Topic: Bitmain AntMiner U1 Tips & Tricks - page 9. (Read 106330 times)

use two dashes for set-device.  it should look like --set-device antminer:clock=x0981

im getting  unexpected command argument when runing  .bat file, what am i doing wrong

bfgminer.exe –set-device antminer:clock=x0981 -o stratum+tcp//us1.ghash.io:3333 -u 1122.ghash -p 123 -S antminer:all

bfgminer.exe –set-device antminer:clock=x0981 -o us1.ghash.io:3333 -u 1122.ghash -p 123 -S antminer:all

i uses these so far and btcguild, i dont think its a url  ,user ,password issue.

Thank you, do not know what the problem is. I even downloaded goat and recreated your settings, worked fine on cgminer (modified for that of course) and still no luck on bfgminer. I am just using p2p or pools for now with said coins, since the hashrate loss on cgminer isn't worth the solo mining gains (altho since I set it up I put goat into my multiminer mix now ^^ )  Have a great day

Sure:
.bat
.conf

hi, would you mind terribly posting your .bat file for bfgminer and your .conf for goatcoin? I am having trouble getting solo mining running with bfgminer (no problem on cgminer) and would like very much to compare my settings to someone who has solo mining running on bfgminer. thank you.

I "sang too early", my two ants work at 5081 but in few hours  one of them every hour delays the response of shares reaching 0.6 Gh really (the third number).
So only one is able to work till 5081 (after that value many HW errors) and my actual configuration is 4F81.

Me too, on my hub i jump the 5V current limitations soldering a wire on it.

http://i58.tinypic.com/2ngg35t.jpg

Changed also the R2 to 4.7K and now I can use 5081 without HW errors, i put also two fans.

Hi, yes, i am mining solo.
Since it is altcoin (goatcoin) it is still possible.
While mining bitcoins on btcguild pool i get something about 9,5 GHS. But other pools can give  much better results.
After running miners for about a hour the hashrate stabilizes around 2,6-2,6 GHS


Did you also  changed the resitors on your miners?

Sometimes the soldering joint is not good, maybe try to resolder  R's on those 2.0 GHS Antminer

chromosoma, you're connected to 127.0.0.1, is that somekind of testmode or are you mining solo?
4C81 is 2.6Gh/s, do the Ants keep that 2.7-2.8Gh/s speed or do they fall to 2.6G in the long run or when connected to a pool/server?
Normally the hashrate is in par or slightly under what the clockspeed is set to, probably because of the workload & network speed.

I have 4 Ants now volt (5v from 30A PSU) & cap modded and stuck to an XBox360 GPU cooler. Three of them can run together at 2.675Gh/s max. with clock=x3506, one can't pass 2.0Gh/s no matter what I do, HW errors will go through the roof...

Success! =)
After  moding my PCI card by   connecting the power directly to the 5V input i managed to get very good results:)
Of course   now i dont  have any  USB protection, so if anything happens, i rely on the internal fuse in PSU:)


 Now i am running  all 4 miners at 4C81 Configuration.
On higher   frequency i get HW.  Maybe if i replace 2,2K resistor with  higher value i could run it at 4GHS:)
But since I dont want to push them to the limits....:9
I will observe them at this  configuration for the next few days:)
The temperatures are now  a bit higher,  like 5°C more then  values in previous post.

Hi,  i measured the Temperature on my 2K2&10K Modes @ 2GHS

Main ASIC: 45 °C
Coil: 38 °C  
DC/DC  converter: 45 °C
Ambient temperature : 22 °C
Fan Speed at 45%

---------------------------------------
 I don't use a hub, i am using USB3.0 PCI  extension  card, which is powered directly through my  PSU.My PSU can provide  up to 25 Amp  on 5V rail.
 Still i  also have voltage drop on my miners to around 4,2-4,1 V if i   connect more then 2 Miners at once. Which is not good.The funy thing is, that  input voltage on my PCI card  remains stable at 5,2V. So the problem is on  PCI Card itself.
I am also thinking  of connecting 5V rail directly to the USB  PORT, avoiding   everything on  the board.
Maybe i will do it today.
Surely it will void all the  safety  like polyfuses and overvoltage protection, but  i am engineer,i am allowed:)))

Do you know how is the circuit of your hub ?
One of mine had 9V 2A p.supply but internally for have 5V use one 7805 that is able to give .5A or 1.5A with a good dissipator, with this is impossible to have more than 1.5A totally.
I bought a new one, in the photo, with 5V 4A ps, without internal regulator, means that the current is regulated only by the usb interface (1A) that i'll try to jump over today.

Yes GREAT in any sense!!

Did you measure the coil, asic and AOZ1021A temperature ?
Are they hot ?
which resistance value do you use for R1 (2.49 K)?

I have only two ant miner actually.
My Hub voltage output is limited to 1A (written on the box) and power supply is 4A for 4 ports.
The usb voltage drop to 4.52, means that ant drain more than 1A, i didn't measure the current.
The next step will be to open the hub and cut + and - wire (the external one of the usb connector) and connect them directly to the main wires of the power supply without the limitation of the chip.
Doing this you can use all the power without limitations and the maximum number of ant become  3 for this hub.

Very nice! I'm still waiting on my other two then I'll upload a shot or two.
Also, can I use any old capacitor on these? I have a bag of 2200µF 25V ones I'm not using, or would that be overkill and possibly fry things?

But my cooling solution is better:))

 Active cooling with  PC controlled Fan:)

Indeed low temp increase stability and performance:)

Hm, interesting. Did you try to run more Antimners on one hub?
I can also achieve about 2,7 GHS  with two mines on one USB Hub, but when i add more they just stop working, or work at much lower ghs rate.

After reading all this thread and learn, now I have 2 AntMiners working at 3.0 Gh.
With out of the box AntMiner, step by step i change the clock parameter to 0881 (1.8Gh), after that value i collect many HW errors, so i add R2 10K and change R1 with 3.3K (dismounted from a cd drive board).
Now all the errors disappear but the ant miner became really hot, i mount it on a dissipator (2.9 °C/W) dismounting the original one leaving the rubber and smearing thermal silicone grease, the coil also was hot (50°C).

http://i58.tinypic.com/34fbjvo.jpg

Now the ant miner works without errors till 0A81 (2.2 Gh).
Pushing up the frequency the coil and the asic increase their heat.
At this point i decide to dissipate also this heat. I cut a piece of aluminium and another squared piece for the asic (8x8 mm and 2mm thick) for equilibrate the different heights of coil and chip.

http://i57.tinypic.com/zvg2d.jpg

Remember to put always thermal silicone grease.

http://i58.tinypic.com/wrc2mw.jpg

Now i try to increase the frequency till 4E81 (3.0 Gh) without HW errors.

http://i62.tinypic.com/359aelc.jpg

After all i discover that also the power chip is hot (70 °C), so decide to put a little piece of aluminium with grease on top of it glued (cyanoacrylate) on the aluminium.

http://i62.tinypic.com/nv4ino.jpg

I use an UsbHub with 4A power supply limited at 1A for each port, but the real voltage on usb ports is 4.52 Volts.

The software is BFGMINER 3.10.0 and the OS is Open Suse.
I insert also inside the conf the icarus-timing with the right Hs value extracted launching miner.php on the web server (Apache 2).

Now for me is enough. Thanks all.

My bfgminer.conf


http://i62.tinypic.com/30w3z48.jpg

I'm 99.9% certain it was a typo in the original bfgminer documentation, as I also had this problem at first.

Yeah, at about 2.2 they get all wonky unless you add a 10K resistor (0.88V). 2.0 should be stable though. You could even add the 2.49K for a full volt but I wouldn't recommend going any higher because the wattage will blow the aoz1021ai.