Bitcoin Forum

Hey.

Me and my colleague are running a mining rig with 2x PowerColor R9 280X, 1x  Sapphire R9 280X OC Dual-X and one MSI 5850 on a motherboard with a FX-8350.
We run all 280X's on 1020/1500, which is the default for the Sapphire card. The PowerColors has been undervolted but the Sapphire wasn't stable at lower volts so it runs completely stock.
If we try to overclock the Sapphire even to 1030 it hangs the rig, so we just run all at 1020/1500 and get about 723Kh/s each and a temperature at about 83°C @ 40-60% fan speed.

So now to the story: Today I noticed that my teamviewer had disconnected and the computer was shut down so the WoL didn't work, I jumped in my car and went to my office to find the computer completely shut down. Nothing looks suspicious so I just press the power button about 4 sparks fly from the back end of the Sapphire card and it smells burned. I take it out of the rig and start it up with just the others and there doesn't seem to be any problems.

So the quetion is, is the Sapphire completely fried? It has only been running for roughly 9 days and all the time it has hade good temperatures.

And an image of the rig before we fitted the 5850 just for the ones that like images :)
https://i.imgur.com/Ci630qjl.jpg (http://imgur.com/Ci630qj)

I would say so! just RMA it, lucky it dident fry any more  :-\

Hope you kept the box, that is RMA time

Same thing happened to me with a Sapphire card that was mining for a couple of days only. They replaced it with no problems.

It's the VRMs on the Sapphire OC and VaporX cards.  They run over 100C (vrm temps not core temps).  They fucked up this time around.  Everyone email them and tell them to get their act together and release a BIOS fix as MSI did.

I have already sent them an earful. If everyone does maybe they will fix it.
tech@althonmicro.com

http://www.techpowerup.com/forums/threads/solution-is-found-to-sapphire-vapor-x-r9-280x-vrm-over-heating-and-throttling-problem.196972/

Hey there,


Noticed your post in January about the Sapphire Dual X 280x burning out.


I woke up this morning to the EXACT.SAME.THING. Computer was off, turned it on and saw sparks. After examining and reseating everything, my machine boots and hashes fine on both cards though, and I didnt have time to look at it in detail as I was on my way to work, but looking closer this evening.

Just unreal. I was mining as well, and was just wondering how your RMA went? Did they send you the same model? I have two of these suckers, and I almost want to RMA both of them. Only lasted a week. FML.


Thanks again for your time.

Guess what? I have a card that does this and still works, It will spark once "shoots about 5 inches like a butane lighter jet" I shit you not... and then works no problem.

Until one day the machine is off and team viewer is disconnected.... just like you.

My rig would run about 5-7 days between randomly shutting off. I would then have a really hard time getting it to come back on. But when it does. Sparks fly.

- Chris

The RMA went bad, they took it in ran 3dmark for an hour and said "we don't have a problem with it" so with got it back and continued to run it with the same problems some times up to 3 times a week.

Then we changed from our old normal powered risers to the usb type risers and it has been running stable for 3 weeks now.

i have 7 of these cards,
2 of them burned exactly like you described,

sapphire is cheap crap!

Horrible Cards. They do have low core temperatures and hash straight up to 760 khash/s but almost 50% have this horrible VRM problem. Usually a rig turns off in case of VRM overheat but I at least once experienced the VRM to fully burn out without the system halting. This wasn't fun at all!

Sapphire has reduced the quality of the components to increase their profit margins .I have RMAed 5 Sapphire cards in 3 months.

Did you guys buy these cards new? Or will sapphire RMA 2nd hand graphics cards?

I had the EXACT same thing happen to my card, worked fine, then it shutdown, then It sparked. I just RMAd it saying that "it didnt work", got one in the mail 2 weeks later. Just RMA the thing.

I have 10 of them and have been runing 6 months and no problem other then the fan dieing. But I only keep my temp around 70C so that might be the reason.

A friend of mine runs about 20 of these cards and he never had any issue with them. He keeps recommending me those :x

Definetely think u didnt connect pciex power to the cards right, either that, or u have a short circuit somewhere on that card.
Ofter ppl dont stick those connectors good enough, so that place gets very hot, and eventualy looses connection.
happened to me few times, but luckily, nothing was destroyed.