Hi everyone this is gonna sound classic but my PC won't start and I am somewhat at a loss as to what has happened. Quite simply a morning it refused to start, no post, no BIOS, no fans nothing. If power is completely removed and then attempted to start after reconnected the fans manage to spin about a quarter of a round, but no indication of anything otherwise. Has anyone tried this before ? and what might be the main problem. My main concern is that the Motherboard has had a bug but I won't by a new one before I am somewhat sure that that's the issue. Would be annoying to find it was the PSU or CPU after buying a new MB. Setup is in the signature.
remove as much ram as possible, all usb excluding usb keyboard. strip the system down the basics as much as possible, Press reset switch on the system board. or connect reset pins. Take a note of any bleep codes on start up.
The annoying thing is that I don't have an extra just lying around, this one has been going strong for about eight years. But I'll see what I can find
I'll try that when I can. But would a RAM problem not allow for it to at least POST/BIOS Have already tried to reset the MB
No, a notebook power supply cannot be used on a PC. So if you have a friend or someone you can borrow one from, I recommend you try that. It is important to at least eliminate your PSU as the problem, even if not an obvious power problem, since a failing PSU will affect everything else.
Its difficult to diagnose things without spares to be honest. With a power supply you could probably put a load on it and test if the voltage is holding using a multimeter. But you would need some kind of load and some kind of multimeter. It wouldn't surprise me to see a power supply give up after 8 years of use.
The problem is, you need a variety of "realistic" loads to do it right and even then, most multi-meters don't measure for ripple and other anomalies. So to test a PSU properly, it must be done by a qualified technician using an oscilloscope or power supply analyzer - sophisticated (and expensive) electronic test equipment requiring special training to operate, and a basic knowledge of electronics theory to understand the results. Therefore, conclusively testing a power supply is done in properly equipped electronics repair facilities. I keep a PSU Tester in my tool bag in my truck. These are handy for verifying a dead PSU, or one that is missing a voltage. But they don't test for ripple either and only provide a small, typically 10Ω load. So for the normal user, swapping in a known good PSU is still best way to conclusively test if a PSU is bad.
Yeah ok, but we're not talking about finding unusual behavior here. Its not even powering on. I think a resistor and a meter would be enough to run a basic check to see if it can maintain voltage under load.
I guess that depends on what a "basic check" is - not to mention the load. That will tell you if the PSU will power on, but that's about it. And who has resistors laying around? And what size is appropriate for a "realistic" load? And how would a "normal user" know what pins to insert the resistor leads into? Anyone with that kind of knowledge probably has a spare, or access to a spare PSU through a trusted friend or another computer. The problem is a PSU can work fine at 20% load when the computer basically at idle. But when tasked and the demand is in the 100s of watts, it can become unstable, or beyond voltage tolerances, and/or overheat. And again, a most meters don't test for ripple. Just powering on, which you can do with a paperclip is not enough either. It tells you something, but not that all 3 voltages are present and within the required ±5% tolerances. There's really no way around this to conclusively eliminate the PSU as the problem. Another, known good PSU needs to be swapped in. Otherwise, any further troubleshooting results will be in doubt since flaky power can affect everything else.
Dude you are way over thinking this. Look at the symptoms. The computer won't power on, nothing is happening when he tries to turn it on. The power supply is old it is the most likely culprit (but it could equally be something else). The guy doesn't have access to a power supply. So lets try and expand his options to help resolve the issue rather than making assumptions. You keep talking about ripple and all of the many issues the power supply can have without focusing on the problem. If a psu is suffering from excessive ripple or the voltages are going out of tolerance will the computer exhibit the symptoms of being dead as a door nail? Run checks on equipment that you can do. The problem may not be revealed in a load plus voltage check, but its also possible that it may high light the issue. Just because you can't check everything precisely or the check won't empirically remove it as a problem doesn't mean you should just give up.
I agree 100%. But you are wrong when you say nothing else happens. The fans try to spin for a second. That suggests the motherboard is signaling the PSU and the PSU is trying to start, but either can't, or is being signaled by the motherboard to immediately shut down again - as may happen with a bad CPU or no CPU fan is detected, among other things. But how likely is it he has a spare CPU to try? He can, and should check to make sure all fan and power connections are securely connected but beyond that, then what? How? With what? With no power to any of the other equipment, what are you going to check? And how? I am not overthinking or making any assumptions at all. This is just basic electronics troubleshooting 101! Is it plugged in? Is it turned on? Is power getting to the equipment? And who suggested giving up? I understand what you are saying and certainly sympathize with the OP. But everything in the computer depends on good, clean power. So when there are apparent power problems that's where you must start. Finding or borrowing a PSU is much easier than a spare CPU or RAM and certainly there are much fewer concerns with compatibility swapping in PSUs too. Even a low wattage PSU will do in this case as there is no need to run Prime95 or max the system out gaming to see if the original PSU is at fault. What else would you have him do? Who has a resistor laying around? And even with one, or a paperclip, and he knew what pins to cross and the PSU starts, that does not tell you all three of the +12V, +5V or +3.3V are present and within tolerances. So with out proper power, the RAM, the CPU, the graphics card, even drives will not work correctly.
I have a number of resistors lying around. Maybe the OP doesn't, but maybe he knows someone who does. Just as he may know someone with a spare power supply, or someone who would let them take apart their PC to check. You're also assuming that he couldn't follow the simple practice of putting a resistor between two connectors and looking at multi-meter. If the power supply voltages were present but not in tolerance would he see the issues he seeing? If there was excessive ripple would he see the issues he seeing?
No I'm not assuming anything and that is why I am sticking to my guns here. As an electronics technician, I can't assume a PSU is good unless it is properly tested. But you are now assuming he has a multimeter. Most households don't. And once again. neither a resistor or a paperclip will conclusively test a PSU. If you have resistors laying around, you are NOT a normal user. And to that, most users cannot read the resistor color coding, or know what size to put in there if they could. And if you have resistors laying around, I would bet you might have a spare PSU, or computer you can pull the PSU out of too. Sure, if the voltages were out of tolerance, he could very easily see these symptoms - depending on which voltage and how far out of tolerance. Ripple probably would not cause this symptom, but again, it depends on how much ripple. The standard is a maximum of 50mV p-p on the +5 and +3VDC and 120mV p-p on the +12V and most motherboard regulator circuits do a good job at compensating for that. But if a suppression filter in the PSU has failed and putting a lot more than that riding the DC, then sure, the motherboard may refuse to start, signaling the PSU to shutdown before the fans have time to fully spin up. So yeah, ripple is not likely the problem but cannot be excluded either. I ask again, what else would you have the user do here? How can he test his RAM, CPU, graphics? Is he to assume his CPU or motherboard are bad?
I would have him swap out the psu. If he doesn't have access to one, I would have him try to test his own psu to the best extent that he can. If he has no means to do that, I would have him remove the motherboard from the case and try and build the PC with as few components as possible, to remove the possibility of shorting or some other device failing. If none of that shows any results I would have him buy a new power supply and try that. Note that it ascends in order of likelihood to show a problem, difficulty/hassle and expense. That is the basic order of things.
Well, that's what I've been saying all along. If none available, I would suggest a tester, as I mentioned before. But after that, we differ significantly. Removing the motherboard can result in damage due to mishandling, or more commonly, ESD. And it can be a daunting task for the less experienced. Instead, I recommend pulling the PSU and taking it to a shop. Many shops will test it for nominal fee, or like my shop, for free - on the spot. This being an Corsair HX1000W - it was not cheap. I would hesitate spending more on a new PSU unless I knew this one was bad, and not under warranty. That adds more weight, in my mind, to have it tested. Unless some foreign object somehow fell under the motherboard, it is not likely something under there is shorting out the power (that typically only happens during computer assembly). Certainly you can pull power from the drives and any other attached devices then attempt to boot. If one of those devices is bad, you should make it though POST (power on self test) and it will stop when it cannot find a boot drive. You can even try pulling all but one stick of RAM and tying, swapping RAM until you get through all. Just remember to ALWAY unplug the PSU from the wall and touch bare metal of the case interior before reaching in. This will discharge any static in your body minimizing ESD (static discharge) potentials.
The guy has 1200 (aprox) posts and probably built the PC himself. He more than likely knows how take a motherboard out of a case and set up a bare bones build Putting a resistor across a voltage source and 0v, turning on the psu and measuring the voltage on the resistor is exactly what I'v been suggesting and is exactly what the bloody psu tester does .
And what ohm resistor would that be? What current value should it be rated at? What pins do you use to measure all 3 voltages? What pins to signal the PSU to turn on? I agree, if you have the proper resistor laying around, and you have a multimeter and you know where to put the resistor and probes and how to short the two pins to start the PSU, and you know what the high and low tolerance range is, then go for it! But not many people, even experienced computer users have those things or know how to use them. Why are you assuming they do? If you don't have the proper resistor (one that won't light up like light bulb - or worse burn up and spew hot carbon bits in your face), a multimeter and know where to put the probes too, and you don't have PSU tester that does all that for you, safely - then take it to someone who can test it properly. You still need to know if you have good power before you can assume anything else is bad. Especially before spending money on anything else.
