[GUIDE] BitFury Miner Support/Tuning

[KNK]

Upon inspecting the card
https://www.dropbox.com/s/sg34hw5u2v4y8rj/2014-01-27%2007.12.17.jpg
Look at the pins above C441 and C44E
The pins are shorted (or whatever the terms for that is)
Upon further inspecting the card at last 4 chips have solder across 2 pads like that.

Those pads are power (VDD) and are internally shorted anyway, so that's not the problem.
If (for the same chip U44) the pins on the left (with no caps) are shorted, then that may be a problem.

[zurg]

Upon inspecting the card
https://www.dropbox.com/s/sg34hw5u2v4y8rj/2014-01-27%2007.12.17.jpg
Look at the pins above C441 and C44E
The pins are shorted (or whatever the terms for that is)
Upon further inspecting the card at last 4 chips have solder across 2 pads like that.

Those pads are power (VDD) and are internally shorted anyway, so that's not the problem.
If (for the same chip U44) the pins on the left (with no caps) are shorted, then that may be a problem.

So when I look at the board with chips facing me and the writing is upright, the flow is going from the left to the right side and those cant be shorted, right?

https://www.dropbox.com/s/amopg7o7ec362ii/2014-01-27%2007.40.52.jpg
(Where the arrows are pointing)
https://www.dropbox.com/s/gy5rrxlk1dv4pci/2014-01-27%2007.41.25.jpg

[KNK]

All of those are power.
Look at each chip with the white circle (on the board near the chip corner - that's pin 1 mark) at top right.

[tom99]

Upon inspecting the card
https://www.dropbox.com/s/sg34hw5u2v4y8rj/2014-01-27%2007.12.17.jpg
Look at the pins above C441 and C44E
The pins are shorted (or whatever the terms for that is)
Upon further inspecting the card at last 4 chips have solder across 2 pads like that.

Those pads are power (VDD) and are internally shorted anyway, so that's not the problem.
If (for the same chip U44) the pins on the left (with no caps) are shorted, then that may be a problem.

So when I look at the board with chips facing me and the writing is upright, the flow is going from the left to the right side and those cant be shorted, right?

https://www.dropbox.com/s/amopg7o7ec362ii/2014-01-27%2007.40.52.jpg
(Where the arrows are pointing)
https://www.dropbox.com/s/gy5rrxlk1dv4pci/2014-01-27%2007.41.25.jpg

 dont worry all of them connect to grd.


edit: all start with Cxxx same and you can check with meter.

[tom99]

Ok guys, I need some assistance

I had an August kit with 2 boards (~47Gh/s)
so the M-Board is V1.0
on Friday I got 10 more boards and placed them 1st in-line then added my 2 pre-existing boards with last board being EOL one.
All this was powered by a 1Kw PS that died on me Sunday evening.
So I had it running about 48 hours.
All the new boards are stock = not overclocked.

Now I placed a spare 430W PS on the BF rig and it runs, however I can only see 96 chips running = 1st 6 cards.

Any ideas why it doesn't see the rest?

Thanks

  you can do test on pcb that mboard did not see if it's working ok by itself.

[zurg]

Ok guys, I need some assistance

I had an August kit with 2 boards (~47Gh/s)
so the M-Board is V1.0
on Friday I got 10 more boards and placed them 1st in-line then added my 2 pre-existing boards with last board being EOL one.
All this was powered by a 1Kw PS that died on me Sunday evening.
So I had it running about 48 hours.
All the new boards are stock = not overclocked.

Now I placed a spare 430W PS on the BF rig and it runs, however I can only see 96 chips running = 1st 6 cards.

Any ideas why it doesn't see the rest?

Thanks

  you can do test on pcb that mboard did not see if it's working ok by itself.

Looks like the card ain't working RMA time

[CCCrypto]

Ok guys, I need some assistance

I had an August kit with 2 boards (~47Gh/s)
so the M-Board is V1.0
on Friday I got 10 more boards and placed them 1st in-line then added my 2 pre-existing boards with last board being EOL one.
All this was powered by a 1Kw PS that died on me Sunday evening.
So I had it running about 48 hours.
All the new boards are stock = not overclocked.

Now I placed a spare 430W PS on the BF rig and it runs, however I can only see 96 chips running = 1st 6 cards.

Any ideas why it doesn't see the rest?

Thanks

  you can do test on pcb that mboard did not see if it's working ok by itself.

Looks like the card ain't working RMA time

What test did you do to determine the card wasn't working?

[zurg]

Ok guys, I need some assistance

I had an August kit with 2 boards (~47Gh/s)
so the M-Board is V1.0
on Friday I got 10 more boards and placed them 1st in-line then added my 2 pre-existing boards with last board being EOL one.
All this was powered by a 1Kw PS that died on me Sunday evening.
So I had it running about 48 hours.
All the new boards are stock = not overclocked.

Now I placed a spare 430W PS on the BF rig and it runs, however I can only see 96 chips running = 1st 6 cards.

Any ideas why it doesn't see the rest?

Thanks

  you can do test on pcb that mboard did not see if it's working ok by itself.

Looks likew the card ain't working RMA time

What test did you do to determine the card wasn't working?

I tried the card just by itself and actually changed it's position in the chain.
Didn't matter where I placed it, as any card after it was not showing at all. (including this card)

[CCCrypto]

I tried the card just by itself and actually changed it's position in the chain.
Didn't matter where I placed it, as any card after it was not showing at all. (including this card)

Okay, thanks. That will be helpful to remember for future troubleshooting.

[cypherdoc]

ok, now it's happening to me; again.

absolutely no hashing with one of my 14 card rigs.  changed out SD cards and raspi.  can login into admin page but no hashing.  took rig apart and checked for any burnt boards (none) and blew out M board connector in case of excess oxidation (none apparent).

what were all the suggestions given for this frequent problem?

[Keefe]

ok, now it's happening to me; again.

absolutely no hashing with one of my 14 card rigs.  changed out SD cards and raspi.  can login into admin page but no hashing.  took rig apart and checked for any burnt boards (none) and blew out M board connector in case of excess oxidation (none apparent).

what were all the suggestions given for this frequent problem?

I've said it many times before, so I'll let you scroll up for details, but you need to check the output of chainminer and the proxies.

[cypherdoc]

ok, now it's happening to me; again.

absolutely no hashing with one of my 14 card rigs.  changed out SD cards and raspi.  can login into admin page but no hashing.  took rig apart and checked for any burnt boards (none) and blew out M board connector in case of excess oxidation (none apparent).

what were all the suggestions given for this frequent problem?

I've said it many times before, so I'll let you scroll up for details, but you need to check the output of chainminer and the proxies.

i only have 2 screens:

Code:

pi@192-168-7-243:~$ sudo screen -r
There are several suitable screens on:
	2069..192-168-7-243	(28/01/14 23:39:27)	(Detached)
	2063..192-168-7-243	(28/01/14 23:39:27)	(Detached)

here's some of the output of 2069:

Code:

JOB 145 PROCESSED 0.005 sec [nonces:0 errors:368 spi:48 miso:67 dup:0] (queue:0)  
JOB 146 PROCESSED 0.006 sec [nonces:0 errors:366 spi:51 miso:67 dup:0] (queue:0)  
JOB 147 PROCESSED 0.011 sec [nonces:0 errors:427 spi:50 miso:67 dup:0] (queue:0)  
JOB 148 PROCESSED 0.003 sec [nonces:0 errors:276 spi:34 miso:68 dup:0] (queue:0)  
JOB 149 PROCESSED 0.002 sec [nonces:0 errors:121 spi:27 miso:56 dup:0] (queue:0)  
JOB 150 PROCESSED 0.005 sec [nonces:0 errors:357 spi:42 miso:65 dup:0] (queue:0)  
JOB 151 PROCESSED 0.007 sec [nonces:0 errors:484 spi:33 miso:64 dup:0] (queue:0)  
JOB 152 PROCESSED 0.005 sec [nonces:0 errors:394 spi:21 miso:68 dup:0] (queue:0)  
JOB 153 PROCESSED 0.007 sec [nonces:0 errors:548 spi:55 miso:61 dup:2] (queue:0)  
JOB 154 PROCESSED 0.005 sec [nonces:0 errors:426 spi:57 miso:60 dup:0] (queue:0)  
JOB 155 PROCESSED 0.006 sec [nonces:0 errors:524 spi:40 miso:68 dup:2] (queue:0)  
JOB 156 PROCESSED 0.005 sec [nonces:0 errors:382 spi:52 miso:65 dup:0] (queue:0)  
JOB 157 PROCESSED 0.006 sec [nonces:0 errors:373 spi:41 miso:63 dup:0] (queue:0)  
JOB 158 PROCESSED 0.004 sec [nonces:0 errors:357 spi:11 miso:58 dup:0] (queue:0)  
JOB 159 PROCESSED 0.006 sec [nonces:0 errors:420 spi:56 miso:64 dup:0] (queue:0)

and some of 2063:

Code:

2014-01-28 23:43:18,070 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:19,099 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:20,026 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:21,051 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:22,077 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:23,104 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:24,031 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:25,057 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:26,084 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:27,109 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:28,035 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:29,061 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:30,088 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:44:06,631 DEBUG protocol protocol.lineReceived # > {u'params': [u'37f1', u'88e467b50f54143fe5be3226344ec74608e47c53696c43740000000000000000', u'01000000010000000000000000000000000000000000000000000000000000000000000000ffffffff2703495104062f503253482f04eb40e85208', u'0d2f7374726174756d506f6f6c2f00000000015d1fe595000000001976a9141157b2fc397d2cf3161313d220017d0dba368db588ac00000000', [u'3f23b91dd29e182079db07b2ba415833c38118042b59ce5b212fca23d27092de', u'66dd511b2faf371bb39c8d93932d028117a97d5295a969757ce7a18bacfd5773', u'9d880202be148bc6c31caddc7562a36ea29676a3fde0200926fd39eae9a2f578', u'0c6d9ee7f9d8854354660349dd09689810db94e36e83d266f767f55267b02642', u'60797fb71da5a528d5244dec03eef24a83006f15eb4f82077de8d048d4c198ac', u'9fd0193634d5b229f850792b0fdebabfc359b5818e01710bed9174002f62a8ca', u'a07dadf4444bda448cc6db75f4060857cae3e0cf04b3bf465876caeb77a88844', u'cccc0a87384a359e2e019f4cb5857c7436c4f599834fb9bd245ddc7f834755b2', u'b8017af8253690aba7b706cfbaa635cc30d4e12736ff8a4518173d713470a8b2', u'bdb8e2711c4a3755bbb36127598da1af92e671ad568928273bbd8177a24fc398'], u'00000002', u'1901f52c', u'52e840e3', False], u'id': None, u'method': u'mining.notify'}
2014-01-28 23:44:06,634 INFO proxy client_service.handle_event # New job 37f1 for prevhash 88e467b5, clean_jobs=False

[daddyfatsax]

[2014-01-29 02:51:56] BSB 1cj: 8 of the last 9 results were bad, reinitializing

Can anyone tell me how to disable that chip in bfgminer? I have tried using 53, 54, etc. and I keep getting the above message. Other than that everything is fine. Thanks!

[Keefe]

ok, now it's happening to me; again.

absolutely no hashing with one of my 14 card rigs.  changed out SD cards and raspi.  can login into admin page but no hashing.  took rig apart and checked for any burnt boards (none) and blew out M board connector in case of excess oxidation (none apparent).

what were all the suggestions given for this frequent problem?

I've said it many times before, so I'll let you scroll up for details, but you need to check the output of chainminer and the proxies.

i only have 2 screens:

Code:

pi@192-168-7-243:~$ sudo screen -r
There are several suitable screens on:
	2069..192-168-7-243	(28/01/14 23:39:27)	(Detached)
	2063..192-168-7-243	(28/01/14 23:39:27)	(Detached)

here's some of the output of 2069:

Code:

JOB 145 PROCESSED 0.005 sec [nonces:0 errors:368 spi:48 miso:67 dup:0] (queue:0)  
JOB 146 PROCESSED 0.006 sec [nonces:0 errors:366 spi:51 miso:67 dup:0] (queue:0)  
JOB 147 PROCESSED 0.011 sec [nonces:0 errors:427 spi:50 miso:67 dup:0] (queue:0)  
JOB 148 PROCESSED 0.003 sec [nonces:0 errors:276 spi:34 miso:68 dup:0] (queue:0)  
JOB 149 PROCESSED 0.002 sec [nonces:0 errors:121 spi:27 miso:56 dup:0] (queue:0)  
JOB 150 PROCESSED 0.005 sec [nonces:0 errors:357 spi:42 miso:65 dup:0] (queue:0)  
JOB 151 PROCESSED 0.007 sec [nonces:0 errors:484 spi:33 miso:64 dup:0] (queue:0)  
JOB 152 PROCESSED 0.005 sec [nonces:0 errors:394 spi:21 miso:68 dup:0] (queue:0)  
JOB 153 PROCESSED 0.007 sec [nonces:0 errors:548 spi:55 miso:61 dup:2] (queue:0)  
JOB 154 PROCESSED 0.005 sec [nonces:0 errors:426 spi:57 miso:60 dup:0] (queue:0)  
JOB 155 PROCESSED 0.006 sec [nonces:0 errors:524 spi:40 miso:68 dup:2] (queue:0)  
JOB 156 PROCESSED 0.005 sec [nonces:0 errors:382 spi:52 miso:65 dup:0] (queue:0)  
JOB 157 PROCESSED 0.006 sec [nonces:0 errors:373 spi:41 miso:63 dup:0] (queue:0)  
JOB 158 PROCESSED 0.004 sec [nonces:0 errors:357 spi:11 miso:58 dup:0] (queue:0)  
JOB 159 PROCESSED 0.006 sec [nonces:0 errors:420 spi:56 miso:64 dup:0] (queue:0)

and some of 2063:

Code:

2014-01-28 23:43:18,070 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:19,099 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:20,026 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:21,051 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:22,077 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:23,104 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:24,031 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:25,057 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:26,084 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:27,109 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:28,035 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:29,061 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:43:30,088 INFO proxy getwork_listener._on_authorized # Worker '243' asks for new work
2014-01-28 23:44:06,631 DEBUG protocol protocol.lineReceived # > {u'params': [u'37f1', u'88e467b50f54143fe5be3226344ec74608e47c53696c43740000000000000000', u'01000000010000000000000000000000000000000000000000000000000000000000000000ffffffff2703495104062f503253482f04eb40e85208', u'0d2f7374726174756d506f6f6c2f00000000015d1fe595000000001976a9141157b2fc397d2cf3161313d220017d0dba368db588ac00000000', [u'3f23b91dd29e182079db07b2ba415833c38118042b59ce5b212fca23d27092de', u'66dd511b2faf371bb39c8d93932d028117a97d5295a969757ce7a18bacfd5773', u'9d880202be148bc6c31caddc7562a36ea29676a3fde0200926fd39eae9a2f578', u'0c6d9ee7f9d8854354660349dd09689810db94e36e83d266f767f55267b02642', u'60797fb71da5a528d5244dec03eef24a83006f15eb4f82077de8d048d4c198ac', u'9fd0193634d5b229f850792b0fdebabfc359b5818e01710bed9174002f62a8ca', u'a07dadf4444bda448cc6db75f4060857cae3e0cf04b3bf465876caeb77a88844', u'cccc0a87384a359e2e019f4cb5857c7436c4f599834fb9bd245ddc7f834755b2', u'b8017af8253690aba7b706cfbaa635cc30d4e12736ff8a4518173d713470a8b2', u'bdb8e2711c4a3755bbb36127598da1af92e671ad568928273bbd8177a24fc398'], u'00000002', u'1901f52c', u'52e840e3', False], u'id': None, u'method': u'mining.notify'}
2014-01-28 23:44:06,634 INFO proxy client_service.handle_event # New job 37f1 for prevhash 88e467b5, clean_jobs=False

Hmm, that hasn't happened to me yet, so I don't know what's the problem. It seems the chips are getting work but are returning tons of bad nonces and other errors. Is this a multi-bank M-board? My first thought is a bad connection, but it's too coincidental for all banks to do that at once. Maybe one of the voltage regulators on the M-board went bad? One should be ~1.8v and the other ~5v. The 5v just powers the rPi, so must be ok. But I can imagine bad 1.8v for chip comm causing weird results like this.

[matt4054]

Hey guys,

I'm sorry if this has been asked a thousand times already, but I have RTFM and OP, and I see nothing about cooling.

My H-cards came with heatsinks. I don't want to OC the chips at this stage, and my ambient temp is around 20°C.

Can I start hashing without active cooling (i.e. fans), just with the heatsinks? The photo in OP doesn't even have heatsinks.
Or is it so obvious that you shouldn't run these without fans, that it's not written anywhere?

Thanks guys! I sold off my CEX.io GHS to buy these, it's just cheaper for more hashrate. Yay!

[bobcaticus]

Hey guys,

I'm sorry if this has been asked a thousand times already, but I have RTFM and OP, and I see nothing about cooling.

My H-cards came with heatsinks. I don't want to OC the chips at this stage, and my ambient temp is around 20°C.

Can I start hashing without active cooling (i.e. fans), just with the heatsinks? The photo in OP doesn't even have heatsinks.
Or is it so obvious that you shouldn't run these without fans, that it's not written anywhere?

Thanks guys! I sold off my CEX.io GHS to buy these, it's just cheaper for more hashrate. Yay!

You'll need to aim a fan at the rig.  Chips run hot and need blown air.

[matt4054]

You'll need to aim a fan at the rig.  Chips run hot and need blown air.

Thanks for your answer!

As I didn't have any spare fan left, I used the one of my BE Blades to create an air tunnel. Air blows to BF first, then the Blades ;-)

It's happily hashing now. I guess there's no thermal probe onboard as I don't see any temp reading in the web interface, right? So I guess I should just do as for the BE Blades, stick an external probe on one of the boards

[zurg]

Well, I am at a loss of whats going on with my setup.
Got replacement card and the rig been mining about 12 hours.
Now I get notification of hashrate drop. I login to see a few of the cards at Zero.
Reboot Pi and only first 3 cards are hashing.
My room is at a constant 60F with fans blowing over the cards.

[Keefe]

Well, I am at a loss of whats going on with my setup.
Got replacement card and the rig been mining about 12 hours.
Now I get notification of hashrate drop. I login to see a few of the cards at Zero.
Reboot Pi and only first 3 cards are hashing.
My room is at a constant 60F with fans blowing over the cards.

Set all chip speeds to something slow like 50 with autotune off, and restart chainmminer. If the problem persists, reseat the cards.

[zurg]

Well, I am at a loss of whats going on with my setup.
Got replacement card and the rig been mining about 12 hours.
Now I get notification of hashrate drop. I login to see a few of the cards at Zero.
Reboot Pi and only first 3 cards are hashing.
My room is at a constant 60F with fans blowing over the cards.

Set all chip speeds to something slow like 50 with autotune off, and restart chainmminer. If the problem persists, reseat the cards.

Just got to work and shut the miner off.
I moved it and realigned the fans etc...to be sure that none of the cards are vibrating.
All the boards started up...
Guess I'll have to keep a close eye on them and have my laser digital thermometer at the ready.

Now praying for this 1 day 14 hours block to be finally solved. lol