Fun with fixing I-Micro power supplies

[Behemot]

Re: Fun with fixing I-Micro power supplies

Can it do labeled power now?

[Dan81]

Re: Fun with fixing I-Micro power supplies

Can it do labeled power now?

I don't have a load tester so I don't know.

[everell]

Re: Fun with fixing I-Micro power supplies

Can it do labeled power now?

Are you kidding? Of course it can. Didn't you check the label on the power supply case? It has UL and CE certification, along with certifiction from a couple of other certification agencies.

[stj]

Re: Fun with fixing I-Micro power supplies

that certifies the case, not the contents.

[Dan81]

Re: Fun with fixing I-Micro power supplies

Are you kidding? Of course it can. Didn't you check the label on the power supply case? It has UL and CE certification, along with certifiction from a couple of other certification agencies.

Am I the only one who thinks that UL logo is fake?

[everell]

Re: Fun with fixing I-Micro power supplies

Back to the story of the last I-Micro I fixed. I was ready to test, hooked it to a Pentium 3 mother board, and watched the fireworks. I repaired it (see previous posts). It ran the Pentium board fine, but after a few hours it quit. I decided to give it the "Uniqu3" modification. I removed the output diodes and installed a Shottky 3045 on the 3.3 volt rail, a Shottky 3045 on the 4 volt rail, and a MBR Shottky 20100 on the 12 volt rail. To improve the input side I installed a pair of C2625 switching transistors. Applied power using the Pentium 3 Biostar M6VLQ motherboard as a load. Ran it many hours with no trouble. At last it is fixed.....I thought.

With boldness, I attached it to a Pentium 4 class motherboard. Pushed PS_on switch.....BANG, fizz, and a small fireball! Both switching transistors shorted, the C945 driver transistors were bad, and the 7500 pwm chip was shorted. Also a couple of bad diodes, and the resistors in one of the switchers had blown resistors. I replaced the switching transistors with 13007. I also replaced the two electrolytic capacitors in the switching transistor circuits. Replaced the driver transformer just in case it shorted. Attached it to the Pentium 3 motherboard and once again it came to life.

Some interesting measurements: with no hard drive attached, the 5V rail reads 5.12 volts and the 12V rail reads 12.58 volts. When I attach the hard drive, a 40 gig Maxtor ATA100, the 5V rail reads 5.01 volts and the 12V rail reads 12.48 volts. Seems to me to be a rather large voltage drop when adding only a hard drive. I think that trying the Pentium 4 motherboard would again farkle its tra-la-la, so I am reluctant to put a heavier load on this 400 watt power supply.

So what do you think?

[Pentium4]

Re: Fun with fixing I-Micro power supplies

Wow, it exploded just connecting it to a P4 board?! What else was it powering?

Some interesting measurements: with no hard drive attached, the 5V rail reads 5.12 volts and the 12V rail reads 12.58 volts. When I attach the hard drive, a 40 gig Maxtor ATA100, the 5V rail reads 5.01 volts and the 12V rail reads 12.48 volts. Seems to me to be a rather large voltage drop when adding only a hard drive.

What else did you have hooked up to it? Was this with the P3 board or just a paperclip? I think it's very normal for the 12V to read very high with low 12V load on half bridge units. Since they were usually designed with low end ultra fast diodes, the schottky seems to bring the "low load" voltage up very high. As we know, these designs like 5V. It seems like the extra 5V from the hdd "levelled" it out a little bit. I'm just guessing. I bet if it were to power a 12V heavy machine, the 12V would drop to something around ~12.3V

I think that trying the Pentium 4 motherboard would again farkle its tra-la-la, so I am reluctant to put a heavier load on this 400 watt power supply.

I don't blame you. Maybe try a different board? I'm not sure why it would power the P3 fine and not the P4 system. It looks beefy enough especially with the new silicon you put in it.

[everell]

Re: Fun with fixing I-Micro power supplies

At present, the power supply is attached to a Pentium 3 Biostar M6VLQ motherboard with onboard video, 512 meg of PC100 memory, PS2 keyboard, PS2 mouse, and 40 gig Maxtor ATA100 hard drive. I don't consider this to be a very heavy load! The voltage measurements were made using this motherboard and hard drive as the load.

The board that blew it up turns out was not Pentium 4. It is a pullout motherboard from a EMachine, a K7MNF-64 made by FIC. It has a Sempron socket 462 AMD K7 2 gig processor (SDA3000DUT4D) and 1.256 gig memory. I am using the onboard video. It also had PS2 keyboard, PS2 mouse, and hard drive attached. I tried using one of the other rebuilt IMicro power supplies on it.....worked OK for several hours. The IMicro that worked had a much larger switching transformer.

The EMachine board was a pullout with some bulging capacitors. I had already replaced capacitors and checked out with blue esr meter. I tried using several of my repaired power supplies on this motherboard, and all worked fine. With a good power supply this motherboard does fine.

I am beginning to think that the small switching transformer is the culprit. It looked like the same size transformer that was in the power supply Uniqu3 rebuilt (from pictures he posted). I would have thought it would be capable of handling the AMD K7 processor mother board. I guess that the motherboard needs more than a 400 watt IMicro power supply.........

[Behemot]

Re: Fun with fixing I-Micro power supplies

That systems takes less than 100 watts so it seems this is really top class PSU lol

[everell]

Re: Fun with fixing I-Micro power supplies

That systems takes less than 100 watts so it seems this is really top class PSU lol

top class? I can think of words much more descriptive!

[Behemot]

Re: Fun with fixing I-Micro power supplies

[everell]

Re: Fun with fixing I-Micro power supplies

Sarcasm??? This whole thread has been nothing but sarcasm. Although it has been fun playing with the IMicro power supplies, I hope no one seriously thinks I like them. I don't even trust any of them to be permanently installed in a computer. i have enjoyed the feedback everyone has given. I also enjoyed the fireworks from those IMicros that didn't make it. And the sarcasm and jokes have been great.

My favorite power supply is the Bestec ATX-250 12E with DM311 mod, and a few other mods. I have one in the computer I am running now. It has been running for some time, and no 5vsb overvoltage problems.

[Behemot]

Re: Fun with fixing I-Micro power supplies

I'd say it was waste of time since you have about hundred units which are 10times better any day of the week waiting to be processed, but that's me…don't really find any fun in making PoS just better PoS and if you want to see it burn, you don't need to do any modifications on that do you

[LDSisHere]

Re: Fun with fixing I-Micro power supplies

This is the first time I saw this thread and I found it very interesting. I understand the sometimes you do things just because you can mentality. It is like mountain climbing, there is no point to it other than some people seem to enjoy it.

[Behemot]

Re: Fun with fixing I-Micro power supplies

Climbing is excercise at least…

[Pentium4]

Re: Fun with fixing I-Micro power supplies

I don't think this one is complete junk, at least it has a 35 transformer. We've seen much more gutless units pump out 250W DC no problem. This one may have a defect, or maybe something is arcing. I have a Sun Pro that looks very similar to this one, and it could power a Prescott and an OC'd 9800GT at full load after the 12V had two schottky in parallel. The unit got really hot but ran for hours no problem.

[everell]

Re: Fun with fixing I-Micro power supplies

I think the transformer is a 33

[Behemot]

Re: Fun with fixing I-Micro power supplies

The unit got really hot but ran for hours no problem.

Yeah another reason not to use these gutless wonders, the efficiency must be in some 60 % region. Decent unit will pay itself on electricity in time. Man srsly, manufacturers are competing in 80 PLUS Titanium (around 95 % peak), there are 550W Titanium units to come on market and you waste your time with this?

[Pentium4]

Re: Fun with fixing I-Micro power supplies

Because man, a Titanium PSU is expensive, and I get these units for free. Just because I was testing it with that system doesn't mean that's the one it was going to be powering. I use a Precott + OCd 9800GT to stress test PSUs. If it can handle that power hungry system for hours, then I know it can safely power a Core 2/Core i3 system with a lower end video card no problem. I have plenty of 80 Plus units in use, I just also have a lot of cheaper units I got for free that I use.

[Behemot]

Re: Fun with fixing I-Micro power supplies

If you count the manpower and material plus that poor efficiency it will become very expensive in the long term no matter how much you payed or did nto pay) for that.

On the other hand 80 PLUS Gold certified Seasonic SSP-350GT or similar quite cheap unit has 100% quality internals and you will definitelly notice 25-30 % higher efficiency as long as you don't steal electricity…