Have I destroyed my motherboard?

[Phil-M]

I have just bought a used Dell OptiPlex 5060 mini tower and after unpacking it, if I remember correctly, it started up as soon as I connected the power cable (but I could be wrong about that). It ran perfectly, and later I tried to see if a GTX 1050 TI graphics card would fit. I wasn’t intending to fit the graphics card with the computer running but I wanted to see if there was enough room from front to back. I tried to get the card near the PCI E slot but there wasn’t a lot of room. Before the card even touched the PCIe slot I noticed it was touching some wires near its back end. Suddenly the fans on the computer stopped.

Some hours later, I returned to it and reconnected the power cable and – immediately (before I touched the on/off switch) the computer started up – there was a screen message saying it hadn’t started properly and offering suggestions. The first was a reboot and I chose that one. It then started perfectly and ran perfectly. After some time, I shut down the computer in the proper fashion.

Some hours later still I returned to the computer and pressed the power button but it wouldn’t start. The fans wouldn’t run and nothing was working. The power supply unit LED was solid green at the back of the case. There is no self-test button.

As an experiment to detect a loose connection, I again put the graphics card near the PCIe slot and wiggled it around. While I was moving it around, the fans started up for a few seconds and then stopped. This made me think there was a loose connection and I had disturbed it by moving the card but I have checked all the wired connections to the motherboard and they all appear to be sound.

As I write this post, the computer is still not working and the power supply unit LED on the back is still solid green. I have checked the output from the PSU with a multimeter and it is delivering power to the motherboard. The wires from the back lead to 2 connectors: a 6-pin (system power connector ATX) and a 4-pin (CPU power connector). I've tested both with my multimeter and the 6-pin connector reads about 4.3 V between the green wire and earth and the 4-pin connector reads 12 V. between green and earth

Any suggestions would be most welcome

 

[Holst1981]

Working on a running PC is a good way to break it.

Please take this to a professional.

I turn mine off before i work on it

 

[Phil-M]

Working on a running PC is indeed a good way to break it. You are absolutely right. I won't be doing it again, I assure you.

I'm hoping somebody on this forum can tell me, from what I've described in my original post, what damage might have been done and to which components.

 

[ANDARIAL]

reset the cmos,maybe

 

[Pulseammo]

Faulty case switch perhaps?

If it's starting when the power cable is connected (this can happen for a few reasons), then one possibility is the case switch is stuck closed, and pressing it has caused it to open but stay open if it's broken. So now when you press it nothing happens.

Use your multimeter to test the case switch perhaps? You can also bridge the contacts on the motherboard pin header where that case switch connects to see if it starts but that might be getting a bit advanced .

 

[Phil-M]

reset the cmos,maybe

I've removed the CMOS battery and tested it with a multimeter and it seems to be completely dead. I replaced it with a new battery but, alas, the PC still won't show any signs of starting up.

 

[sx_turbo]

Working on a running pc is like putting a knife in a toaster.

As has been stated already, look at the front panel connections specifically the power button, and make sure the power button is working correctly,

It's unscrupulous, but I might be tempted to try and return it to the purchaser in exchange for a replacement.

 

[Phil-M]

Faulty case switch perhaps?

If it's starting when the power cable is connected (this can happen for a few reasons), then one possibility is the case switch is stuck closed, and pressing it has caused it to open but stay open if it's broken. So now when you press it nothing happens.

Use your multimeter to test the case switch perhaps? You can also bridge the contacts on the motherboard pin header where that case switch connects to see if it starts but that might be getting a bit advanced .

I'm not sure how to test the case switch but, yesterday, following instructions online, I did bridge various contacts on the power switch pins on the motherboard (there are 5) – all to no avail.

 

[Phil-M]

There's advice online showing how to test the PSU by bridging a couple of contacts with a paperclip but all the examples show a 24-pin connector whereas my PC (Dell OptiPlex 5060) has two groups of wires coming from the back of the power supply – one leading to a 6-pin connector (ATX_SYS) and the other leading to a 4-pin connector (ATX_CPU). My multimeter shows just over 3 V at the 6-pin connector and 12 V at the 4-pin connector. Does that suggest the power supply is working properly? Or should I bridge a couple of the contacts – if somebody can tell me which ones to bridge?

 

[Pulseammo]

I'm not sure how to test the case switch but, yesterday, following instructions online, I did bridge various contacts on the power switch pins on the motherboard (there are 5) – all to no avail.

There shouldn't be 5. This might work differently on a Dell PC but on a standard motherboard the pin header for the case switch is 2 pins, one will be ground and the other will be a signal wire, by shorting the signal wire to ground the PC should prompt the PSU to start. The pin header is usually next to other stuff for the front IO though such as reset switch, case speaker, power LED and HDD LED, sometimes all the front IO has a combined connector to help with wiring, but the power switch its self will only have 2 distinct pins.

Unless you have absolute certainty of which pin is which you shouldn't be bridging anything at random at all, you'll end up damaging something.

The way you'd test a case power switch is the same as any other momentary switch, disconnect its header from the motherboard, put your meter probes each on the metal contacts (may be difficult because they'll be recessed in plastic), put the multimeter into continuity mode (usually comes under resistance then you push a button so it beeps when probes are touched), then operate the case switch and if the meter beeps you know the switch is closing properly.

Honestly I wouldn't go faffing about with this any more, you should have someone look at this for you.

 

[Phil-M]

Wise words, Pulseammo.
I've just been into town to the PC and laptop repair centre, described the problem, and they say it'll cost £100 -£150 to diagnose it and a new motherboard will be another £100-£200.
Sod that! I've now sent off to the USA for a guaranteed functional motherboard on eBay (yes, it's an exact match for the existing one) and I'll install the bugger myself.
Needless to say, I shall disconnect the PC from the mains while I'm doing this.

 

[ANDARIAL]

You can get another same pc complete from ebay you know?
you would have xtra ram etc if you did that

 

[Phil-M]

Yes, that's true. I shall look into that if my replacement motherboard doesn't do the trick. I'm told it should arrive sometime within the next month.

 

[Phil-M]

I have now replaced the motherboard with a board from a reputable company that had more than sixty 5060 tower motherboards in stock.
The only change is that, with the original board, when I connected the mains power cable to the PC power supply, the green LED on the power supply came on and stayed on permanently but now, with the replacement board in place, when I connect the mains power cable to the PC power supply, the green LED comes on for a few seconds and then goes out again. According to Dell, this 'LED illuminates for three seconds then goes off' response is the correct response to connecting the mains power cable if the PSU is delivering power to the motherboard.
But the fans still do not work and PC won't switch on.

 

[Retro6]

Disconnect the GPU with the machine off and power cord unplugged and remove the CMOS battery for 10 seconds or longer and push the power button "again while the machine is off" then re-insert the CMOS battery then plug in the power and start up the machine using the onboard graphics if your board has it and then see how it runs.

Give the connections on your GPU a wipe down with some IPA if you have any... if not you can get a bottle on online, its very handy for cleaning components, switches and connectors etc. Re-insert your GPU "with the computer off/power cord unplugged" then go from there.

It might also be worth checking your RAM just encase. You can create a G-Parted boot CD and run MEM Test.