Bitcoin Forum

I have been having nothing but dramas with my antminer s7's. For the most part it's just been heat sinks falling off.

The current problem I am having with them is that one by one over time they just die. It starts with a miner mining for a few days and then randomly it will just stop mining.

I restart it, then it will happen the next day again. The the day after that it will happen about 3 times. Then the day after that I am lucky to get it to mine maybe 2 times for an hr or 2 and then after that it's dead for good.

I have lossed a total of 5 hashing boards this way.

It seems like I am the only on having this trouble.

I have tried the hashing boards on another controller with the same result. I have tried a different PSU. The temperatures are fine.

When I attempt to start the miner after this, this is what it looks like.

http://i164.photobucket.com/albums/u9/Skum_bucket/drams1_zpsemesq8cc.jpg

your photo shows a fan not two fans.

why ?  if your fans don't show the gear won't hash.

i had this issue as one of my fans was not plugged in.

once i plugged both fans in the gear worked.

we need more concrete information, like temperatures, maker and model of psu, fan speed settings....

Also showing temps at 0 and the wrong number of chips?

Isn't it still under warranty?

Ah that will be the rare 144 chip version of the S7 Sir...  :)

Rich

what firmware do you have ?
have you installed any firmware ?
phil is right, there is something wrong about your fans, just 1 working :P

I am working on the fan problem at the moment. I have 3 fans plugged in. When it works it says 54 asic per board, but when not working on this controller it says 48 and 0 temp... On another controller it says 30 asic and 0 temp and all dashes instead of 0's


I am using the bitmain PSU am using the october firmware.

Will get back to you guys ASAP, thanks for your replies.

Its possible spaghetti code doesn't like more than 2 fans plugged into the controller.

I ended up getting the 3 fans to show by plugging the 3rd fan in between the other two plugs. so plugs 1,2 & 3.

It is hashing now... but I am a bit sceptical because it doesn't explain how they just drop out randomly...

Fingers crossed... I have sent back a total of 7 hashing boards now... Not keen on another 3..

Cheers, I will keep you posted

doggie- do not write another BS comment.

I have eight double size S7 and they all show  four fan speeds.

FAN 40%

Good for you.

I have ha a day or 2 now to see if the problem is solved but it is not.

The miner still is dropping out. It has dropped out 2 times in 2 days. Both times I can't get it to hash again unless leave it turned off for a while before starting it up again.

Over night I also had another hashing board on another miner drop some heat sinks. Am I the only one with this problem? Out of the 15 hashing boards I purchased, 7 have had heat inks fall off. I swear all I have to do is look at these miners the wrong way and they will drop heat sinks.

It is very frustrating also because desperately want to just leave the miners alone. Every time you touch them, the more likely they will drop heat sinks it would seem.

Can You provide pictures of fallen heatsinks? I'm interested does only bigger heatsinks are falling or smaller also. Thx.

Edit: wrong positioned question mark.

Small radiators are tinned. They can not easy way fall off.

So if S7 is positioned in a way so big heatsinks are at the top side, there will be no "falling heatsinks" anymore.

my 1 s7 first batch, has the same problem, randomly going offline and losing speed,
i applied RMA beacause of warranty.

I have now If I want the noise of a small jet fighter 1,5 TH/s from 1 hashbord the rest is being recognized.
but no hash at all.

There is no rare 144 chip version, there is only a  162 chip version and a 135 chip version.  This is not an image of an S7.  This is the single section of an S5+ which had has 144 chips on it (obviously the one with the controller card on it which can be moved). 

I think you missed the  :) on my comment...

So you are suggesting that the OP has the centre section from an S5+ and not an S7? Got to say that seems very unlikely and I do not think the S5+ has 600MHz available?

Rich

Well I am actually running a center piece of an S5+ right now in my farm so that is how I know on the chip count.  Yes, I did miss the  :), I thought you might be referring to the the 135 chip unit that bitmain is putting out now(sorry about that).  They put out 135 chip units in batch 6 as well batch 8 which has not shipped yet.  I can tell you one thing, the image that the OP shared is not that of an S7 card.  When I get to my office here in a couple hours I will look at the frequency options to see if 600 is available.  What I do know is that if he was have issues with chips not working then they would simply be x'd out but I do not believe that they would completely disappear.  If he had a 135 chip S7 you would see 45 chips on each line, and one 162 you would see 54.  Like I said, 48 chips is not an S7 board.

With S5+ software max frequency can be only 500 not 600

You can see the S7 image with less chips when the temperature sensor shows 0 and even then, if one of the fan does not work.

I am having the same issue, out of 22 Antminer s7 only 18 are up now while 4 of Antminer s7 are out mining.

Well I just checked and with the firmware that is on the controller card that is my S5+ it only goes to 500 Mhz.  Which leads me to my next question, could an S7 controller card work on an S5+ card.  I sent the OP a PM to see if he got this straight from bitmain or 2nd hand.  I have a funny feeling about this but kind want to leave it at that until I get more information.  If he tells me 2nd hand, I am 99% certain I know who it is.

The original S7 Controller and the S5+ controller were identical. I think later S7 had a new controller?  Tupsu is the expert on this, let's see what he has to say?


Rich

My batch 5 controller is labeled on the upper part of the metal piece
Model S7 Version 2.0
but bottom an the lower plate is  Model S5 Version V1.71

BBB  is not labeled only Q.C. PASSED

I have made some progress. At first I thought it may have something to do with heat because 2 days in a row miner stopped at the hottest time in the day but the miners temp was approx 60C at the time and I thought it may just be some sort of hardware error shutting it down at about 60 instead of 80C.

I directed some more AC at it and it seemed to work well but eventually died a little later in the day. Every attempt to restart it failed and I got the notorious 48 asic screen.

Now I am able to get it mining again at lower frequencies. 575 worked for a little while and now I am trying 568... The loss of hash rate hurts though...

So dropping the frequency din't work. I was able to isolate one hashing board which is causing the trouble. Every day it would cause the PSU to shut down so I have it connected to it's own PSU. This helps by preventing the whole miner dropping out which has bought me some time.

I am hoping someone can shed some light on what signals to the PSU to shut down that the hashing board is triggering. One thing I do notice is when I attempt to probe the 6 pin connectors connected to the hashing board to test the voltage, the PSU shuts down. Visibly I cannot see anything which could cause a short. Bare in mind that it only happens when the ambient temp gets to about 38C.

I will clean the contacts up with some contact cleaner to see if that helps.

Cheers

The only things that would cause a PSU to shut down would be if you take too much current or it gets too hot. Have you swapped PSU's to confirm that it is not a PSU fault? The fact that previously you had to leave things turned off for a while does indicate a heat related problem.

Is it possible that you have a heatsink that is partially detached that when things heat up it is causing a short?


Rich

you are not alone in this.  i'm searching everywhere for an answer to the same problem.

Both my S7s i purchased on batch 5 are doing the same thing.  I have them running with 100% fan speed, and there is no way they are overheating as the whole area is very well ventilated (warehouse space).  I'm running the miners on the EVGA supernova 1600 G2.  And i'm not getting a temp reading on any of the 3 boards.  Then the miner drops off, keeps beeping and the red light's on.  WHen i check the miner status, i now get a temp reading of 127.  I don't know if the heatsinks are falling off of not, everything looks in perfect condition inside.  At this point now, i'm not capable of turning one of the miners back on unless i unplug the 3rd board.  Then i get about 3TH out of it instead of 4.8.

I don't know what to make of it, customer service is slow on email, but i'm waiting to hear back from them after submitting as many screenshots as i could of the miner status.

I'm having the same shutdown problem, and after seeing this post i checked, and i'm have the same thing.  Both my S7s come with a board that says S5+ control board Version 1.6

The hashing board eventually stopped working. Good luck with your troubleshooting and thanks for your input. I hope you have better luck than I did.