[auglocqnuk]

I have a desktop computer that will not completely boot. When I press the power button, it will power up for a split second, then the processor's fan will quit (which makes me assume that maybe the processor fails as well) and go into hibernation mode (I make this assumption because the power led turns off, but the sleep led remains lit). However, when I remove the ram, the computer will power up and function properly without any hardware giving out. But when I return the RAM to the system, the computer acts the same as it did before I took out the RAM. What is the cause of this?

[ScottGem]

Bad RAM chips. Are you removing the all the memory modules? I don't see how a PC can boot with no RAM installed at all.

[auglocqnuk]

That's what I thought as well so I replaced the RAM chips and tried again, but the was no improvement. It doesn't boot, per-say, without the RAM because that is virtually impossible. Actually what happens, is it gets stuck between power-up and system boot when it does not have any Ram present. However, when the RAM is present, the system bypasses the system beet and enters a non-returnable hibernation (that is to say, it enters hibernation in such a way that you must shut down the system entirely, in order to start the system again). When this happens, the processor shuts down, but the hard drive remains active. Yet, this only occurs when the RAM is present, and replacing the RAM does not work. My suspicion is that either the processor or the motherboard has become faulty, because, as I said before, I know the RAM is not the culprit.

[carefreelayner]

I have a desktop computer that will not completely boot. When I press the power button, it will power up for a split second, then the processor's fan will quit (which makes me assume that maybe the processor fails as well) and go into hibernation mode (I make this assumption because the power led turns off, but the sleep led remains lit). However, when I remove the ram, the computer will power up and function properly without any hardware giving out. But when I return the RAM to the system, the computer acts the same as it did before i took out the RAM. What is the cause of this?

Sounds to me as though you have a bad ram

[Tj3]

I have a desktop computer that will not completely boot. When I press the power button, it will power up for a split second, then the processor's fan will quit (which makes me assume that maybe the processor fails as well) and go into hibernation mode (I make this assumption because the power led turns off, but the sleep led remains lit). However, when I remove the ram, the computer will power up and function properly without any hardware giving out. But when I return the RAM to the system, the computer acts the same as it did before i took out the RAM. What is the cause of this?

Assumptions are:

- That the computer previosuly worked normally and is not a new computer
- That when you removed and replaced the RAM, that you removed and replaced all the RAM

A few possibilities comes to mind:

1) You stated that the RAM is good - are you certain that the RAM is fully seated?

2) Check the electrolytic capacitors on the motherboard (usually cylindrical components perpendicular to the motherboard which show the + and - polarity markings and a value of XX uF). If these show signs of leaking or expansion (i.e. expanding around the middle), then the motherboard needs to be replaced (Note: These capacitors are involved in supplying power to the computer)

3) The symptoms cause me to wonder question whether the power supply may be on its last legs. It may have all the voltage rails still putting out power, but it could be that it cannot handle the load and shuts down, or that perhaps one of the sensors intended to prevent overload is not functioning properly and causing the motherboard to put the system into the sleep mode to save power.

4) It is possible that the thermal paste to transfer the heat from the processor (CPU) to the heatsink is not providing proper protection and the processor may be overheating. A repaste job may be necessary - this should be done by someone with appropriate expertise/experience.

If it were my computer, I would check (1) and (2), then try (3), and lastly do the repaste (4). If none of those worked, then I would think that another motherboard problem is quite likely, and would replace the motherboard.

[auglocqnuk]

Tj3:

Thank you for your input. I had already considered all possibilities that you mentioned except Possibility 4. All preceding possibilities had turned out to be false. However, I will now check the thermal compound on the processor. I have already ordered a new motherboard and processor, but I am sure you input will come in handy in the future. If it turns out that the thermal compound is not the problem, do you have any more suggestions?

Thanks again,

Auglocqnuk

[ScottGem]

If you replaced the RAM with modules that you are are good, then the problem is probably somewhere in the connection of the sockets to the motherboard. However, the details of that are out of my league. But since there isn't a practical way to fix that, the solution is replacing the motherboard.

[Tj3]

If you replaced the RAM with modules that you are are good, then the problem is probably somewhere in the connection of the sockets to the motherboard. However, the details of that are out of my league. But since there isn't a practical way to fix that, the solution is replacing the motherboard.

I would consider this unlikely if the motherboard was previously working. For the socket to be an issue only when RAM was installed, there would have to be a short circuit between two or more pins at the interface to the MB, and it is very unlikely for that to have occurred after the computer has been assembled and proven functional. The most likely source of such a fault would be either a fault in the MB material itself which caused an unwanted connection, or a solder hair between pins.

We may see problems such as this in the future because of the imposition of the RoHS (Restriction of Hazardous Substances) requirements in the European Union, China and several US states which, in part. Requires elimination of lead or products including solder. Without lead, solder tends to form "tin whiskers" over time which could cause shorts, and these could and probably would occur on the MB surface between any points where solder joints exist (sockets, chips, etc.). There is an exemption for IT products, but that does not mean that some computer and/or connector/socket manufacturers will not go lead-free regardless.

If it was a connector issue, then the issue would be likely on the RAM board itself, but since the RAM was replaced, that would eliminate this as a cause.