The Chili – 30+GH/s BFL based Bitcoin Miner Assembly

[nexus99]

Just wanted to put it out there that I am using Ubuntu 12.04 CGMiner 3.5.0 and all I had to do was plug it in to an open USB port and my Chili has been hashing away at 34 GH/s with temp at 70-72C. I really like this board....

This is how it should be.

Ill have mine hashing by Friday with an ARCTIC Accelero TT II and press on heatsinks under the chips and voltage regulator.

I will post my cgminer results Friday.

Please post up some pics when you have it all built. I'm interested to see how you are attaching the heatsinks.

[nwfella]

Two out of them is harder to get them start mining, at the very first beginning. and it become harder and harder with each reboot. These two eventually cease to operate yesterday, until now (but still viewable in cgminer console)

I got 2 boards that won't boot up (2nd led keep blinking) but once I turn on the AC to cool the room they boot up in about 1 minute. After that they will hash just fine even with AC off.

When a board first starts up, the firmware makes sure the board and chips are cooled down before it starts the self test.  This helps ensure the maximum number of engines (cores) will pass self test.

LED decoder ring

The LEDs are numbered 1 to 8 with #1 closest to the USB connector and #8 closest to the power and fan connectors.  During first power up, LEDs 5, 6, 7, and 8 will light up and LEDs 1, 2, 3,  and 4 will indicate a failure code if anything is wrong with the hardware.  This happens VERY quickly - in less than a second.

If the initial hardware checks are good, #7 will blink indicating it is waiting for the board to cool down.  This will always blink for a few seconds even if the board is cold.  

Then #7 and #8 will blink indicating ASIC self test.  

After that, 1, 2, 3, 4 will indicate how many jobs from cgminer are waiting to run in the input queue.  LED 5 is just a debug output for me but it roughly indicates a job has completed - but sometimes blinks too fast for the human eye to see.  Just ignore LED 5.

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 am having some serious difficulty getting either of my two chilis to run properly. After replacing my BFL drivers with Zadig v6.1.7600.16385 driver(s) and loading cgminer v3.6.6 I am getting the device to detect but getting the following output:
"AVA looking for AVA 0403:6001 but found 05e3:0608 instead" <---which varies at times
Any idea what may be causing this.

*Also, LED's 7 and 8 are flashing at different speeds (seven faster than eight), while LED's 2,3,4, and 5 are all solid green.

Any suggestions you might be able to offer would be greatly appreciated.

[MrTeal]

I am having some serious difficulty getting either of my two chilis to run properly. After replacing my BFL drivers with Zadig v6.1.7600.16385 driver(s) and loading cgminer v3.6.6 I am getting the device to detect but getting the following output:
"AVA looking for AVA 0403:6001 but found 05e3:0608 instead" <---which varies at times
Any idea what may be causing this.

*Also, LED's 7 and 8 are flashing at different speeds (seven faster than eight), while LED's 2,3,4, and 5 are all solid green.

Any suggestions you might be able to offer would be greatly appreciated.

The AVA message is what cgminer puts out? I've never seen anything like that before, I'll have to download 3.6.6 and look into it.

[Bargraphics]

Looking forward to that Firmware update guys Got two dead chilis I'd like to potentially resurrect (Happy Halloween)

[jelin1984]

From what the chilli died?

[CrazyGuy]

Two out of them is harder to get them start mining, at the very first beginning. and it become harder and harder with each reboot. These two eventually cease to operate yesterday, until now (but still viewable in cgminer console)

I got 2 boards that won't boot up (2nd led keep blinking) but once I turn on the AC to cool the room they boot up in about 1 minute. After that they will hash just fine even with AC off.

When a board first starts up, the firmware makes sure the board and chips are cooled down before it starts the self test.  This helps ensure the maximum number of engines (cores) will pass self test.

LED decoder ring

The LEDs are numbered 1 to 8 with #1 closest to the USB connector and #8 closest to the power and fan connectors.  During first power up, LEDs 5, 6, 7, and 8 will light up and LEDs 1, 2, 3,  and 4 will indicate a failure code if anything is wrong with the hardware.  This happens VERY quickly - in less than a second.

If the initial hardware checks are good, #7 will blink indicating it is waiting for the board to cool down.  This will always blink for a few seconds even if the board is cold.  

Then #7 and #8 will blink indicating ASIC self test.  

After that, 1, 2, 3, 4 will indicate how many jobs from cgminer are waiting to run in the input queue.  LED 5 is just a debug output for me but it roughly indicates a job has completed - but sometimes blinks too fast for the human eye to see.  Just ignore LED 5.

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 am having some serious difficulty getting either of my two chilis to run properly. After replacing my BFL drivers with Zadig v6.1.7600.16385 driver(s) and loading cgminer v3.6.6 I am getting the device to detect but getting the following output:
"AVA looking for AVA 0403:6001 but found 05e3:0608 instead" <---which varies at times
Any idea what may be causing this.

*Also, LED's 7 and 8 are flashing at different speeds (seven faster than eight), while LED's 2,3,4, and 5 are all solid green.

Any suggestions you might be able to offer would be greatly appreciated.

Vendor and product ID should be 0403:6014, not sure what that one is. Second id might be a hub(maybe try plugging directly into your controller). WinUSB driver installation can be a real pain when you already have a default set, like FTDI. Were you using bfgminer beforehand? I would suggest completely removing the existing drivers and then reinstalling with winusb. Once Winusb is the default driver for that device type, you won't have to go through the process again with each new device. Don't forget to unplug the device and re-plug it in after installing WinUSB drivers.

[daemonfox]

Just wanted to put it out there that I am using Ubuntu 12.04 CGMiner 3.5.0 and all I had to do was plug it in to an open USB port and my Chili has been hashing away at 34 GH/s with temp at 70-72C. I really like this board....

This is how it should be.

Ill have mine hashing by Friday with an ARCTIC Accelero TT II and press on heatsinks under the chips and voltage regulator.

I will post my cgminer results Friday.

Please post up some pics when you have it all built. I'm interested to see how you are attaching the heatsinks.

Looking for some feedback before I make any permanent additions.

Here is the Chili...



Here is what I plan on doing with the heatsinks on the bottom... please let me know if this is worth doing or needs better placement... was going for base of mosfets and capacitors not the inductors.



Here is whaat I might do with the remaining heatsinks... is this even worth doing?

[nexus99]

Are you going to use the glue that came with the heatsinks?

[MrTeal]

I just got my production boards in the mail today. It's been a long wait for these, it's shocking that I'm not more prepared.

I mounted a few with the Antec Kuhler Shelf's, but I'm not terribly happy with how much board flex I got with those stock back brackets. I was going to go out to the garage to cut up a better backplate for it, but before I did I thought I'd try a quick experiment and just used another one of the Kuhler Shelf's on the back side essentially just as a backplate. I just used the included paste as well.
Results were pretty spectacular, especially considering it is hot (and loud) in my furnace room with 3 Avalon Minis, a BFL Single and LS, along with now two Chilis. It ramped up to 39GH/s pretty much right away. The power supply got Hot Hot Hot though. It definitely needed to have some press on heatsinks added there.

I'll bring it in to the office tomorrow so I can hook some diagnostic equipment up to it and take a look at the current draw of the board and the ASICs at those power levels. For $10 though, I'm tempted to just double up the coolers on every unit and call it a day. It will save me a lot of time compared to actually making a backplate.

I can't wait to get my PDU and other equipment set up so I can run it properly where it's meant to be, with some nice 0C ambient instead of 25C ambient.

[nwfella]

Two out of them is harder to get them start mining, at the very first beginning. and it become harder and harder with each reboot. These two eventually cease to operate yesterday, until now (but still viewable in cgminer console)

I got 2 boards that won't boot up (2nd led keep blinking) but once I turn on the AC to cool the room they boot up in about 1 minute. After that they will hash just fine even with AC off.

When a board first starts up, the firmware makes sure the board and chips are cooled down before it starts the self test.  This helps ensure the maximum number of engines (cores) will pass self test.

LED decoder ring

The LEDs are numbered 1 to 8 with #1 closest to the USB connector and #8 closest to the power and fan connectors.  During first power up, LEDs 5, 6, 7, and 8 will light up and LEDs 1, 2, 3,  and 4 will indicate a failure code if anything is wrong with the hardware.  This happens VERY quickly - in less than a second.

If the initial hardware checks are good, #7 will blink indicating it is waiting for the board to cool down.  This will always blink for a few seconds even if the board is cold.  

Then #7 and #8 will blink indicating ASIC self test.  

After that, 1, 2, 3, 4 will indicate how many jobs from cgminer are waiting to run in the input queue.  LED 5 is just a debug output for me but it roughly indicates a job has completed - but sometimes blinks too fast for the human eye to see.  Just ignore LED 5.

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 am having some serious difficulty getting either of my two chilis to run properly. After replacing my BFL drivers with Zadig v6.1.7600.16385 driver(s) and loading cgminer v3.6.6 I am getting the device to detect but getting the following output:
"AVA looking for AVA 0403:6001 but found 05e3:0608 instead" <---which varies at times
Any idea what may be causing this.

*Also, LED's 7 and 8 are flashing at different speeds (seven faster than eight), while LED's 2,3,4, and 5 are all solid green.

Any suggestions you might be able to offer would be greatly appreciated.

Vendor and product ID should be 0403:6014, not sure what that one is. Second id might be a hub(maybe try plugging directly into your controller). WinUSB driver installation can be a real pain when you already have a default set, like FTDI. Were you using bfgminer beforehand? I would suggest completely removing the existing drivers and then reinstalling with winusb. Once Winusb is the default driver for that device type, you won't have to go through the process again with each new device. Don't forget to unplug the device and re-plug it in after installing WinUSB drivers.

I was indeed using bfgminer beforehand. I'll go ahead and give your suggestion a try.

*Although I thought the LED's 7 and 8 where supposed to stop flashing at some point.

[Keefe]

I think you need to wait for 7 and 8 to stop flashing before running the software.

[shade88]

I think you need to wait for 7 and 8 to stop flashing before running the software.

Yes, every led must stop flashing before you can run the software.

[jelin1984]

Any new firmware mr teal
Please

[daemonfox]

Are you going to use the glue that came with the heatsinks?

Yes I was planning on using the included glue.

Mr Teal... any advice on the placement of the heatsinks in my pics above?

Also will the glue itself be ok for use on the bottom... I do not want to short anything...

[Keefe]

Are you going to use the glue that came with the heatsinks?

Yes I was planning on using the included glue.

Mr Teal... any advice on the placement of the heatsinks in my pics above?

Also will the glue itself be ok for use on the bottom... I do not want to short anything...

Glue? You probably mean thermal paste, which won't keep things stuck on reliably.

[DeathAndTaxes]

There are thermal adhesives.

Good stuff:  http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm

It is likely the heatsinks came with thermal adhesive not thermal paste.

[DeathAndTaxes]

Here is whaat I might do with the remaining heatsinks... is this even worth doing?

Probably not.  Those don't appear to be high power components.  If you don't have a IR thermometer I would try the "finger test".  If you can put a finger on it for >1/4 second without burning yourself the component isn't getting hot enough to benefit from a heat sink.  The key is the time like checking a grill temp with your hand.   If you can hold your finger there for 2-3 seconds before it eventually gets hot enough you have to remove it that isn't "hot enough".

It won't hurt to add the heatsinks if there isn't enough power usage but it won't do much good beyond "nerd bling".

[daemonfox]

There are thermal adhesives.

Good stuff:  http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm

It is likely the heatsinks came with thermal adhesive not thermal paste.

Exactly... the tube that came with my Arctic Accelero Twin turbo II is listed as Thermal Adhesive and is referred to as "glue" in the paperwork demonstrating how to use all the heatsinks for RAM and VRMs (the picture saying use pliers to get ALL the glue out of the tube lol).

I have not even fired this up yet... sadly my TIM is still in USPS and has not arrived yet.

[MrTeal]

Any new firmware mr teal
Please

Are yours having an issue of some kind?

[Keefe]

Any new firmware mr teal
Please

Are yours having an issue of some kind?

One of mine is. It often goes into spew-errors mode, requiring a power cycle.