Resolved Power Up Question

I have a PC that was hit by an electrical outage (although there was a UPS, it apparently failed when I needed it most) and zapped something that would not even let it power up. My first logical choice (also because I'm cheap) was to replace the PS. That did nothing. It then became obvious to me (wrongly I guess) that something happened to the mainboard. So I then replaced that. Much to my chagrin, still nothing. That leaves only one component (of course the most expensive) in my mind, the CPU. However, it seems inconceivable to me that a bad CPU can prevent the machine from even powering up. And by that I mean no fans, no mainboard LEDs, etc. Before I throw away more money, is that possible or is there something I overlooked that could cause this problem. Other than PS, mainboard and CPU, what else is there?

 

A fuse in the plug to the wall?

 

If it were only that easy.

 

Sometimes the blindingly obvious is the hardest to see.

You need to start from the beginning. Are you sure the PSU is functioning properly? Do you have a PSU tester?

 

Hi gw1500se, Please take the time to fill in your System Details in User CP and don't forget to scroll to the bottom and Save.
We can only assume at this point that you have checked to make sure you have AC power from the wall socket - power lead to PSU and that any Fuses have been checked as Don suggested.
Case switch may have been zapped - pay to check.
Any loose/bad connections may have been fried if the voltage was high enough. (Meaning internal cables/leads). Neil.

 

Since one of the first things the BIOS does during POST (power on self-test) is to check for the presence of a central processor, yes a bad CPU can prevent the computer from starting. Or rather, it will start, but notice the bad CPU and immediately shut down. However, you typically see the fans just start to spin for a 1/2 second before they too shut down.

Note that bad RAM can prevent the system from booting too. I would try with one stick at a time - this after verifying you have good power.

And don't give up on using an UPS. I am sorry yours failed you when you needed it most, but the fact remains, every computer should be on a "good" UPS with AVR - automatic voltage regulation. Note, like PSUs, there are cheap UPS and good UPS. A perfectly functioning cheap UPS may work great for a light bulb, but not be near quick enough (at cutting over to battery power or at blocking extreme anomalies) to "hold" power to the high-speed digital electronics of today's computer systems. Note the ATX Form Factor Standard for computer power supplies only calls for the PSU to deliver power for just 17 (might be 19 - too lazy to verify this morning) milliseconds in the event of power loss. That's just 19/1000th of 1 second. Not very long.

Also note UPS batteries MUST be replaced about every 3 years.

 

Thanks for the replies. Let me clarify a few things. Yes, the PSU is verified good, I put it on another system and it works fine. The fans do not spin even for a fraction of a second. The machine is completely dead, no lights anywhere, not even a flash or a click. I put a Vohm on the case switch leads and verified that it works.

I think I read somewhere, but I cannot find it any more, that there are 2 pins on the main power connector that must make a complete circuit for the system to power up. I have not checked connectivity of that since I cannot find which pair of pins it was.

As for the UPS, I do recognize the need and have not abandoned it. Its just a bad bit of luck that the batteries failed 2 months after the 1 year warranty expired (I won't be using that brand again).

In any case, I think you have answered my main question in that this problem is not a bad CPU based on the symptoms. It is beginning to look like I have a bad mainboard out of the box.

 

These are the two pins from the motherboard's front panel's power button. The button provides a "momentary" short between those two pins, which signals the PSU to power up. Note this is a momentary circuit. The two pins are shorted only when the button is pressed.

Is this the exact same motherboard? Note all PC cases are designed to support 1000s of motherboards. So there is almost always more motherboard mounting holes in the case than the motherboard. A common mistake is to ***** in more standoffs in the case than the motherboard has mounting holes. This causes the motherboard to short out. If lucky, once the extra standoff is removed, all works. If not lucky, permanent damage occurs.

If me, I would pull the board and assemble the system outside the case. Then verify standoffs match motherboard mounting holes exactly.

 

Thanks again. Yes, the standoffs matching holes in the mainboard are the only ones present as I had to rearrange them all for this new board (thus it obviously is a different board as the old one was obsolete any way).

As for the pins I am almost certain I read there were another pair of pins independent of the case switch that also had to make a complete circuit, at all times, for it to power up when the case switch is pressed.

 

Nope. No such pins or circuit. As long as all of the necessary PSU connections (and most current motherboards have at least two) are made, and solid. that's all you need.

 

****, this is driving me crazy. There is absolutely no reason why this will not power up or do something. It defies all logic unless the brand new, unused, mainboard is bad.

 

OK. So you put a Voltmeter on the switch on the case.

2 things here.

1. Did you use the voltmeter to check the voltage at the switch or the connectivity?

2. When you checked the PSU on another machine, was it connected to the same socket in the wall? Or did you take the PSU to the other machine?

 

1) Connectivity.

2) Same wall socket.

 

TBH, thats now becoming more likely.

Is it possible to borrow a compatible CPU from somewhere to verify it or take your CPU to another compatible machine?

Like Bill's already said, even if a major component was faulty, you'd at least get something while it was going through POST although it would then hit the buffers.

 

Alas, no. I this is the only socket 775 board I have. I do have other 775 CPUs none of which are known quantities but they all behave the same.

 

IMHO then, its unlikely that they are all exhibiting the same fault.

Therefore that leaves the MOBO.

 

It is hard to disagree. OTH, this is exactly the same symptom after the original power failure with the old mainboard. Like the CPUs it is difficult to imagine a brand, spanking new, mainboard will fail the same way out of the box.

 

Disconnect all drives any extra USB devices besides the mouse and keyboard, pull all cards (except graphics if not using integrated), pull all the wires from the case's front panel, and pull all but one stick of RAM. Make sure both motherboard power connections (I am guessing 2, there could be 3) are firmly connected. And of course your monitor is powered on and connected properly too. Then plug in and boot. You should see on the monitor the full POST, then it will halt when it cannot find a boot drive.

I use a small flat tip screwdriver to short the two motherboard pins to signal the PSU to boot. Whatever you use, just have a steady hand and remember, it is a momentary "circuit" you are shorting. Meaning, once you short them you can immediately back out as the circuit will ignore any more "input" until the boot cycle is complete.

And the monitor does work, right? Might try another monitor to be sure.

Are you observing good ESD precautions? That is, before reaching in, do you unplug from the wall to remove the +5Vsb standby voltage? Every time? Do you touch bare metal of the case interior to discharge any static in your body (putting you and the case at the same potential)? Every time?

Proper ESD precautions are essential when handling (or just getting within a couple inches) of any ESD sensitive device - including CPUs, GPUs, memory modules and the many many integrated circuits (ICs, "chips ") mounted on the motherboard - all so sensitive they can be totally destroyed by a discharge (arcing electricity) so tiny we (humans) aren't even aware an arc occurred.

While DOAs are rare, until Man can create perfection 100% of the time, there will always be some brand new, never used products that fail right out of the box. But sadly, when it comes to motherboards, CPUs, and RAM sticks, I suspect ESD is the real cause of "apparent" DOAs, rather than an actual DOA.

 

Thanks. I'll give it a shot but I have little hope at this point. Yes, the monitor is fine and I always use an anti-static wrist band so the possibility of ESD is low. I don't think there is much doubt the MB is bad since the PSU works. If it were memory I would expect some lights, fans and/or beeps, even if briefly.

 

Just a question about the PSU connector to the board. How many pins - 20 or 24? Neil.
ps. Please enter your "System Details ", it really helps in answering your questions.