Topic: ZTEX USB-FPGA Modules 1.15x and 1.15y: 215 and 860 MH/s FPGA Boards - page 29. (Read 182450 times)
March 30, 2012, 08:58:24 AM
Also catfish: please post the rig in FPGA Photo thread later, can't wait to see this thing!
March 28, 2012, 02:20:44 PM
@RUPY
Please can you post a picture of how you've mounted the Zalman flower heatsinks to your 1.15x boards please - specifically how you've connected the fiddly pin-post connectors to the heatsink?
I'm in the process of sending Stefan a wad of money for 25 boards and I have 25 Zalman heatsinks to assemble. And they're tricky and fiddly.
I *really* don't want to assemble 25 of these damn things and then find I've got the 'hooks' pointing the wrong way!!!!
I'd *really* appreciate it
Please can you post a picture of how you've mounted the Zalman flower heatsinks to your 1.15x boards please - specifically how you've connected the fiddly pin-post connectors to the heatsink?
I'm in the process of sending Stefan a wad of money for 25 boards and I have 25 Zalman heatsinks to assemble. And they're tricky and fiddly.
I *really* don't want to assemble 25 of these damn things and then find I've got the 'hooks' pointing the wrong way!!!!
I'd *really* appreciate it
Good luck catfish. Stefan's latest batch is fast. ~210 average, the fastest runs at 228Mhz (stock cooler). I hope you can get them up and running soon.
March 28, 2012, 09:53:16 AM
They go with the hook through the board and the spring on the metal wing end. Also you have to turn one of the wings so it doesn't poke the big black square component (not because it's hot, but because you can't properly center the heatsink otherwise), but you'll notice; it only goes one way.Also I used pliers to gently narrow the hook a bit so I didn't have to force the hooks through with too much pressure.
I would use some quality grease though, not the zalman one.
How are you going to cool these, passively?
March 28, 2012, 09:22:52 AM
@RUPY
Please can you post a picture of how you've mounted the Zalman flower heatsinks to your 1.15x boards please - specifically how you've connected the fiddly pin-post connectors to the heatsink?
I'm in the process of sending Stefan a wad of money for 25 boards and I have 25 Zalman heatsinks to assemble. And they're tricky and fiddly.
I *really* don't want to assemble 25 of these damn things and then find I've got the 'hooks' pointing the wrong way!!!!
I'd *really* appreciate it
Please can you post a picture of how you've mounted the Zalman flower heatsinks to your 1.15x boards please - specifically how you've connected the fiddly pin-post connectors to the heatsink?
I'm in the process of sending Stefan a wad of money for 25 boards and I have 25 Zalman heatsinks to assemble. And they're tricky and fiddly.
I *really* don't want to assemble 25 of these damn things and then find I've got the 'hooks' pointing the wrong way!!!!
I'd *really* appreciate it
Wish i had a wod of cash for 25 boards
Good luck with that venture. I hope to some day, Maybe next gen FPGA's
March 22, 2012, 08:55:14 PM
I have tested this according to your instructions and it worked without problems in my 3 Ztex (+5 Icarus) setup). It recognized the Ztex boards automatically, while I had to add my Icarus COM ports manually (as expected I guess). I only had 5 minutes so I didn't set up my own blockchains (pools), but everything worked as expected.
Once I have more time I will test a bit more and give you feedback in your thread, I have only been using the old version of MPBM so far. One thing I noticed, using Firefox the display on the top left (menu) was messed up. With Chrome it worked fine. Could be one of my Firefox add-ons though.
Once I have more time I will test a bit more and give you feedback in your thread, I have only been using the old version of MPBM so far. One thing I noticed, using Firefox the display on the top left (menu) was messed up. With Chrome it worked fine. Could be one of my Firefox add-ons though.
Yes, I haven't found a way to autodetect Icarus boards yet, their protocol doesn't really make that easy. It's kinda fire and forget.
What do you mean with "old version"? The old version (master branch) doesn't have ztex board support (except for nelisky's branch, but I'm not sure if that one even works yet, haven't tested it). The web user interface is the new (testing) branch.
The firefox issues are known, seems like it has a problem with sizing of scrollable divs. Need to look into that when I find time to do that.
March 22, 2012, 06:24:04 PM
Would love to test this.... But:
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
1. The overclocking code tries to exactly match what ZTEX BTCMiner is doing. And I see no other way how it could be blown up, if at all.
2. Here are some general instructions:
- Check out the testing branch (git clone git://github.com/TheSeven/Modular-Python-Bitcoin-Miner.git -b testing)
- Run "python run-mpbm.py"
- Open http://localhost:8832
- Log in with user name "admin", password "mpbm"
- Configure the WebUI frontend to set your own password
- Add your own work sources to the "user work sources" group
- Right click on the created work sources and assign them to the "Bitcoin" blockchain (or whichever blockchain they belong to), to enable long poll aggregation
- When you're done, hit "save configuration" (will be done during a clean shutdown anyway, but it can't hurt to do it manually after lots of changes to reduce the risk of data loss)
The boards should be detected automatically. I haven't tested this on Mac OS at all though, so you might hit some OS dependent issues. If that happens, just let me know (ideally on IRC, so we can try to figure things out interactively).
If you leave the demo (and donation) work sources enabled (which I would of course appreciate), and don't have a hashrate of multiple gigahashes per second, the work distribution might not work correctly and allocate more than 1% of your hashrate to those demo/donation work sources. To fix that, just change the number of long poll connections on the BTCMP (demo) work source to zero.
I have tested this according to your instructions and it worked without problems in my 3 Ztex (+5 Icarus) setup). It recognized the Ztex boards automatically, while I had to add my Icarus COM ports manually (as expected I guess). I only had 5 minutes so I didn't set up my own blockchains (pools), but everything worked as expected.
Once I have more time I will test a bit more and give you feedback in your thread, I have only been using the old version of MPBM so far. One thing I noticed, using Firefox the display on the top left (menu) was messed up. With Chrome it worked fine. Could be one of my Firefox add-ons though.
March 22, 2012, 05:44:06 PM
Would love to test this.... But:
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
1. The overclocking code tries to exactly match what ZTEX BTCMiner is doing. And I see no other way how it could be blown up, if at all.
2. Here are some general instructions:
- Check out the testing branch (git clone git://github.com/TheSeven/Modular-Python-Bitcoin-Miner.git -b testing)
- Run "python run-mpbm.py"
- Open http://localhost:8832
- Log in with user name "admin", password "mpbm"
- Configure the WebUI frontend to set your own password
- Add your own work sources to the "user work sources" group
- Right click on the created work sources and assign them to the "Bitcoin" blockchain (or whichever blockchain they belong to), to enable long poll aggregation
- When you're done, hit "save configuration" (will be done during a clean shutdown anyway, but it can't hurt to do it manually after lots of changes to reduce the risk of data loss)
The boards should be detected automatically. I haven't tested this on Mac OS at all though, so you might hit some OS dependent issues. If that happens, just let me know (ideally on IRC, so we can try to figure things out interactively).
If you leave the demo (and donation) work sources enabled (which I would of course appreciate), and don't have a hashrate of multiple gigahashes per second, the work distribution might not work correctly and allocate more than 1% of your hashrate to those demo/donation work sources. To fix that, just change the number of long poll connections on the BTCMP (demo) work source to zero.
March 22, 2012, 05:33:42 PM
Would love to test this.... But:
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
1. How Safe is it ? (i don't want my boards to blow up .....)
2. Is there a "how to" aviable (for a windows guy that moved to MAc OSX)
March 20, 2012, 04:50:01 PM
ZTEX board support has finally hit the MPBM testing branch!
Available here: https://github.com/TheSeven/Modular-Python-Bitcoin-Miner/tree/testing
I appreciate any feedback!
If you run into any issues, please contact me on IRC: #mpbm on irc.freenode.net
Available here: https://github.com/TheSeven/Modular-Python-Bitcoin-Miner/tree/testing
I appreciate any feedback!
If you run into any issues, please contact me on IRC: #mpbm on irc.freenode.net
March 20, 2012, 04:26:00 PM
Totally agree... One big silent fan instead of lots of humming small ones. I need something like this:
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
I've been thinking about mounting three to six of them inside of the hard drive tray in my case. It has hot swap trays with 3.5 and 2.5 mounts already designed into them. The front of the case has a 230mm fan that blows directly on it.
March 20, 2012, 02:48:34 PM
So the entrance is circular and the exit is a very thin and long rectangle.
Maybe you should have a look at cross flow fans?
Hm, no but thanks for the suggestion.
I'm actually thinking about aquarium air pumps now for the "octopus" solution... but I know the final solution will be convection, even if it means wearing my chips a little.
March 20, 2012, 02:25:46 PM
So the entrance is circular and the exit is a very thin and long rectangle.
Maybe you should have a look at cross flow fans?
March 20, 2012, 12:28:37 PM
What I have noticed using 10 board vertical stacking is that a thermal stratification occurs inside the case. Not the air, but material and pcb. Higher boards have higher temps. Horizontal might be a more efficient method if running in a high ambient temp area. I put fastest of the boards in top slot and installed the rest in order going down. 10 boards running in cluster mode have had a combined speed of 2077.9 MH/s for weeks now on P2Pool.
Yes, It's best to not stack vertically or inline (Turbor). Best is bottom up, heat rises.
Another solution to the above is to build a web of small pipes, one to each heatsink. Like an inverted octopus vacuum cleaner!
The point here is that you don't need a large area of air if your air is fast and/or cold.
March 20, 2012, 11:28:18 AM
What I have noticed using 10 board vertical stacking is that a thermal stratification occurs inside the case. Not the air, but material and pcb. Higher boards have higher temps. Horizontal might be a more efficient method if running in a high ambient temp area. I put fastest of the boards in top slot and installed the rest in order going down. 10 boards running in cluster mode have had a combined speed of 2077.9 MH/s for weeks now on P2Pool.
March 20, 2012, 10:28:59 AM
Totally agree... One big silent fan instead of lots of humming small ones. I need something like this:
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
That wouldn't work but something like this would:
______
F FPGA
A FPGA
N FPGA
-------
and 2-3 side by side. Depending upon the fan size, this would create a 2x2 or 3x3 box, depending upon spacing ofc.
That funnel would fail to work because too small pressure differentiation and airflow speed, for the funnel to force the airflow to spread. Pressure being the more important factor.
Yes but the funnel would slim out on the other diagonal, to even the airflow. Like so:
Topview:
/ FPGA
- FPGA
FPGA
FAN FPGA
FPGA
- FPGA
\ FPGA
Sideview:
-
\
FAN FPGA
/
-
So the entrance is circular and the exit is a very thin and long rectangle.
March 20, 2012, 09:07:01 AM
Totally agree... One big silent fan instead of lots of humming small ones. I need something like this:
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
That wouldn't work but something like this would:
______
F FPGA
A FPGA
N FPGA
-------
and 2-3 side by side. Depending upon the fan size, this would create a 2x2 or 3x3 box, depending upon spacing ofc.
That funnel would fail to work because too small pressure differentiation and airflow speed, for the funnel to force the airflow to spread. Pressure being the more important factor.
March 20, 2012, 03:33:46 AM
In my opinion an optimal cooling solution for a larger cluster would be as follows:
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
I can confirm that this will work. I used this method for cooling a couple of 1.15d FPGA boards (with smaller heat sinks).
March 20, 2012, 03:30:15 AM
Totally agree... One big silent fan instead of lots of humming small ones. I need something like this:
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
/ * FPGA
/ * FPGA
FAN= * FPGA
\ * FPGA
\ * FPGA
Like a big vacuum cleaner piece. Wonder if it exists?
March 19, 2012, 08:10:20 PM
In my opinion an optimal cooling solution for a larger cluster would be as follows:
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
I will do some experimenting with this in the near future.
Does anybody know a "formula" of what kind of airflow (amount of air) we need to cool X watt of heat dissipation? I know this probably depends on the heat sinks too and the surface, but I am looking for some kind of textbook formula to get a ballpark figure - so that I don't have to start off a good guess.
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
I will do some experimenting with this in the near future.
Does anybody know a "formula" of what kind of airflow (amount of air) we need to cool X watt of heat dissipation? I know this probably depends on the heat sinks too and the surface, but I am looking for some kind of textbook formula to get a ballpark figure - so that I don't have to start off a good guess.
You can start by looking at the air thermal transfer rate + surface area formulas. I guess nothing ready as every custom case is... well custom. I'd say just build it and test it.
Just remember that air is actually very poor in thermal transfer, you can actually use air as an insulator if you can stop it from moving ...
I've been thinking of doing the same thing, designed a case, based on single 3W 230mm fan after i get my first boards and got a bit of time on my hands, going to design in 3d and then print the casings. Testing will show how well it works.
But i'm still of a month or couple before doing that, other things to do in regards of mining first.
March 19, 2012, 07:06:46 PM
In my opinion an optimal cooling solution for a larger cluster would be as follows:
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
I will do some experimenting with this in the near future.
Does anybody know a "formula" of what kind of airflow (amount of air) we need to cool X watt of heat dissipation? I know this probably depends on the heat sinks too and the surface, but I am looking for some kind of textbook formula to get a ballpark figure - so that I don't have to start off a good guess.
- put a number of boards in a single case
- establish proper airflow through case
- individual boards using proper heat sinks
- no fan on individual boards
- airflow forced through heat slinks
(ie. space around them tight so air can't go around)
I will do some experimenting with this in the near future.
Does anybody know a "formula" of what kind of airflow (amount of air) we need to cool X watt of heat dissipation? I know this probably depends on the heat sinks too and the surface, but I am looking for some kind of textbook formula to get a ballpark figure - so that I don't have to start off a good guess.
Jump to: