Those temps are PERFECT for an S7. They actually crash more when the cards get under 60c and the max temp before the software shuts them down is 80c so you have plenty of headroom.

One thing of note. The current 13.5th s9 bitmain estimates 1323w + 10% which is actually around 1450w. This is about the actual usage as I have tested them on a 94% efficient PSU.


Total power available in watts:

32A * 220 = 7040w

You run breakers at 80% of max for safety:

7040 * .80 = 5632w

Available power divided by per unit draw:

5632 / 1450 = 3.884 units

So 4 would be the max I would put on that circuit.

Just to give you an idea of how common that bug is here is a screenshot of 30 miners with 8 of them showing the bug.

There is no serial number embedded in the software. It is on the stickers they put on the hash board as well as the top of the machine itself.

Hosting a bitcoin miner in a traditional tier style datacenter is not really financially viable. A traditional server fits into a rack and uses much less power than a bitcoin miner would. You need to find a place that specializes in hosting crypto rigs if you go the datacenter route.

Herp Derp

If you the avalon lost power and had no lights or fan that is 100% a power supply issue.

What color are the lights on the unit and on the AUC serial adapter?

These machines will run safely for months at 100c chip temps, you have nothing to worry about.

My magic 8 ball says "reply hazy, please try again"

Any S9 sold this year has a 180 day warranty. If you are still in warranty just go to your bitmain account and submit an RMA. Otherwise you would have to contact a place like bitmainwarranty.com. They have you send the system in, they diagnose it, then they give you a repair price.

Avalon literally produced the first bitcoin ASIC that was put on the market. Nobody can claim to be around longer.

One of your boards is bad. You will have to pay if you want it repaired.

MSI boards are well known to not want to run past 4 GPUs. I have seen this problem on a number of test boards direct from MSI. You might get lucky and figure a way around it but the true answer is dump that MSI board for something useful

Were the miners off for any length of time? With the way you describe it I feel like condensation would be the most likely culprit

I am pretty sure we will all be riding our S9s for a good while. Id love a little competition but it just doesnt seem to be going that direction. Bitmain is still the big dog and everyone else is putting out hardware that cant even come close to touching it.

Ive had this problem on many Avalon controllers over the years. I just keep a few spare flashed SD cards and replace as needed. There is no fix that is worthwhile, its much easier just to replace.

The first of the s9s I have on site have been running since January.

With 60f intake temps pretty much all of my machines are in the 70-80c range. IMO 70-80c is ice cold for these machines. I ran these things all summer over 100c with no issues. These kinds of chip temps are the new norm for 14/16nm tech. For example my water cooled GTX 1080 still hits 70c at full load.

Sadly sending them off for repair is the only answer.


With so few miners there couldnt you just cool them with a cold air intake? Running AC units is the number one way to kill your profits :/

In the summer my S9s run over 100c at the chip level. The thermal cutoff is 115c, thats when they stop mining. 80 degrees on a chip is relatively cold, dont worry about it.