I will do some other clocks freq 190 and freq 260. I would like to be able to leave this alone and feel pretty safe when I travel then pump it up to an oc when I am home.
I can keep this piece cool when I am home so an oc of 250 or 260 with my current setup seems okay.
When I leave the house I will try to find a dc I feel safe about.
The temps rose from 80 f to 105 f with the garage door shut.
The watts rose from 998 to 1045 with the garage door shut.
Clock at freq 200. I don't quite feel safe about this.
If freq 190 does the trick for unattended I will be happy.
If freq 260 works while I am in house I will be happy.
More testing to come.
I'd get that thing away from anything combustible, whether you're home or not. If it blows a gasket, you run the risk (whether you're home or not) of lighting all that nice kindling that surrounds it. In contrast, if that S3 fries crispy, the ignition source stays inside it's case and likely would just be a smoke bomb. Blocking that square hole in the back also helps force all the air across the heat sinks.--------------- second fan does this it covers the center hole very nicely and pulls on the outer heatsink ribsI used aluminum foal tape (not cloth, fiberglass, or plastic duct tape). As in my previous pics. Cheap at the hardware store. Remember 70% of the heat is dissipated by the heat sinks (internal), and the remaining 30% is dissipated from the visible components and board. Assuming you're clocked @ 200 and drawing ~1000W and assuming the power supply is 92% efficient. The Prisma is dissipating 920 watts total. That means 644 watts are being dissipated by the heat sinks (internal) and 276 watts are being dissipated component side. BTU = watts * 3.15. The downward facing hashing board is therefore dissipating 217.35 BTU (276W/4 * 3.15) continuously 24/7 into the wood shelf it's sitting on. And is why we mount ours on metal racks vertically. I betch'a the bottom board on your Prisma is running 2x-3x hotter than the other 3 (top and sides). Just sayin'. yep the bottom board is hottest , but that board is not wood it is a laminate much more heat resistant then wood. I am going to use some ceramic tiles for more heat protection. ----------- I have a lot of kevlar but can't find it. My problem with this unit is I don't think I can leave it unattended.----garage door shut for a few days on end. I have some s-3's that I leave alone for weeks. I just check to see if they are hashing. So far I can't keep the unit cool enough without have the garage door open. even with my lower clocks I shut the door the cold air does not come in the hot air stays in I get thermal creep and power creep. Still trying to get it to stay under 1000 watts without an open garage door. So far it reminds me of the gpu's running on bitminters client . I would get thermal creep and power creep. Then shut downs since they were pc's. |
|
|
|
@PlanetCrypto
I have an electricians clamp meter I got at Lowes that works well. If you have access to one side of your 240V feed, just measure the amps and multiply
times voltage measured with an ohmmeter. What I measured for some of my gear was 12A X 248V =2976W. I measured this right at the circuit breaker.
Roger the P=I*E Have a clamp-on for larger loads (100A+), she's not terribly accurate below 50A. I'll check out the one at Lowes. Thanks this is decent gear http://documents.ekmmetering.com/EKM-25IDS-Spec-Sheet.pdfhttp://www.ekmmetering.com/ekm-metering-products/electric-meters-kwh-meters.htmlyou can build a test box with 2 sockets and 1 plug. We had a few diy boxes a while back in an old amp / recording studio. We like to know what an amp would pull on transient loads so we did not trip breakers when giving shows. But that was a while back. |
|
|
|
With 2.3 days left numbers are : btc = 377usdhttps://bitcoinwisdom.com/bitcoin/difficulty >>>>>>>>>>>>>>>>> (+2.22%)Bitcoin Difficulty: 39,603,666,252 Estimated Next Difficulty: 40,482,069,778 (+2.22%) Adjust time: After 341 Blocks, About 2.3 days Hashrate(?): 291,765,602 GH/s Block Generation Time(?): 1 block: 9.9 minutes 3 blocks: 29.8 minutes 6 blocks: 59.5 minutes http://www.bitcoincharts.com/ >>>>>>>>>>>>>>>>>>>>>>>>> (+1.36%)Blocks 330282 Total BTC 13.507M Difficulty 39603666252 Estimated 40144276705 in 342 blks Network total 271776.341 Thash/s Blocks/hour 5.75 / 626 s This looks to be under 5% jump and may occur on the 18th of nov not the 17th |
|
|
|
|
I will do some other clocks freq 190 and freq 260. I would like to be able to leave this alone and feel pretty safe when I travel then pump it up to an oc when I am home.
I can keep this piece cool when I am home so an oc of 250 or 260 with my current setup seems okay.
When I leave the house I will try to find a dc I feel safe about.
The temps rose from 80 f to 105 f with the garage door shut.
The watts rose from 998 to 1045 with the garage door shut.
Clock at freq 200. I don't quite feel safe about this.
If freq 190 does the trick for unattended I will be happy.
If freq 260 works while I am in house I will be happy.
More testing to come.
|
|
|
|
last set of photos shows what I do to keep this very cool at freq 200 I open back door of the garage. let a fan pump in some cool air  I made a slide and or platform to bring the unit closer for exhaust of hot air  with 43 f air coming in you can keep your hand on the unit.  side fans help it  and I get power under 1000 watts freq 200 hash 1202gh so this is .83 watts per hash at the plug  |
|
|
|
OOPS on the previous post, just noticed your temps are in F not C.
MY BAD!!
happens . I have other points besides safety. I have run this overnight and I am at 104f-105f in a closed space. ( a 235 square ft garage ) in terms of 105 f that is 40 c once I open a door temps drop to 75f or 23.89c my power draw is 985 watts that is 985/1201 0.8201 watts per gh this power number include all my extra fans if I remove the extra fans I drop to 975 watts that is 0.811 watts . I don't knock oc but if your oc gives you .9 watts vs my .811 watts ( due to the more power costly fans) besides heat which looks to be no problem what about power cost? not to mention my error rate is .88% and your error rate is 6.62 % My main reason for the first post was trying to figure if a 260 oc is better then a 200 uc . 1)power 2)heat + safety 3)profit Now when I set this up last night power draw was 985-990 watts. It ran 11 hours with the garage door closed temps rose from 75 f to 105 f. temp seems to get stable at 105f I did find that my power went from 985-90 to 1033-1049 watts. the meter jumps back and forth with the readings. So temps may play a big part in power draw. I just opened my garage door and I will see if the temp drop down lowers the power down. I keep getting the feeling this gear will need too much care to be safe. For your farm with multiple units emmersion cooling may be best.  this is the unit with garbage door open letting in 43f or 6c air in 2 hours time temps dropped a lot and power dropped a lot  |
|
|
|
luck seems quite good --- makes me wonder about some larger pools and if they are being 'attacked' with selfish mining/block witholding.....
It's mathematically impossible to get more than 100% over a long period of time. It is incredibly lucky that this pool is doing so well ... at some point luck will turn. M EDIT: That's assuming someone didn't find a way to magically find blocks. That's highly unlikely (presumably impossible), of course. If so, it's the end of Bitcoin as we know it. M You can send 500th of dead hash (due to poorly written code) at a pool and hurt it. Was done to btcguild was supposed to be an accident. So if I was a huge attacker I could attack the bigger pools with false hash. This would benefit the pools I don't attack. Thus if I had 100ph of false hash on the network right now spread between every pool but kano's kano would have great luck. Would cost me a fortune to do unless I have a better hashing machine then anyone in the world. I don't . The only true protection against this attack is to solo mine. I do some solo mining at ck's solo pool. I mine here because Kano and CK deserve our support. |
|
|
|
at planetcrypto what is the watts for that. >>>> if you get 1600gh and I get 1280gh my way has to be a lot safer since the watts are low. I am working on low clock right now I am at freq 200 great hash numbers over 1280gh and error numbers under 1% >>>>>>>>>> I have 2 watt meters and they vary so much I suspect they are both wrong. one is too low and one is too high. my lower meter read 1025 watts. I run extra fans so this unit is pulling about 1000 watts to do 1250 gh . if accurate I am getting .8 watts per gh. the high meter reads 1100 which is 1075 after extra fans if acurate I am getting .86 watts a gh. it is about 40f outside I am going to open the garage door which will drop the temp from 104 to 80 I want to see if the cool air helps get better numbers.  |
|
|
|
edit----------the post above may be your answer.
okay try just using my pool info and none of yours.
and what sd card image are you using?
if my pool alone does not work I have to think you have a bad sd card image and / or a bad usb adapter
back to my under clock tests.
this is freq 210 the garage down is open gear is cool gives me 1260-1280
but power is lousy 1180 watts .925 watt per gh .. I am beginning to not believe the meter
I'm using CG minera build on a sdcard that I've been using for months without issue on the r-pi. I tried using the Rockminer, AM switch and reboot and get the same no hash. I've tried it multiple times with different boards and no change at all. The usb adapter sees the miner fine. The errors I'm getting appears that the BE chips have wrong addresses: my guess is that is out of date the tubes came out in aug if that card is pre aug the software is wrong. |
|
|
|
edit----------the post above may be your answer. okay try just using my pool info and none of yours. and what sd card image are you using? if my pool alone does not work I have to think you have a bad sd card image and / or a bad usb adapter back to my under clock tests. this is freq 210 the garage down is open gear is cool gives me 1260-1280 but power is lousy 1180 watts .925 watt per gh .. I am beginning to not believe the meter  |
|
|
|
All right. Testing now:
Pool shows alive with your info.
Changed r-pi to 1.1A power supply
changed to different white cable.
No hashing so far
* changed boards and no hashing on different board.
with full rebooting of the minera any luck?
|
|
|
|
Legitmate Question about calculations in the pool:
Can someone explain to me about shares?
One would think the S4 would have more shares than the S3's? I'm curious why not? Yes, the block percentage is higher than the rest but not the shares. Just curious why? Trying to learn.
The S4 is worker_17...
It's called variable difficulty. The S4 has a higher share difficulty than your S3s, ending up in fewer, yet higher value shares, submitted. M Hence, the higher block % based on variable difficulty of the shares [not hash rate], thank you for the answer. Appreciate it! it is based on both ( shares hashed) x (difficulty of the share ) 3 x 4 = 12 vs 3 x 1 = 3 in your case 1721 shares x 3,500,000 = .148 % of the block for the s4 about .00148 x 25 = .037 coins 1875 shares x 850,000 = .036% of the block for the s3 about .00036 x 25 = .009 coins all very rough numbers due to rounding this assumes they both hashed the entire block . |
|
|
|
I think I am going to put two up for sale in the market place. I don't have time to baby sit them.
Atleast you can get yours to run... What's wrong with yours? Here is my post in the other thread: https://bitcointalk.org/index.php?topic=791827.msg9552765#msg9552765Minera is running fine but none of the boards will hash either together or individually. When I tried running the boards individually, I made sure all of the switches were 'off'. Minera appears to be detecting the miner fine. Check the wiring. A to B Inside being b never used Pi but it's the same Set up from what folks have posted. Also check your pin settings. Have you done work on LED boards? No one board can have the same pin configuration. I'm trying only 1 board at a time. When I try the single board, I set the 5 dip switches all 'off'. I get no hashing from any of the boards. Minera appears to recognize miner though. (atleast the uart adapter) okay lets take it a step at a time. the one board you test has the fan plugged into it? so when you power your psu on the fan turns on. next you only need 1 white wire attached it goes direct from the board that is powering the fan to the little usb adapt which plugs into the rasp pi Ok. Good idea. The board that I'm trying right now has the fan on it. It is being powered by a corsair rm850 - so plenty of power. I have the 3 white wire plugged into the left white adapter on the board. The dip switch on that board is set all to 'off'. the 3 white wire is plugged into a r-pi via uart. and the pi is running crazyguy minera build. cgminer in minera set for am build. minera detects the uart / miner whever power is on the board. i have 2 pools set on minera, the default one and slush. both pools are alive. No hashing the powered board right now. I've tried these steps with all 4 boards and I get the same result. okay humor me on this go to settings copy my pool info in make mine first choice leave your choices 2nd and third. hit save and restart miner. if none of that works. use another white try a different power source for the rasp pi if nothing happens and you have a dead miner go to system menu hit reboot if none of that works. use another white cable. try every cable. if none of that works. try a different power source for the rasp pi. if none of that works you should check the one board to see if your all off are all on. after that i have no more tests to try.   |
|
|
|
I think I am going to put two up for sale in the market place. I don't have time to baby sit them.
Atleast you can get yours to run... What's wrong with yours? Here is my post in the other thread: https://bitcointalk.org/index.php?topic=791827.msg9552765#msg9552765Minera is running fine but none of the boards will hash either together or individually. When I tried running the boards individually, I made sure all of the switches were 'off'. Minera appears to be detecting the miner fine. Check the wiring. A to B Inside being b never used Pi but it's the same Set up from what folks have posted. Also check your pin settings. Have you done work on LED boards? No one board can have the same pin configuration. I'm trying only 1 board at a time. When I try the single board, I set the 5 dip switches all 'off'. I get no hashing from any of the boards. Minera appears to recognize miner though. (atleast the uart adapter) okay lets take it a step at a time. the one board you test has the fan plugged into it? so when you power your psu on the fan turns on. next you only need 1 white wire attached it goes direct from the board that is powering the fan to the little usb adapt which plugs into the rasp pi |
|
|
|
The Tower of Prisma
Solved a lot cooling issues this way and freed up some rack real estate in the process.
Mounted this way the boards are running about 60F cooler.
On to OC'ing them.
Sorry about the blur, was standing on a chair and I'm an old fart.
You said in this configuration they are running 60F cooler? What were your numbers before and after? My setup looks very similar with the box fan under it. Plus I have two fans on each prisma, the one stock drawing air in and the one philipma1957 suggested to pull air http://www.amazon.com/gp/product/B00A460TK6/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1 Those Silverstone Tek 140mm x 38mm Fans are really quite even in power mode. I have them all hooked up to a fan controller that GrapeApe recommended http://www.amazon.com/gp/product/B00DN3IT7M/ref=oh_aui_detailpage_o02_s01?ie=UTF8&psc=1 I have them all running at full power, the stock fans are loud, like a hairdryer as some people say. But if that setup you have going is actually running 60F cooler and the stock fans are cycling you just maybe on to something. With my setup minera is reporting 85-86F running @ 240mhz. Im just curious what your numbers were before and after. Also im not sure where minera is getting the temp numbers from, does anyone know? I saw a post on this thread where someone said it could be the RPI temp, so I don't have a clue. Now I have to go give the leaning tower of prismas a try, will report back. Also im a little worried of them tipping/falling over, so maybe secure them somehow from the top. I think the temp on minera is the rpi temp. I am running into an interesting problem. I need to be away from the miners 2 days at a time out of every 10 ten days for the next 100 days. roughly 20% of the overall time. based on the cap popping issue. which can be mitigated with cooling and a smoke alarm . I have been trying to downclock the miner to see minimum power / freq numbers. if I do freq 240 i use 1250 watts I get about 1417 gh = 0.882 watts at the plug per gh. if I do freq 230 i use 1215 watts I get about 1366 gh = 0.890 watts at the plug per gh. if I do freq 220 i use 1150 watts I get about 1263 gh = 0.91 watts at the plug per gh. if I do freq 210 i use 1085 watts I get about 1074 gh = 1.01 watts at the plug per gh. has anyone done the freq 210 and had the really surprising power efficiency loss? I am going to try this again since I can not understand it. |
|
|
|
A Prisma cost 1.47 BTC or $579, an SP20 cost over $900 with shipping. The SP20 is $258 away from being the same cost of 2x 1.4T Prismas.
My sp20 cost 197 usd more then my asic miner prisma. I over paid for the prisma as I used paypal. I like the prisma but I don't think I can run it when I am out of my home. I have a sick relative and spend 2 days at a time with him in NY about 80 miles from my NJ house. I tried to downclock the prisma to use low power when it is unattended and I have found that no matter how low I clock it. It uses 1000 watts. So at freq 210 it hashs at 1000 gh and uses 1000 watts or 1 watt per gh not good. So I bumped it back up to freq 240 I get 1400gh and 1245 watts. or about .889 watts per gh I can tell I do not want to run it if I am not home for 2 days or so. This is going to be the case for the next 3 months. If it would scale down and use 890 or 880 watts when down clocked to freq 210 hashing at 1000gh I would keep it. The sp20 should be safer to leave unattended. The sp20 should be able to downclock and not use more power per gh. If I leave it unattended and it scales down at 1000gh I am looking at 640 watts. vs 1000 for the asic miner Since my price difference is 197 usd. I think I can make up the price in the long run. |
|
|
|
with about 3 1/2 days to go to next jump todays numbers are: BTC = 387 usd at coinbase http://www.bitcoincharts.com/ >>>>>>>>>>>>>>> (+1.66%)Blocks 330130 Total BTC 13.503M Difficulty 39603666252 Estimated 40263173704 in 494 blks Network total 287388.314 Thash/s Blocks/hour 6.08 / 592 s https://bitcoinwisdom.com/bitcoin/difficulty >>>>>>> (+3.01%)Bitcoin Difficulty: 39,603,666,252 Estimated Next Difficulty: 40,793,804,930 (+3.01%) Adjust time: After 492 Blocks, About 3.4 days Hashrate(?): 287,734,941 GH/s Block Generation Time(?): 1 block: 9.9 minutes 3 blocks: 29.7 minutes 6 blocks: 59.4 minutes Updated: 8:50 (1.5 minutes ago) |
|
|
|
go to https://bitcoinwisdom.com/bitcoin/difficultyput in your hash amount it gives you what that should earn in a day. 1th should earn .0127 btc a block is 25 btc divide it by the .0127 you get 1968.5 not quite right due to rounding but easy to do. so a s-3 at 450gh should earn .005714 a day or 4375.218 chance for a block today. I am running 8 s-3's and 1 asicminer prisma or 4.7th this is .05968btc in a day or a 418.9 chance at a block today |
|
|
|
well my unit is in my garage. on a woodbench near the garage door. it is 40 f outside in NJ door is open with a fan puling in air  note temps below not sure it that sensor is in the rasp pi or on the miner. I will now shut the garage door. it is 11:11:11 am in NJ  door open fan blowing in.  |
|
|
|
|
Show Posts
