Power Supply tester doesn't mean a %*#@

[meanderer]

A customer brought in a HP desktop that's totally dead. I used a power supply tester and the fan spins, the voltage level is fine. But it still won't boot the computer.

Replaced it with a new power supply. Everything is fine.

$16 bucks investment down the drain.

 

[ATTech]

A power supply tester tells you if it's bad, not if it's good.

 

[MrUnknown]

A power supply tester tells you if it's bad, not if it's good.

???

The PSU was bad, not sure where you are going with this.

The real issue is a PSU tester doesn't test under load. I just use a multimeter to test PSUs.

 

[14049752]

The PSU was bad, not sure where you are going with this.

The idea is that it confirms a bad power supply, it can't confirm good. Meaning...if you get NO power, it's bad. If you get power, it means nothing. The extra voltage readouts and everything are just a waste of your time, because all you need is an LED to do the same thing.

This topic has come up at least three times in the past couple of weeks, in various threads. The only way to test a power supply correctly is to put a load on it, then hold it for a while. Realistically, just swap a power supply and never trust those cheap little pieces of junk to tell you if a supply works.

 

[norm1320]

???

The PSU was bad, not sure where you are going with this.

The real issue is a PSU tester doesn't test under load. I just use a multimeter to test PSUs.

The point is to save time in diagnosing PSUs. If the tester says it's bad, no need to try anything else, just replace the PSU. If not, then go ahead with further testing.

 

[ATTech]

???

The PSU was bad, not sure where you are going with this.

The real issue is a PSU tester doesn't test under load. I just use a multimeter to test PSUs.

What 14049752 said is what I meant by it. If the tester doesn't light up, then the power supply is bad. If it does light up, then that means nothing.

 

[kozman]

Now that is a good idea. Test with the stupid tester first (stupid testers put someones kid through college). Then meter if needed. But thanks for explaining checking something under load.
BTW: do computer power supplies have auto resets for overload of the unit? I have seen psu with this in manufacturing areas.
Peace,
kozman

 

[14049752]

Now that is a good idea. Test with the stupid tester first (stupid testers put someones kid through college). Then meter if needed.

If you mean multi"meter", that won't do anything, either. Those power supply testers are just little multimeters, just with very specific connections. A regular multimeter doesn't put any kind of load on the power supply, either. Unless you build or buy a power supply tester that puts a sustained (and sufficient) load on the supply, and then gives you voltage readouts, you may as well be testing it with a paperclip. That's why, realistically and practically, you just swap a known good part in.

 

[CD Computer Services]

Power Supply Tester

???

The PSU was bad, not sure where you are going with this.

The real issue is a PSU tester doesn't test under load. I just use a multimeter to test PSUs.

That is an interesting point.I prefer multimeter to power supply tester because ATX power supply must be connected to the motherboard to function properly.AT power supply,on the other hand,doesn't call for that.I test a power supply when all the motherboard connectors are inserted and providing power to all the components.

 

[Alan22]

A customer brought in a HP desktop that's totally dead. I used a power supply tester and the fan spins, the voltage level is fine. But it still won't boot the computer.

Replaced it with a new power supply. Everything is fine.

$16 bucks investment down the drain.

What make & model power supply tester did you use? $16 sounds very cheap. Does it have an LCD readout showing voltages, or just little lights that show green & red?

Also, was the system upgraded to the point where the current PSU did not have the proper wattage to power all the devices upon boot?

 

[gazza]

A customer brought in a HP desktop that's totally dead. I used a power supply tester and the fan spins, the voltage level is fine. But it still won't boot the computer.

Replaced it with a new power supply. Everything is fine.

$16 bucks investment down the drain.

I use a Antec PSU tester with great success. If a voltage rail is suspect it will either light up yellow for insufficient power or red for overvoltage. I think they are a better tester than those ones sold on Ebay.

http://www.antec.com/pdf/flyers/power_tester.pdf

 

[16k_zx81]

I use a Antec PSU tester with great success. If a voltage rail is suspect it will either light up yellow for insufficient power or red for overvoltage. I think they are a better tester than those ones sold on Ebay.

http://www.antec.com/pdf/flyers/power_tester.pdf

Ive had a few of the cheapies and they have all been garbage. How long have you used the Antec item for?

 

[gazza]

Ive had a few of the cheapies and they have all been garbage. How long have you used the Antec item for?

I have had it for about 8 years and have found the lighting indicators for each voltage rail works well.

 

[norm1320]

BTW: do computer power supplies have auto resets for overload of the unit? I have seen psu with this in manufacturing areas.

I think you're referring to self-healing fuses. If so, good quality PSUs have them, some of the cheap ones don't.

 

[joydivision]

I use my PSU tester all the time. I had a job last week would not boot up. I suspected the PSU straight away but it could have been a dead motherboard. I used my PSU tester and in a process that took less than one minute I it confirmed it was the PSU. I could have just swapped out a PSU but that would have taken a bit longer.

However I never use it to test for a good PSU, e.g if a computer has some odd symptons the PSU tester would be useless. They do not tell you the PSU is working properly, they only tell you its not dead.

The only way to test a PSU is with a multimeter and under a load even that method is not always fool proof.

 

[cdbrehm]

I just bought an "Ultra" PSU tester, which are supposed to be among the better ones. For the same reason that it is a pain, and often unnecessary, to hook up a known good PSU. A PSU tester is like any other diagnostic utility--you can trust the positive results--"There's a problem", but they often yield false negatives--"There is no problem." To give an example, I had someone bring over a $4000 custom-built gaming PC that was freezing during certain games. I made the mistake of ruling out bad RAM because the RAM had passed no less than 20 cycles of Memtest, no errors. Now a week and many hours of labor later, and endless swapping of other components, I have finally confirmed it is a RAM problem after all . . . a vexingly intermittent one. Memtest can be trusted if it yields a positive result but not a negative one.