Rig will no longer mine

[Malach]

Setup
Built mining rig with parts from newegg.
Bought 5970 off ebay.
Installed BAMT.
Was mining at around 730 hash for a couple months before I came home to find the gpu's fan was making loud noises.
Replaced fan and started having the problem listed below.
Lowered hash rate to 700 seemed to fix the problem for a couple weeks.

Now the below has happened again.

Problem
Whenever the rig starts to mine it just locks up. Doesn't matter how I start mining, the screen freezes and the monitor goes black a couple seconds later telling me there's no connection. However the computer is still on and running.

I can access the gui and ssl before it starts mining.

However after it attempts to start mining the gui locks up, monitor goes blank and tells me there's no connection, and I can't access gpumon through ssl.

I could lurk around the computer sometimes through ssl after it started mining in the past, but whenever I tried to access gpumon, the connection locked up.

Can't really tell if its mining after it locks up but deepbit does say it connected at some point.

Likely Outcome
I'm pretty sure the gpu was damaged somehow when the fan went out because I had problems mining and had to reduce the hash rate to fix the problem. However I have heard that the newer gpus will shut down if they get too hot.

Just wanted to make a quick post and maybe get some more experienced answers. 

[1715477856]

1715477856

[1715477856]

1715477856

[zvs]

Setup
Built mining rig with parts from newegg.
Bought 5970 off ebay.
Installed BAMT.
Was mining at around 730 hash for a couple months before I came home to find the gpu's fan was making loud noises.
Replaced fan and started having the problem listed below.
Lowered hash rate to 700 seemed to fix the problem for a couple weeks.

Now the below has happened again.

Problem
Whenever the rig starts to mine it just locks up. Doesn't matter how I start mining, the screen freezes and the monitor goes black a couple seconds later telling me there's no connection. However the computer is still on and running.

I can access the gui and ssl before it starts mining.

However after it attempts to start mining the gui locks up, monitor goes blank and tells me there's no connection, and I can't access gpumon through ssl.

I could lurk around the computer sometimes through ssl after it started mining in the past, but whenever I tried to access gpumon, the connection locked up.

Can't really tell if its mining after it locks up but deepbit does say it connected at some point.

Likely Outcome
I'm pretty sure the gpu was damaged somehow when the fan went out because I had problems mining and had to reduce the hash rate to fix the problem. However I have heard that the newer gpus will shut down if they get too hot.

Just wanted to make a quick post and maybe get some more experienced answers. 

Well, if the card feels hotter than it should, then it's still doing something.  If it doesn't, then it isn't.

It's typical for a card to blackscreen and reacquire after failing w/ a too high clock setting, overheating, or w/e.  I've always had problems getting 5970's going again without restarting my computer, but 5830's and 5870's when I was testing mine, I'd just let it crash, drop the clock speed 5mhz and try again.

Apparently, the card actually boots up and works fine (no artificing or anything?) so it's still worth decent money to sell as 'parts'.  If it was a card I had though, I'd make sure to find out what the problem was before taking 50% value or w/e.

I had a card that did the same thing, the fan died while I was away and when I came back it was throttling itself, getting ~120mhash on the GPU closest to the fan (the fan that was laboring to spin at like 800rpm) and ~50mhash on the other.  The GPU temps were over 100o, I imagine the VRM's were over 120 or 130o.  I replaced the fan, put the card back in, and haven't had any problems with it since.  I've actually been running it at (I believe) 920 core and 1.112 volts lately since winter started (i'm in texas, so it rarely gets super cold here, but maybe this weekend I'll try 1.125).  I've got all the 'stable' settings on a sheet of paper in the other room....  It's one of my better 5970's though.  They range from about 860-950 stable at that voltage..

Uhm...  All my cards used to run stable in the summer at 0.99v,... err.. I think it was 725/300.... but you could just see if it works at 0.99, 600/300 or something similar.

[Hippie Tech]

Have you ruled out the power supply, mobo and system ram ?

[Malach]

Well, if the card feels hotter than it should, then it's still doing something.  If it doesn't, then it isn't.

It's typical for a card to blackscreen and reacquire after failing w/ a too high clock setting, overheating, or w/e.  I've always had problems getting 5970's going again without restarting my computer, but 5830's and 5870's when I was testing mine, I'd just let it crash, drop the clock speed 5mhz and try again.

Apparently, the card actually boots up and works fine (no artificing or anything?) so it's still worth decent money to sell as 'parts'.  If it was a card I had though, I'd make sure to find out what the problem was before taking 50% value or w/e.

I had a card that did the same thing, the fan died while I was away and when I came back it was throttling itself, getting ~120mhash on the GPU closest to the fan (the fan that was laboring to spin at like 800rpm) and ~50mhash on the other.  The GPU temps were over 100o, I imagine the VRM's were over 120 or 130o.  I replaced the fan, put the card back in, and haven't had any problems with it since.  I've actually been running it at (I believe) 920 core and 1.112 volts lately since winter started (i'm in texas, so it rarely gets super cold here, but maybe this weekend I'll try 1.125).  I've got all the 'stable' settings on a sheet of paper in the other room....  It's one of my better 5970's though.  They range from about 860-950 stable at that voltage..

Uhm...  All my cards used to run stable in the summer at 0.99v,... err.. I think it was 725/300.... but you could just see if it works at 0.99, 600/300 or something similar.

Well my problem is whenever I try to mine, even without overclocking, the whole system will still freeze up. Without OC it mines at around 640 hash and I tried mining at around 600 and it ran for about 15 seconds before locking up.

Have you ruled out the power supply, mobo and system ram ?

Like I said, everything works fine when I'm not using the card. I can easily modify files, connect to the internet, ect. The gpu is the only thing in the rig that's has given me trouble so thats the only thing that comes to mind. Unless, there's something I'm missing.

[Hippie Tech]

I'm thinking that the +12v rail(s) might be shot and/or barely able to handle it. Thus why it crashes when engaged.

Check the capacitors on the gpu, mobo and psu for signs of rust or swelling.

Have you tried it on another rig ?