This cabling looks really amateurish. Sorry for their loss, but if power lines were done like this Ethernet cabling no wonder problems occurred.
The post-fire pictures are taken from the back, and the power cabling does seem to be better organised (though bunching them together neatly may be worse, since it puts all the warm conductive cables tight together so if one melts they all burn together) I have no clue why the pre-fire picture couldn't have the ethernet cabling nicely clustered and numbered, running in straight lines with the shelving rack - it would be tidier, safer, and easier to service. |
|
|
|
This project is proven to been feel interesting. Yes, right now my cabinet is 3.5kw capacity. The question is how capacity should I upgrade my design. Talking from point that I have pictures shown as working production for another application already. So I like to find a spot in bitcoin mining. But the one I have does not fit the capacity requirements yet.
Thank you for your advice mate
I am racking my brain to understand what situation someone would ever use a 20kW contained-cooling rack. Usually these are used with 1-2kW of gear as a sound-dampening enclosure, not meant to actually cool a large amount of power. OPTION 1: You put 20kW of equipment in an open rack or on shelving. It costs <$400. 20kW of heat is blown out the back of the cabinet. OPTION 2: You put 20kW of equipment into a sealed enclosure with a 1.3 PUE AC built in. It costs >$5000 and weighs >300lbs when empty. 26kW of heat is blown out the back of the cabinet. oh, and option 2 is a little quieter, you only have the noise of a 20kW-capable air conditioner. Seriously - this is a product with no market. Liquid Immersion cooling tanks are the obvious alternative, and are quieter, FAR more power efficient, and can move the heat outside of a building by pumping the hot liquid to a radiator located on the roof. |
|
|
|
I think that would be awesome. I'm already envisioning a tank-less water heater and hour long scolding hot showers. Powered by bitcoin.
Didn't Gavin talk about something similar about a year ago? He was saying that if/once bitcoin becomes popular and Moore's Law is reached or close, that people would buy a hot water heater, furnace, heater, etc with some ASIC chips built in to earn a few dozen satoshi. = ROI on electric bills. free energy for everyone. Bitcoin's end game. awesome. not happening anytime soon. The cost difference between building a mining system into an appliance vs a basic metal heating element likely exceeds what will be mined in its lifespan. $500 computer wedged into the heater, that runs maybe 8-16hrs/day, or a $20 metal coil? Think you'll mine $480 worth of BTC? +hazardous materials at disposal +if you lose your internet, mining stops right? then what - no heat? |
|
|
|
also think antminer s4 cost $1,250 without shipping and taxes, so that is around $9,000 for 6 units
then if 30kw per rack support 20 antminer s4, where you can put in max (42/3) = units
for Spondoolies SP35 would need 3.5kw per unit of 2U. then one rack can fit 21 x SP35 = 73.5kw which miner would cost $83,580
what the hell are you talking about? The specs in the first post state that the maximum cooling capacity of the upgraded-level cabinet is only 3.5kW. Thats a single SP35 sitting in the cabinet, and add about 30% to that to cover the costs of running an AC unit to handle the heat. ADD ANOTHER 30% to that to cover the cost of running an AC in the building that contains the cabinet, since the building is now dealing with both the SP35 heat PLUS the heat created by running an AC in the cabinet. 3.5kW*(1.3^2) = 5.9kW of electricity used per cabinet that holds only 1 SP35, and dumps all the heat directly out the back (which is what an unenclosed SP35 does anyways) |
|
|
|
looks a lot like the TPS53355 regulator in this picture, with two resisters right by it - if so should be quite easy to pencil mod http://www.ti.com/lit/ds/symlink/tps53355.pdf - look at the bottom image on page 7, the reference on page 20, and figure 38. TLDR; Pin 1 has two resistors on it, R2 goes to ground and R1 goes through a small circuit. The choice of values for these gives the output voltage. This is how the Bitfury overvolts, and how the S1 can over/undervolt. looking at the rockminer PCB (would love a high-def image of the section), this is the solution to undervolt for efficiency:  some pencil lead on R2 increases the output voltage, while lead on R1 would decrease it. Presumably the efficiency could be brought closer to the 0.7w/GH achieved by the prisma this way *note*: Im basing this on a blurry image and dont have a PCB in front of me to confirm that Pin 1 is the right corner of the chip, but the positionng of the two resistors like that seems logical. hopefully ill try myself on monday evening update: I brought a unit home, and am unable to determine any markings on the regulator component. looks an awful lot like the TPS5355 though, and as there arent many >20A alternatives I would assume that it is. however, my expectations were not correct. pencil modding either R1 or R2 (or any other resitor in the vicinity of the regulator) has no visible effect. It appears that R1 and R2 in my above image are both brought to ground together. HOWEVER - I made an interesting discovery - pencil modding BOTH R1 and R2 results in the output voltage increasing, with 1.15V achieved quite easily using a 2B pencil (stock is 0.75V). Obviously this is the opposite of my intentions though. UPDATE 2: Partial destruction of R1 and R2 (chipping the sides of the resitors with a sharp tool) so that each is roughly 3.3kOhm when measured in circuit results in a regulator output voltage of 0.6V Testing now -> preliminary results are mediocre at best - 16GH. However, its worth noting that this unit has always been a source of problems, and that two other boards in it report 25GH and 58GH, neither of which were modified in anyway. Not sure if its a USB hub issue, RPI, or maybe even something to do with the server PSU (providing 12.03V) UPDATE 3: replaced a controller on a different board that was causing reboots and often not functional -> 110GH,90GH, 105GH, 70GH. Obviously there is quite a spread here but it seems like the modified board is functional. Because of difficulties operating (or lack of) at 0.6V, pencil mod was added to increase the voltage to 0.7V on the modified (70GH) board. Reject rate is 6.5% at 300MHz UPDATE 4: 12hr stats: 112,112,111,66 GHash across the 4 boards, with the 65GH board being the one i modified. Total reject % is 4.3. TLDR; Pencil mod is not easily implemented on the RK-Box. reduction of chip voltage to 0.7V significantly impacts hashrate, and voltage below 0.65V is often non-functional. No efficiency numbers available yet |
|
|
|
at this point, USB miners are worthless other than conversation pieces
Imagine a phone or device with an ASIC chip that mines continually and are embedded in billions of consumer devices. USB miners are worthless. The way we break the farms and get 'decentralization', as indicated in the thread title, is not with USB's but with embedded ASIC miner chips in devices already cheap and in mass production. Think coffeemaker / miner that is wifi / LAN enabled. Waste miner heat could help keep coffee warm. ^not going to happen. The heating unit in a coffee pot is a $2 metal coil with an on/offr switch. to put an ASIC in there would cost $50 of ASICs, controller, and wifi gear. operating for maybe an hour a day - you will make an absolute pittance mining, FAR FAR less than the costs of the specialty coffee pot (which would stop working if the internet connection is lost!) |
|
|
|
It's unfortunate that they called it account, because it connotes with bank account.... but, a bank account is really a bank deposit account. Account really means some numbers that you track, and deposit means that you have given your money away. So it is not really bad to call it an account (as in BIP-0032). It is a way to arrange your numbers.
Gives us a splendid pretext to argue what an account really is, and what a deposit really is.
de·pos·it verb: deposit; 3rd person present: deposits; past tense: deposited; past participle: deposited; gerund or present participle: depositing 1. put or set down (something or someone) in a specific place, typically unceremoniously. 2. store or entrust with someone for safekeeping. |
|
|
|
Another thing that's wrong with these devices is that they should be able to be run from a laptop power supply brick - so 12-20V input or 12 x root(2) V (12v lighting transformer)
I'm not sure what you mean, but they do run from a power supply brick? My criticism of this device from Bitmain is that the cost is such that it can never be recouped. This is a coffee table device, a conversation piece, a piece of art, but it will never make the purchaser money. If Bitmain really wants to support the home miner, the design has to be drastically simplified so that people can make a working miner with parts from their junk box. To wit, no power supply should be included plus the purchaser provides a heat sink and cooling. It should be a barebones device which can run from a range of voltages (12V from an old ATX power supply or 18.5V from a laptop brick - in other words whatever is to hand), it should be available with an order quantity of one and the price should be such that no VAT or duty is payable. Delivery should be low cost. I don't know what the market for such a device would be, but considering that the Raspberry Pi has sold 3 million, there is evidently considerable demand for products for hackers. sounds like an S3+... at this point, USB miners are worthless other than conversation pieces, and for a reasonable $200 you can get a machine that is still quite small, very quiet, and has a fighting chance to produce a positive return. |
|
|
|
PBmining is the best priced offering by far but it's also a ponzi scheme. Before they implemented a coinmixer it was easy to see the coins go directly from new customer payments to the "mining earnings" wallet. There is no way that they can give you a 5 year contract on any hashing power right now with no maintenance fees. The power bill alone would sink them several times over. I believe "Ponzi scheme" is putting it mildly. Also, it is located in Canada but will not sell to Canadians. Go figure. ^is it really? I saw that disclaimer on the site but never considered it was actually a canadian company. That just ranks of the mentality that someone scammed outside of canada will have little chance of retirbution, whereas if a canadian is wronged could easily take them to court |
|
|
|
any way to undervolt the RK-box for better power efficiency? I would prefer to be around 1w/GH even if i lose 10% hashrate
nothing? I might pull my R3 unit out of hosting to do some tickering and see if undervolting is possible Not that I know of anyway, I didn't think BE200 boards have voltage control. I'm still having issues with my pair of rk-boxes (rockminer was extremely helpful though and gave me a cubie to try and a r3 compensation), so i think i might try to put all the troublesome boards on a single cube (or on the r3 frame) and bring it home from my hosting spot in order to look at closely and try tweaking. I imagine it uses voltage regulators, and not just chips in series, in which case something may be viable - or reducing the 12V source to 10V?  looks a lot like the TPS53355 regulator in this picture, with two resisters right by it - if so should be quite easy to pencil mod http://www.ti.com/lit/ds/symlink/tps53355.pdf - look at the bottom image on page 7, the reference on page 20, and figure 38. TLDR; Pin 1 has two resistors on it, R2 goes to ground and R1 goes through a small circuit. The choice of values for these gives the output voltage. This is how the Bitfury overvolts, and how the S1 can over/undervolt. looking at the rockminer PCB (would love a high-def image of the section), this is the solution to undervolt for efficiency: some pencil lead on R2 increases the output voltage, while lead on R1 would decrease it. Presumably the efficiency could be brought closer to the 0.7w/GH achieved by the prisma this way *note*: Im basing this on a blurry image and dont have a PCB in front of me to confirm that Pin 1 is the right corner of the chip, but the positionng of the two resistors like that seems logical. hopefully ill try myself on monday evening |
|
|
|
Just about the start wiring the PCI-e plugs for the Server PSU. Could someone just quickly on confirm the plug wiring layout below is correct for the miner, thanks Black = Ground Yellow = 12v  thats correct. |
|
|
|
Thank you. Going to give it some thought, BTC's low price being the only thing that makes me hesitate. That SP35 would be nice.  yeah i agree these guys have been good to me (gave me a refund even thou technically they did not need to do so at least legally) feel I owe them a product ...just at 15c/kwh and hosting fees can't get the numbers to work at these btc prices (or the prev units not really good for my home use) well maybe the next lot then ... another.generation or some such I can figure out at least how to kinda float a miner from them.....the NO pre-order plan is helpful on me jumping the shark on this....pretty much no more pre-orders of asic equip for me (no refund no roi Titan) Searing 15c/kwh alone virtually prices you out of the mining game - hosting fees atop that must hurt. I pay $0.15/kwh but have free hosting in the back of a small business, and I still can't justify most current hardware. @SP - will there be a price break in the SP20 soon? at $700/TH at 0.7w/GH its almost twice as expensive as antminer S3 units. I understand its much smaller and the software platform is much better, but that really prices it out of the market for most home users. |
|
|
|
Recheck the resistance. It is a giant pain in the ass, but you need to adjust it every once in a while.
yep, that did it, back into a good range now - thx I find it takes about 3 days to fully settle as the graphite oxidizes/laminates/whatever chemistry occurs. Generally the resistance will drift downwards by a few percent |
|
|
|
why small fans? so loud
If you went with two (maybe 3) large fans, the failure of ANY of those fans could cause catastrophic overheating as BE200 doesn't have temperature sensors. By having 6, a fan can fail without problem. tldr, redundancy. quality fans dont crap out in my experience - and the AM Prisma only uses 1 fan itself anyways |
|
|
|
Looks like my flash card got corrupted when the power went out recently. Won't reboot. This happened once before but it was a long time ago so I don't remember the procedure to recover. Thanks for any help!
use a class 10 card when reimaging. It will survive power blips that often crap out the class 4 cards |
|
|
|
I feel like they could do advertising alot better, ex tell sig campaign members to have two adds one for the prisma one for amhash, ive seen it in someone profile....and they arent even getting paid
because seeing every other forum member with an advertisement in thier signature already wasn't at all annoying. I see around 5 links for the Prisma on almost every single page of every single thread lately - its extremely annoying and IMO should be banned by the forum moderators (not that they do anything) |
|
|
|
24 hours left...will the SP20 tracking number be received? I'd bet NO.
We are still working hard to ship out all the SP20s by tomorrow and will continue to update you. You are one angry miner... You mean about the fact that 3 days prior to their shipping deadline they had no clue if they could ship or not? And 24 hours before there is no additional news? I may be angry, but it's better than being a pushover who just accepts these poor business practices. think it was reference to the fact that spondoolies just promised that they will likely ship out all the SP20 orders by the end of the month - 1 hour before you said you dont beleive them |
|
|
|
any way to undervolt the RK-box for better power efficiency? I would prefer to be around 1w/GH even if i lose 10% hashrate
nothing? I might pull my R3 unit out of hosting to do some tickering and see if undervolting is possible |
|
|
|
Not the answer I was looking for unfortunately, but thanks for your reply.
I will add that our estimations on RB and systems preformance were honest. It wasn't a deliberate tactic. I did learn a lesson here. We also found the root cause, which we'll make sure not to repeat on future gens. I'm sorry that I'm vague here. We'll start another round of refunds (and complete the first one, we have problems with small number of international wires) soon. Guy should work on the underpromise/overdeliver model a bit. By comparison, almost every single bitmain product has been capable of acheiving 10% over spec S1: 180->200GH S2: 1TH->1.15TH S3: 440GH->500GH S4: 2TH->2.2TH |
|
|
|
so, whats the noise with these watercooled miners?
Depends on what you are using to cool your radiators. If you have good fans, it shouldn't be too loud. could you give me an order of magnitude plz? im quite interested to just have one of these to show of in my one room appartment .. ^^ With the stock fans, it will probably be the same as the noise of 2 s3 units - which is pretty reasonable, around the same volume as a desktop pc playing video games. Noticeable but not excessive for a living room If you swapped some quieter fans, it would be pretty quiet, just a light whisping noise of airflow through the rad |
|
|
|
|
Show Posts
