Author

Topic: Fault Finding - Dual PSU Rig - Second PSU Turns Off (Read 134 times)

member
Activity: 99
Merit: 10
Basically it's intermittant where the second PSU would switch off after mining for a little while or days. I switched the PSU's around to see if it was PSU specific however this is not the case. I reinstalled drivers, reinstalled windows and now I'm starting to think it could potentially be one riser or GPU.

I am suspecting a Riser seeing this, do some basic checking, like connections from USB to board PCI-e x1 or Riser x16slot to GPU if properly in place and doesn't have any loose connections. otherwise try to replace them if you have extras but ensure replacements is reliable and solidly working.


Quote


What is the best way to find a faulty riser/gpu  or is there anything else which it could potentially be?

basically, you can do swapping/interchange/replacement method.

I see perhaps it's best to change all 4 risers and then do testing on them one by one. this will eliminate downtime on rig. I'll await till my new batch of risers arrives. Thanks!
sr. member
Activity: 1008
Merit: 297
Grow with community
Basically it's intermittant where the second PSU would switch off after mining for a little while or days. I switched the PSU's around to see if it was PSU specific however this is not the case. I reinstalled drivers, reinstalled windows and now I'm starting to think it could potentially be one riser or GPU.

I am suspecting a Riser seeing this, do some basic checking, like connections from USB to board PCI-e x1 or Riser x16slot to GPU if properly in place and doesn't have any loose connections. otherwise try to replace them if you have extras but ensure replacements is reliable and solidly working.


Quote


What is the best way to find a faulty riser/gpu  or is there anything else which it could potentially be?

basically, you can do swapping/interchange/replacement method.
sr. member
Activity: 588
Merit: 335
Steady State Finance
the fan is working but when it's I find it after I lose 4 cards, it's not running. I.e. protection mode I think.

Yes, it's Over-voltage protection, under-voltage protection, short circuit protection, over power protection, and over temperature protection.
Maybe one of that protection has happened to your PSU.

so the splitter could be issue? I had saame setup on another rig without any issues. hmmm

Not sure, But it can be. I just inform you about the Y splitter power issue.
member
Activity: 99
Merit: 10
the fan is working but when it's I find it after I lose 4 cards, it's not running. I.e. protection mode I think.

so the splitter could be issue? I had saame setup on another rig without any issues. hmmm

i'll check doing device manager and see.
sr. member
Activity: 588
Merit: 335
Steady State Finance
I'm using Y-Split on the 8Pin to Dual 8pin (gpu requires 8pin + 6pin)

If you were splitting 8 Pins to dual 8 Pins, you would get max 150 Watt with combined those two connectors on Y splitter. 8 Pins + 6 Pins means 150 + 75 = 225 Watt and 8+6 Pins on Y splitter only produce max 150 Watt.

Any suggetions where to start? I have a feelng it's gpu or riser.... What's the best way to get to the bottom of this? Should I test the gpus on bench mark one by one? Is there a good testing tool or easier way to find gpu fault? I'm worried that benchmark will run okay and still not find the root cause.

You can try to disable cards one by one (via device manager), and see which card causing.  

Don't forget to check physical PSU, check its fan and make sure your PSU didn't produce much heating.
member
Activity: 99
Merit: 10
Hi All,

I've been having an issue with one of my rigs.

- H270 Prime Plus
- 4gb ddr4
- 120gb ssd
- Pentium
- 8x 1070ti
- 2x Corsair TX850M PSU
- 2x M.2 Risers
- 6x Ver008s Riser

Basically it's intermittant where the second PSU would switch off after mining for a little while or days. I switched the PSU's around to see if it was PSU specific however this is not the case. I reinstalled drivers, reinstalled windows and now I'm starting to think it could potentially be one riser or GPU.

I believe the PSU is going into protection mode as a switch cycle restores PSU to operation.

I don't believe it's power consumption as I'm reading around 540-560 at the wall. Second PSU is running 4 risers and 4 gpu. sata-6pin adapter powering the m2 riser, two seperate molex cables running another two ver008 risers and another seperate sata cable running one ver008 riser. I'm using Y-Split on the 8Pin to Dual 8pin (gpu requires 8pin + 6pin)

Settings is 75 tdp / 100 gpu / 450 mem. The readout on miner is saying 126-140w per card. I'm mining RavenCoin x16r.

I have another idential setup which doesn't have any issues.

Any suggetions where to start? I have a feelng it's gpu or riser.... What's the best way to get to the bottom of this? Should I test the gpus on bench mark one by one? Is there a good testing tool or easier way to find gpu fault? I'm worried that benchmark will run okay and still not find the root cause.

What is the best way to find a faulty riser/gpu  or is there anything else which it could potentially be?

I was using a 2 psu adapter cable but ended up shorting the second psu 24pin just incase it was the adapter cable causing issue but the issue still presisted.

Thanks guys
Jump to: