ZTEX USB-FPGA Modules 1.15x and 1.15y: 215 and 860 MH/s FPGA Boards

[Inspector 2211]

Heatsink placement is a bitch.

Heat sink installation on USB FPGA Modules 1.15d should not be difficult if you follow the instructions at http://www.ztex.de/usb-fpga-1/usb-fpga-1.15.e.html#hs.

Only two additional things have to be considered (in comparison to the 1.15x board):

• Heat sink removal is critical. Always unmount it by twisting (see http://www.ztex.de/usb-fpga-1/usb-fpga-1.15.e.html#hs ) tilting may damage the PCB (IMHO that happened to Turbors board)
• Some airflow has to be ensured, either by case fans or by using the active cooling upgrade in the shop

The board is working fine. It just needs some pressure from above. I can live with that It's the faster board of the two btw...

I've always wondered what happens when you PULL a heat sink off a BGA chip.
Could some of the solder balls come undone?
The solder ball would still be there, but its contact to the PCB (or its contact to the chip itself) would be something like what happens in a "cold solder joint".
That it works when you apply pressure, and ceases to work when the pressure is gone, fits this picture 100%.

[Turbor]

I didn't pull the heatsink ! That much i know. When i removed the pad the first time i could see that it did not made enough contact with the chip. The heatsink is perfectly flat. The latest fail could be too much of thermal adhesive. I tried to put on as little as possible. Either way, i had my chances. Now there is no way to remove the heatsink without serious damage.

[Inspector 2211]

Module 1.15x fan failure during mining
============================

Let me start out with pointing out that it was not ZTEX's luxury fan (retail cost $29.99 or $19.99 or something like that) that failed,
but my cheap fan from China.
Which has 2mm pitch connectors and which I had attached to the 1/10" pitch connectors by a crudely made adapter cable.
 
It failed on Wednesday.

Here's what happened in  detail, starting with a time when it is still mining along happily:

ztex_ufm1_15d3-04A32DC6E1: Set frequency to 216.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=198.0MH/s,  submitted 0 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=198.0MH/s,  submitted 0 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=198.0MH/s,  submitted 1 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=0.23%,  maxErrorRate=0.29%,  hashRate=197.5MH/s,  submitted 1 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=0.39%,  maxErrorRate=0.49%,  hashRate=197.2MH/s,  submitted 0 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 216.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=204.00MHz,  errorRate=3.67%,  maxErrorRate=3.67%,  hashRate=196.5MH/s,  submitted 0 new nonces,  luckFactor=1.02
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 216.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=210.00MHz,  errorRate=7.43%,  maxErrorRate=7.43%,  hashRate=194.4MH/s,  submitted 0 new nonces,  luckFactor=1.03
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=192.00MHz,  submitted 0 new nonces,  luckFactor=1.05
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=192.00MHz,  errorRate=0.50%,  maxErrorRate=0.50%,  hashRate=191.0MH/s,  submitted 0 new nonces,  luckFactor=1.05
ztex_ufm1_15d3-04A32DC6E1: f=192.00MHz,  errorRate=0.85%,  maxErrorRate=0.92%,  hashRate=190.4MH/s,  submitted 0 new nonces,  luckFactor=1.05
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=3.73%,  maxErrorRate=3.73%,  hashRate=190.6MH/s,  submitted 0 new nonces,  luckFactor=1.05
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 210.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=192.00MHz,  errorRate=1.47%,  maxErrorRate=1.62%,  hashRate=189.2MH/s,  submitted 1 new nonces,  luckFactor=1.06
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=192.00MHz,  errorRate=2.49%,  maxErrorRate=2.61%,  hashRate=187.2MH/s,  submitted 0 new nonces,  luckFactor=1.07
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  submitted 1 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.00%,  hashRate=186.0MH/s,  submitted 1 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.00%,  hashRate=186.0MH/s,  submitted 0 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=186.0MH/s,  submitted 1 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=186.0MH/s,  submitted 3 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=186.0MH/s,  submitted 1 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=198.00MHz,  errorRate=6.29%,  maxErrorRate=6.29%,  hashRate=185.6MH/s,  submitted 0 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=0.47%,  maxErrorRate=0.57%,  hashRate=185.1MH/s,  submitted 0 new nonces,  luckFactor=1.08
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 204.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=204.00MHz,  errorRate=9.74%,  maxErrorRate=9.74%,  hashRate=184.1MH/s,  submitted 1 new nonces,  luckFactor=1.09
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=1.16%,  maxErrorRate=1.23%,  hashRate=183.8MH/s,  submitted 1 new nonces,  luckFactor=1.09
New block detected by long polling
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=1.51%,  maxErrorRate=1.63%,  hashRate=183.2MH/s,  submitted 2 new nonces,  luckFactor=1.10
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=1.63%,  maxErrorRate=1.64%,  hashRate=183.0MH/s,  submitted 1 new nonces,  luckFactor=1.10
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=2.36%,  maxErrorRate=2.40%,  hashRate=181.6MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=186.00MHz,  errorRate=3.13%,  maxErrorRate=3.13%,  hashRate=180.2MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.00%,  hashRate=180.0MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.00%,  hashRate=180.0MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.47%,  maxErrorRate=0.47%,  hashRate=179.2MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.29%,  maxErrorRate=0.47%,  hashRate=179.5MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.19%,  maxErrorRate=0.47%,  hashRate=179.7MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.13%,  maxErrorRate=0.47%,  hashRate=179.8MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.09%,  maxErrorRate=0.47%,  hashRate=179.8MH/s,  submitted 1 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.07%,  maxErrorRate=0.47%,  hashRate=179.9MH/s,  submitted 1 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.05%,  maxErrorRate=0.47%,  hashRate=179.9MH/s,  submitted 0 new nonces,  luckFactor=1.11
New block detected by long polling
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.04%,  maxErrorRate=0.47%,  hashRate=179.9MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.28%,  maxErrorRate=0.47%,  hashRate=179.5MH/s,  submitted 0 new nonces,  luckFactor=1.11
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 198.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.73%,  maxErrorRate=0.73%,  hashRate=178.7MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=0.98%,  maxErrorRate=1.23%,  hashRate=178.2MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.19%,  maxErrorRate=1.23%,  hashRate=177.9MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.47%,  maxErrorRate=1.54%,  hashRate=177.3MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.81%,  maxErrorRate=1.88%,  hashRate=176.7MH/s,  submitted 1 new nonces,  luckFactor=1.13
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.78%,  maxErrorRate=2.06%,  hashRate=176.8MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.76%,  maxErrorRate=2.06%,  hashRate=176.8MH/s,  submitted 0 new nonces,  luckFactor=1.12
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=1.96%,  maxErrorRate=2.11%,  hashRate=176.5MH/s,  submitted 1 new nonces,  luckFactor=1.13
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=2.34%,  maxErrorRate=2.37%,  hashRate=175.8MH/s,  submitted 0 new nonces,  luckFactor=1.13
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=2.90%,  maxErrorRate=2.93%,  hashRate=174.8MH/s,  submitted 0 new nonces,  luckFactor=1.13
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 174.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  submitted 2 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  hashRate=174.0MH/s,  submitted 2 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.00%,  maxErrorRate=0.00%,  hashRate=174.0MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.20%,  maxErrorRate=0.25%,  hashRate=173.7MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 192.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 174.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.41%,  maxErrorRate=0.43%,  hashRate=173.3MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.54%,  maxErrorRate=0.60%,  hashRate=173.1MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.41%,  maxErrorRate=0.60%,  hashRate=173.3MH/s,  submitted 2 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.30%,  maxErrorRate=0.60%,  hashRate=173.5MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.23%,  maxErrorRate=0.60%,  hashRate=173.6MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.17%,  maxErrorRate=0.60%,  hashRate=173.7MH/s,  submitted 2 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.34%,  maxErrorRate=0.60%,  hashRate=173.4MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 186.00MHz
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 174.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.73%,  maxErrorRate=0.80%,  hashRate=172.7MH/s,  submitted 0 new nonces,  luckFactor=1.15
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 180.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=180.00MHz,  errorRate=4.15%,  maxErrorRate=4.15%,  hashRate=172.5MH/s,  submitted 0 new nonces,  luckFactor=1.15
ztex_ufm1_15d3-04A32DC6E1: Set frequency to 174.00MHz
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=1.03%,  maxErrorRate=1.26%,  hashRate=172.2MH/s,  submitted 0 new nonces,  luckFactor=1.15
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.77%,  maxErrorRate=1.26%,  hashRate=172.7MH/s,  submitted 0 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=0.80%,  maxErrorRate=1.26%,  hashRate=172.6MH/s,  submitted 1 new nonces,  luckFactor=1.14
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=1.09%,  maxErrorRate=1.26%,  hashRate=172.1MH/s,  submitted 0 new nonces,  luckFactor=1.15
ztex_ufm1_15d3-04A32DC6E1: f=174.00MHz,  errorRate=1.05%,  maxErrorRate=1.26%,  hashRate=172.2MH/s,  submitted 0 new nonces,  luckFactor=1.15
Warning (try 1): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 2): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 3): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 4): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 5): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 6): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 7): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 8 ): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 9): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


Warning (try 10): libusb0-dll:err [control_msg] sending control message failed, win error: The device does not recognize the command.


ztex_ufm1_15d3-04A32DC6E1: Error: bus=bus-0  device=\\.\libusb0-0002--0x221a-0x0100: Read hash data: libusb0-dll:err [control_msg] sending control message failed, win
 error: The device does not recognize the command.

: Disabling device


So, it kept throttling down, the error rate went up, it throttled down some more and finally it seems the USB microcontroller got too hot from the radiated/conducted heat and stopped working properly, which in turn prevented the FPGA from getting more work (well, the FPGA's flip-flops would still be toggling, right?)

This morning, I re-attached the Xilince "luxury" cooler and restarted the mining software, whereupon...
(scroll down to read the shocking facts)




























































...IT WORKS AGAIN!  

[Turbor]

From the thread it appears that I'm going to need to ensure that my PSU doesn't kill the boards - it's a switching supply for CCTV units, rated at 12V DC - sadly I only have a decent multimeter and no oscilloscope, but Stefan said 'if it's switching, it's regulated' and I'll be testing with one board first anyway.

I think a multimeter should be good enough to check. I run my x's @ 12V and the d's at 13.8V.

[CA Coins]

Module 1.15x fan failure during mining

Interesting to see what happens when a fan fails during mining.  I guess the FPGA doesn't completely clock down to 0 right away due to the heatsink, but the USB controller poops out first.  Good to see that no serious damage to the board though.  Anybody suspect that it causes any damage to the USB controller/board components when the fan dies?

[Inspector 2211]

Module 1.15x fan failure during mining

Interesting to see what happens when a fan fails during mining.  I guess the FPGA doesn't completely clock down to 0 right away due to the heatsink, but the USB controller poops out first.  Good to see that no serious damage to the board though.  Anybody suspect that it causes any damage to the USB controller/board components when the fan dies?

>clocks down to 0

No, Stefan set the minimum at 174 MHz, but at this clock rate a PASSIVE heat sink (which my active heat sink turned into after the fan failure) seems to completely suffice (read the log that I posted).

>damage to the USB controller/board components

I think it was just operating out of spec, and thus failing to execute its program properly.
As I said, there is no evidence of any damage, and it's mining just fine.

That said, prolonged exposure to out-of-spec temperatures may cause electrolyte capacitors, but also other components, to age more rapidly than usual. But this is typically not a "bang, you're dead" event, more like "instead of an anticipated life span of, say, 6 years, we're now moving closer to 3 or 4 years".

[CA Coins]

No, Stefan set the minimum at 174 MHz, but at this clock rate a PASSIVE heat sink (which my active heat sink turned into after the fan failure) seems to completely suffice

I see.  I didn't know Stefan set the minimum at 174.  It doesn't look like the error rate was too high either at 174 from your log.  I was initially concerned that in a cluster, some fans are bound to fail at one point.  I wouldn't want to have to check on the boards every so often, but I guess you will be able to pick up that a fan failed based on the block rates.   

[Inspector 2211]

For the 1.15d boards - worth sticking a passive heatsink on the USB chip then? Small heatsinks intended for RAM (on GPU boards and the like) are easily found and small enough to sit on that second IC on the board.

On the 1.15x you probably couldn't do this because the main heatsink covers part of the second chip, but the 1.15d has more room, and the USB chip is a couple of centimetres from the FPGA.

I've got plenty of these little RAM heatsinks... could easily equip each 1.15d board with one... reckon it'd cause harm rather than good?

Ultimately I'd like a thermistor somewhere on the board to monitor temps, but I've found a 25mm fan to sit on my 1.15d heatsink for testing, and it has a three-wire connector intended for PC logic boards, so the speed should be easily monitored with extra logic or a separate fan controller...

>passive heat sink on the USB chip

I don't believe in trying to fix SYMPTOMS as opposed to CAUSES.
The USB chip got too warm by conducted [via the PCB] and radiated heat - putting a black heat sink on it
might actually cause it to pick up MORE radiated heat. Thus, I'd say "no" to your heat sink idea.

>speed should be easily monitored with extra logic

Stefan does feed the fan's RPM signal into the USB microcontroller, but AFAIK it's not [yet] being monitored by the microcontroller's firmware. While that would be easy enough to do, it'll then thwart people like me who intentionally want to use [low-profile] 2-wire fans. 2-wire meaning, they don't have an RPM signal output.
(I have this vision of slotting several 1.15x boards into some kind of "card cage", vertically, and the 2 inch or 3 inch tall Xilence fan would not be compatible with that.)

Btw., I don't think it is my low-profile fan that failed - quite likely it was just my makeshift fan power wiring that came undone.

[Inspector 2211]

No, Stefan set the minimum at 174 MHz, but at this clock rate a PASSIVE heat sink (which my active heat sink turned into after the fan failure) seems to completely suffice

I see.  I didn't know Stefan set the minimum at 174.  It doesn't look like the error rate was too high either at 174 from your log.  I was initially concerned that in a cluster, some fans are bound to fail at one point.  I wouldn't want to have to check on the boards every so often, but I guess you will be able to pick up that a fan failed based on the block rates.   

Actually, on second thought, you may be right. During one of the "downclock of death" episodes I encountered, it went as low as 126 MHz before I stopped it. Thus, I now think you're right - it simply stabilized at 174 MHz, worked at 174 MHz for a while, before the microcontroller crashed.

[Turbor]

Actually, on second thought, you may be right. During one of the "downclock of death" episodes I encountered, it went as low as 126 MHz before I stopped it. Thus, I now think you're right - it simply stabilized at 174 MHz, worked at 174 MHz for a while, before the microcontroller crashed.

Thats correct. I had several DCODs with my 1.15d and the controller never crashed !

[Inspector 2211]

Actually, on second thought, you may be right. During one of the "downclock of death" episodes I encountered, it went as low as 126 MHz before I stopped it. Thus, I now think you're right - it simply stabilized at 174 MHz, worked at 174 MHz for a while, before the microcontroller crashed.

Thats correct. I had several DCODs with my 1.15d and the controller never crashed !

Apples and oranges.

DCOD...caused by the Bitstream not doing something 100% right when programming the DCM (digital clock manager) on the FPGA, turning DCM programming into something  like a lottery.

Cypress controller crash...believed to be caused excessive conducted and/or radiated heat from a PASSIVELY cooled FPGA.

[ztex]

Yup - though the Xilinx SDK *is* required if I want to change the 'code' the FPGA runs, or write a new app for the FPGA as hinted at above - am I correct?

Yes, Xilinx ISE is required if you want to do FPGA development.

[ztex]

For the 1.15d boards - worth sticking a passive heatsink on the USB chip then?

The power disspiation of the USB controller during bitcoin mining should be less less 100mW.

[ztex]

Interesting to see what happens when a fan fails during mining.  I guess the FPGA doesn't completely clock down to 0 right away due to the heatsink, but the USB controller poops out first.  Good to See that no serious damage to the board though.  Anybody suspect that it causes any damage to the USB controller/board components when the fan dies?

If the original heat sink is used and if there is at least free convection the board is still save if the fan fails. This only costs about 5-10 MH/s of performance. (With some well placed case fans and good airflow the 40mm fans are obsolete.)

AFAIR Inspector2211 did not use the original cooler. The board was either shut down by a USB controller overheating (there is only a small gap between heat sink and USB controller) or the overcurrent protection was triggered (more likely) by the overheating of the FPGA or a disfunction caused by overheating.

No, Stefan set the minimum at 174 MHz, but at this clock rate a PASSIVE heat sink (which my active heat sink turned into after the fan failure) seems to completely suffice (read the log that I posted).

There is no such limit. (There is an internal limit in the Firmware at 100MHz bit this limit is not visible to the host software).

I don't agree that your (non-original) heat sink without fan was sufficient. There must have been temperatures around 100°C.

But it is an good idea to shut down the board if an frequency drop (of 10%) occurs. I will add this feature to BTCMIner.

[Turbor]

Actually, on second thought, you may be right. During one of the "downclock of death" episodes I encountered, it went as low as 126 MHz before I stopped it. Thus, I now think you're right - it simply stabilized at 174 MHz, worked at 174 MHz for a while, before the microcontroller crashed.

Thats correct. I had several DCODs with my 1.15d and the controller never crashed !

Apples and oranges.

DCOD...caused by the Bitstream not doing something 100% right when programming the DCM (digital clock manager) on the FPGA, turning DCM programming into something  like a lottery.

Cypress controller crash...believed to be caused excessive conducted and/or radiated heat from a PASSIVELY cooled FPGA.

I never had a DCOD because of something else than passive or bad cooling, ever. But you are right that the layout of the boards is different.

[ztex]

A new BTCMiner version with improved overheat protection has been released. Read https://bitcointalk.org/index.php?topic=40047.msg761071#msg761071 for details.

[Inspector 2211]

catfish, beside the obvious (FPGA), it would be interesting to see what temperatures the two buck controllers and the MOSFET reach after, say, an hour of continuous mining. Especially the AOZ1025 8-amp buck controller in its tiny DFN-16 case, which has one MOSFET built in. I have a 1.15x, but no infrared thermometer.

[ztex]

What temperature should I consider 'something is wrong - panic' and cut the power?

Provided that the heat sink is installed properly you should make sure that the temperature is less then 60°C. For optimal operation the heat sink temperature should be less than 40°C.

I'm not sure whether IR thermometers work on Al surfaces.

These FPGAs are rated to 70˚C according to Stefan's website - is this a sensible working temperature for the IC, or a 'maximum, not recommended continuous' temperature?

This is the board limit and defined by other components.

The absolute maximum junction temperature of the FPGA is 125°C the maximum operation junction temperature is 85°C .

[CA Coins]

The software clocks up/down based on the error rate.  Use the latest version (120221.jar) and it should shut the board down automatically if the rate drops significantly (ie. fan failure, etc).  It's running fine for me but I haven't tested the auto-shutdown feature yet. 

[CA Coins]

Hey Catfish, just curious, does OS X support running VMs?  I would think the java process/drivers should have no problem running inside a VM (as opposed to GPUs).  It would be nice to have an OS X software, but maybe run linux inside a VM if you run into more roadblocks.