Voltage spikes when powering on PSU

Testing my 7yr old Corsair since I had a HDD die after I powered on my PC, I noticed I had voltage spikes for a fraction of a second after turning on the PSU when testing with a multimeter. All voltages were within tolerance but the 5v jumps up to 7v and 12v jumps up to 17v. Is this just an anomaly with my DMM? I don't normally use a DMM so maybe this is normal for them when calibrating voltage, I don't know. The PSU's in a non-mission critical PC and I'm using a backup PSU at the moment. I'll check to see if the DMM spikes with the backup PSU and update in a moment.

Edit: My other PC seems to have those same spikes unless I do a quick power on/off with the PSU switch. Maybe it's just an anomaly with my DMM or perhaps something common to most PSU's? Maybe the VRM's in the PSU take a fraction of a second to kick in?

Edit 2: After reading a couple Tech Powerup articles and the section where they review turn-on transience of the PSU's, I'm going to assume it's anomalous behavior of the DMM. Antec may not be what it used to be but the Antec Trio I have, I read, was made when they used quality Seasonic components so I'm assuming both wouldn't spike this much.

Many cheap DMMs will overshoot on the initial reading.

I suppose that could be the issue too. It's a Mastech MS8268, not expensive but still a good one for my needs. I replaced the PSU just in case but nothing seems to indicate that it was my PSU that killed the HDD's so I guess my old Corsair was still holding up well, same can be said with the Antec.

You can't reliably see transient effects with a multimeter.

You need an oscilloscope for that.

Try measuring a battery with the DMM because batteries don't overshoot.

I just did and you're right. I'm not sure I ever noticed before or just assumed it's a property of most electrical components but it seems to be an issue with my DMM. Is there a DMM someone can recommend that won't spike like that? It'd be great when diagnosing parts sometimes. Or maybe cyberfish is right and you need something more intricate like an oscilloscope.

usp8riot said:

I just did and you're right. I'm not sure I ever noticed before or just assumed it's a property of most electrical components but it seems to be an issue with my DMM. Is there a DMM someone can recommend that won't spike like that? It'd be great when diagnosing parts sometimes. Or maybe cyberfish is right and you need something more intricate like an oscilloscope.

Click to expand...

DMM's will all overshoot to some degree, that's why the oscilloscope was made.
It allows you to see a time plot of what's going on with the voltage.

usp8riot said:

I just did and you're right. I'm not sure I ever noticed before or just assumed it's a property of most electrical components but it seems to be an issue with my DMM. Is there a DMM someone can recommend that won't spike like that? It'd be great when diagnosing parts sometimes. Or maybe cyberfish is right and you need something more intricate like an oscilloscope.

Click to expand...

I have the Fluke 115 and highly recommend it (so do many other professionals), but that won't solve your problem. Multimeters are designed for highly accurate measurements of constant or very slowly varying quantities. Oscilloscopes are designed for high speed measurements, at the expense of accuracy (they are much less accurate than multimeters at measuring DC for example).

A multimeter is not the right tool for the job.

If you have some gumption and can fashion a voltage divider, you can use a sound card on another computer with software to act as a scope.

It won't be calibrated, but you could see the spike.

Well, with more ambition you could calibrate it, too.

7 year old PSU?
New PSU time.

Bobnova said:

7 year old PSU?
New PSU time.

Click to expand...

True, wise and logical - it's about the limit of time.

Yet....for those of us with electronics skills, a recap will make an golden oldie like that come back for another 7 years.

I have a PC Power & Cooling that's 10 now....wouldn't BE 10 if I hadn't recapped it 2 years ago, but it's as clean now as when it was 2.

Just had it on the scope last month.

Those of you that have the knowledge to open up a box that can kill you... great idea. That leaves it to about 3 people on this forum of well over 100K. LOL!

If you are left with a choice, buy another PSU people, unless you have the skills and know how to re-cap a PSU.

EarthDog said:

That leaves it to about 3 people on this forum of well over 100K. LOL!

Click to expand...

I was hoping for at least 10 or 15

In practice, it's highly unlikely for a working PSU to hold a charge for more than a few minutes, though checking for voltage on the mains caps is always a good idea. (The only PSU fault I have seen that would cause the caps to retain charge for a very long time is a bad startup circuit, thus removing the load.)