ANTMINER S4 Discussion and Support Thread

[MyRig]

@bittalc1...

may be our response didn't get send or made it to you...

PM us your shipping address please with a telephone number and protac's order confirmation

[bittalc1]

@bittalc1...

may be our response didn't get send or made it to you...

PM us your shipping address please with a telephone number and protac's order confirmation

I send pm with the requested info.

Thanks Bitmain!!

[bittalc1]

yes 61...  a bit too high but every customer have different deployment method.  Looking at the running up time, at least the high end is stable or managed some how, so it's not crushing the unit.

For some reason my chain#1 on my units are the hottest not chain #4.

because chain1 is connected to J7 and fpga cannnot set your proper voltage value to J7 connected hashboard. chain1 is always working 0750V. i told a lot but no ones listen...

Hi cypher
as i can see you say that chain1 is connected to J7 and in previous posts i saw that chain1 is the top hashing board, but in my case J7 is connected to the bottom hashing board and that is chain#3 right? and what did you meant when you said connect to the neighbor board? is that to top the botom? please clarify
Thanks in advance

[bhanu545]

I observed reduced hashrate and rebooted the miner, i see -------- -------- -------- -------- -------- in chain 1.

Is the chain 1 dead?

I just upgraded the firmware to see if it can fix the issue. But NO.

Any solution for this?

[Prelude]

I observed reduced hashrate and rebooted the miner, i see -------- -------- -------- -------- -------- in chain 1.

Is the chain 1 dead?

I just upgraded the firmware to see if it can fix the issue. But NO.

Any solution for this?

With the old firmware I had a chain turn off every day or 2. With this new firmware, it hasn't happened yet.

You need to physically power down the machine, and turn it back on. Should be smooth sailing afterwards with the new firmware. Be aware that no amount of soft resets will fix --------- chains, has to be turned off.

[bhanu545]

I observed reduced hashrate and rebooted the miner, i see -------- -------- -------- -------- -------- in chain 1.

Is the chain 1 dead?

I just upgraded the firmware to see if it can fix the issue. But NO.

Any solution for this?

With the old firmware I had a chain turn off every day or 2. With this new firmware, it hasn't happened yet.

You need to physically power down the machine, and turn it back on. Should be smooth sailing afterwards with the new firmware. Be aware that no amount of soft resets will fix --------- chains, has to be turned off.

That worked. Thanks bro.
Also this miner became unresponsive yesterday and the LCD display showed weird too. Did you see such scenario happening ever?

[Prelude]

Yes, I've seen that before on the old firmware. Hasn't happened yet with the new firmware.

[Bazeman]

Is your miner not behind a firewall? It should be impossible to SSH into it from outside your network.

To be certain, I recommend pull it's micro SD card out, and re-imaging the firmware. You can download a copy of the image in the second post of this thread, provided by dogie. Use windisk32 to write the image to the SD card.

It wasn't behind a firewall, now it is.

re-imaging the firmware worked, as you discribed, thank you!

To be fore sure, is it possibel to change the SSH password? How can i do it?

[Cyper_BLC]

yes 61...  a bit too high but every customer have different deployment method.  Looking at the running up time, at least the high end is stable or managed some how, so it's not crushing the unit.

For some reason my chain#1 on my units are the hottest not chain #4.

because chain1 is connected to J7 and fpga cannnot set your proper voltage value to J7 connected hashboard. chain1 is always working 0750V. i told a lot but no ones listen...

Hi cypher
as i can see you say that chain1 is connected to J7 and in previous posts i saw that chain1 is the top hashing board, but in my case J7 is connected to the bottom hashing board and that is chain#3 right? and what did you meant when you said connect to the neighbor board? is that to top the botom? please clarify
Thanks in advance

no, doesnt matter which hashboards is top-bottom or right-left. J connectors are define which hashboard is chainX.
you can easily change ones to another... if you can move j7 connectecor to top hashboard, you will see 0750 V (you cant set your own value to J7)

on fpga board :
J7 means Chain1
J9 Chain2
J2 Chain3
J6 Chain4

[nokati]

hi,

I have tryed to update the firmware

https://www.bitmaintech.com/support.htm?pid=00720140930114518599JXGHWWD80660
https://www.bitmaintech.com/files/download/initramfs.bin.SD-20150129.tar.gz


I get this error;
tar: invalid magic tar: short read cp: can't stat '*': No such file or directory

Any idea?


Is there also a way to get the antminer s4 to restart after a power failure?


Thanks,
Sam

[Prelude]

hi,

I have tryed to update the firmware

https://www.bitmaintech.com/support.htm?pid=00720140930114518599JXGHWWD80660
https://www.bitmaintech.com/files/download/initramfs.bin.SD-20150129.tar.gz


I get this error;
tar: invalid magic tar: short read cp: can't stat '*': No such file or directory

Any idea?


Is there also a way to get the antminer s4 to restart after a power failure?


Thanks,
Sam

You aren't unpacking the file, right? You have to upload the firmware as it was downloaded.

No way to automatically power on with the stock PSU. Replacing it with a different PSU (or 2) is the only way.

[nokati]

hi,

I have tryed to update the firmware

https://www.bitmaintech.com/support.htm?pid=00720140930114518599JXGHWWD80660
https://www.bitmaintech.com/files/download/initramfs.bin.SD-20150129.tar.gz


I get this error;
tar: invalid magic tar: short read cp: can't stat '*': No such file or directory

Any idea?


Is there also a way to get the antminer s4 to restart after a power failure?


Thanks,
Sam

You aren't unpacking the file, right? You have to upload the firmware as it was downloaded.

No way to automatically power on with the stock PSU. Replacing it with a different PSU (or 2) is the only way.

No I am not unpacking, I just select and upload as I have downloaded. I have try this on 2 antminers s4

Miner Type   Antminer S4
Hostname   antMiner28
Model   GNU/Linux
Hardware Version   3.4.3.0
Kernel Version   Linux 3.8.13 #12 SMP Mon Sep 29 11:13:50 CST 2014
File System Version   Fri Oct 3 18:17:17 CST 2014
Cgminer Version   4.6.1
Uptime   3:38
Load Average   2.05, 2.11, 2.13

[Prelude]

Damn, you weren't even using the Oct. 26th firmware. Try re-download the file, maybe it got corrupted. If that doesn't work either, try upgrading to Oct. 26th firmware, and then to the latest. 

[bittalc1]

yes 61...  a bit too high but every customer have different deployment method.  Looking at the running up time, at least the high end is stable or managed some how, so it's not crushing the unit.

For some reason my chain#1 on my units are the hottest not chain #4.

because chain1 is connected to J7 and fpga cannnot set your proper voltage value to J7 connected hashboard. chain1 is always working 0750V. i told a lot but no ones listen...

Hi cypher
as i can see you say that chain1 is connected to J7 and in previous posts i saw that chain1 is the top hashing board, but in my case J7 is connected to the bottom hashing board and that is chain#3 right? and what did you meant when you said connect to the neighbor board? is that to top the botom? please clarify
Thanks in advance

no, doesnt matter which hashboards is top-bottom or right-left. J connectors are define which hashboard is chainX.
you can easily change ones to another... if you can move j7 connectecor to top hashboard, you will see 0750 V (you cant set your own value to J7)

on fpga board :
J7 means Chain1
J9 Chain2
J2 Chain3
J6 Chain4

Thanks that explains alot. So can i take off the cable from J7 and connect to one of the free slots on the top hashing board? Will that work?

[Cyper_BLC]

yes 61...  a bit too high but every customer have different deployment method.  Looking at the running up time, at least the high end is stable or managed some how, so it's not crushing the unit.

For some reason my chain#1 on my units are the hottest not chain #4.

because chain1 is connected to J7 and fpga cannnot set your proper voltage value to J7 connected hashboard. chain1 is always working 0750V. i told a lot but no ones listen...

Hi cypher
as i can see you say that chain1 is connected to J7 and in previous posts i saw that chain1 is the top hashing board, but in my case J7 is connected to the bottom hashing board and that is chain#3 right? and what did you meant when you said connect to the neighbor board? is that to top the botom? please clarify
Thanks in advance

no, doesnt matter which hashboards is top-bottom or right-left. J connectors are define which hashboard is chainX.
you can easily change ones to another... if you can move j7 connectecor to top hashboard, you will see 0750 V (you cant set your own value to J7)

on fpga board :
J7 means Chain1
J9 Chain2
J2 Chain3
J6 Chain4

Thanks that explains alot. So can i take off the cable from J7 and connect to one of the free slots on the top hashing board? Will that work?

Use FPGA board J connecectors for change, dont use hashboard connectors.

[wajira]

hi pls somebody tell me antminer s4 underclock settings for better power consumption and low fan noise. Thank you!!!

[starsoccer9]

For S4, if the miner powers up, the back light of the small LCD comes on but screen shows blank, highly likely it is the MicroSD card either got loose inside, or data on the microSD card got corrupted.

You will find the MicroSD card inside the case, behind the LCD panel and next to the PSU.  There is a Black Colored Beagle Bone Black and there is a little MicroSD card slot on it.  Lightly push the card and it will pop right out and push it back in.

Then please try to power it back on and see if it boots up.  If not, you may have to use this software to reimage (Write) the factory default data of S4.

Image Software (Windows):
https://bitmaintech.com/files/download/SD-partition-tool.zip

S4 Factory Default Image:
https://bitmaintech.com/files/download/initramfs.bin.SD-20141016.tar.gz

Any issues, please PM or contact our live support at 844-248-6246 Option 2

Thank you!

That link to Image Software doesn't appear to be a Windows app.

M

Using windows, is there software to flash the S4 image? I cant seem to find any software. I looked at windows disk imager but it only takes .img files

[MyRig]

Please go to http://bitmain.zendesk.com  and go to S4 Download Section.  There are several version on bootable images you can use to restore S4 MicroSD card.

you can use the free software called HDD Guru Raw Disk Copy

http://hddguru.com/software/HDD-Raw-Copy-Tool/

http://hddguru.com/software/HDD-Raw-Copy-Tool/HDDRawCopy1.10Setup.exe

Using windows, is there software to flash the S4 image? I cant seem to find any software. I looked at windows disk imager but it only takes .img files

[starsoccer9]

Please go to http://bitmain.zendesk.com  and go to S4 Download Section.  There are several version on bootable images you can use to restore S4 MicroSD card.

you can use the free software called HDD Guru Raw Disk Copy

http://hddguru.com/software/HDD-Raw-Copy-Tool/

http://hddguru.com/software/HDD-Raw-Copy-Tool/HDDRawCopy1.10Setup.exe

Using windows, is there software to flash the S4 image? I cant seem to find any software. I looked at windows disk imager but it only takes .img files

I tried this, it still did not work. I cant connect to it at all, but both LED are Green and the LCD screen is lit up just blank

[mdude77]

It seems there's a problem with the extranonce.subscribe implementation in the latest S4 firmware.  I was able to get it working with westhash by putting #xnsub at the end of the URL.  However, when the price dropped below my parms to make it worth mining, the pool didn't switch to dead.  Furthermore, the speed (5s) changed to 0.  All rebooting the unit did was change the speed (avg) to 0 as well.  Lastly, it happened to both of my S4s simultaneously.  Only once I removed the #xnsub from the URLs did both start behaving normally again.

M