Blue Fury Support Thread.

[bitterdog]

PENCIL MOD +. Heatsink removed, swapped for a CPU hsf unit. measured at @ 29deg C
R15 = 0.85 (measured at 20k)
HW's = 9.89%
2.732GH/s (5s avg)

You happen to have a picture of this you can post>?

[nicksmith]

PENCIL MOD +. Heatsink removed, swapped for a CPU hsf unit. measured at @ 29deg C
R15 = 0.85 (measured at 20k)
HW's = 9.89%
2.732GH/s (5s avg)

You happen to have a picture of this you can post>?

Also removing the heat sink, its just on there with some sticky tape and you pried it off?

[Taugeran]

alright ladies and gents here is a quick spreadsheet summarizing the approximate voltages achieved with various resistor mods on my BPMC BF1 ( first batch run with green board and full size silver heatsink. WARNING: you can use this at your own risk. i assume no responsibility for any damages incurred. and as always YMMV  

Modsheet


Edit: this sheet includes some odd resistor values that are linear extrapolations between commonly available smd components

[bitterdog]

alright ladies and gents here is a quick spreadsheet summarizing the approximate voltages achieved with various resistor mods on my BPMC BF1 ( first batch run with green board and full size silver heatsink. WARNING: you can use this at your own risk. i assume no responsibility for any damages incurred. and as always YMMV  

Modsheet


Edit: this sheet includes some odd resistor values that are linear extrapolations between commonly available smd components

Thank You

[Taugeran]

alright ladies and gents here is a quick spreadsheet summarizing the approximate voltages achieved with various resistor mods on my BPMC BF1 ( first batch run with green board and full size silver heatsink. WARNING: you can use this at your own risk. i assume no responsibility for any damages incurred. and as always YMMV  

Modsheet


Edit: this sheet includes some odd resistor values that are linear extrapolations between commonly available smd components

Thank You

YW, ill also update it with temps if i get my FLIR cam as a holiday gift from family


UPDATE: i should face palm. i used the incorrect calculation in the Vin column. DONOT use those values yet. updating now

[Taugeran]

spread sheet is fixed. sorry bout that. simultaneously working on this one and one for the nanofury NF1 and got em mixed up 

[jay1111]

If you check your 'devices and printers' do you get the BF to shows as a com:xx device?  Maybe your configuration follows the cgminer one, where these USB miners use WinUSB drivers, rather than (serial) Com port drivers?
If you don't get a Com:xx assignment for your BF, you'll need to reinstall the special BF driver provided in the early part of this thread.
Note that there is a newer version of bfgminer (3.6.0) which supports the BF's, so you may want to try that as well (it seems to handle the HW errors a bit better, and doesn't 'hide' them like the 3.2.0 does).
I've been using a batch file similar to
bfgminer.exe -S bigpic:all -o stratum+tcp://stratum.btcguild.com:3333 -u xxxxx -p xxx


Good luck

Thanks, I eventually got it working, each bit fury showed up as "Bitfury BF1 USB Miner (Com#)" but after i reinstalled the driver, it seems to be working now...

[induktor]

Hello
I just received my 4 units from ssinc, installed one in windows 7 x64, installed the driver, it shows as COM3

but when I try to load BFGMINER (the one supplied in the first post in this thread) i get bfgminer has stopped working.

I tried to run as administrator, run as XP compatibility mode.

I tried:
bfgminer.exe -S "BF1:ALL" -o stratum+tcp://mint.bitminer.com:3333 -u user -p password
bfgminer.exe -S "BF1:all" -o http://mint.bitminter.com:3333 -u user -p pass
bfgminer.exe -S "BF1:all" -o stratum+tcp://stratum.btcguild.com:3333 -u user -p password

also
bfgminer.exe -S "bigpic:all" and -S "BF1:COM3"

nothing worked it still gives the windows error that stopped working, what can I do?
thanks
Indtk

[OleOle]

Hello
I just received my 4 units from ssinc, installed one in windows 7 x64, installed the driver, it shows as COM3

but when I try to load BFGMINER (the one supplied in the first post in this thread) i get bfgminer has stopped working.

I tried to run as administrator, run as XP compatibility mode.

I tried:
bfgminer.exe -S "BF1:ALL" -o stratum+tcp://mint.bitminer.com:3333 -u user -p password
bfgminer.exe -S "BF1:all" -o http://mint.bitminter.com:3333 -u user -p pass
bfgminer.exe -S "BF1:all" -o stratum+tcp://stratum.btcguild.com:3333 -u user -p password

also
bfgminer.exe -S "bigpic:all" and -S "BF1:COM3"

nothing worked it still gives the windows error that stopped working, what can I do?
thanks
Indtk

I use Win7x32 but some other comments earlier on this thread indicate that it will probably work for Win7x64, so try this in the target line after the executable:

-o [your pool and port] -u [your username] -p [your password] -G -S bigpic:all

That's working for me on BFGMiner 3.6.0 and also worked on 3.5.1.

Good luck.

[Atomar]

Any difference in the Heat put out?

When I pencil modded I really had to put a fan on it because they get pretty hot. That cheap ass heat sink they put on the blue fury was A) TOO THICK B) Not Adequate for 2.6 GH/s

No. Sorry. I do not have any temp. messurment at home.
Here at the Office I do have an IR meter. Messurement arround the CPU is +/- 60° .
But the allmost hottest part on the board are the 2 resistors for the LED ;-)

And yes. The heat sink coulb be bether ...

[Atomar]

Hello
I just received my 4 units from ssinc, installed one in windows 7 x64, installed the driver, it shows as COM3

but when I try to load BFGMINER (the one supplied in the first post in this thread) i get bfgminer has stopped working.

I tried to run as administrator, run as XP compatibility mode.

I tried:
bfgminer.exe -S "BF1:ALL" -o stratum+tcp://mint.bitminer.com:3333 -u user -p password
bfgminer.exe -S "BF1:all" -o http://mint.bitminter.com:3333 -u user -p pass
bfgminer.exe -S "BF1:all" -o stratum+tcp://stratum.btcguild.com:3333 -u user -p password

also
bfgminer.exe -S "bigpic:all" and -S "BF1:COM3"

nothing worked it still gives the windows error that stopped working, what can I do?
thanks
Indtk

I'm usind this line to start the bfgminer:

bfgminer.exe -S bf1:all -o stratum+tcp://mint.bitminter.com:3333 -O USERNAME_WORKER --disable-gpu

[Atomar]

Ok. For the record:

Pencil-Mode: 2.618 GH/s
9.1k resistor Mode: 2.655 Gh/s

Both on 1 powered hub with airflow.
RaPi rev. 2 with Minepeon 0.2.4.2 / bfgminer 3.6.0

Similar results to me, can I ask what the speed before mod was?

Arround 2.3 - 2.4 GH/s

[LockUnlock]

Is the recently released Red Fury basically the same thing as the Blue Fury? Just different color?

[kabopar]

If you check your 'devices and printers' do you get the BF to shows as a com:xx device?  Maybe your configuration follows the cgminer one, where these USB miners use WinUSB drivers, rather than (serial) Com port drivers?
If you don't get a Com:xx assignment for your BF, you'll need to reinstall the special BF driver provided in the early part of this thread.
Note that there is a newer version of bfgminer (3.6.0) which supports the BF's, so you may want to try that as well (it seems to handle the HW errors a bit better, and doesn't 'hide' them like the 3.2.0 does).
I've been using a batch file similar to
bfgminer.exe -S bigpic:all -o stratum+tcp://stratum.btcguild.com:3333 -u xxxxx -p xxx


Good luck

Thanks, I eventually got it working, each bit fury showed up as "Bitfury BF1 USB Miner (Com#)" but after i reinstalled the driver, it seems to be working now...

Great...

[kabopar]

Is the recently released Red Fury basically the same thing as the Blue Fury? Just different color?

I got some Red Furies today, and they are basically the 'same thing' as the Blue Fury.  The main difference that I can spot is that the heatsink fins are shorter (which allows these to be placed on 'standard' hubs on adjacent USB sockets, while the Blue Fury unit that I have has longer fins that preclude that).  Also the heatsink is over the full length of the PCB, and is red in color (as is the heatsink).  The performance is similar and the same pencil mod applies to R15.

Cheers

[btmtb]

PENCIL MOD +. Heatsink removed, swapped for a CPU hsf unit. measured at @ 29deg C
R15 = 0.85 (measured at 20k)
HW's = 9.89%
2.732GH/s (5s avg)

You happen to have a picture of this you can post>?

Also removing the heat sink, its just on there with some sticky tape and you pried it off?

I used a snap-off craft knife (for blade lengths) and slid it down the HEATSINK side of the thermal tape, it came away very easily once you broke the initial adhesion. The tape is surprisingly thick so you could safely get a knife down the middle and never even touch the heatsink or the board. I'll edit this post and include my pictures of the HSF mod later tonight.

My experience is that the temperature went from 42deg at heat sink without a fan to about 80deg at heatsink. I would recommend that you DONT pencil mod without increasing the cooling from the stock setup or it's likely you'll dramatically shorten the lifespan.

*** EDIT ***

For those who are interested, this is my CPU HSF mod.



Stock heatsink removed. CPU Heatsink applied with thermal compound. Fixed with screws through PCB fixing holes, between fins of HSF unit to hold securely (but not over tight). Cut yellow and black wires to HSF 4pin connector. Strip these wire ends. Strip a USB micro cable. Safely cap the data wires. Connect the yellow and black HSF wires to the USB 5V power lines, and tape/shrink-tube safely. You now have a usb powered, extraordinarily efficient heat sink fan that runs nearly silently, plug this mutant creation into a usb extension lead, and same with the miner and away you go. The HSF is designed for up 12v feed from the motherboard to run at full speed, so it works fine with usb's 5v, just not as fast as normal, which has the pleasing side effect of being very quiet. Yellow insulation tape was my temporary low-tech way of lowering the brightness of the LED temporarily, it serves no other functional purpose.

[kabopar]

Combined BF and RF (Red Fury) USB miners with pencil mods

Initially had R15 at 0.77K on one of them, but it got too hot to touch, so wound it down to 0.89 which seems more stable (and less HW errors)

[quasicoltd]

High error rate for my taste, but respectable throughput IMO considering the circumstances. I'm most keen to see if I can reduce the HW with sufficient cooling, will report back when I've had time to play properly this weekend.

That's by design in these chips. An error rate around 10% is to be expected. I can make it read zero artificially in cgminer if you like...

Thanks for the fixes in the 3.8.x series. Seeing my 2 RedFurys go from and average 1.8 each to 2.2 each was just jaw-dropping.  I too was one of those people that freaked out about the massive jump in reported HW errors when I upgraded from 3.7.2 Very Reassuring that this is normal.

However, I also seem to have experienced a large increase in reported rejects in cgminer as well, about 11% overall. This combined with a reported 7-10% DOA rate in my local p2pool node also has me wondering if this is just accurately reporting what was already happening and I was living in a dreamworld of .02% DOA before.

[induktor]

Hi
I end up downloading from the bfgminer repository this: bfgminer-3.5.1-win64  and it worked perfectly (in W7 x64)

command:
bfgminer.exe -S "BF1:all" -o stratum+tcp://mint.bitminter.com:3333 -u User_Worker_1 -p 1234

I am getting 2.2 GH on one device and 2.4 on the other.
The 2.4 GH one is a lot hotter than the 2.2GH device.
also the 2.4GH have a blinking red light, and a steady ambar.
the 2.2GH only the blinking red light

odd... hehe.

BTW: I didn't flash any firmware yet, dunno if it is necessary.
thank you all
indtk

[-ck]

Thanks for the fixes in the 3.8.x series. Seeing my 2 RedFurys go from and average 1.8 each to 2.2 each was just jaw-dropping.  I too was one of those people that freaked out about the massive jump in reported HW errors when I upgraded from 3.7.2 Very Reassuring that this is normal.

However, I also seem to have experienced a large increase in reported rejects in cgminer as well, about 11% overall. This combined with a reported 7-10% DOA rate in my local p2pool node also has me wondering if this is just accurately reporting what was already happening and I was living in a dreamworld of .02% DOA before.

That's cause the code was also accidentally throwing out results that were already there to be processed when a work restart hit so it was unintentionally acting like enabling no-submit-stale (note you should never enable this on p2pool).