Bitmain AntMiner U1 Tips & Tricks

[deagel]

Guys, I have a strange Problem. Hope someone is able to help me. 0.05BTC bounty for getting my miners to atleast 2,6 Ghash each

I got 16 Antminer U1. All are running at 1,088 V. I have a 49 Port USB Hub and a 450W PSU exclusively for those 16 U1. I have additional coolers attached and a fan blowing cool air. The Miners get definitely not too warm.

I tried Bitmain´s Cgminer, different BFGminer version on Windows, Minepeon and Raspcontrol. Cgminer gave me too much errors and for BFGminer I have the following Problem:

While I should be able to get 48 or atleast 44,8 Ghash out of the 16 U1, I get far less. Recently did some testing with BFGminer 3.10 and Raspcontrol

At 4D81 I get 6 Ghash
At 4C81  I get  10 Ghash
At 0B81  I get  12 Ghash
At 0A81  I get  15 Ghash
At 0981  I get  14 Ghash
At 0781  I get 10 Ghash

The Problem here is, that most Miners run somewhere under 500 Mhash. No Hardware Errors. Those U1 who do more Ghash are getting warmer then those who only do some Mhash. From test to test the variance is very big.

Currently they run with 4C81 under BFGminer 3.9.0 and Minepeon with 25Ghash. For some reason the hashrates with bfgminer 3.9.0 are similar among all sticks. The best result I got so far was about 30Ghash (also BFGminer 3.9.0 and Minepeon). The funny thing is, that I didn't give any frequenzy to BFGminer in this test so that the Miner should have run with 1,6Ghash. For some reasen they did nearly 2Ghash. This was absolut stable (ran for <24 hours). Again: I do not get any errors (or very very few on some sticks <.1) beyond 4E81.

Sorry for the bad english  

Thanks
Muto

Hi,
take a look at my solution:

https://bitcointalk.org/index.php?topic=390929.msg5155376#msg5155376


i needed the y usb adapter without my antminer run slow as yours.

[Jay_Pal]

Posted cgminer 4.0.1 with fixes for these.

Hi ckolivas

Running 4.0.1 and it's running great!
Thank you!!!

[forbun]

Hi all,

I have a Bitmain Antminer U1 USB miner (1.6 GH/s) which seems to be dead. When I start cgminer, it doesn't seem to find any USB device attached. Here's the output:

Code:

 cgminer version 3.8.5 - Started: [2014-03-08 15:57:01]
--------------------------------------------------------------------------------
 (5s):0.000 (avg):0.000h/s | A:0  R:0  HW:0  WU:0.0/m
 ST: 2  SS: 0  NB: 1  LW: 7  GF: 0  RF: 0
 Connected to stratum.bitcoin.cz diff 1 with stratum as user ...
 Block: c8c977e6...  Diff:3.81G  Started: [15:57:01]  Best share: 0
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
--------------------------------------------------------------------------------

 [2014-03-08 15:57:38] Bmsc send golden nonce
 [2014-03-08 15:57:38] Bmsc recv golden nonce timeout
 [2014-03-08 15:57:38] -----------------start freq-------------------
 [2014-03-08 15:57:39] Send frequency 82078106
 [2014-03-08 15:57:39] Send freq getstatus 84000400
 [2014-03-08 15:57:41] ------recv freq getstatus no data finish------
 [2014-03-08 15:57:42] -----------------start nonce------------------
 [2014-03-08 15:57:42] Bmsc send golden nonce
 [2014-03-08 15:57:42] Bmsc recv golden nonce timeout
 [2014-03-08 15:57:46] Pool 0 difficulty changed to 1
 [2014-03-08 15:57:46] Stratum from pool 0 requested work restart
 [2014-03-08 15:57:48] -----------------start freq-------------------
 [2014-03-08 15:57:48] Send frequency 82078106
 [2014-03-08 15:57:49] Send freq getstatus 84000400

These messages mostly just repeat. After I hit "q", I see:

Code:

 [2014-03-08 15:58:24]
Summary of runtime statistics:

 [2014-03-08 15:58:24] Started at [2014-03-08 15:57:01]
 [2014-03-08 15:58:24] Pool: http://stratum.bitcoin.cz:3333
 [2014-03-08 15:58:24] Runtime: 0 hrs : 1 mins : 22 secs
 [2014-03-08 15:58:24] Average hashrate: 0.0 Kilohash/s
 [2014-03-08 15:58:24] Solved blocks: 0
 [2014-03-08 15:58:24] Best share difficulty: 0
 [2014-03-08 15:58:24] Share submissions: 0
 [2014-03-08 15:58:24] Accepted shares: 0
 [2014-03-08 15:58:24] Rejected shares: 0
 [2014-03-08 15:58:24] Accepted difficulty shares: 0
 [2014-03-08 15:58:24] Rejected difficulty shares: 0
 [2014-03-08 15:58:24] Hardware errors: 0
 [2014-03-08 15:58:24] Utility (accepted shares / min): 0.00/min

 [2014-03-08 15:58:24] Work Utility (diff1 shares solved / min): 0.00/min

 [2014-03-08 15:58:24] Stale submissions discarded due to new blocks: 0

 [2014-03-08 15:58:24] Unable to get work from server occasions: 0

 [2014-03-08 15:58:24] Work items generated locally: 9
 [2014-03-08 15:58:24] Submitting work remotely delay occasions: 0

 [2014-03-08 15:58:24] New blocks detected on network: 1

 [2014-03-08 15:58:24] Summary of per device statistics:

Does Bitmain have any kind of warranty, repair, or replacement?

I have another (separate) Antminer U1 and it seems to be working great.

[chromosoma]

"nonce timeout" means  your antminer is not responding.:

Try to change its frequency to the lowest value.

[-ck]

Hi all,

I have a Bitmain Antminer U1 USB miner (1.6 GH/s) which seems to be dead. When I start cgminer, it doesn't seem to find any USB device attached. Here's the output:

Code:

 cgminer version 3.8.5 - Started: [2014-03-08 15:57:01]

 [2014-03-08 15:57:38] Bmsc send golden nonce

Does Bitmain have any kind of warranty, repair, or replacement?

I have another (separate) Antminer U1 and it seems to be working great.

Try the official version of cgminer instead of this unofficial fork just to be sure.

[jedimind]

Hey guys, I'm having a lot of trouble getting my antminer U1 (2) overclocked.
I guess the frustrating part for me has been I had them overclocked and now after having to restart my cgminer I am not able to get the CMD prompt to read the command correctly.

Here is the command line I'm using cgminer.exe --bmsc-options 115200:20 -o poolname -u username -p password --bmsc-freq 0A81

The result is cgminer.exe: --bmsc-options: unrecognized option

Does cgminer need to be running/not running while I enter the command in CMD?
I tried making a .bat file with this command with no luck.
I uninstalled cgminer and downloaded a new version...
Command cgminer.exe runs the miner no problems.
I'm thinking its something with the command line and I had trouble with it the first time I overclocked.
I keep finding command lines that are similar and pasting them in (editing pool user and pass) with no luck.
Do I need to be using a specific version of cgminer?

Thank you for your help

[Dunkelheit667]

Some examples for 2.0 GHash/s:

Code:

cgminer.exe --bmsc-options 115200:20 -o pool_url -u user -p password --bmsc-freq 0981

Should work with Bitmain's cgminer 3.8.5 fork, found here.

Code:

cgminer.exe --anu-freq 250 -o pool_url -u user -p password

Should work with the latest official cgminer, found here.

Code:

bfgminer.exe -S antminer:all --set-device antminer:clock=x0981 -o pool_url -u user -p password

Should work with the latest bfgminer, found here.

And last but not least, BitMinter's (Beta) Client will run them at 1.6 GHash/s.

[jedimind]

Code:

[code][quote author=Dunkelheit667 link=topic=390929.msg5754189#msg5754189 date=1395097842]
Some examples for 2.0 GHash/s:

[code]cgminer.exe --bmsc-options 115200:20 -o pool_url -u user -p password --bmsc-freq 0981

Should work with Bitmain's cgminer 3.8.5 fork, found --anu-freq 250 -o pool_url -u user -p password[/code]
Should work with the latest official cgminer, found [url=http[Suspicious link removed] -S antminer:all --set-device antminer:clock=x0981 -o pool_url -u user -p password[/code]
Should work with the latest bfgminer, found [url=http://luke.dashjr.org/programs/bitcoin/files/bfgminer/latest/]here.

And last but not least, BitMinter's (Beta) Client will run them at 1.6 GHash/s.
[/quote]

Thank you so much!!!
I used the latest version command line.

[GenTarkin]

Messing with these U1's today, found an easy way to overvolt them... its a complete shot in the dark but....

I pencil modded the contacts where the R2 resistor usually goes (stock there is no resistor). Have no idea what the resistance is that I got to cuz the voltmeter I used to measure is shit as hell on resistance measurements...

But, I got the asic up to 1v doing this

Clocks seem stable at 300mhz @ hashrate of 2.4GH and HW of 0 for the last 8hrs =)
Probably could clock it higher, I tried 325mhz but it stops hashing after like 20mins... think the USB may not be up to par in supplyin enough power to it.

I have a fat old P4 chipset NB heatsink on the heatspreader, its definitely runnin hotter than before ...

[swampyness]

Hi.

Three have been working fine for a few weeks now.  

# 1.6GH/s       0781 (DEFAULT)
# 1.8GH/s       0881 (GOOD)
# 2.0GH/s       0981 (FINE) <1% HW errors (I use this with fan)
# 2.2GH/s       0A81 (UNSTABLE) >5% HW errors

Sample of bfgminer on Linaro Cubietruck and hash rate looks as expected:
ST:6  F:1  NB:748  AS:0  BW:[ 19/  9 B/s]  E:96.68  I: 0.00 BTC/hr  BS:127k
 3            |  6.30/ 6.19/ 5.85Gh/s | A:4382 R:104+1(2.9%) HW:2/.00%
--------------------------------------------------------------------------
 AMU 0:       |  2.06/ 2.06/ 1.96Gh/s | A:1447 R: 26+1(2.1%) HW:1/.00%
 AMU 1:       |  2.06/ 2.06/ 1.95Gh/s | A:1442 R: 37+0(3.1%) HW:1/.00%
 AMU 2:       |  2.06/ 2.06/ 1.94Gh/s | A:1493 R: 41+0(3.3%) HW:0/none

[Jay_Pal]

I quite don't remember seeing an answer for my following question, so excuse me if somehow I'm duplicating it but, what is the real impact in the hashrate of those hardware errors?
Does it mean that hashing at 2.2 with 5% HW is the same as hashing at 1.8 with 0%?
Basically, is it worth?

[bitcoiner49er]

I quite don't remember seeing an answer for my following question, so excuse me if somehow I'm duplicating it but, what is the real impact in the hashrate of those hardware errors?
Does it mean that hashing at 2.2 with 5% HW is the same as hashing at 1.8 with 0%?
Basically, is it worth?

IF it was a straight 5% HW loss, you'd be looking at 2.09GH/s. I doubt this is the case, so yeah, I'd say worth it. The life span of these (really any ASIC miner) is pretty short. Run 'em while you can!!

[Jay_Pal]

I quite don't remember seeing an answer for my following question, so excuse me if somehow I'm duplicating it but, what is the real impact in the hashrate of those hardware errors?
Does it mean that hashing at 2.2 with 5% HW is the same as hashing at 1.8 with 0%?
Basically, is it worth?

IF it was a straight 5% HW loss, you'd be looking at 2.09GH/s. I doubt this is the case, so yeah, I'd say worth it. The life span of these (really any ASIC miner) is pretty short. Run 'em while you can!!

Uh, my mistake...
My brain runs math with 90% errors at 10am :p (believe me, it gets worse along the day!)

Thank you!

[chibi]

Trying to repair some AntMiner U1 that failed due to overvoltage issue.

Having trouble identifying the U5 IC chip.

I have:
Y1 25 MHz Oscillator (that’s the 25 in the frequency formula)
U1 CP2102 USB to UART Bridge
U2 BM1380 ASIC
U3 503B or 5038 (hard to read) WTF?
U4 AOZ1021A Voltage Regulator
U5 5B402? WTF?

On four of these that I have looked at; the U5 shows "EB=TOW", does anyone else have different part numbers on U5?
http://s18.postimg.org/sys3nw34p/EB_TOW.png
http://s18.postimg.org/sys3nw34p/EB_TOW.png

FYI: Here are the part numbers I found and links to a source for them:

U1 CP2102 USB to UART Bridge
http://www.digikey.com/product-detail/en/CP2102-GM/336-1160-5-ND/696598

U2 BM1380 ASIC
? custom to AntMiner?

U3 503B TVS Diode {voltage spike protection}
http://www.digikey.com/product-detail/en/SP0503BAHTG/F2715CT-ND/1154322

U4 AOZ1021A Voltage Regulator
http://www.digikey.com/product-detail/en/AOZ1021AI/785-1155-1-ND/1856090

U5 5B402 ??? EB=TOW ???

[BoogieMan82]

Messing with these U1's today, found an easy way to overvolt them... its a complete shot in the dark but....

I pencil modded the contacts where the R2 resistor usually goes (stock there is no resistor). Have no idea what the resistance is that I got to cuz the voltmeter I used to measure is shit as hell on resistance measurements...

But, I got the asic up to 1v doing this

Clocks seem stable at 300mhz @ hashrate of 2.4GH and HW of 0 for the last 8hrs =)
Probably could clock it higher, I tried 325mhz but it stops hashing after like 20mins... think the USB may not be up to par in supplyin enough power to it.

I have a fat old P4 chipset NB heatsink on the heatspreader, its definitely runnin hotter than before ...

Pencil mod?

[BRADLEYPLOOF]

Changing resistance by rubbing the graphite of a pencil on connections...I think...

[brush242]

Messing with these U1's today, found an easy way to overvolt them... its a complete shot in the dark but....

I pencil modded the contacts where the R2 resistor usually goes (stock there is no resistor). Have no idea what the resistance is that I got to cuz the voltmeter I used to measure is shit as hell on resistance measurements...

But, I got the asic up to 1v doing this

Clocks seem stable at 300mhz @ hashrate of 2.4GH and HW of 0 for the last 8hrs =)
Probably could clock it higher, I tried 325mhz but it stops hashing after like 20mins... think the USB may not be up to par in supplyin enough power to it.

I have a fat old P4 chipset NB heatsink on the heatspreader, its definitely runnin hotter than before ...

Pencil mod?

Yes. Basically, you take a standard lead (graphite) pencil, and make a line between the two contacts you want to connect/lower the resistance of(?). This was bandied about for the BE USBs if I remember.

JUST KEEP IN MIND, you can ruin yer equipment with this type of trick. Be sure you don't mind frying one, and if you need more explanation than I gave, then you probably shouldn't try.

Don't let me stop you though:  https://bitcointalk.org/index.php?topic=524358.msg5814070#msg5814070  I hack the hell out of this stuff.

[Raptor2213]

I am not an electronics guru. I am a merely man with some soldering equipment and a willingness to injure myself and damage sensitive electronic equipment.

So pardon me if this is a silly question...

Am I to take it that by replacing the resistors per this chart we can increase the voltage and thus reach the hash rates beyond the published overclockable speed? (Which would presumably require increasing the ability of the U1 to dissipate heat.)


 

src: https://github.com/AntMiner/AntGen1/blob/master/manual/AntMiner-U1%20user%20guide.pdf

I am having a hard time believing it would be that easy but the ASICMINER upgrades were (after a huge thread of experiments by many talented individuals https://bitcointalk.org/index.php?topic=241652.0)

The Antminer U2 does not appear to have the same resistor configuration.  Has anyone looked into what would be required to increase the voltage?

[Jay_Pal]

I am not an electronics guru. I am a merely man with some soldering equipment and a willingness to injure myself and damage sensitive electronic equipment.

So pardon me if this is a silly question...

Am I to take it that by replacing the resistors per this chart we can increase the voltage and thus reach the hash rates beyond the published overclockable speed? (Which would presumably require increasing the ability of the U1 to dissipate heat.)


 

src: https://github.com/AntMiner/AntGen1/blob/master/manual/AntMiner-U1%20user%20guide.pdf

I am having a hard time believing it would be that easy but the ASICMINER upgrades were (after a huge thread of experiments by many talented individuals https://bitcointalk.org/index.php?topic=241652.0)

The Antminer S2 does not appear to have the same resistor configuration.  Has anyone looked into what would be required to increase the voltage?

Did you meant "U2?
I read somewhere those already have the resistors soldered on.

[Raptor2213]

I am not an electronics guru. I am a merely man with some soldering equipment and a willingness to injure myself and damage sensitive electronic equipment.

So pardon me if this is a silly question...

Am I to take it that by replacing the resistors per this chart we can increase the voltage and thus reach the hash rates beyond the published overclockable speed? (Which would presumably require increasing the ability of the U1 to dissipate heat.)



src: https://github.com/AntMiner/AntGen1/blob/master/manual/AntMiner-U1%20user%20guide.pdf

I am having a hard time believing it would be that easy but the ASICMINER upgrades were (after a huge thread of experiments by many talented individuals https://bitcointalk.org/index.php?topic=241652.0)

The Antminer S2 does not appear to have the same resistor configuration.  Has anyone looked into what would be required to increase the voltage?

Did you meant "U2?
I read somewhere those already have the resistors soldered on.

Yes, the U2.  They do, but they seem to have OC issues at anything higher than 2.4Ghash.  Seems to be a voltage issue.