PSU paper clip trick

[educating fire]

I think my enermax 620 might be faulty. I tried the paper clip trick, green wire to ground wire and the wires produced a short fan cycle (less than 5 seconds) then it would pause for about the same length and then repeat the cycle. The other black ground wires did not produce any fan spinning. I'm just wondering if the wires that did produce a short fan spinning cycle, pause and repeat is indicative of a bad power supply? Shouldn't the fan be running full force, with no pausing???? After today, I shall invest in a power supply tester... HEadaches...


It's the ground wire next to the green wire that produces these short cycles. I hooked a piece of elwire via 4 pin molex up and the light would come on when the fan was cycling and it would turn off it stopped.

 

[[Spectre]]

Try it with a load as well.

 

[Spartacus]

I see this jumper test advised all the time and I have to say it's not a very valuable test in many cases if not most. In my many years of field work, I've seen many PSUs that will pass a test like this yet still be bad. Likewise, I've seen many PSUs that will fail this test, yet still be good.

A PSU can pass this test yet still fail under load. During the jumper test, it may even have voltages well in spec if you put a meter on it. Yet plug it into the mobo, and you get nothing. The natural assumption then for noobies is that the mobo is bad. They buy a new mobo and it still won't power up. Then they start looking at RAM, video card, RMA everything, etc. when the only problem was the bad PSU that passed the jumper test. I've seen it many times.

The PSU may also fail the jumper test if the design must be loaded in order to power up.

Put these facts together with advising a noobie to stick a paper clip in a connector and I just don't think it's a good idea.

The PSU may power up and it may not. Either way, it doesn't prove much of anything.

The load testers are a much better idea (though not always perfect either).
If you work on computers, it's a great idea to have a reliable spare to test with.

 

[educating fire]

I'm a noobie in the sense of the forums but I have some eletrical background. Enough to know what I shouldn't do. Anyway, the PSU is obviously bunk because a new one fired the system right up. Shame, I had very high hopes for the Enermax and now it has to be RMA'd.

 

[jonnyGURU]

I see this jumper test advised all the time and I have to say it's not a very valuable test in many cases if not most. In my many years of field work, I've seen many PSUs that will pass a test like this yet still be bad. Likewise, I've seen many PSUs that will fail this test, yet still be good.

A PSU can pass this test yet still fail under load. During the jumper test, it may even have voltages well in spec if you put a meter on it. Yet plug it into the mobo, and you get nothing. The natural assumption then for noobies is that the mobo is bad. They buy a new mobo and it still won't power up. Then they start looking at RAM, video card, RMA everything, etc. when the only problem was the bad PSU that passed the jumper test. I've seen it many times.

The PSU may also fail the jumper test if the design must be loaded in order to power up.

Put these facts together with advising a noobie to stick a paper clip in a connector and I just don't think it's a good idea.

The PSU may power up and it may not. Either way, it doesn't prove much of anything.

The load testers are a much better idea (though not always perfect either).
If you work on computers, it's a great idea to have a reliable spare to test with.

Zzzzzzzzzzzzzz.........

The point of the test is not to pass/fail a PSU, but to at least narrow the scope of a problem down.

For example: If a PSU doesn't even power up via the mobo switch on the motherboard, but does power up with a paper clip, then odds are the PSU isn't dead. Not to say that it "passes" because it could still fail with a load, but if the PSU doesn't even TRY to turn on when hooked up to a motherboard with only a CPU in it, you know the problem may very well be elsewhere.

 

[Spartacus]

Zzzzzzzzzzzzzz.........

The point of the test is not to pass/fail a PSU, but to at least narrow the scope of a problem down.

For example: If a PSU doesn't even power up via the mobo switch on the motherboard, but does power up with a paper clip, then odds are the PSU isn't dead. Not to say that it "passes" because it could still fail with a load, but if the PSU doesn't even TRY to turn on when hooked up to a motherboard with only a CPU in it, you know the problem may very well be elsewhere.

The point of a hardware test is absolutely to pass/fail the part.

Anything less is NOT a test.

I fully understand what people are saying about it being a cheap test when you don't have a spare. However, it's a false logic that leads to trusting results that are not trustworthy.

I could have filled my garage with bad power supplies that will pass the paper clip test over my years of field service. I used to get them on a pretty regular basis. They will all power up but they are bad.

To me, telling somebody to do a paper clip test on a PSU is like telling them to shine a flashlight on a stick of RAM to see if it's good or not. Both tests give about the same reliability.

 

[jonnyGURU]

So you still don't get my point.

And telling someone to use a paper clip to test a PSU is more a kin to telling someone to use MEMTEST to test memory. Not shine a flash light on the stick. That statement right there just shows your arrogance.

 

[Spartacus]

No, Memtest will actually provide fairly accurate results.

I was comparing an invalid test to an invalid test.

>>shows your arrogance

Ahhh, there we go.... a personal attack when backed into a corner with logic. That's your ego that can't face realizing you are arguing a false argument.

What? Admit you are wrong?
No way..... attack!

Not arrogance at all sir, experience.

Please don't paper clip test your toaster, you'll hurt yourself.

 

[jonnyGURU]

No, Memtest will actually provide fairly accurate results.

I was comparing an invalid test to an invalid test.

No, because Memtest often does pass bad RAM. Therefore: Invalid test, or at least just as invalid as a paper clip to a power supply.

>>shows your arrogance

Ahhh, there we go.... a personal attack when backed into a corner with logic. That's your ego that can't face realizing you are arguing a false argument.

What? Admit you are wrong?
No way..... attack!

Not arrogance at all sir, experience.

Please don't paper clip test your toaster, you'll hurt yourself.

"Shining a flashlight on a stick of RAM" is just a smart ass reply since you couldn't come up with a better analogy.

You didn't even bother to read my post. I never said that a paper clip would test a power supply. But if you have a SYSTEM that doesn't turn on, not even a twitch of a fan, then a paper clip on the green and black will give you some direction to go to further troubleshoot the system. If the PSU doesn't turn on with the PS_ON pins on the motherboard, but will with the paper clip, then it's most likely the fault is in between. If you're going to argue otherwise, you're arguing for arguing sake.

I never said a PSU that passed the paper-clip would also pass with a load. You seem to think I did and then use the pissy little "flashlight on the stick of RAM" argument to make your point.

 

[Providence]

I suppose I'm slightly hijacking this thread, but with this at top it's not worth making another one.

Why even jump the power supply anyway, other than to test it? I just bought a water cooling system and a guy I talked to said I should "jump" my PSU...something about always keeping the pump running or some such. Looking back I guess I didn't catch all of that.

Will jumping my PSU with a paper clip help me in any way as well? Or did I just completely misunderstand?

 

[Spartacus]

Ok, point taken.

If you are testing for a bad power switch, then the paper clip test may or may not tell you anything depending on if the switch and PSU are both bad.

At least we are in agreement that the paper clip test is not any kind of reliable PSU test.

 

[Spartacus]

I suppose I'm slightly hijacking this thread, but with this at top it's not worth making another one.

Why even jump the power supply anyway, other than to test it? I just bought a water cooling system and a guy I talked to said I should "jump" my PSU...something about always keeping the pump running or some such. Looking back I guess I didn't catch all of that.

Will jumping my PSU with a paper clip help me in any way as well? Or did I just completely misunderstand?

The issue of jumping out the PSU connector to power it up in order to power a water pump is a completely different issue.

In that case it's not any kind of test, it's a way to provide power to the pump.

 

[Providence]

Oh, okay.

*does more google searches*

 

[[Spectre]]

Paper clip jumping is fine and does help eliminate possible issues.


This thread is done.