I am assuming the -1C temperature is really false.
My guess is the power supply. If the ground on your "free hanging" power supply (at the card) goes above 0.7V (relative to the rest of the computer) due to high current draw, you can likely cause all kinds of problems.
You should have the ground on the two power supplies connected together. In another thread, the method suggested for starting a pair of power supplies is to tie the green wires together as well. If it was me, I would want the Power good signals tied together with an AND gate, but that requires some research (is 'Power good' a digital signal?) and circuitry.
Topic: Unique Problem! 4x5830: Temp drops to -1C w/ freeze (Read 888 times)
Hi,
I have a new mining rig I'm setting up.
It has
Win7 Ultimate 64bit
AMD XII 245 2.9Ghz Dual Core
MSI 890fxa-gd70
2GB RAM
4x Sapphire 5830
2x600W Diablotek DA Series PSDA600 600W ATX12V v2.2 Power Supply
Catalyst 11.6, SDK 2.4
One PSU is using paper clip trick
I have three dummy plugs and one card plugged into a monitor.
This has been driving me nuts for 24 hours now.
I made this new rig with 4x5830.
When I try to start mining usually two of the cards work fine and mine no problem, but then one of the remaining 2 cards goes haywire.
As soon as I start the problem card mining the screen gets very stuttery, the temperature in MSI Afterburner drops to -1C for that card, and then the computer hangs. The temperatures of the cards never even get hot, it's almost instantaneous (as seen in the pic below)
I started troubleshooting by disabling two of the cards, and running two... the top two cards worked fine on their own, no problems. These are connected to the PSU thats on the MoBo. That meant it had to be one of the two bottom cards.
So then I tried the two cards connected to the free hanging PSU, and one of them consistently locked up, while the other worked fine.
I figure I have found the culprit, have a bad GPU and need an RMA. I take out the bottom card (which seems to be the problem).
Here's where it gets weird.
Now I have 3 cards left in my rig, supposedly all good working cards.
I run the 3x5830 overclocked to 900Mhz, mem down to 500Mhz for about 6 hours... seems about stable right?
So I decide to move the machine to another room that has a better airflow, and when I boot it back up, the 3rd card now does the -1C, stutter and freeze!!
As seen in the screencap the GuiMiner doesn't even have time to update the image from "Start" to "Stop" before the machine locks up entirely, although MSI Afterburner does show the -1C right before it all goes to hell.
This card is the one plugged into the free hanging PSU also! But when I tested it with it's brother earlier, it was fine and the other card was bad!
I have a feeling that it's to do with my two cards plugged into the second PSU that's not hooked up to the MoBo but I don't see why.
What could possibly cause this? I can't see anyone else having this problem anywhere..
So really:
Is it the free-hanging PSU?
Is it the dummy plugs?
http://img189.imageshack.us/img189/6608/unledviv.png
I have a new mining rig I'm setting up.
It has
Win7 Ultimate 64bit
AMD XII 245 2.9Ghz Dual Core
MSI 890fxa-gd70
2GB RAM
4x Sapphire 5830
2x600W Diablotek DA Series PSDA600 600W ATX12V v2.2 Power Supply
Catalyst 11.6, SDK 2.4
One PSU is using paper clip trick
I have three dummy plugs and one card plugged into a monitor.
This has been driving me nuts for 24 hours now.
I made this new rig with 4x5830.
When I try to start mining usually two of the cards work fine and mine no problem, but then one of the remaining 2 cards goes haywire.
As soon as I start the problem card mining the screen gets very stuttery, the temperature in MSI Afterburner drops to -1C for that card, and then the computer hangs. The temperatures of the cards never even get hot, it's almost instantaneous (as seen in the pic below)
I started troubleshooting by disabling two of the cards, and running two... the top two cards worked fine on their own, no problems. These are connected to the PSU thats on the MoBo. That meant it had to be one of the two bottom cards.
So then I tried the two cards connected to the free hanging PSU, and one of them consistently locked up, while the other worked fine.
I figure I have found the culprit, have a bad GPU and need an RMA. I take out the bottom card (which seems to be the problem).
Here's where it gets weird.
Now I have 3 cards left in my rig, supposedly all good working cards.
I run the 3x5830 overclocked to 900Mhz, mem down to 500Mhz for about 6 hours... seems about stable right?
So I decide to move the machine to another room that has a better airflow, and when I boot it back up, the 3rd card now does the -1C, stutter and freeze!!
As seen in the screencap the GuiMiner doesn't even have time to update the image from "Start" to "Stop" before the machine locks up entirely, although MSI Afterburner does show the -1C right before it all goes to hell.
This card is the one plugged into the free hanging PSU also! But when I tested it with it's brother earlier, it was fine and the other card was bad!
I have a feeling that it's to do with my two cards plugged into the second PSU that's not hooked up to the MoBo but I don't see why.
What could possibly cause this? I can't see anyone else having this problem anywhere..
So really:
Is it the free-hanging PSU?
Is it the dummy plugs?
http://img189.imageshack.us/img189/6608/unledviv.png
Jump to: