I continually get this warning and associated action when using Claymore on ETH at Sparkpool even though every rig is tuned with -ttli, -tt, and -tstop values much higher. In fact, I set it anywhere from 100 to 1000 just to see what would happen and I STILL get my rigs offline more than they are online because of it. This is making Hive OS impossible to use. Any ideas?
P.S. Also, I live in the low desert with air temps around 110F. Most of my cards are high 60sC so please stick with solutions that don’t include “Use A/C.” or “You need more airflow” because the infrastructure is already engineered.
Follow-up: I’m going to leave this here for other searching for a solution to this problem.
I discovered that the Red Temp °C setting on the monitor page is a hard stop temperature limit. I did not see this mentioned ANYWHERE and incorrectly assumed that it was a display limit for high temperatures. I set it to 100C and everything continues to run as expected.
OMG, thanks for that information. Had this problem and didn’t know how to fix it!
No. It’s not work for me. I set Red Temp = 95 °C and I set ttstop=95. But I am still have GPU “reached 85°C, mining stopped”. What’s wrong. How can I change it? 85C is absolutely normal temperature for my cards. But I can’t use HiveOS for this cards.
It’s not a normal temperature and it should never be above 75C. It is not a safe temperature for any electronic device on the long run.
Fix your undervolt first, then try to fix a working system.
It’s probably related to the autofan. Try increasing the CRITICAL_TEMP value in /hive-config/autofan.conf (create it if it doesn’t exist), but you really shouldn’t be doing it.
Sorry for that, but you are wrong. Different devices has defferent temperature modes. I agree, that RX470 (for example) should be never above 75C. But for cards R9 280x even 85C is a normal temperature (and critical temperature about 110C). I don’t use fan software fan control - BIOS controls fans. When the temperature of the card is 75-80C, the fans spin only by 50-60%. Even AMD engineers do not consider it necessary to rotate the fans faster at this temperature. Right now one of my card at 85C and fans works at 77%. The other card right now is 87C and fans working at 81%. But RX470 at this temperature will spin at 100%.
About your recomendation: I don’t use autofan (because of HiveOS or Claymore can’t control fans for R9 2xx cards). I increases red temp in montor. I increased tstop temperature, but whole mining stops at 85C and never restarting. Don’t know what’s wrong. Eralier all was great.
Well yes but i would really recomment changing the R9 Fan speed in Bios cause the R9s make your other cards hot. So lower the Voltage and increase fan speed in the R9 bios.
@Brizovsky: Yeah, I just saw the messages again even though claymore is set to the correct temperatures. Oh well.
@brnfex: That is a completely false statement. Look up the data on a GPU or call the manufacturer. 85C is completely fine. My OC/UV is completely fine and works completely fine on all cards and on SMOS. Only one card is 85C+, the rest are 68C-72C. But this one card shuts down everything.
@BitForge @Brizovsky it’s acceptable for short loads such as gaming, but not for 24/7 mining. If the GPU chip is 85-90C, the entire power circuitry (mosfets, inductors and caps) would be hovering around 100C or more, which is VERY bad for their lifespan. No GPU is rated at 85C 24/7 load.
But of course - do as you wish. Your card’s power delivery is more susceptible to a critical failure when exposed to high temperatures. And remember that it takes just 1 out of the 100-120 power delivery components to die and your card will be toast. Not here to argue, just giving you a heads up.
Okay, well this is a deal breaker. I’m away from my farm and I have to sit here and fight with rigs because of this idiotic error. Fine you want to hard code something like this temperature limit, I’ll go back to SMOS fully tonight.
Fan control comletly NOT working in claymore dual and etherminer in MANUAl mode