Desktop won't power on

[Steven R]

Yes, the results sound like it is a short somewhere and in all test so far you've had it in the case right? You know the PSU shuts down for protection if it detects too high of a draw (a short).

Maybe the new motherboard was bad when you got it or even worse maybe something damaged the old board and now it has damaged the new board also.

OK, here's what I see. I took the mobo out of the case and now nothing powers on at all. I also see that the case is designed with a metal riser that the board attaches to so that some wiring can be fed through the case. Each attachment point on that riser is metal. There is nothing between it and the board. Having pulled the whole case apart now, I also see what looks like small plastic peach colored beads. As if something got spilled into the computer. There's only a couple of them though and nothing inside would be made out of that material. I am wondering now if the case has shorted out both boards now.

 

[Steven R]

Latest test:

I took the old motherboard and the brand new power supply. Placed the old motherboard on the antistatic mat, hooked up the 24-pin and the power button. Nothing else except ram. Not even the CPU.

Result: Turns on/shuts off.

 

[Steven R]

My anti static mats are conductive Hope you had an insulating layer between MB and mat

Ha. Yes I did

 

[B Trevathan]

My anti static mats are conductive

Yep, they should be, and they should be grounded so that any electric static charge can be allowed to safely discharge through a path of least resistance to ground and not fry (hopefully) a sensitive motherboard component. This is why you also hook anti-static wrist straps to the a grounded part of a computer so that the charge even if you don't feel it or see a spark can go to ground instead of shocking a component. When doing out of case testing it would be good to place boards on a non-conductive material such as a counter or wooden table as I mention above.

 

[Steven R]

Yep, they should be, and they should be grounded so that any electric static charge can be allowed to safely discharge through a path of least resistance to ground and not fry (hopefully) a sensitive motherboard component. This is why you also hook anti-static wrist straps to the a grounded part of a computer so that the charge even if you don't feel it or see a spark can go to ground instead of shocking a component. When doing out of case testing it would be good to place boards on a non-conductive material such as a counter or wooden table as I mention above.

Everything was properly set up. I do everything possible to minimize the risk of damage to the components or me.

 

[Steven R]

I have given up on this. I put everything back in the case, hooked it up and closed it. There is no power at all now when the power button is pressed so I something shorted out both the original and the replacement boards. With all the testing I've done, I'm pretty certain there is an issue with the prewired case (prewired for the power button, the sleep button and all the lighting). Obviously I could be wrong but I can't see anything else that is causing these issues.

And swinging the chicken over my head didn't work but that might be because I couldn't stomach the idea of spitting perfectly good whisky out

I really do appreciate everyone's help. I wish it turned out better but sometimes machines fail.

 

[B Trevathan]

I have given up on this.

Sometimes you got to just let it go.. don't let it get to you, you tried, maybe you gained something little from all our postings.

You did test the new motherboard out of the case?

Did you say you tried another CPU fan or just tested the old one? Fans like PSU should be swapped out, you can test them with a meter and they show good, you can blow on them and they spin, you can turn them and they spin, then you hook them up and the computer only turns on for half a second and shuts off.

Better luck next time.

 

[mr m]

Like B Trevathan said don't beat yourself up over it. You did everything possible. Take a shot of that whiskey, you earned it!

As far as anti static mats go I was taught to use them religiously and still use them all the time. It was mentioned earlier but is worth repeating. The mat must be grounded to the buildings ground. The wrist strap is grounded via the mat. There should be a 1 M Ohm resistor in the wrist strap to protect you.

All surfaces hold a static charge in varying degrees. Wood, paper, plastic or an anti static mat. Any molecule rubbing against another creates static. They might not make good conductors they do hold a charge. The anti static mat is designed to safely dissipate the static charge to an earth ground. The idea is to remove that charge before it builds up to dangerous levels and zaps sensitive components like semiconductors.

Some of the static shielding bags, at least back in the day, that protect your modules were called nickel bags because of the high nickel (metal) content. Now you gray hairs out there know what I'm talking about but you youngsters are paying way more today.

 

[Steven R]

Like B Trevathan said don't beat yourself up over it. You did everything possible. Take a shot of that whiskey, you earned it!

As far as anti static mats go I was taught to use them religiously and still use them all the time. It was mentioned earlier but is worth repeating. The mat must be grounded to the buildings ground. The wrist strap is grounded via the mat. There should be a 1 M Ohm resistor in the wrist strap to protect you.

All surfaces hold a static charge in varying degrees. Wood, paper, plastic or an anti static mat. Any molecule rubbing against another creates static. They might not make good conductors they do hold a charge. The anti static mat is designed to safely dissipate the static charge to an earth ground. The idea is to remove that charge before it builds up to dangerous levels and zaps sensitive components like semiconductors.

Some of the static shielding bags, at least back in the day, that protect your modules were called nickel bags because of the high nickel (metal) content. Now you gray hairs out there know what I'm talking about but you youngsters are paying way more today.

I'm gray but young too

I don't beat myself up over something like this. I just don't like telling a customer who was hopeful that it's dead but they had an idea it might be when they called me.

No matter whether I succeed or fail, I always learn something and I've learned a lot from the responses I've gotten today from everyone. And that makes this effort worth it.

 

[Steven R]

Sometimes you got to just let it go.. don't let it get to you, you tried, maybe you gained something little from all our postings.

You did test the new motherboard out of the case?

Did you say you tried another CPU fan or just tested the old one? Fans like PSU should be swapped out, you can test them with a meter and they show good, you can blow on them and they spin, you can turn them and they spin, then you hook them up and the computer only turns on for half a second and shuts off.

Better luck next time.

Yes, I tested both boards out of the case. The CPU fan I didn't test with a known good one. But I had unplugged the fan when I was doing all that testing so if the fan was the culprit, I would've thought it would work without the fan. Then again, this has been the strangest job I've ever had come across my bench so I guess I shouldn't have ruled it out.

 

[B Trevathan]

The mat must be grounded to the buildings ground. The wrist strap is grounded via the mat.

I think it would be a little better if the wrist strap was grounded to the computer and then the computer grounded to the mat and then the mat grounded to an earth ground (if possible).

Believe it or not I know techs that think the way to ground a computer is to have the power cord plugged into a wall outlet.

 

[mr m]

I think it would be a little better if the wrist strap was grounded to the computer and then the computer grounded to the mat and then the mat grounded to an earth ground (if possible).

Believe it or not I know techs that think the way to ground a computer is to have the power cord plugged into a wall outlet.

The thinking is that the metal chassis or module is grounded to the mat just by sitting on it.

I carry a mat in my truck but in a pinch I've left the AC power cord in the wall also.

From what I see most techs don't bother to use a mat in the field or bench. ESD doesn't always destroy a component right away buy may weaken it greatly. Every time you power a component on it dies a little bit. Think of a semiconductor as a bucket of sand. Each power cycle kicks that bucket and a few grains fall out. Over time you lose enough sand and the component fails.

This is why I leave my PC's on 24X7. I also do this with my TV's if I'm going out for a bit. It's the surge of current into a relatively cool component that burns out electronics. Think of a light bulb. They rarely fail while on but usually pop when you first energize them.

 

[B Trevathan]

The thinking is that the metal chassis or module is grounded to the mat just by sitting on it.

My thinking is the part that would be sitting on the mat might be painted or plastic coated or just not making a good enough contact. A grounding wire fasten to the mat on one end and an alligator clip on the other end clipped to a ground area on the computer would make a better connection don't you think than it just sitting on the mat?

but in a pinch I've left the AC power cord in the wall also

Not me, if I have to "in a pinch" I try to dis-charge myself to a ground and then while working I try to keep in contact with the computers frame.
But to each his own.

Every time you power a component on it dies a little bit

That is just sad...

 

[B Trevathan]

Yes, I tested both boards out of the case. The CPU fan I didn't test with a known good one. But I had unplugged the fan when I was doing all that testing so if the fan was the culprit, I would've thought it would work without the fan. Then again, this has been the strangest job I've ever had come across my bench so I guess I shouldn't have ruled it out.

Get yourself some known good parts, RAM, PSU's, HDD, burners, fans, etc. it will help you in the long run.

Speaking of CPU fans I replaced one in one of my older computers not too long ago and I think the cord was probably two feet long, I was thinking why in the world was it so long because most fan headers are no more than a few inches away.

You don't really fail if you learn from the things you do.

 

[mr m]

My thinking is the part that would be sitting on the mat might be painted or plastic coated or just not making a good enough contact. A grounding wire fasten to the mat on one end and an alligator clip on the other end clipped to a ground area on the computer would make a better connection don't you think than it just sitting on the mat?

Good point.

 

[brandonkick]

http://www.technibble.com/forums/showthread.php?p=395060#post395060

My advice is to start there. It seems like you have made peace with just calling it a day on this one, and that is fine but I feel like you could have done yourself just a bit more justice here.


That CPU could be bad and causing the entire issue. If I recall correctly, a motherboard will not fire without a CPU inserted? I could be wrong on that though. I guess I've never really tried?

But the starting point here is always bare minimum hardware, and you can go a step further then the guide that I've started by removing the motherboard and power supply from the case. You can set them up outside of the case and power the board up manually by shorting out the pins that the power switch was on.

 

[westom]

Then again, this has been the strangest job I've ever had come across my bench so I guess I shouldn't have ruled it out.

Recently had a system that did exactly what you are seeing. A problem was identified in but a minute by using numbers from a multimeter. And by not disconnecting even one wire. (Any changes can only exponentially complicate the problem.)

No numbers are provided. So the fewer who know much more cannot comment. Also missing are discussions about what causes a computer to power on, power off, and do as you have seen: a power controller. Even the switch does not power on a machine. A front panel power switch only requests the power controller to act. Other inputs also cause a power "system" to power on then quickly power off. All that can only be analyzed by numbers from a meter.

Why is a PSU tester so useless? It also does not provide numbers. Tests in a manner that cannot see many PSU defects. And does not report on the many other components in a power "system" - including the power controller.