How would you recommend I test my psu?

MumboJumbo

MumboJumbo

MumboJumbo

Soldato
Joined
2 Dec 2006
Posts
8,204
Ok then here is the drill I am fully upgrading my pc and need to know if my psu is dodgy or not the rest of the parts have already been bought and are to arrive on monday (yay). I've had 2 gpu's die in 4-5 years in this pc. They are the only components that have ever been changed. I've had other problems which could all have been symptoms of a dodgy psu. Note, the psu is no longer in warranty.

The culprit? A seasonic m12 600w psu. An unlikely culprit for a faulty psu if you ask me. It is however not completely dead if it is damaged at all, thus I must test it rigorously as I do not think it has a problem while idle. The question is how?

Options:

1) Pay to have a shop test it, but I question just what my local shops might actually do to test it. I'm not convinced simply using a multimeter is sufficient and I suspect that might be all they do.

2) Buy a multimeter and possibly attempt to test while under load? I'm suspecting highly dangerous and just stupid.

3) Buy a psu tester that can test the psu while under load.

Next question. Do such psu testers exist and are they going to be reasonably priced? I've been looking at psu testers but I can't find a lot on them. Most of them just have a light which says good or faulty and some have claimed that they are hideously inaccurate.

Any other suggestions?

:EDIT:

I have found a psu tester with an lcd screen that will test everything but its £36, which is quite steep. There is an unbadged version available which I can't link you to which looks to be the same for £25 though.

I'm thinking that the only way to test this psu under load is to actually stress test it on a stable pc which I'm not willing to risk as it might fry my parts and I'm thinking no shop is going to bother doing either.
 
Last edited:
Yeah I know but I spent £120 to get a psu to last me the years and have peace of mind. I don't want to simply throw £120 away and spend another £70 odd ontop of that.

Further research has shown the only way to do this properly is with a multimeter and the pc under load. It doesn't look as dangerous as I had of thought.

My only real choice is realistically to get a shop to do it as I have no 100% stable pc to really test it on.
 
Last edited:
What's wrong with the one from Antec at £25.99.

Currently out of stock, however you could buy elsewhere.

There's a good review of it on Youtube. Seems to do the job very well and is very detailed. A lot better than a multimeter for the job in hand.
 
I have one of those PSU tester with the blue LED.. didnt pay £36.. only £8.

They are OK but dont tell you anything about how it will perform underload.
 
Buchanan0204 said:
What's wrong with the one from Antec at £25.99.

Currently out of stock, however you could buy elsewhere.

There's a good review of it on Youtube. Seems to do the job very well and is very detailed. A lot better than a multimeter for the job in hand.
Click to expand...

Didn't even know ocuk stocked them. Didn't think to look tbh.

Also as stated above, they don't tell you anything about under load and I think my psu needs to be under load as it's clearly slowly killing my components if it is taking years and thus functional to some extent.
 
8igdave said:
Didn't even know ocuk stocked them. Didn't think to look tbh.

Also as stated above, they don't tell you anything about under load and I think my psu needs to be under load as it's clearly slowly killing my components if it is taking years and thus functional to some extent.
Click to expand...

Can you ive us a run down of some of the problems you think to be PSU related.

also may i ask if you have lived in the same house all this time?
 
If the psu were giving high voltage spikes (unlikely but just offering a scenario based on the possibility of the psu being the culprit) when a load is placed or removed from it, then the spikes would likely be far too quick for a multimeter to detect. I have never used a psu tester so couldn't say how often they sample the voltages. At that price though, I suspect a psu tester is a very basic device.

Assuming psu is ok: Do you observe static precautions when working with circuit boards. Static damage usually shows up as a fault several months after occurring. Hence, lots of folk say they have never static damaged a component. When in fact, they have, they just don't make the connection.

Not saying it is. Just posing possibilities.
 
Last edited:
Baldbloke said:
If the psu were giving high voltage spikes (unlikely but just offering a scenario based on the possibility of the psu being the culprit) when a load is placed or removed from it, then the spikes would likely be far too quick for a multimeter to detect. I have never used a psu tester so couldn't say how often they sample the voltages. At that price though, I suspect a psu tester is a very basic device.

Assuming psu is ok: Do you observe static precautions when working with cuircuit boards. Static damage usaually shows up as a fault several months after occuring. Hence, lots of folk say they have never static damaged a component. When in fact, they have, thay just don't make the connection.

Not saying it is. Just posing possibilities.
Click to expand...

Most multi meters detect 2-3 changes per second, power spikes can be in the milli second range, you'd really need a Oscilloscope to measure these sorts of things.

I would always follow my rule for this sort of thing "If in doubt change it out"
 
Last edited:
bigjimmyauk said:
Can you ive us a run down of some of the problems you think to be PSU related.

also may i ask if you have lived in the same house all this time?
Click to expand...

Well right from start I've had stability issues. Some were caused by the fact my motherboard was always a bit dodge to boot so It's a mix between the two I think:

1) Difficulty overclocking on a board that was ment to be very overclockerble followed by a sharp decline in overclockability from 3.4ghz to 3ghz over a course of about a year and a half, maybe less.

2) Cold boots, restart cycles

3) Moving the pc on its side would sometimes fix problems yet I could n ever identify any unprotected cables or loose connections, thus the problem was suspected to even be a loose conection in the psu possibly.

4) Screen would randomly turn black for a second or two - gpu/psu issue

5) Two gpu's dieing where on their last legs they gave pretty colourful displays on screen.


While the list of symptoms is bad, they were never all persistent at teh same time and for the large part could all of a sudden dissapear for months at a time. Changing gpu fixed my problems and thus it was concluded the gpu and dodgy mobo were to blame. My pc was tested with a spare psu to find there were no changes. Now I think otherwise and instead my psu has been slowly killing parts.

Also my first gpu died at home, my second gpu died at uni.

Baldbloke said:
Assuming psu is ok: Do you observe static precautions when working with circuit boards. Static damage usually shows up as a fault several months after occurring. Hence, lots of folk say they have never static damaged a component. When in fact, they have, they just don't make the connection.

Not saying it is. Just posing possibilities.
Click to expand...

Having worked with the insides of pc's for many years and only ever gone inside my pc when I had to I do not feel I have caused any static damage.
 
You must log in or register to reply here.
Back
Top Bottom