Pages:
Author

Topic: [Setup & Troubleshoot] Bitmain AntMiner S1 180GH/S miner - page 7. (Read 452359 times)

newbie
Activity: 13
Merit: 0
Fluxuation in miner speed. As I watch the miners. I have noticed a 30 to 50 percent fluctuation in the speed on the miners. They all do it, and at different times. Miner 1 will be running at 180 GH/s Miner 2 Will be at 185 GH/s and Miner 3 Will be at 195 GH/s. Then one of them will drop to 135 GH/s. The rest stay close to what they were. Then a few minutes later one of the others will drop to 120 GH/s and the one that was at 135 GH/s will be back to 185 or so GH/s Any ideas as to why? This goes on all day long..
member
Activity: 61
Merit: 10
Tell me why not drop in speed devices, if the temperature is 36 rpm and 3600

The fan speed of 3600 RPM at temp 36D is very high.
The fan connectors may be not be placed properly, try unplugging the connect and check the pins in the blade.
Also check for any cracks or cuts in the controller wires from the fan to blades.
If the above two things are not the culprit then, the two possibility are fan rpm control or some controller issues in firmware of your ant.
Do the troubleshooting this way;
Fan cable check
Fan connector check
Firmware reset or update
Try checking with a spare fan
Temp sensor problem in Ant.

But when it turns 49 temp drop to 3350 for some reason, everything in turn.
full member
Activity: 229
Merit: 100
We support you for as long as you need.
They have on the instruction manual here http://www.coinminer.ch/webshop/wp-content/uploads/Setup-and-Installation-Antminer-S1.pdf



If you are lucky to have the PCI ATX and PCI Express Extender cables, do all the cuts on
the extenders, so you don’t chop the cables on the PSU itself.
WARNING: An 8 pin 12 volt cable has yellow wires on the same side as the
connector clip. An 8 Pin PCI Express cable has black wires on the clip side. (See
Above Pinout Diagram)
We need at least 2 available +12 V line. (Yellow and Black. 6 black and 6 yellow are
recommended) The miner has 2 hashing PCBs and each needs own power source. We If
you just use 1 +12V to each PCB, it will more than likely over load the line and we DO
NOT RECOMMEND it due to the HAZARDOUS CONDITION.
PCI ATX plug is usually +12 V line. These are the plug with 4 wires into 1 plug and you’ll
see 2 plugs as a pair. PCI-E (PCI Express) plug can be +12 V line as well. By cutting the
PCI ATX plug, you’ll have total of 4 yellow lines and 4 black lines.
Cut off the PCI-E (If it’s +12V) connector and you’ll have additional 2 yellow lines and 2
black lines.


So with that being said, the 4 yellow are on the clip side, opposite of the diagram.  So correct me if i'm wrong but:
I cut the 8 pin off, then since this is reverse of the instructions, what color do i put to the negative of the antminer and the positive.  Then do the other two wires just not do anything?

Then for the 4 pin:  The one in their pic has 2 black wires and 2 yellow wires.

Mine has 2 Black, 1 Red, 1 Yellow

They show that they have two on their power supply, this one only has one.  But I can grab the other 4 pin plug from the other power supply and that should plug right in to the open 4 pin port if I'm not mistaken.  Do I need to do that to have two 4 pin plugs?

And even if I do that, what wires am i using?  I take it that this means that I can use these power supplies?

Thanks for the help


hero member
Activity: 784
Merit: 504
[im]http://i00.i.aliimg.com/wsphoto/v0/1601493210/Free-Shipping-font-b-RACKABLE-b-font-SYSTEMS-450W-2U-1U-SERVER-POWER-SUPPLY-3F27-45.jpg[/img]

Now I have another of these so I can prob snag the 4 pin from that one and make it so this one has two 4 pins if needed?

As per my understanding the PCI-E power connector ports yellow cables are +12VDC.
The black ones are ground. Join all the +12V yellow wires into one and connect to your S1 like wire for the black one too.
You may use the second PSUs PCIE pin connector to make another PCI-E connector for the first PSU.
Make sure that you are not exceeding the limit of +12V rail that's 12V*30A = 360Watts for your PSU in image.
Better, buy a cheap digital multimeter for less than $10 and you yourself check polarities are correct.
NB: Also check the wires are of correct 12V rails with a multimeter.
I think the one you are saying is a CPU connector cable and it may not have the full ampere support.
/Edited
full member
Activity: 229
Merit: 100
We support you for as long as you need.


Now I have another of these so I can prob snag the 4 pin from that one and make it so this one has two 4 pins if needed?
hero member
Activity: 784
Merit: 504
Alright so just got another one here, hooking it up to a server power supply I had laying around that was Brand new.  

The 8 pin has the yellow wires on the clip side, so do those wind up being the negative?  So just so i'm straight, clip the plug and put the yellow wires into the ground, and black wires into the positive and just don't use the other 2 wires that are hanging.

And for the 4 pin, I have 2 blacks, a red and a yellow?  What do I do with those?

Thanks

Putting an image will help understand better and save 1000 words.
hero member
Activity: 784
Merit: 504
Tell me why not drop in speed devices, if the temperature is 36 rpm and 3600

The fan speed of 3600 RPM at temp 36D is very high.
The fan connectors may be not be placed properly, try unplugging the connect and check the pins in the blade.
Also check for any cracks or cuts in the controller wires from the fan to blades.
If the above two things are not the culprit then, the two possibility are fan rpm control or some controller issues in firmware of your ant.
Do the troubleshooting this way;
Fan cable check
Fan connector check
Firmware reset or update
Try checking with a spare fan
Temp sensor problem in Ant.
full member
Activity: 229
Merit: 100
We support you for as long as you need.
Alright so just got another one here, hooking it up to a server power supply I had laying around that was Brand new. 

The 8 pin has the yellow wires on the clip side, so do those wind up being the negative?  So just so i'm straight, clip the plug and put the yellow wires into the ground, and black wires into the positive and just don't use the other 2 wires that are hanging.

And for the 4 pin, I have 2 blacks, a red and a yellow?  What do I do with those?

Thanks
member
Activity: 61
Merit: 10
Tell me why not drop in speed devices, if the temperature is 36 rpm and 3600
legendary
Activity: 3892
Merit: 4331
I'm not an EE but as I asked before whats the temperature around your device.  My ambient temp of my s1 room is around 80F (26C)and the last time I check I'm sitting at 40 C with the default fan and a titan fan (not enough cfm but enough to make a difference without buying an antminer case cover off amazon).

Mine seems to be running fine [knock knock] but I occasionally try to see if there's a dust nest but not yet and blow it out.

That's my non engineer expert advise.

When doing blowing or vacuuming, especially the boards, watch out for ESD (electrostatic discharge), keep the unit connected to the ground.
I tried to do as little as possible re cleaning of working units, but got better A/C filters, hence less dust.
member
Activity: 61
Merit: 10
Hello tell me the reason for my fan speed does not reset.  Thats when cold then immediately grows And does not reset when the temperature is 35 or 45 shows in 3720 within 3480  Thank you  if daste answer.
sr. member
Activity: 291
Merit: 250
Scam-Busting PSA: Beware of Black Arrow Software
I'm not an EE but as I asked before whats the temperature around your device.  My ambient temp of my s1 room is around 80F (26C)and the last time I check I'm sitting at 40 C with the default fan and a titan fan (not enough cfm but enough to make a difference without buying an antminer case cover off amazon).

Mine seems to be running fine [knock knock] but I occasionally try to see if there's a dust nest but not yet and blow it out.

That's my non engineer expert advise.

Just for reference, with the extra fan and heatsinks, and an ambient of 20-22c, the chips were running in the mid 30's c normal, and late 20's undervolted.
hero member
Activity: 546
Merit: 500
I'm not an EE but as I asked before whats the temperature around your device.  My ambient temp of my s1 room is around 80F (26C)and the last time I check I'm sitting at 40 C with the default fan and a titan fan (not enough cfm but enough to make a difference without buying an antminer case cover off amazon).

Mine seems to be running fine [knock knock] but I occasionally try to see if there's a dust nest but not yet and blow it out.

That's my non engineer expert advise.
sr. member
Activity: 291
Merit: 250
Scam-Busting PSA: Beware of Black Arrow Software

If the machine seems to mine but light always flashing, no matter what you do, order a new ECB, always handy to have a spare anyways.

Waiting a week for a replacement is agonizing  Angry

Thanks for the detailed suggestions, but just a quick note on that last point. As of last week Bitmaintech have no more stock of S1 controllers. I paid BTC0.08 for one, and had to get refunded for this reason. This is the main reason I'm hoping someone more EE orientated than myself might be able to establish what is causing so many blades to fail at the moment so we can do at-home repairs.
hero member
Activity: 1666
Merit: 513
Leading Crypto Sports Betting & Casino Platform
Since I last posted. I upgraded my power supplies. Added a new Antminer S1. Now that I have done that. The new one is running fine. The old one was running fine until I added the new one. I changed the address on them both so there would not be a conflict. I will call it antminer2 I can log in I can do everything. I have eth1 connected and and I have a wifi connection. But when I run a diagnostic. It times out. Any ideas?

It happend me once with 3 devices. I had to clear cookies of browser or use different browser to ssh. Iam not sure if it helps you but give it a shot

I tried that. Like I said. I can login and look at all the settings. I can run the graphs. I can change settings. It just will not mine. I ran the diagnostic. Ping return loss data. Route trace returns the address to my avalon miner and nslookup returns 78.24.191.177 openwrt.org  It will let me go local on the network, just not on the internet.

Something else I have learned. I have two Avalon2 miners. One is running the other I am working on. When I shut down the Avalon2 miner. The last antminer s1 comes up and mines no problem. Not sure as to why. Anyone?

The blades seem to dropping like flies, lots of people are reporting similar problems. Web gui/ssh all fine, full interface access, but blades just stop firing up. Mine has died, and it was very well looked after. I thought it must be a faulty controller as both blades stopped at once, but I tried another controller and its definitely the blades. I've ask people to add posts here so we can keep track: https://bitcointalk.org/index.php?topic=677110.0;topicseen

I've asked Bitmain to chime in, but they haven't yet. Do we know anyone else with any electrical engineering knowledge of the S1 who could help us work out which component on the blades is failing? Any suggestions anyone?

Something I discovered this morning. Yesterday I got all 3 miners running. Sometime during the night I lost power. One miner came back up the other two didn't. They were running but not mining. I ran system log. Looked at the difference between the 3 miners. The one difference between the one running and the other two was this. /tmp/resolr.config.auto. The one running had an address of 192.168.1.1 the other two had an address of 192.168.1.101. The miners with the192.168.1.101 address will not mine. I did a reset not a reboot, on one miner. It can up and I did a system log and the address was 192.168.1.1 and it has started to mine. The last mine is not mining at the moment. It's system log is 192.168.1.101. Any idea? Is the resolr.config.auto corrupted? How to I see this file and or edit it? Any ideas?

Here is what I normally do when that happens....

Start from scratch, with a clear conscience and a clear mind.

1- Turn everything off - PSU, Internet, Your Computer (which you use to config your miners) and switch if you use one and/or router if you use one, then wait 3 minutes.
2- Turn on your internet and let it fully boot itself; if you have a router, turn it on next or your switch and finally your computer.
3- Wait about 1 minute (You want to take your time restarting during the day) for the network to get to idle (finish start up process)
4- Turn on one (or two) machines at a time; waiting until both are connected and mining before continuing.
5- If one machine don't mine and the rest do, perform a hard reset (button) on it - reconfigure as you know it.  Make sure it's not an internet problem before!!
6- If a machine doesn't turn back on after a network-wide hard reset, you may need a new ECB, which are about $50, simply message Bitmain Tech.  If none turn on, it's likely a network or internet issue.

I use a rather low subnet range (1.11,12,13,14,15) than most because no personal devices you buy in stores use that range and I have a few dozen internet connected devices in my lab aside from bitcoin miners.  Some network experts around would recommend against using 192.168.1.1 as some router/modem manufacturers use 192.168.1.1 as default IP and they don't tell you often times.  It's probably something stupid conflicting with your network.  That's why many would say to start at .254 and work your way down.  It could also be the ECB which isn't TOO bad, not as bad as changing a whole blade.

I had that same problem over a month ago.

I found adding a 24 port switch and changing my subnet/IP starting at 192.168.1.11 worked wonderfully.

If you screw around too much with the reset button, it will increase the risk of needing a new ECB which sucks when you need one.

That said the S3 upgrade kits should come out by moose hunting season I'd assume.

If the machine seems to mine but light always flashing, no matter what you do, order a new ECB, always handy to have a spare anyways.

Waiting a week for a replacement is agonizing  Angry
newbie
Activity: 14
Merit: 0
Since I last posted. I upgraded my power supplies. Added a new Antminer S1. Now that I have done that. The new one is running fine. The old one was running fine until I added the new one. I changed the address on them both so there would not be a conflict. I will call it antminer2 I can log in I can do everything. I have eth1 connected and and I have a wifi connection. But when I run a diagnostic. It times out. Any ideas?

It happend me once with 3 devices. I had to clear cookies of browser or use different browser to ssh. Iam not sure if it helps you but give it a shot

I tried that. Like I said. I can login and look at all the settings. I can run the graphs. I can change settings. It just will not mine. I ran the diagnostic. Ping return loss data. Route trace returns the address to my avalon miner and nslookup returns 78.24.191.177 openwrt.org  It will let me go local on the network, just not on the internet.

Something else I have learned. I have two Avalon2 miners. One is running the other I am working on. When I shut down the Avalon2 miner. The last antminer s1 comes up and mines no problem. Not sure as to why. Anyone?

The blades seem to dropping like flies, lots of people are reporting similar problems. Web gui/ssh all fine, full interface access, but blades just stop firing up. Mine has died, and it was very well looked after. I thought it must be a faulty controller as both blades stopped at once, but I tried another controller and its definitely the blades. I've ask people to add posts here so we can keep track: https://bitcointalk.org/index.php?topic=677110.0;topicseen

I've asked Bitmain to chime in, but they haven't yet. Do we know anyone else with any electrical engineering knowledge of the S1 who could help us work out which component on the blades is failing? Any suggestions anyone?

Something I discovered this morning. Yesterday I got all 3 miners running. Sometime during the night I lost power. One miner came back up the other two didn't. They were running but not mining. I ran system log. Looked at the difference between the 3 miners. The one difference between the one running and the other two was this. /tmp/resolr.config.auto. The one running had an address of 192.168.1.1 the other two had an address of 192.168.1.101. The miners with the192.168.1.101 address will not mine. I did a reset not a reboot, on one miner. It can up and I did a system log and the address was 192.168.1.1 and it has started to mine. The last mine is not mining at the moment. It's system log is 192.168.1.101. Any idea? Is the resolr.config.auto corrupted? How to I see this file and or edit it? Any ideas?
newbie
Activity: 14
Merit: 0
Since I last posted. I upgraded my power supplies. Added a new Antminer S1. Now that I have done that. The new one is running fine. The old one was running fine until I added the new one. I changed the address on them both so there would not be a conflict. I will call it antminer2 I can log in I can do everything. I have eth1 connected and and I have a wifi connection. But when I run a diagnostic. It times out. Any ideas?

It happend me once with 3 devices. I had to clear cookies of browser or use different browser to ssh. Iam not sure if it helps you but give it a shot

I tried that. Like I said. I can login and look at all the settings. I can run the graphs. I can change settings. It just will not mine. I ran the diagnostic. Ping return loss data. Route trace returns the address to my avalon miner and nslookup returns 78.24.191.177 openwrt.org  It will let me go local on the network, just not on the internet.

The blades seem to dropping like flies, lots of people are reporting similar problems. Web gui/ssh all fine, full interface access, but blades just stop firing up. Mine has died, and it was very well looked after. I thought it must be a faulty controller as both blades stopped at once, but I tried another controller and its definitely the blades. I've ask people to add posts here so we can keep track: https://bitcointalk.org/index.php?topic=677110.0;topicseen

I've asked Bitmain to chime in, but they haven't yet. Do we know anyone else with any electrical engineering knowledge of the S1 who could help us work out which component on the blades is failing? Any suggestions anyone?

Something I discovered this morning. Yesterday I got all 3 miners running. Sometime during the night I lost power. One miner came back up the other two didn't. They were running but not mining. I ran system log. Looked at the difference between the 3 miners. The one difference between the one running and the other two was this. /tmp/resolr.config.auto. The one running had an address of 192.168.1.1 the other two had an address of 192.168.1.101. The miners with the192.168.1.101 address will not mine. I did a reset not a reboot, on one miner. It can up and I did a system log and the address was 192.168.1.1 and it has started to mine. The last mine is not mining at the moment. It's system log is 192.168.1.101. Any idea? Is the resolr.config.auto corrupted? How to I see this file and or edit it? Any ideas?
hero member
Activity: 546
Merit: 500
Whats the temps on these blades. Its summer in the U.S so that could be an issue.

Also have to hope they are "new" if stated and not an s1 over clocked and sold as new.
sr. member
Activity: 291
Merit: 250
Scam-Busting PSA: Beware of Black Arrow Software
Since I last posted. I upgraded my power supplies. Added a new Antminer S1. Now that I have done that. The new one is running fine. The old one was running fine until I added the new one. I changed the address on them both so there would not be a conflict. I will call it antminer2 I can log in I can do everything. I have eth1 connected and and I have a wifi connection. But when I run a diagnostic. It times out. Any ideas?

It happend me once with 3 devices. I had to clear cookies of browser or use different browser to ssh. Iam not sure if it helps you but give it a shot

I tried that. Like I said. I can login and look at all the settings. I can run the graphs. I can change settings. It just will not mine. I ran the diagnostic. Ping return loss data. Route trace returns the address to my avalon miner and nslookup returns 78.24.191.177 openwrt.org  It will let me go local on the network, just not on the internet.

The blades seem to dropping like flies, lots of people are reporting similar problems. Web gui/ssh all fine, full interface access, but blades just stop firing up. Mine has died, and it was very well looked after. I thought it must be a faulty controller as both blades stopped at once, but I tried another controller and its definitely the blades. I've ask people to add posts here so we can keep track: https://bitcointalk.org/index.php?topic=677110.0;topicseen

I've asked Bitmain to chime in, but they haven't yet. Do we know anyone else with any electrical engineering knowledge of the S1 who could help us work out which component on the blades is failing? Any suggestions anyone?
newbie
Activity: 14
Merit: 0
Since I last posted. I upgraded my power supplies. Added a new Antminer S1. Now that I have done that. The new one is running fine. The old one was running fine until I added the new one. I changed the address on them both so there would not be a conflict. I will call it antminer2 I can log in I can do everything. I have eth1 connected and and I have a wifi connection. But when I run a diagnostic. It times out. Any ideas?

It happend me once with 3 devices. I had to clear cookies of browser or use different browser to ssh. Iam not sure if it helps you but give it a shot

I tried that. Like I said. I can login and look at all the settings. I can run the graphs. I can change settings. It just will not mine. I ran the diagnostic. Ping return loss data. Route trace returns the address to my avalon miner and nslookup returns 78.24.191.177 openwrt.org  It will let me go local on the network, just not on the internet.
Pages:
Jump to: