• 20.07.2025, 19:26
  • Register
  • Login

  • You are not logged in.

 

Quadro preventing system power on (was previously working)

Corren

Junior Member

Quadro preventing system power on (was previously working)

Thursday, September 2nd 2021, 4:42pm

New 2020 build: 3900XT, Asus ROG Strix x570-e
Everything went well installing the Quadro and fans originally. 2 ambient sensors, 5 fans split across 2 radiators and connected to Fan headers 1 and 2. EKWB D5 PWM connected to header 3. Had everything tuned the way I liked it. Was seeing sporadic hard-resets and situations where the computer would refuse to power on. There was a brief flash of power but then nothing other than ambient Mainboard lights.
I decided to try a different PSU (brand new Seasonic Titanum) and replace. Plugged everything back in, same symptoms, could not get it to power on at all now. Started removing components. As soon as I removed the 5V Molex power from the Quadro, everything booted perfectly. Tested with different cables and ports on the PSU, different USB headers on the mainboard. No dice.

Don't know how to further diagnose if there is a hardware problem on the Quadro and what might have necessitated it in the first place. Looking for guidance, I have a multi-meter.

Hufeisen

Senior Member

Thursday, September 2nd 2021, 5:22pm

Remove all Sensors and Fans from the Quadro, plug the Molex into the Quadro and try to start your computer.
If the Computer refuses to start, the Quadro is probably broken. (maybe a short circuit to ground)

You can try to get an Ohm reading on your Multimeter, between the Molex 5V Pin and the Ground Pin on the Quadro.

This post has been edited 3 times, last edit by "Hufeisen" (Sep 2nd 2021, 5:34pm)

Corren

Junior Member

Thursday, September 2nd 2021, 5:43pm

Confirmed that with only the molex power plugged in and no sensors, fans or usb, it still does not power on the computer.
Ohms across the 5v and ground next to it (right side of the power port with mitered corners up) reads ~12k ohms which is the same as the bottom of the circuit board solder locations. I don't know if that's good or bad.

Speedy-VI

Senior Member

Thursday, September 2nd 2021, 8:13pm

Sounds like you had an intermittent short somewhere in the Quadro Molex power wiring, and that short became permanent. Have you tried replacing the Molex power cable to the PSU? Does that cable power anything else?

Corren

Junior Member

Thursday, September 2nd 2021, 9:21pm

I've tried two different power supplies and two different cables. I don't think it's that side.

Corren

Junior Member

Friday, September 3rd 2021, 12:55am

Some additional information:
I used a bridging clip on a second power supply to enable it to be used without a mainboard, and then a molex connector to the pump and Quadro, and it works perfectly. I can even connect the USB back to the mainboard and it works completely.
However, if I connect power to the quadro from the same PSU as the motherboard, it does not work.

Corren

Junior Member

Friday, September 3rd 2021, 10:23pm

After confirming that the quadro functioned fine, I decided that was a good excuse to completely re-run every cable in the system, and work my way back up, and this time, after reconnecting everything, it works exactly as expected.
I'm truly baffled where the problem was and hate not discovering the root cause, but re-connecting everything from the ground up solved the problem.

Speedy-VI

Senior Member

Friday, September 3rd 2021, 10:48pm

Glad to hear you got it working. Sounds to me like there was a short somewhere in the Molex cable that was powering the Quadro. You didn't say how the Quadro is physically installed. I read a post on this forum from a guy who used the included screws to mount his Quadro (might have been an Octo) and he ran the screws through whatever surface he was mounting it to up into the controller. The screws were long enough that they contacted the back of the circuit board and shorted a trace to the chassis. It blew up the controller. Not saying that is what happened here but if your Quadro is mounted with screws, take a look at whether they can touch the back of the circuit board. If this is not the issue, then I would closely inspect the Molex power cable - maybe a loose strand of wire hanging out of the crimped pin? Too tight wire tie cutting into insulation? Wire pinched in the chassis somewhere? That's all I can think of.

Similar threads