Testing Power Supply Rails-

[tufkal]

So I have a reallly wierd PC im trying to fix, where the DVD-ROM and HD both died at the same time, but everything else seems fine. I have a new power supply to replace the old one, but im not convinced it was his PS. I have a multimeter and according to the mutlimeter, and the bios readout, all volt rails are within range.

What I want to test is the draw on those rails, as related to the amps lsited on the side of the power supply. I know very little about using a multimeter to test anything except volts and I dont want to damage the PS if it is indeed functioning properly.

Can anyone tell me what settings I should have the multimeter set at and where to poke the cords to get the readings I am looking for?

 

[oneshot47]

Most mm's that ive seen cant handle the amount of amps a typical psu puts out. Youll have to spend a bit more money for that.

 

[Geniere]

Yeah ? All of the DVMs I?ve used only allow current measurements up to 10amps. You can use three DVMs in parallel to read up to 30amps or 5 for fifty amps if you have access to several meters. Just total the readings.

If the rails have the proper voltage, your probably ok if the readings were taken when a particular rail was near full load, i.e. HD?s were writing continuous data.

Use an ?O?scope or switch the DVM to AC volts (low volt position) to measure and/or look at the ripple voltage.

 

[beer]

Originally posted by: Geniere
Yeah ? All of the DVMs I?ve used only allow current measurements up to 10amps. You can use three DVMs in parallel to read up to 30amps or 5 for fifty amps if you have access to several meters. Just total the readings.

Yea uh...I wouldn't recommend that. If one of your contact leads slip you have a bunch of very dead multimeters. (at least until you can fix the fuse)

 

[damonpip]

Most multimeters only will test up to 10 or 20 amps. Even if you parallel several multimeters, you will have trouble getting a useful reading because the multimeter will only measure the current draw at that point in time, not the maximum current output of the power supply.

 

[Geniere]

Indeed, thats why I added:

"If the rails have the proper voltage, your probably ok if the readings were taken when a particular rail was near full load..."

 

[DungeonMaster]

OMG!!!!!!!!! You NEVER Connect a Multimeter directly to the source to read the max current. The use o the Amp function is to put it in serial with the load (motherborad or a resistor) and then see the current. Thats because a multimeter in amp mode has ) Ohms resistance and the current that is going through the MM is theoreticaly not countable. The mathematical explanation is...
I (Amps)=V (Voltage) / R (Resistance) <=>

I = 12Volt / 0Ohms = Countless. Do it in the windows calculator. By using a mm in that way you burn its inside fuse and you put a great stress on the PSU.

 

[xenos500]

Originally posted by: Geniere
You can use three DVMs in parallel to read up to 30amps or 5 for fifty amps if you have access to several meters. Just total the readings.

Testing your powersupply by shorting it out with a meter would be like testing a smoke alarm by setting your wall on fire.

 

[Red Squirrel]

You need to to "add" it in the circuit, in series... at least that's how I do it to test household amperage, but I was later told if it's not true RMS it's not accurate but for DC RMS is not an issue.

So it would go yellow wire -> multi meter ->load (ex: hard drives etc) -> black wire.

But like what has been mentioned most have a 10 amp limit. Though I'm not sure if this limit applies to low voltages such as 12 volts, since I think the limit is mostly heat related, a 120v circuit that is drawing 15 amps will heat up small wires, but a 12 volt circuit drawing 100's of amps won't. So small components inside such as the fuse won't melt. Though I would not take a chance anyway since I'm probably wrong, there must be other factors related to the limit.

 

[Geniere]

Some posters obviously have poor reading skills. I earned a 6-digit income because of my knowledge of electronics and used that knowledge in the proper application of test instrumentation. One certainly can insert a parallel group of ammeters in series within a circuit. My previous post is exactly correct.

 

[PlasmaBomb]

Originally posted by: RedSquirrel
You need to to "add" it in the circuit, in series... at least that's how I do it to test household amperage, but I was later told if it's not true RMS it's not accurate but for DC RMS is not an issue.

So it would go yellow wire -> multi meter ->load (ex: hard drives etc) -> black wire.

But like what has been mentioned most have a 10 amp limit. Though I'm not sure if this limit applies to low voltages such as 12 volts, since I think the limit is mostly heat related, a 120v circuit that is drawing 15 amps will heat up small wires, but a 12 volt circuit drawing 100's of amps won't. So small components inside such as the fuse won't melt. Though I would not take a chance anyway since I'm probably wrong, there must be other factors related to the limit.

A fuse is melted (fused) by the current going through it not the voltage of the system. Thats why they are 1A,3A,5A fuses etc. not 1V...120V fuses.

 

[beer]

For a highly technical forum, the lack of knowledge here seems a tad scary