MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 14, 2013, 07:38:05 PM |
|
Got a part number for the mini-heatsink?
An electronics store would be best bet and/or a massive techie computer parts web store... Ebay has some but have not found a fit but not done an extensive search.
|
|
|
|
2weiX
Legendary
Offline
Activity: 2058
Merit: 1005
this space intentionally left blank
|
|
June 14, 2013, 07:45:51 PM |
|
Got a part number for the mini-heatsink?
An electronics store would be best bet and/or a massive techie computer parts web store... Ebay has some but have not found a fit but not done an extensive search. ICKSMDF8SA http://www.tme.eu/de/details/icksmdf8sa/kuhler/fischer-elektronik/is that the one? how does one put such sink on such chip?
|
|
|
|
FiatKiller
|
|
June 14, 2013, 07:54:12 PM |
|
heatsink compound - there an old zinc based one and some newer ones
|
|
|
|
2weiX
Legendary
Offline
Activity: 2058
Merit: 1005
this space intentionally left blank
|
|
June 14, 2013, 07:56:33 PM |
|
anyways - I am excited to hear some results.
|
|
|
|
MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 14, 2013, 08:01:57 PM |
|
Got a part number for the mini-heatsink?
An electronics store would be best bet and/or a massive techie computer parts web store... Ebay has some but have not found a fit but not done an extensive search. ICKSMDF8SA http://www.tme.eu/de/details/icksmdf8sa/kuhler/fischer-elektronik/is that the one? how does one put such sink on such chip? IMO-An Exact Match!
|
|
|
|
FiatKiller
|
|
June 14, 2013, 08:09:30 PM |
|
Now where can Americans get it?? lol
|
|
|
|
razorfishsl
|
|
June 14, 2013, 11:25:47 PM |
|
I'll take 10 of those please. I plan on testing one out later today - let you know how it goes. Stick a little peltier cooler on it and freeze it then crank up the v's! No don't..... it is a STUPID idea...... 1. Peltiers consume a MASSIVE amount of power for the cooling you get. 2. There is a significant cooling down time. 3. They fail silently. 4. You STILL need a fan....
|
|
|
|
MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 14, 2013, 11:34:09 PM |
|
I'll take 10 of those please. I plan on testing one out later today - let you know how it goes. Stick a little peltier cooler on it and freeze it then crank up the v's! No don't..... it is a STUPID idea...... 1. Peltiers consume a MASSIVE amount of power for the cooling you get. 2. There is a significant cooling down time. 3. They fail silently. 4. You STILL need a fan.... was jk...
|
|
|
|
|
LaserHorse
|
|
June 15, 2013, 05:19:11 AM |
|
Hmm - after a few hours of testing with the added heatsink, I'm seeing no significant reduction in error rate. (This was w/ a 120mm ThermalTake fan, btw)
I'm guessing we might see some benefit from replacing the large stock heatsink w one that has a greater surface area (such as the one seen on the emerald prototypes) - but again, just a guess. No doubt that friedcat did plenty of testing before finalizing this design - always fun to experiment tho.
|
|
|
|
buckus69
Newbie
Offline
Activity: 27
Merit: 0
|
|
June 15, 2013, 05:25:37 AM |
|
As cool looking as these are, and as much as I want to get into ASIC, it looks like these will probably never earn you your 2BTC. At current difficulty, it would take something like 240 days to recover your investment, but we all know difficulty is shooting through the roof. Considering the basic chip is around $10 in bulk, the whole shebang should be going for around $50. I know you didn't set the price Canary, but that's my .02 BTC.
|
|
|
|
byronbb
Legendary
Offline
Activity: 1414
Merit: 1000
HODL OR DIE
|
|
June 15, 2013, 06:32:03 AM |
|
I take it error rate % is determined by HW/A * 100 (cgminer)? Also mine is holding at 330MHs. Anyone getting better? No fans, plugged into my desktop.
|
|
|
|
LogicalUnit
|
|
June 15, 2013, 06:56:10 AM |
|
I take it error rate % is determined by HW/A * 100 (cgminer)? Also mine is holding at 330MHs. Anyone getting better? No fans, plugged into my desktop.
Mine are holding at 335.5MHash/s. Error rate about 4.5/m but this is without any cooling.
|
|
|
|
MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 15, 2013, 08:55:13 AM |
|
Hmm - after a few hours of testing with the added heatsink, I'm seeing no significant reduction in error rate. (This was w/ a 120mm ThermalTake fan, btw)
I'm guessing we might see some benefit from replacing the large stock heatsink w one that has a greater surface area (such as the one seen on the emerald prototypes) - but again, just a guess. No doubt that friedcat did plenty of testing before finalizing this design - always fun to experiment tho.
I do think the Emerald type heat sinks protrude so much they heat up the Erupter next to it in the Hub. Hub spacings are too tight.
|
|
|
|
FiatKiller
|
|
June 15, 2013, 12:24:25 PM |
|
Maybe it would help to have a cross piece added to the heatsink that hangs over the sides like a telephone pole's crosspiece.
|
|
|
|
MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 15, 2013, 12:54:23 PM |
|
Maybe it would help to have a cross piece added to the heatsink that hangs over the sides like a telephone pole's crosspiece.
I have a couple ideas but need to implement first...
|
|
|
|
razorfishsl
|
|
June 15, 2013, 10:43:37 PM |
|
Hmm - after a few hours of testing with the added heatsink, I'm seeing no significant reduction in error rate. (This was w/ a 120mm ThermalTake fan, btw)
I'm guessing we might see some benefit from replacing the large stock heatsink w one that has a greater surface area (such as the one seen on the emerald prototypes) - but again, just a guess. No doubt that friedcat did plenty of testing before finalizing this design - always fun to experiment tho.
I suspect that adding a heat-sink on the 'back' of the chip is not going to achieve much..., the thermal resistance is too high. What is the temp on the inductor next to the chip and the DC/DC control chip? Also looking at the board & it's PSU... it could do with a couple of tants. NOT ecaps... also a bit surprised that the 'crypto' chip does not appear to be 'decoupled'... But first point would be the temp on the inductor....
|
|
|
|
FiatKiller
|
|
June 16, 2013, 12:39:58 PM |
|
Ironically, I started playing around with the infrared temp gun yesterday and the ambient temp of the room seems to be the biggest factor. In the afternoon, the highest temp was ~120F. This morning it is ~110F. The inductor(along with the support chips) is about 5 degrees lower. I'm waiting on another Anker hub and I have an extra one plugged straight into a PC port vertically and it has a spot near the LED in the ~135F range. But the same area on the 9 in the hub is fine.
|
|
|
|
pazor
Legendary
Offline
Activity: 966
Merit: 1000
|
|
June 16, 2013, 12:56:22 PM |
|
try this as an alternative exsys ex-1177hmv
|
treuhand-Dienst gewünscht? - frag per PM an BTC 174X17nR7vEQBQo4GXKRGMGaTmB49Gf1yT
|
|
|
MikeMike (OP)
Legendary
Offline
Activity: 2254
Merit: 1020
★ Ⓚ KORE TEAM Ⓚ ★
|
|
June 16, 2013, 01:11:18 PM |
|
try this as an alternative
exsys ex-1177hmv
Nice. Has surge protection. Do you know of anyone using them?
|
|
|
|
|