I've got this old computer, with the orders cheap fix or throw away.
So I plug it in, and by pressing the power button, nothing's happening. I immediately suspect the power supply, but decided to test the actual button itself first. I've tried turning on the computer using a screwdriver. No go. My friend forgot to mention it's dead, lol. Well, ok, it has to be the PSU.
When everything is plugged in, you could hear a small buzzing sound coming from the inside, but it's hard to tell from where. I unplug everything, so only the PSU and motherboard are connected. The buzzing sound is still there.
I finally test the PSU using the paper clip trick, but it's completely dead. Ok, makes sense. Plug it back into the motherboard, and the buzzing sound returns.
Now I know there's a great possibility that the motherboad is damaged aswell, but my question is, how is it still producing the buzzing sound even if the PSU that's powering it is dead?
EDiT: Typos.
Last edited by iNViSiBiLiTY at 2013-05-16 22:58:05
The buzzing is - most likely - caused by the 5v stand-by section/circuit, INSIDE the power supply.
It's a separate circuit in the power supply using one of the small transformers you see inside, which gives the computer a few watts of power on 5v.
The motherboard uses that to power a couple of usb ports in the back for keyboard and mouse for "wake on keypress", wake on lan stuff, and also to tell the power supply to start and give the rest of the voltages when you press the power on button.
When you took the power supply out of the case, it's running without anything connected to 5v standby so there's no buzzing.. duh.
When you plug it in, there's some power used by the board hence the noise, it's the magnetic field of the transformer expanding or shrinking, or some inductors (coils) inside the psu..
Though, in some cases, buzzing could mean that the 5v stand by circuit is overloaded (motherboard has some short circuit or something demands more power than what the stand by circuit can provide).
In rare cases, the standby circuit is toast, and no longer gives 5v...for example a capacitor used by the standby circuit may have gone bad inside and now the motherboard sees 4v or something like that, or even worse an IC inside may have blown and you now have something like 10-15v instead of 5v (with potential to have something now killed on the motherboard)
google the atx pinout, find the 5v standby wire, get a multimeter and check the voltage there while psu is plugged into the wall. Should be 5v give or take 0.2-0.3v
plug the atx connector into the board... measure the 5v standby again, if it's 0 or goes way below previous reading... the board is toast, it's overloading the 5v circuit..
If all's good in that department, then something is probably bad inside the power supply outside the 5v stand by.
Could be sometimes as simple as a capacitor leaking and causing 12v to drop 11v or some minimum threshold which would trigger the under-voltage protection and stop the psu from starting up.
Could also be a blown resistor or transistor in which case it's harder to fix.
tldr : test the psu in another system, test the system with another psu, flip a coin.
That's a great idea. Why didn't I think of it myself? Oh wait, I did. It would have been the first thing I'd do if I had one.
It is customary to point out the obvious when dealing with questions on the internet, as often the person asking hasn't considered it - there's no need for you to act like a child about it, diddums.
EDIT: Probably worth noting too that, while accompanied by a lot of supporting information, Angelitta's suggestion is basically the same as mine:
Angelitta wrote:
test the psu in another system, test the system with another psu, flip a coin.
test the psu in another system, test the system with another psu, flip a coin.
Chill, Winston.
the tldr is short for too long; didn't read... and in some circles it's customary to write a joke/something sarcastic or obviously incorrect when poster wants readers to actually read the whole post.
the line was a poor attempt of a something witty/joke.
the tldr is short for too long; didn't read... and in some circles it's customary to write a joke/something sarcastic or obviously incorrect when poster wants readers to actually read the whole post.
the line was a poor attempt of a something witty/joke.
I know - the point is though that even with all of the info you gave, he's still in the same position really (bar tearing down the PSU itself, which I highly recommend *against* unless you know exactly what you're doing, iNViSiBiLiTY - Angelitta obviously does, but it's another case entirely for a novice to go poking around in there) - test the PSU in another system (inadvisable) or test the system with another PSU (better). If it was intended as a joke then it is a little ironic, because those are the only realistic options to the novice.
While were pointing out the obvious lest not forget the kettle lead
Buzzing sound, could be as simple as what i said above or as much as what Angelitta said.
Going inside a PSU....ive watched a dodgy capacitor discharge from its top, so don't assume you have to be touching the bottom for that and with the size and power of those in a PSU it will certainly make for an interesting outcome if it does discharge into you.
Also TiberiusUK is right, not only is this the internet this isnt exactly a noobless forum (go read the first page of the torrentbytes.net subforum), starting at the basics is the obvious way to go, if you don't like that, state exactly what you've done already rather than making people assume and in most cases, finding another and testing it with that or in that is the best and quickest route. Having one or not is not the point, find one test it, or buy a new one and test. Guessing is just that.
So a bit of an update. I got a new board for this old pc, and the system finally booted up. Only to die 30-45 seconds later. It doesn't matter if it goes to Windows, or if it just sits in the BIOS, it dies eventually. And when it does, you have to unplug it if you want to power it up again. I know this usually means a faulty power supply unit, but this one is brand new. So I tested it using the paperclip trick, and it kept running for multiple hours. So the next in line are either a faulty cpu or ram.
I've decided to take everything out of the case, and put it on my table with only the essentials connected: the power supply, the motherboard, cpu and ram. I turn it on using a screwdriver, but it dies again.
Then I took out the ram stick and the thing kept running for 17 hours straight, with only the motherboard, cpu and psu connected. So I guess there's no doubt, the ram is faulty aswell.
My question is, would a faulty ram cause the system to shut down in such a way, you'd have to unplug it from the power socket if you want to start it up again? Usually this indicates a psu problem...
Just because something is brand new doesn't mean its not faulty, things arrive DoA, partially working and more. How you ran it for 17hrs without any RAM is a curiosity to me, it means it didn't boot at all.Thats not a test.
Well it did die instantly when the ram was there, so atleast it ran without dying. Of course it didn't boot, no hdd or ram, but atleast it kept running.
Tested the power supply unit in my own pc, without the gfx card. Works like a charm.
(Power User) at 2013-05-17 01:53:55 (2 months ago) - [
(Mad Ultra Mod) at 2013-07-19 21:58:26 (1 week ago) - [
