Yet another computer question?

[TopCatGr58]

Hey there fellow geeks!

Okay, nephew had old eMachines unit that "died". Looked inside case, saw swollen capacitor thinking that the culprit, ordered and replaced it, still no go. Then realized possibly power supply. Put in "known good" PWS, still no go! Ordered replacement used Mobo from eBay, reputable seller pretests all before shipping, put it in case, still no go. On one occasion after having all PWS connections connected, plugged AC cord into PWS case, CPU fan came on, but still no HD spin, no video out, no boot error beeps!

Got another known good PWS from another computer, put it in, same thing! So theorized replacement Mobo bad...was loosely packaged in shipping box so easily slid around in box banging corners....although no signs of damage to corners of PCB of Mobo!?

Then put the previously working PWS back into my backup slave computer, and would NOT turn on! Also put other PWS from the other known working old computer back into it's case, and it also would NOT boot/start! The power LED would blink on the Mobo, but that's it! I then tried one of the recently "damaged" PWS and the CPU fan came on, and power led stayed lit, but still no boot!

SO...I did google search, several places said that bad Mobo can not damage PWS...that PWS's have built in protection to prevent that from happening, that a person could short ALL of the wires in the 20 pin connector together and it would not hurt the PWS!?

Have any of you geeks had a Mobo damage a known good PWS???

I've found out how to paperclip switch the PWS to turn on without being plugged into the mobo by shorting the green and proximal black wire in the 20pin connector together. I'll be finding a diagram for the 20 pin output to be able to try to test it with my multimeter? Any other thoughts?!

T.C.

[ViperRon]

TC I have worked on many PC's different venders and 3 Emachines and each of the Emachines had both the PS and mother board bad. The bad mother boards had shorts and may have possibly damaged a Power supply. If I had plugged it in but not sure as I was not going to plug one in to find out. Most times only a PS or mother board failes and most times it is because a fan is not working or someone pushed them in a corner to cook. Dial up connections in thunderstorms are bad however a PC that runs 30 days with no problem usually runs good till its overheated, trash gets in it, or a fan overheats it.

Another issue is Emachines puts in a low watt power supply and you can definately put in too many optional parts drives, memory, or cards to blow one.

[Jerry]

Have only 1 case to go on from personal experience. Stock HP quit working. Replaced power supply under warranty. Installed new pws, unit ran for 3 days and quit again. PWS was bad. Sent entire unit in to HP and they replaced both the motherboard and power supply. So yes, I think a motherboard error can cause a power supply to go bad.

However, if there are other acccessories plugged into the slots, such as a higher-end video board with an on-board fan, these can also cause a problem. We had a digital video recorder that uses XP for an OS and the video card fan locked up causing the power supply to fail, we think. Replaced the video card, installed new pws, and unit fired right up.

Don't think you can say "never" in cases like this.

[TopCatGr58]

Update

Thanks Ron and Jerry,

Yeah, I know that Emachines are the cheap units, it was my nephew's, and he had installed a slightly better graphics card in it for gaming, but it just had a regular heatsink, no cooling fan. It had worked for them prior to them bringing it to me to install XP's SP3, but it was dead when I tried to turn it on. SO...good chance the PWS died and could have fried the OEM mobo at the same time.

I got word back from the eBay seller that received the alleged defective Mobo, they tested it, and it would NOT Post! They found that a small capacitor that was soldered onto the BOTTOM of the board had gotten broken off....probably during the rough shipping, but they said it could also have occured during my installation into the case?? Anyways, they are Honoring their guarantee and shipping out a replacement Mobo so that I can get the eMachine back up and running.

But that still leaves me with concerns that possibly the fried OEM mobo damaged the other PWS's that I had tried testing them out along with trying to get the "bad" new/used Mobo working???

What's bad is that I didn't see any of the "MAGIC SMOKE" escape from any of the other previously good PWS's that NOW won't work, and that 2 previously working systems now will not work possibly due to bad PWS's! So now I have to get some replacement PWS's as well! $$$$$ I know I've gotten some decent deals both on eBay as well as NewEgg.com...just urks me that the ones I had harvested from other old mobo/dead machines are now toast!

T.C.

[CaptonZap]

Just in case it is hardware related


http://www.badcaps.net/forum/

The trouble with being old school is, I don't trust software, and I don't trust hardware.

CZ

[DiverRay]

T.C., I am repairing radio's at the moment for income. I have seen a LOT of things that do NOT let the smoke out, but kill the part, or at least one or two pins of the part. That could be part of your problem, the sensor chip could have lost part of the control pins, and now it will not work.
Just a thought from the sunset side of the country.

[TopCatGr58]

Hey Ray,

Thanks for that! Well, after researching the PINOUT diagrams for 20 and 24 pin PSU connectors, plus the +4 plug, the Molex and Sata and such, I got out my voltmeter and fired up the various now suspect PWS's.

The Original Emachines PWS would NOT even power on=DEAD.
The other previously working PWS in the other HP/Compaq D530T Case now showed below Specs for the 12 volt rail....~9-10V! Dead

The PWS for my slave system powered up and measured well within specs, but the actual slave Mobo that was previously working till I tried the BAD PWS in it, now will not boot...so the bad PWS probably killed the Mobo!

I think I had also tried the Bad Hp/Cpq PWS in the other HP/Cpq D530 before, so now when I put the "Tested Good" ATX Pws from my slave unit into the 530 case, the mobo power indicator light will come on solid...but no response from the Power On button, so that Mobo may have gotten damaged by the Bad emachines PWS!?

So...looks like I might have killed two otherwise good Mobos by "testing" them with the bad PWS!? LESSON LEARNED the HARD WAY!

I know they make/sell a PWS testing device...cheap($5.00) simple green/red LED types, as well as fancier ($20.00) LCD readout ones, but I have a MultiMeter, and it's relatively simple to get the PWS to power on and then just probe the connectors for readings. I don't think the cheap testers put an actual LOAD on the PWS, and even the more expensive one may not either, so I would think that my MM testing would be as good as the testers in this case, and I'll know NOW to test a PWS that I suspect rather then "try-test" it out by putting it in another case/mobo!

I just ordered a 400 watt "inexpensive" PWS off eBay from a POWER Seller, it does have a basic 1 Year warranty and at least FCC listed/rated, not just UL!

Then if it doesn't work on the now recently non-functioning Mobos then I'll know that the Mobos are fried. I'm getting a replacement Mobo for the eMachines case, so that with the new PWS should then fire up and work. Then I'll have to see about getting another SLAVE Mobo for my slave case!
I like having a spare/Slave desktop system for both my music and just to be able to use when my MAIN system is being used...ie. video editting and such!

T.C.