Dead Computer

[westom]

Well that didn't work. Without the MB connector connected the PSU won't turn on.

Nobody said anything about disconnecting anything. Push the probe into the nylon connector as instructed. In fact, good diagnostic procedure is to never disconnect or remove anything until all facts are first collected.

Again, view numbers in your diagnostic report. The error is not related to any bit location or any address line. Numbers suggest you are chasing ghosts.

Follow instructions for the meter. If it does not work, then you have probably misinterpreted the instructions.

 

[mfenn]

Well that didn't work. Without the MB connector connected the PSU won't turn on. I can probe the back plane, but that will require I remove the MB from the chassis and increases my risk of shorting signals. Any ideas on other ways to test the output from the PSU? I'm sure I'm missing something here.

What westom is asking you to do is to stick the probe into the back of the ATX connector (top if the motherboard is lying flat) while it's still attached. There should be enough free space in order to get the probe to touch the contact from the back.

 

[Martimus]

What westom is asking you to do is to stick the probe into the back of the ATX connector (top if the motherboard is lying flat) while it's still attached. There should be enough free space in order to get the probe to touch the contact from the back.

The fit was too tight to fit a probe, but I snipped a couple pieces of used solder wick since that is quite skinny. I stuck that in the ground and 3.3V DC line and checked the voltage. It was a solid 3.383V DC. However, it was very difficult to keep the posts still, so I stopped before I accidentally shorted the lines.

 

[mfenn]

The fit was too tight to fit a probe, but I snipped a couple pieces of used solder wick since that is quite skinny. I stuck that in the ground and 3.3V DC line and checked the voltage. It was a solid 3.383V DC. However, it was very difficult to keep the posts still, so I stopped before I accidentally shorted the lines.

You might want to pick up some fine tips for your probe. That'll make it easier to probe the back of the connector.

 

[westom]

However, it was very difficult to keep the posts still, so I stopped before I accidentally shorted the lines.

Then use a paper clip or sewing needle. Never worry about shorts. Any power supply output can be shorted to any other without PSU damage. That was required even in Intel ATX specs some twenty years ago. One of the so many functions in a PSU that, for example, make it impossible for a PSU to cause motherboard or disk drive damage.

Measure what was aked for. About one minute of labor - plus time to read the instructions. First measured wire was a purple one. And when the computer is off - no voltage on the 3.3. Black probe should have been connected to the chassis - not to any ground wire on that connector. Please follow instructions. Without all numbers taken as requested, then nobody can even say if 3.3 volts is stable - and other important information. Please stop making assumptions and deviating from those instructions since how measurements are taken can adversely affect a conclusion.

You don't need to be a jerk to get your point across. Tone it down a bit.

mfenn
General Hardware Moderator

 

[Martimus]

You might want to pick up some fine tips for your probe. That'll make it easier to probe the back of the connector.

I decided to set it up so that when the solder wick fell out it would land on a piece of paper instead of my video card. Then I tested the pins I could get to. Here is what I recorded:
Pin number; color; voltage spec; actual voltage
12; orange; +3.3VDC; 3.384V
11; yellow; +12VDC; 12.34V
10; yellow; +12VDC; 12.34V
9; purple; +5VSB; 5.106V
8; Gray; PWR_OK; varied voltage
7; black; COM; 0.000V
6; red; +5VDC; 5.038V
5; black; COM; 0.000V
4; red; +5VDC; 5.038V
3; black; COM; 0.000V
2; orange; +3.3VDC; 3.378V
1; orange; +3.3VDC; 3.379V
13; orange; +3.3VDC; 3.378V

All of these readings stayed within 1 significant digit while reading the voltage.
The 12V line was 3% high, the 3.3V line was about 2% high and the 5V line was less than 1% high. The purple line was labeled +5VSB but I don't know what SB means.

The line regulation tolerance is +-1% for all rails and voltages
The load regulation tolerance is +-3% for all rails and voltages

 

[westom]

Here is what I recorded:

Numbers do not report what I requiested. The instructions stated what pin must be measured when. For example a purple wire must be measured only when powered off with AC power cord still connected. Green and gray wires measured both before and as the power button is pressed. As noted in the previous post, varying from instructions means I cannot provide a useful reply.

Measurements should have been only on six wires. If voltage varies on a gray wire, those variations must be listed numerically. Numbers for each wire are important.