Pages:
Author

Topic: The Chili – 30+GH/s BFL based Bitcoin Miner Assembly - page 25. (Read 137986 times)

legendary
Activity: 1274
Merit: 1004
I am working on some firmware updates that will fix most (or all?) of the problems the boards are having.  For example, one failure I think I have figured out is occasionally a board will jump to more than 100 GH/s but have 100% hardware errors.  I have one board that does this once per day or so and needs to be rebooted.  After chasing this for a while, I believe I have finally figured out what is going on and will have a fix in the next release.

I have a semi-dead board I recently bought from Keefe that does exactly this. It has always been prone to freezing. I received it from him last Friday, then went out of town overnight. On Saturday it was hung hard, and no amount of restarting/cooling would help it. It has been thoroughly inspected and has no visible flaws. There was an apparent short 3v3->ground, but I found some TIM under the edge of an ASIC, so overnight cleaning in an ultrasound bath should have removed all debris. (The TIM had an electrical conductivity I could measure with my ohmmeter.)

Now, it starts seemingly normally, then reports a ridiculously high speed, 100% error rate for 10 proofs, and stops. BFGminer reports it as sick, 27C. I don't know how to tell what firmware it has, etc.

Any suggestions welcomed.
Keefe had batch 1 boards (-0XXX serial numbers) so it's likely an older version. It couldn't hurt to update.
newbie
Activity: 16
Merit: 0
I am working on some firmware updates that will fix most (or all?) of the problems the boards are having.  For example, one failure I think I have figured out is occasionally a board will jump to more than 100 GH/s but have 100% hardware errors.  I have one board that does this once per day or so and needs to be rebooted.  After chasing this for a while, I believe I have finally figured out what is going on and will have a fix in the next release.

I have a semi-dead board I recently bought from Keefe that does exactly this. It has always been prone to freezing. I received it from him last Friday, then went out of town overnight. On Saturday it was hung hard, and no amount of restarting/cooling would help it. It has been thoroughly inspected and has no visible flaws. There was an apparent short 3v3->ground, but I found some TIM under the edge of an ASIC, so overnight cleaning in an ultrasound bath should have removed all debris. (The TIM had an electrical conductivity I could measure with my ohmmeter.)

Now, it starts seemingly normally, then reports a ridiculously high speed, 100% error rate for 10 proofs, and stops. BFGminer reports it as sick, 27C. I don't know how to tell what firmware it has, etc.

Any suggestions welcomed.
legendary
Activity: 3164
Merit: 2258
I fix broken miners. And make holes in teeth :-)
Quick heat question: My Chili is expected tomorrow, but my heat sink is not coming till Tuesday. Would it be suicide to put an older IBM X series big copper heat sink on top of it and try it out? Will the chips burn out, or take damage?

C
They should throttle down if contact is crappy, but I still wouldn't. Up to you though.
And it looks like my NewEgg Cooler-master 120 just arrived this morning, so I should have something to play with today.

This will be fun.

C
legendary
Activity: 1274
Merit: 1004
Got the Chili + Evo 212 cooler. But I cant seem to make the cooler fit.....did anyone make a Youtube video on how to Mount ? Or pics ?

Lots of posts early in this thread

Edit:-   https://bitcointalksearch.org/topic/m.3355048

The feet do not work(holes in the board are too small)

Didnt notice the screws didnt fit Sad

What to do to get it running today when in Denmark......
There's a couple mods you can do to use the included hardware and the longer standoffs. Remove the screws from the cross brace and toss a couple #10 nuts (or some other space) on them, then put the spring and retaining clip back on.



The less annoying and higher performance method (the backplate is in the better orientation) is to use some M4x50mm screws to hold it in place though.

sr. member
Activity: 476
Merit: 250

Getting 33 to 35 ghash per card on low fan.  37+ on high.

H80 waterblocks

Cases were about $5.00/piece in raw material ( I have my own laser cutter) and the coolers were refurbished so I picked them up for a song.  Power supplies are just 500 watts so only 2 cards per supply. But they were pretty cheap too.

The slowest par of the setup was the cases took an hour each to burn on the cutter. It was a nice weekend project.

Nice job Hands
newbie
Activity: 24
Merit: 0
Got the Chili + Evo 212 cooler. But I cant seem to make the cooler fit.....did anyone make a Youtube video on how to Mount ? Or pics ?

Lots of posts early in this thread

Edit:-   https://bitcointalksearch.org/topic/m.3355048

The feet do not work(holes in the board are too small)

Didnt notice the screws didnt fit Sad

What to do to get it running today when in Denmark......
hero member
Activity: 868
Merit: 1000
Got the Chili + Evo 212 cooler. But I cant seem to make the cooler fit.....did anyone make a Youtube video on how to Mount ? Or pics ?

Lots of posts early in this thread

Edit:-   https://bitcointalksearch.org/topic/m.3355048

The feet do not work(holes in the board are too small)
newbie
Activity: 24
Merit: 0
Got the Chili + Evo 212 cooler. But I cant seem to make the cooler fit.....did anyone make a Youtube video on how to Mount ? Or pics ?
member
Activity: 67
Merit: 10
Holy crap Hands.

You love those chilis.  What model is that corsair water block?  What hashrate are you getting?

Getting 33 to 35 ghash per card on low fan.  37+ on high.

H80 waterblocks

Cases were about $5.00/piece in raw material ( I have my own laser cutter) and the coolers were refurbished so I picked them up for a song.  Power supplies are just 500 watts so only 2 cards per supply. But they were pretty cheap too.

The slowest par of the setup was the cases took an hour each to burn on the cutter. It was a nice weekend project.
donator
Activity: 686
Merit: 519
It's for the children!
I'm debating selling my Chili farm.  10 Chili's at 340Ghash total.
donator
Activity: 686
Merit: 519
It's for the children!
Holy crap Hands.

You love those chilis.  What model is that corsair water block?  What hashrate are you getting?
member
Activity: 67
Merit: 10
So the only room I could find in the house right this second is our "wreck" room so ignore the mess and dust on the desk... But I got 9 of my 10 cards going in there Quesos




Front


Back



The mosfet heatsinks I am using are kind of crappy so the extra 120mm fan on them helps ALOT.. going to have to upgrade that..

Running the H80 water coolers at their lowest settings still sees 33GH per card with some cards as high as 37GH.

Once I have more space I can lay them out a bit better... I think right now there is a fair bit of hot air co-mingling rather than nice fresh air.
legendary
Activity: 1274
Merit: 1004
I found ZcX which adds the internal chip temperatures behind the processor output:

PROCESSOR 0: 9 engines @ 137 MHz -- MAP: E07E 27 C
PROCESSOR 1: 14 engines @ 198 MHz -- MAP: DFF7 26 C
PROCESSOR 2: 7 engines @ 188 MHz -- MAP: 0BF0 27 C
PROCESSOR 3: 15 engines @ 201 MHz -- MAP: FF7F 29 C
PROCESSOR 4: 13 engines @ 159 MHz -- MAP: E7FE 27 C
PROCESSOR 5: 8 engines @ 168 MHz -- MAP: 0FF0 27 C
PROCESSOR 6: 8 engines @ 179 MHz -- MAP: 0FF0 26 C
PROCESSOR 7: 15 engines @ 179 MHz -- MAP: F7FF 32 C


are these commandline flags?
No, you can use Z-Commands by connecting a terminal program to the serial port.
I tried to use minicom but it never works for me so I used tail -f /dev/ttyUSB0 on one console and echo -n ZcX >/dev/ttyUSB0 on another console to get the result on my linux box.

I found only working 5 commands: ZCX, ZcX, ZLX, ZlX and ZGX. Maybe the're more available...
That's pretty much it.
legendary
Activity: 1274
Merit: 1004
Quick heat question: My Chili is expected tomorrow, but my heat sink is not coming till Tuesday. Would it be suicide to put an older IBM X series big copper heat sink on top of it and try it out? Will the chips burn out, or take damage?

C
They should throttle down if contact is crappy, but I still wouldn't. Up to you though.
legendary
Activity: 3164
Merit: 2258
I fix broken miners. And make holes in teeth :-)
Quick heat question: My Chili is expected tomorrow, but my heat sink is not coming till Tuesday. Would it be suicide to put an older IBM X series big copper heat sink on top of it and try it out? Will the chips burn out, or take damage?

C
full member
Activity: 128
Merit: 100
I found ZcX which adds the internal chip temperatures behind the processor output:

PROCESSOR 0: 9 engines @ 137 MHz -- MAP: E07E 27 C
PROCESSOR 1: 14 engines @ 198 MHz -- MAP: DFF7 26 C
PROCESSOR 2: 7 engines @ 188 MHz -- MAP: 0BF0 27 C
PROCESSOR 3: 15 engines @ 201 MHz -- MAP: FF7F 29 C
PROCESSOR 4: 13 engines @ 159 MHz -- MAP: E7FE 27 C
PROCESSOR 5: 8 engines @ 168 MHz -- MAP: 0FF0 27 C
PROCESSOR 6: 8 engines @ 179 MHz -- MAP: 0FF0 26 C
PROCESSOR 7: 15 engines @ 179 MHz -- MAP: F7FF 32 C


are these commandline flags?
No, you can use Z-Commands by connecting a terminal program to the serial port.
I tried to use minicom but it never works for me so I used tail -f /dev/ttyUSB0 on one console and echo -n ZcX >/dev/ttyUSB0 on another console to get the result on my linux box.

I found only working 5 commands: ZCX, ZcX, ZLX, ZlX and ZGX. Maybe the're more available...
hero member
Activity: 711
Merit: 532
Spotswood made me this excellent Chili rack (apologies for the poor photo):



I love this rack for many reasons:
• It holds up to 7 Chilis (I'm just waiting on a few more -- come on, Lucko!).
• Entirely configurable -- adjust everything so it fits just how you want, space them as close or as far apart as you like.
• Everyone has their own cooling solution, I went with the Accelero Mono, and Spotswood configured this to my needs (awesome).
• Holds everything together and lets me move it around, reining in the Chili octopus.
• Came with extra bolts, brackets, even tools.
• Great price! He probably has different prices for different needs, but especially considering he made this custom and it only took ~ 1 week, I think it was a great price (less than $100).

So, if you've got a bunch of stuff that needs to be bolted down: http://richchomiczewski.wordpress.com/

(I do not know Spotswood, nor do I get any kickbacks here -- I just think he's doing great work and wanted to share it.)
member
Activity: 67
Merit: 10
BTW- Not super useful for most people since you can't use it while hashing since the mining software has the port open, but the ZlX command (lowercase "L") will return the temperature of all the chips. If you're compiling your own cgminer, you might want to add in a way to read that.
I found ZcX which adds the internal chip temperatures behind the processor output:

PROCESSOR 0: 9 engines @ 137 MHz -- MAP: E07E 27 C
PROCESSOR 1: 14 engines @ 198 MHz -- MAP: DFF7 26 C
PROCESSOR 2: 7 engines @ 188 MHz -- MAP: 0BF0 27 C
PROCESSOR 3: 15 engines @ 201 MHz -- MAP: FF7F 29 C
PROCESSOR 4: 13 engines @ 159 MHz -- MAP: E7FE 27 C
PROCESSOR 5: 8 engines @ 168 MHz -- MAP: 0FF0 27 C
PROCESSOR 6: 8 engines @ 179 MHz -- MAP: 0FF0 26 C
PROCESSOR 7: 15 engines @ 179 MHz -- MAP: F7FF 32 C



are these commandline flags?
full member
Activity: 128
Merit: 100
BTW- Not super useful for most people since you can't use it while hashing since the mining software has the port open, but the ZlX command (lowercase "L") will return the temperature of all the chips. If you're compiling your own cgminer, you might want to add in a way to read that.
I found ZcX which adds the internal chip temperatures behind the processor output:

PROCESSOR 0: 9 engines @ 137 MHz -- MAP: E07E 27 C
PROCESSOR 1: 14 engines @ 198 MHz -- MAP: DFF7 26 C
PROCESSOR 2: 7 engines @ 188 MHz -- MAP: 0BF0 27 C
PROCESSOR 3: 15 engines @ 201 MHz -- MAP: FF7F 29 C
PROCESSOR 4: 13 engines @ 159 MHz -- MAP: E7FE 27 C
PROCESSOR 5: 8 engines @ 168 MHz -- MAP: 0FF0 27 C
PROCESSOR 6: 8 engines @ 179 MHz -- MAP: 0FF0 26 C
PROCESSOR 7: 15 engines @ 179 MHz -- MAP: F7FF 32 C

sr. member
Activity: 280
Merit: 250
Helperizer
This happens to me sometimes and unplugging power for a few seconds helps.
That was the second thing I tried after resetting the USB bus.  But thanks anyway.  ;-)

The only thing that seems to work (mostly) reliably is to blow compressed air towards the contact between the hashing chips and the cooler.  So, somethiing is horked with the temperature sensor that lets it get into the boot sequence - either it's got too much leftover thermal paste on it or it has an offset of some sort.  I did make it almost spotless when I cleaned up the paste and converted to thermal pad, but...
Pages:
Jump to: