Astec AA16990 PSU failure

[PeteS in CA]

Re: Astec AA16990 PSU failure

I'm not sure what your purposes are at this point. If you want to see if it can deliver full rated power, be sure to notice that if all 3 outputs are at maximum current the output power will be 94W, well above the rated 86W.

With that noted, a 10A load on the 5V would, nominally, be 50W and the load resistance would be .5 ohm. Two 1 ohm, 50W resistors in parallel would be an easy way to do that.

[powermax]

Re: Astec AA16990 PSU failure

I'm not sure what your purposes are at this point. If you want to see if it can deliver full rated power, be sure to notice that if all 3 outputs are at maximum current the output power will be 94W, well above the rated 86W.

Oh yes, you're right! That shows that I can't fully trust those (bogus) values on the PSU's case.

With that noted, a 10A load on the 5V would, nominally, be 50W and the load resistance would be .5 ohm. Two 1 ohm, 50W resistors in parallel would be an easy way to do that.

Thank you very much, Sir! I'll do the tests and report the results.

[Per Hansson]

Re: Astec AA16990 PSU failure

You don't need to load the rails down to 100%
A 10% load on each rail should be enough to tell if the voltages become stable then or not.

[PeteS in CA]

Re: Astec AA16990 PSU failure

Oh yes, you're right! That shows that I can't fully trust those (bogus) values on the PSU's case.
...

They aren't bogus values. It just means that if you use the full load current of the +12V, you have to use less than full load on the +5V. I would expect this to be the case with any PSU model from a good power supply vendor (e.g. Delta, Lite On). Astec was a top power supply vendor (Are they sold under the Emerson name now? Emerson has owned them for decades, so it's probably just a name change.).

[momaka]

Re: Astec AA16990 PSU failure

You don't need to load the rails down to 100%

Exactly!

This being an old PSU, though, I would recommend loading the 5V rail more than anything else. 2 to 3 Amps would be very good to keep the PSU stable. As for the 12V rail - an old hard drive should do (just take one that you don't mind sacrificing *if* something goes bad... which it probably won't, since this is an Astec PSU).

Lamps from a car can also be used but be careful because the inrush current is extreme and could actually damage the PSU.

I have not damaged a PSU by doing that yet. Then again, I've done that only with 2-3 bulbs max attached to the whole PSU just as a minimum load. If you try to load down a PSU to the maximum load with 12V auto bulbs... yeah, you're going to have a problem with inrush.

Yes, such a "testing" circuitry built using incandescent bulbs has been already mentioned several times in this forum. Unfortunately, I couldn't find anything concrete.

Well, there's two, actually.

One is the incandescent light bulb primary (line-side) series test, which is used when you rebuild a PSU that had a blown fuse and want to make sure it doesn't blow again.

The other test involving incandescent bulbs is for testing the output (as mentioned above already). For this one, you use 12V car bulbs or the 12V bulbs found in some home track lightning (Ikea has a few of these). A bulb in the range of 20-40 Watts should give 1.5-3 Amps of load on the 12V rail and about 1-2 Amps on the 5V rail.

[powermax]

Re: Astec AA16990 PSU failure

I've built a "load device" that consist of two 2 Ohms, 25W Ohmite tubular resistors connected in parallel and two 12 Ohms, 25W Vishay wirewound resistors in parallel mounted on an improvised heatsink. This device loads the 5V rail with 5A and the 12V rail - with 2A. The -12V rail remains unloaded.

Now I get the following voltages:

+5V rail ----> +4.98V
+12V rail ---> +11.86V
-12V rail ----> -12.08V

Everything seems to be okay now. The voltages are stable too, although the test lasted only about 2 Min. because I didn't want to "fry an egg" with my load device (PeteS in CA was right - the resistors will get pretty warm very quickly and the fan spins very fast).

Thank you all for your precious technical help! I think it's the time to test the PSU with the computer, isn't?

[Per Hansson]

Re: Astec AA16990 PSU failure

Thank you all for your precious technical help! I think it's the time to test the PSU with the computer, isn't?

Absolutely!
And well done on the load test

[PeteS in CA]

Re: Astec AA16990 PSU failure

I'm going to venture a guess that the -12V output is regulated by a 7912 or LM320T-12, so being unloaded is no big deal. I think the 3-T might be on the large heatsink near the output wires.

Looks like you got it fixed. In real use, I doubt that computer loads the power supply much higher than 75%.