1

IMPORTANT

The problem turned out to be that the CMOS battery was exhausted, so changing it was enough. Make this test as the first test on the motherboard, because you could give for dead the whole mobo just because of the battery!

Original question

So suddenly, overnight, my computer doesn't turn on anymore: not a single sound or light when I hit the power button.

This is the system:

  • psu: Cooler Master V850
  • case: anidees AI-CL-Cube-W-RGB Cubo Case
  • mobo: Gigabyte X399 AORUS Gaming 7
  • cpu: AMD Ryzen Threadripper 1950 X 3.4GHz 32MB L3
  • ram: Corsair Vengeance RGB LED 32 GB, 2 x 16 GB, DDR4 2666 MHz C16 XMP 2.0
  • vga: Gigabyte AORUS GeForce GTX 1070Ti 8G

(I have omitted other components, like the HD, as it being broken is not compatible with a no-sound-nor-light-anywhere, as far as I know; but if more info is needed, please, let me know.)

My first thought was the PSU, so I tried the paper clip test, and I hear the fan noise from the PSU. Is this test enough to assume that the PSU is working, or at least that the whole system should give a minimal sign of life?

If the answer to the above question is yes, then my understanding is that it could be a mobo failure and or something wrong with the power button of the case, the reason being that since cpu ram and vga are connected to the mobo, my understanding is that if the mobo was ok, I should at least see some light on it, for instance the one telling me the OS is not standing yet.

So how can I make sure the power button is ok, and, if it is, that the mobo is totally dead?

In hindsight, is it possible that brownouts and "the opposite of brownouts" can break a mobo? I thought that they mostly kill the PSU, which I way cheaper than my mobo :(

Improve this question
2
  • "or something wrong with the power button of the case" You'll have a lead from the switch to the headers on the motherboard. Use a small flat-headed screwdriver to short the 2 pins the power lead header connects onto. This will rule out the switch if the system still won't power up. Pins 3 & 4 on the F_Panel header bottom right on the board.
    – spikey_richie
    Commented Aug 7, 2020 at 7:11
  • Yes; Brownouts could damage your hardware
    – Ramhound
    Commented Aug 7, 2020 at 13:33

2 Answers 2

Reset to default
1

While brownouts (loss of power) are not likely to damage your computer, power surges when the power comes back on, definitely can. Electrical devices that you do not want damaged from surges should be plugged into a surge protector, at minimum. PSUs generally have fuses in them, but they are not a guarantee that your computer will be safe.

The first step you need to do is check if your power supply unit is functioning properly. The "paperclip test" is not a useful test. All it does is show that PSU will turn on, nothing more. While knowing it will turn is good, it doesnt tell you if the PSU itself is damaged and outputting the wrong voltages, which can destroy your computer. These voltages can be read with a multimeter. However, based on your post, I gather you lack knowledge in electronics. I would suggest purchasing an ATX power supply tester. You do not need a fancy or expensive one. This will tell you if the PSU will turn on and far more importantly, if it is outputting the proper voltages.

Once you know your PSU is working properly, then you can move on to troubleshooting other issues.

Improve this answer
6
  • 1
    If you have another setup.. you can also go the other route and pull things out one by one and test them in a different box. As they succeed.. put them aside and text the next component.
    – Señor CMasMas
    Commented Aug 7, 2020 at 18:54
  • @SeñorCMasMas, that would definitely be the best solution, but I don't have another setup, unfortunately.
    – Enlico
    Commented Aug 8, 2020 at 18:56
  • However, I have had the PSU tested and it works; and the mobo has been tested and seems completely dead.
    – Enlico
    Commented Aug 8, 2020 at 18:56
  • Mala suerte para ti. ¡Lo siento! Espero que las cosas mejoren.
    – Señor CMasMas
    Commented Aug 8, 2020 at 19:18
  • @SeñorCMasMas while that is definitely an option, I doubt it is one he has. He could have just used to PSU from the other machine to test the dead one. Also, the machine has to use the same RAM, CPU, etc.
    – Keltari
    Commented Aug 8, 2020 at 20:58
0

Gigabyte tech support solved my issue, which was: the CMOS battery was probably exhausted, or close to that.

This is the main part of their message:

Please try a full reset of the motherboard:

  • Switch off power supply and wait 30 s
  • Remove CMOS battery.
  • Short-circuit the contacts in the battery holder for about 2s.
  • Remove short-circuit.
  • Reinstall CMOS battery.
  • Now short-circuit Clear CMOS jumper for 10 s.
  • Remove short-circuit.
  • Turn on the PSU switch.
  • Start system and immediately boot into the bios setup.

I will now accept my own answer, because, in hindsight, anybody with good hardware experience (unlike me), upon reading the title PC does not boot (no sound, no light, nothing) should be able to make the hypothesis that the CMOS battery (which I didn't even know what it is) is exhausted, along with other hypothesis

Improve this answer
3
  • Weird enough that replacing the CMOS battery did the trick. The battery itself is just to prevent data loss from the CMOS, a volatile memory storing e.g. BIOS configuration parametres. One can usually identify a dead CMOS battery by recognising that the displayed date and time (no NTP-sync enabled) is reset to a date/time far in the past. However - apart from that a dead CMOS battery does nothing more - your PC will boot just normally, but with default BIOS parameters (as any custom once are stored in the CMOS).
    – Yannik Z.
    Commented Aug 12, 2020 at 19:52
  • @YannikZ., in this case would you say that it was the short-circuit Clear CMOS jumper step that did the trick, and not the new battery I put on?
    – Enlico
    Commented Aug 12, 2020 at 19:56
  • Frankly unlikely, since any changes you made on the BIOS settings did not prevent your computer from booting before that faithful night. Albeit - I am not really into the physics behind CMOS on cell-level, but maybe a significant voltage drop of the CMOS battery (as it occurs while being discharged over time) could lead to a unpredicable state of the CMOS itself, e.g. some cells were still powered and retained data while the voltage threshold for others was too low, resulting in a partial lost of data. As a result, the PC won't start anymore. But that's just a wild guess
    – Yannik Z.
    Commented Aug 12, 2020 at 20:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .