edited:
What are we saying, to tie the input leads together with a resistor? If people think that will help, i could try it, but this power supply worked alright inside this commercial sound board for two years, without exibiting this response, so I don't know if I follow what we're trying to correct...
 
Are we all in agreement that the caps are the problem, if the circuit is oscilating? If so, then I'll probably just replace all of them, I'm supposed to use this board for work this friday... :S

No we're not, as no tests have been done how could we be?.

But by FAR the largest cause of failure these days is high-ESR electrolytics.
 
I have a power supply on a sound board that will blink on and off when it is turned on, about once a second, I saw a couple caps that looked popped, and replaced them. No dice. Any ideas what might make a 110ac to 15 vdc power supply oscilate? even if i knew generally what might be wrong, i could replace a couple of different things... Thanks for any insight!

-Parker

Other than the power supply capacitors is there a possibility the power supply is OK but trying to come up under an excessive load? Maybe another component in the system has failed placing an excessive load on the power supply causing the symptom you are seeing. I doubt there is an easy way to isolate the power supply but I woud be looking at devices like the sound card output devices and circuits for something shorted. Without seeing all of this up close and personal and how the power supply is designed it is hard to say but maybe what you are seeing is by design if the power supply has an excessive load?

Just another thought beyond a power supply capacitor problem.

Ron
 
Okay, I replaced all the caps, and now it starts normally and works perfectly fine, but only for maybe two minutes, then shuts off, after 30-45 seconds, you can power cycle the unit and it will start again, and do the same thing. It restarts if it is being used, or at an idle, and it seems to work perfectly fine, while it is on. Ideas?
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top