[Setup & Troubleshoot] Bitmain AntMiner S1 180GH/S miner

[klondike_bar]

Bogart
Hes (SVK) asking can two miners or better to say two PDU be powered from single power elctricy socket (AC 220 or 110V)

I say yes, it can.
I suggest to use multiextension power socket cord with protection and fuse.

Exactly 

I have 220/240V elec. connection(normal voltage). Each miner has its own PSU CX750M and each blade has its own PCI 6 pin cable.

I was worried if two miners are not too much for one wall socket.

Thanks everyone for a help 

naw. a fridge or microwave can easily pull the same 1500W

[nairb131]

Bogart
Hes (SVK) asking can two miners or better to say two PDU be powered from single power elctricy socket (AC 220 or 110V)

I say yes, it can.
I suggest to use multiextension power socket cord with protection and fuse.

Exactly 

I have 220/240V elec. connection(normal voltage). Each miner has its own PSU CX750M and each blade has its own PCI 6 pin cable.

I was worried if two miners are not too much for one wall socket.

Thanks everyone for a help 

The amount that can be drawn on each outlet is a matter of the circuit. I am running 4 of these miners on one outlet. they are split 2 to a surge protector which is connected to a wattage meter and then connected to the wall. There is no heat or problems with mine. I do however know that there is nothing else on this circuit. Each set of miners is pulling ~850w.

[allinvain]

I have one unit that seems to be underperforming. After leaving it to mine for 30+ minutes the AVG hashrate is 155 to 160:

See screenshot below:



Uploaded with ImageShack.us

Using the formula posted by giletto it came out to 0.54. Can someone just double check my math please in case I did something wrong. If 0.54% is correct then I'm wondering why the miner reports the average as being lower than what the other units report. Also why so many HW when the other units have a significantly less amount.

All units have the latest firmware as according to the kernel log:

Linux version 3.10.12 (xxl@ubuntu) (gcc version 4.6.4 (OpenWrt/Linaro GCC 4.6-2013.05 r38031) ) #7 Thu Dec 12 12:53:57 CST 2013

[allinvain]

I checked on the poolside and it's definitely lower than the other miners..~150 gh.

sushi, any ideas what I can do?

[giletto]

The miner are offered with 180GH/s +/-10%

So my first opinion is, maybe yours run at the bottom.

You could Overclock to reach at least 180GH/s you temperate look low enough so it should be easily for your miner.

[Bismarckbkk]

I had a miner that ran about the same. Around 160gh/s .

Ive played around with the clocking and its now running 400mhz 199-201 gh/s . Still about 4 -5 gh/s lowe then other miners but not nearly as bad.

You should try to play around with the 375 ./ 400mhz settings ( void warranty ). Do a couple of reboots. see if it helps.

[allinvain]

Here is what worries me. All of a sudden the hashrate dropped to 89 GH. This makes me think one of the blades is faulty. I tried rebooting it 2 min ago; still at 89 GH.

This IMHO justifies a warranty claim. I haven't power cycled it as I'm at work but I will try that later on.

[allinvain]

2
32
350
0
37
xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx

shit..all x's on the second blade.

*sigh* here I was hoping I'd get lucky and get perfect working miners.

[Bismarckbkk]

Here is what worries me. All of a sudden the hashrate dropped to 89 GH. This makes me think one of the blades is faulty. I tried rebooting it 2 min ago; still at 89 GH.

This IMHO justifies a warranty claim. I haven't power cycled it as I'm at work but I will try that later on.

I had the exact same thing happen to my miner....dont give up yet.

[allinvain]

Here is what worries me. All of a sudden the hashrate dropped to 89 GH. This makes me think one of the blades is faulty. I tried rebooting it 2 min ago; still at 89 GH.

This IMHO justifies a warranty claim. I haven't power cycled it as I'm at work but I will try that later on.

I had the exact same thing happen to my miner....dont give up yet.

How did you fix yours?

Not that I think this makes a diference but this miner is running on the same PSU as another one. It's a 1000W psu. Each blade has it's own dedicated 6pin PCIE cable.

[Andrey]

2
32
350
0
37
xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx

shit..all x's on the second blade.

*sigh* here I was hoping I'd get lucky and get perfect working miners.

try to replace power supply, or at least measure the voltage on blade power screws. at 11V and lower these units are underperforming greatly and a lot of cheaper PSUs drop the voltage greatly under load.

[allinvain]

2
32
350
0
37
xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx

shit..all x's on the second blade.

*sigh* here I was hoping I'd get lucky and get perfect working miners.

try to replace power supply, or at least measure the voltage on blade power screws. at 11V and lower these units are underperforming greatly and a lot of cheaper PSUs drop the voltage greatly under load.

Hmm, I'm using an OCZ EliteXtream 1000W unit which should be a fairly top end PSU. But I think your idea about trying a different PSU is a good one. I am thinking placing this unit on it's own dedicated PSU and see if that makes a difference. I have another exact same PSU that I will try. If that fails I will have to try a completely different one.

[allinvain]

Now half the chips on the second blade are dead:

2
16
350
0
36
xxxxxoxo xoxoxoxo xxxxxxxx xxxxxxxx

edit: this is after a power cycle

[klondike_bar]

Now half the chips on the second blade are dead:

36
xxxxxoxo xoxoxoxo xxxxxxxx xxxxxxxx

edit: this is after a power cycle

how do you have these powered? (PSU brand, model, wattage, and # of 12v rails and what cables you are using to run 12V)
alternatively, turn off and let it cool for a few minutes before starting up again

[allinvain]

Now half the chips on the second blade are dead:

36
xxxxxoxo xoxoxoxo xxxxxxxx xxxxxxxx

edit: this is after a power cycle

how do you have these powered? (PSU brand, model, wattage, and # of 12v rails and what cables you are using to run 12V)
alternatively, turn off and let it cool for a few minutes before starting up again

PSU: OCZ EliteXstrem 1000W
Rails: Single rail.
Cables: 4 pci-e 6pin connectors, 2 per unit. This is not a modular PSU so I have no chance of somehow running the unit off of the same PSU connector. I made sure the pci-e leads do not seem to originate from the same PSU strand but it's not really all that easy to tell.

The unit is hosted with a family member so sadly I have to wait until I can do the turning it off and letting it cool trick.

I think though that either putting the unit on it's own dedicated PSU will fix it, and if that does not then it's clearly a fault with the miner itself.

[Bismarckbkk]

So youre running 2 miners on the PSU? If thats the case this could def be cause of the problem.

At the time I had this problem I bought a new PSU as well but ultimately the performance improved after I changed the clocking setting.

I was super sleep deprived when I did it and dont remember what I changed but if you could post your current clock settings I might be able to give you a suggestion.

Also I highly recommend reading this thread : https://bitcointalk.org/index.php?topic=348327.0 .

When the miners first ran I had tons of x too but they were firmware errors. My miners worked fine, I do think I had some faulty clock settings that caused it but again I dont remember which they were.

[allinvain]

So youre running 2 miners on the PSU? If thats the case this could def be cause of the problem.

At the time I had this problem I bought a new PSU as well but ultimately the performance improved after I changed the clocking setting.

I was super sleep deprived when I did it and dont remember what I changed but if you could post your current clock settings I might be able to give you a suggestion.

Also I highly recommend reading this thread : https://bitcointalk.org/index.php?topic=348327.0 .

When the miners first ran I had tons of x too but they were firmware errors. My miners worked fine, I do think I had some faulty clock settings that caused it but again I dont remember which they were.

Here is the weird thing, on the web front end it says the clockrate is 350 MHz, but if I ssh into the box and vi /etc/config/cgminer it says the following:

config cgminer 'default'
        option chip_frequency '282'
        option miner_count '24'
        option api_allow 'W:127.0.0.1'
        option more_options '--quiet'
        option target '60'
        option overheat '70'

BUT why would that be an issue cause the second miner has the SAME config.

Errr...duh, nevermind. I see that /etc/config/asic-config is where it stores the actual chip frequency config table. There is says 350.

[allinvain]

Should I change " option 'timeout'        '40' " to something else perhaps? It's at 40 now, goxed in the thread suggested 36 if you set the clock to 375. I want to avoid overclocking the unit just yet for warranty purposes, so I'm looking for some suggestions besides just bumping the clock higher.

[Bismarckbkk]

what is the timout number?

[Bismarckbkk]

can you copy the entire settings here?