Can you give me instructions for pencil mod?
...
By the way - if anyone does indeed do some pencil mods - would you please share your observations? Some feedback about what works the best would be more than welcome
Topic: [CLOSED] NanoFury NF1 USB stick - Group Buy + Product Assembly - page 4. (Read 74360 times)
November 14, 2013, 03:58:32 AM
...
By the way - if anyone does indeed do some pencil mods - would you please share your observations? Some feedback about what works the best would be more than welcome
November 13, 2013, 08:12:01 PM
Does the NF1 worth with cgminer as well or only bfgminer?
We started working on a cgminer port but it turned out bfgminer somehow gets 10-15% more GH/s, so we've sort of given up on fiddling with cgminer (at least for now). I can send you what we have if you want - I've gotten it to work on Windows only though. If there are any volunteers that know their way around Linux and can take care of that part - just let me know and I'd be glad to assist.
For a proper port into cgminer it will have to be done by Con Kolivas - but at the moment it is unclear when and if at all that's going to happen. In my last email exchange with him he said "I'm overwhelmed with other code tasks at the time (mostly paid which means things like your code gets low priority I'm afraid)."
And since Luke did such an awesome job with bfgminer I haven't had to bother Con again.
November 13, 2013, 07:32:14 PM
Does the NF1 worth with cgminer as well or only bfgminer?
November 13, 2013, 06:39:42 PM
@vs3
can you elaborate on the req's to pencil mod the NFY similar to the BF1?
if the voltage setup circuits are anything similar: on BF1
rub graphite pencil across R15 until resistance measured across R15 and R16 is ~0.89Kohm
can you elaborate on the req's to pencil mod the NFY similar to the BF1?
if the voltage setup circuits are anything similar: on BF1
rub graphite pencil across R15 until resistance measured across R15 and R16 is ~0.89Kohm
Yes - the procedure is almost identical - I'll repost what I wrote about that (on Marto74's thread) - let me know if that covers it or I missed anything:
Can you give me instructions for pencil mod?
See this image:
R2 and R3 determine the voltage:
V = 0.8*(R2+R3)/R3
(and factory settings are R2=100 and R3=2.4k for about 0.833V)
If you want to increase the voltage you can either increase R2 or decrease R3 (pencil mod decreases resistance).
And to lower the voltage - you can lower R2 or increase R3.
My 2 cents - start by first lowering the voltage. At least that's the safer option
November 13, 2013, 06:34:28 PM
@vs3
can you elaborate on the req's to pencil mod the NFY similar to the BF1?
if the voltage setup circuits are anything similar: on BF1
rub graphite pencil across R15 until resistance measured across R15 and R16 is ~0.89Kohm
can you elaborate on the req's to pencil mod the NFY similar to the BF1?
if the voltage setup circuits are anything similar: on BF1
rub graphite pencil across R15 until resistance measured across R15 and R16 is ~0.89Kohm
November 11, 2013, 08:54:19 PM
I wonder if they're not getting enough power?
Still unclear....
I have rosewill's and dlink hubs
I inserted one into hub:
Code:
[2013-11-10 00:56:43] Pushing work from pool 1 to hash queue
[2013-11-10 00:56:43] mcp2210_get_gpio_input: Failed to get current GPIO input values
[2013-11-10 00:56:43] NFY 0: checkport failed
[2013-11-10 00:56:43] NFY 0 failure, exiting
Still not sure of power... is there a light on these things?
I alternated between all 10 units and plugged them in with no lights on the nano fury's
The clue for me in the log above is the "checkport failed" message.
The program does "checkport" while it first tests all devices and later at regular intervals when it sends work to the device.
The fact that it passed the first "checkport" means that there was good communication with the device at that time and everything worked as expected.
The line immediately preceding the "checkport failed" message says "mcp2210_get_gpio_input: Failed to get current GPIO input values" which indicates that the device is no longer plugged into the USB hub. Or in other words - the host can no longer communicate with the device. As for why did the communication get disrupted - there are many possible answers.
My guess is that the USB host (or hub) did not provide enough power to the device and either the hub got entirely dropped or just the device got dropped.
You can test whether that is the case by plugging the device directly into your PC - all motherboards will provide sufficient power. Note however that Raspberry Pi ports are not USB2.0 compliant - they barely provide enough power for a mouse and a keyboard - so plugging it directly into RPi's USB port may or may not work.
It seems that wherever you had plugged the device was good enough for the initial communication, and had it not been for the power issues everything else worked.
November 11, 2013, 04:40:25 PM
Wouldn't they be the same for all USB miners on the Pi then?
Proof is in testing and what I posted is what we know so far.
I was also an Admin not a programmer but maybe commands are used that the Broadcom BCM2835 used in the Pi as USB host doesn't understand?
November 11, 2013, 04:32:30 PM
Would love to hear from some additional linux users.
We know they work:
On windows with USB2 and USB3 hubs.
On "normal" linux box with USB3 hubs.
We know it doesn't appear to work w/ Pi and USB2 hubs.
November 11, 2013, 04:22:55 PM
Would love to hear from some additional linux users.
We know they work:
On windows with USB2 and USB3 hubs.
On "normal" linux box with USB3 hubs.
We know it doesn't appear to work w/ Pi and USB2 hubs.
November 11, 2013, 04:18:02 PM
Would love to hear from some additional linux users.
We know they work:
On windows with USB2 and USB3 hubs.
On "normal" linux box with USB3 hubs.
We know it doesn't appear to work w/ Pi and USB2 hubs.
November 11, 2013, 03:54:32 PM
Would love to hear from some additional linux users.
We know they work:
On windows with USB2 and USB3 hubs.
On "normal" linux box with USB3 hubs.
We know it doesn't appear to work w/ Pi and USB2 hubs.
November 11, 2013, 12:06:16 AM
Can't see the image at work.
So works on your Dlink, What about your Pi?
Again, sorry if it shows in the pic.
not currently running on a pi. having lockup issues (investigating...)
How is Pi powered? maybe lockup is power feed back from powered hub?
As far as running nanos on Pi, I wonder if it has something to do with the Broadcom BCM2835 chip used.
My hubs are powered but no go on nanos
its not a lockup of the whole device. i can ctrl-c bfgminer to death just fine.
its just bfgminer seems to hang/not produce anymore debug output after a seemingly random number of MCP2210 transfers temporarily rejected....
anyway ttyal must rest
pi was powered through power micro usb and probably some back feed from hubs
November 10, 2013, 11:57:34 PM
Can't see the image at work.
So works on your Dlink, What about your Pi?
Again, sorry if it shows in the pic.
not currently running on a pi. having lockup issues (investigating...)
How is Pi powered? maybe lockup is power feed back from powered hub?
As far as running nanos on Pi, I wonder if it has something to do with the Broadcom BCM2835 chip used.
My hubs are powered but no go on nanos
November 10, 2013, 11:25:35 PM
I'm impressed. You have all the types of USBs working in one miner and on one hub?
2 hubs the top screen was before i set up the second
command used
Code:
bfgminer -c windows.conf -D -S BPM:\\.\COM8 -S BES:\\.\COM9 -S BES:\\.\COM10 -S BES:\\.\COM11 -S BES:\\.\COM12 -S BES:\\.\COM13
though i could condense it into my config file as:
Code:
"scan-serial":
[
"BPM:\\.\COM8",
"BES:\\.\COM9",
"BES:\\.\COM10",
"BES:\\.\COM11",
"BES:\\.\COM12",
"BES:\\.\COM13"
]
and unfortunately ill probably never have a klondike k1 to add to the collection :,< silent tear
November 10, 2013, 11:19:11 PM
I'm impressed. You have all the types of USBs working in one miner and on one hub?
2 hubs the top screen was before i set up the second
November 10, 2013, 11:17:02 PM
If someone that has them working can try them in a USB 2.0 hub that would be awesome.
i have them working in the hub with all the asicminer usbs and the BF1 pictured below. pic is from when i first unwrapped them
snip
Hub is a Dlink DUB-H7 powered usb2.0 hub i can get them going at 56 bits in the charging ports but they don't have enough cooling to be stable so 54 bits for now
screen of 11 hrs of them running on that hub
snip
Can't see the image at work.
So works on your Dlink, What about your Pi?
Again, sorry if it shows in the pic.
not currently running on a pi. having lockup issues (investigating...)
running on windows but should run just fine on a normal linux box
the dlink is running 2 NFY and all my other hardware is on a cheap 7 port hub from MCM electronics (i live just down the street from their warehouse)
in all 5 BES, 1 BPM, 2 NFY totaling ~8-9 ghs depending on the timeframe
goin to try few different linux distros to see fi the lockups are raspian specific
November 10, 2013, 10:25:55 PM
If someone that has them working can try them in a USB 2.0 hub that would be awesome.
i have them working in the hub with all the asicminer usbs and the BF1 pictured below. pic is from when i first unwrapped them
Hub is a Dlink DUB-H7 powered usb2.0 hub i can get them going at 56 bits in the charging ports but they don't have enough cooling to be stable so 54 bits for now
screen of 11 hrs of them running on that hub
Can't see the image at work.
So works on your Dlink, What about your Pi?
Again, sorry if it shows in the pic.
November 10, 2013, 10:02:21 PM
I'm impressed. You have all the types of USBs working in one miner and on one hub?
I have two four port hubs, but yea.
November 10, 2013, 09:59:15 PM
I'm impressed. You have all the types of USBs working in one miner and on one hub?
November 10, 2013, 09:16:26 PM
If someone that has them working can try them in a USB 2.0 hub that would be awesome.
i have them working in the hub with all the asicminer usbs and the BF1 pictured below. pic is from when i first unwrapped them
Hub is a Dlink DUB-H7 powered usb2.0 hub i can get them going at 56 bits in the charging ports but they don't have enough cooling to be stable so 54 bits for now
screen of 11 hrs of them running on that hub
Jump to: