 Show Posts
|
|
Pages: [1] 2 »
|
That would (continue to) be outstanding! Yes, please.  So we are within the 72 hour window to go gangbusters? Production heatsink is 83x100x40mm. Board dimensions: https://www.dropbox.com/s/gkuayv84tr63x70/BitBurnerXX_rev3.pdfI could provide a iges/step files of the assembled board if that helps. Todays status: Found a nasty bug that produced the flood of nonce errors. And I fried one part of that chip that i planned to replace tomorrow. - just another reason to change it. But it meant that i could only run it with one chain today. It was pretty stable though, about 2% HW errors. But with the replacement chip that should hopefully go down to 0%. And then I can pull the trigger on the website and the PCB Orders. |
|
|
|
I think Sophokles answered your question. Links: http://tinyurl.com/m4zy7p3 <-- this was posted elsewhere, maybe in this thread IIRC. do the chips really transfer more heat into the PCB than to the upper side?
Yes, they are designed this way. And you shouldnt put a heatsink on top. Yifu said this will have a negative effect instead. Can anybody verify this? On computer mainboards. (NB/SB , MOSFETS) or RAM you can cool with out problem from the top. The differential temperature in one chip should be less than 1°C ... i can not image that this would have a negativ effect. The only can explain the negativ effect throuhg wrong/to high mounting force. (so that the chip will break) |
|
|
|
No that's not the Production heatsink:
Am working on a case/cooling solution, are dimensions of the production heatsink known? |
|
|
|
I hear mineral oil brought up in discussion so many times, but these sort of solutions are surely just more trouble than they're worth.
Edit: I'll see your tad of unqualified FUD and raise you an iota of qualified optimism: Read the links that discuss this. Cooling electric/electronic systems by direct oil contact has been in use for decades. Computer users have historically not done so, likely because there hasn't been a need. The likes of Intel have done studies spanning a year of nonstop server operation. Follow up failure analysis indicated hardware was in fine shape. The benefit was significant operational cost reductions, primarily in the form of reduced air conditioning costs. There are others who have had similar results, these studies are interesting reads. The heat captured still has to go somewhere - if air cooling isn't enough, then waterblocks are a far more sensible and proven solution if you insist the stock cooling is inadequate for your needs.
Yes, the heat capacity of water is the best. Cheap, widely available. It is one alternative. Indeed, thermodynamics tells us that the heat must go somewhere. In this particular case, away from the ASIC chip is all that matters. There are many ways to do this. Water is one, oil another. The combination of the two... or just air. Creativity is the only limiting factor here. Yes, but what about overclocking I hear you say. 450MHz dare I say it, is a fantasy figure. Theoretically possible according to Avalon but based on pretty much all real world overclocking experiences the absolute maximum is typically impractical for either stability, safety, energy consumption, chip longevity etc..
From what I have seen on the forum, the OC attempts were performed under air cooled conditions. I may have missed a thread, if I am mistaken I would be interested in reading it. If all these OC studies were under air cooling, these are not the conditions that Yifu (BitsynCom) identified that 450MHz was achieved at Avalon. Your point is taken, I would like to see more info from Avalon regarding that study (duration, conditions, errors thrown, etc.). The only info offered was that "450" was not achieved under air cooled conditions. If the majority of Avalon chips were capable of this, for what conceivable reason would they have limited their clock to 285 MHz in their own units.
[...] What about the fools that try 450Mhz on air cooling?
I think you answered your own question there. I also agree with you: we need to get these boards in hand as soon as the PCB designers have achieved their design aim. |
|
|
|
Thanks Silv0r. Are there any reliable estimates for amp/power requirements at that speed (within a tolerable range)? Because I have a small brain, I am not clear what this means: was wondering how long it'd take people to notice ( and more importantly share the constant that we've released on github.) the number you are all aiming for is 450  of course, that's not really possible on just air cooling. Is the 450 value clock speed, MH/s, GH/s??? Is that value for a 3, 4 card Avalon unit, a per chip over clocking value or a per chip hash value? The discussion in this thread seems unclear. Per Chip. It is the frequency. 450 MHz. But from overclocking 282 MHz to 450 MHz this would be a sick overclock. Even with active watercooling i.e. O.o But good luck! |
|
|
|
Because I have a small brain, I am not clear what this means: was wondering how long it'd take people to notice ( and more importantly share the constant that we've released on github.) the number you are all aiming for is 450 of course, that's not really possible on just air cooling. Is the 450 value clock speed, MH/s, GH/s??? Is that value for a 3, 4 card Avalon unit, a per chip over clocking value or a per chip hash value? The discussion in this thread seems unclear. |
|
|
|
Hopefully the power requirements for 450MHz clocking is being considered for both forms of the board? burnin the XX model is looking great, is the X version coming along?
My personal guess is that he will use the same PCBs but only put 10 chips. Controller will then try to run both lines, fail and only run one stack of chips. That would cut development costs and he would get cheaper PCBs. |
|
|
|
[...]
Given the overclocking the 1U I propose will need 700W and the oil submerged system, with heat exchange and an external cooling tower. Given the heat in Indonesia this might just be the most cost effect system in the longer term.
[...]
I do love some of the case ideas here... especially the CUBES.
Interesting info on oil choices: IEC 60422, IEC 60296 (Google search should turn up some useful links) and http://www.power-eng.com/whitepapers/2012/shell.whitepaperpdf.render.pdfI think oil and cubes design could be integrated. If you couple the heated oil to water, you can generate hot water and save on that as well. Some ideas to ponder from the world of power transformers http://www.electrical4u.com/transformer-cooling-system-and-methods/Not to be contentious, but Sebastian I think you are mistaken regarding the operation of equipment 24/7. Intel and many other data centers have done exactly that. Intel did a one year pilot study and follow up failure analysis indicated nothing of mention. There are certainly other issues in terms of maintenance, which is where good design can make a significant difference. Modular mining units (the blade designs appeal too for this reason) that have snap on feature seem like a good idea IMHO. If you search for the Intel pilot study, it is also interesting to note the significant decrease in power consumption. Considering that these miners will have a carbon footprint the likes of Godzilla rampaging through downtown Tokyo, this is a good thing *financially* for miners as well. Less cost up front. That being said, there is the oil and despite the videos demonstrating ease of service http://www.youtube.com/watch?v=U5zoIEjo1Zk there's still the oil. The fewer small parts to drop, misplace, recover from the bottom of the oil reservoir, the better. Lots of design opportunities here... I don't think ordering a GRCooling Carnojet system is necessarily the most prudent solution for everyone... as pointed out previously the shipping costs are likely crippling. |
|
|
|
See page one of this thread. Two bullets of interest. First is about halfway down describing how/when website will go up. The second is a link at the bottom so you can register for newsletter. is the website from burnin online ?
|
|
|
|
ah.  I was mistaken and thought there were more pages to the pdf, not that you were referring to the thread page. Am looking at oil cooling, some interesting options to consider. This size info will help for the enclosure. I am wondering whether the 2 year warranty burnin is offering will be voided by this... Regardless, thanks! bitcoiner49er: If that link is for p. 29, where might one obtain the entire doc? Page 29 of this thread |
|
|
|
bitcoiner49er: If that link is for p. 29, where might one obtain the entire doc? |
|
|
|
If CC/Paypal was added, you would have a large advantage over competition and questions of legitimacy would probably fall off substantially.
But then we'll have to fix price in USD, something we don't want to do. Only initially, whilst you build reputation and trust... Think of it as an insurance policy due to the volume and frequency of previous scams. People are justifiably wary and don't want their fingers burnt a first, or a second time. We are willing to trust you, but you need to build it. We're not denying that dollars aren't the favourite medium of exchange, but in this initial instance they are preferred with a VISA credit card 120 day protection. Also most miners are new currently, don't have the coins mined to pay you, and exchanging is messy right now. Terrahash: The rationale to not escrow through John K. remains unaddressed. Simply declining to escrow without offering a rationale seems amiss, especially since it would actually benefit you in terms of credibility. It could still remain in BTC, it would offer buyers a degree of confidence in an unproven product/service and I am sure the small fee that John K. charges is not a barrier to purchasing for those preferring fair terms of trade. |
|
|
|
|
SUP; FullFathom5; flyonwall
|
|
|
|
In their defense I emailed them a question and they responded within1 day.
If you think the content of that exchange would make a meaningful contribution to their credibility, please share it. Otherwise your statement is, like many in this thread, purely sentimental. Not to their defense: they cannot respond to legitimate questions about their business. |
|
|
|
But more importandly as the opening of the website draws closer i created an email newsletter to alert you people when there are important Project Updates. Register here: http://eepurl.com/zNH7vAttempted to sign up for this list but did not work for me. Is this list still up & active? Worked for me, just now... Hah! Worked for me twice apparently, but each time the web page gave error that the list had "disappeared". All good. Thanks for feedback. |
|
|
|
Amir has 13 posts and he is providing a service ... better be careful.. he does not accept Escrow ..
While I am not keen on FUD (however well intended) I do have a couple questions that Terrahash should address. You are offering assembly, which means you will need a certain amount of resources (space, equipment, personnel). How are you going to accomplish this when the last two business addresses you have claimed for Terrahash are vacant? The first address you listed, 1261 Birchwood Drive, Sunnyvale, CA 94089 has been vacant to over 6 months, confirmed by realtor. The second address, 1800 Wyatt Drive, Suite 15, Santa Clara, CA 95054 is also vacant, confirmed by realtor. Edit: When I took these photos, I also called the numbers listed below to confirm the history/vacancy of these properties. Links to realtor contact info for both properties: http://imgur.com/iBIzscYhttp://imgur.com/nPc93SYTerrahash please explain. You offer no escrow which John K. on this forum offers for reasonable fee. This starts to look like a questionable enterprise. |
|
|
|
But more importandly as the opening of the website draws closer i created an email newsletter to alert you people when there are important Project Updates. Register here: http://eepurl.com/zNH7vAttempted to sign up for this list but did not work for me. Is this list still up & active? |
|
|
|
In his thread he only said he contacted him. Is this the plan now official or only checking out? Fair questions. From what I read on the announcement thread: >snip< 1. There is no definite physical location as yet. We are scouting for a location that has the lowest cost supply of reliable electricity and 24/7 security service. Right now it's looking like ASIC Hosting Service has the best option. This should lower electricity costs drastically. We are signing up with this ASIC Hosting Service. 2. Once a service agreement with the ASIC Hosting Service is ironed out, we will publish here the recurring costs. Some people I'm sure are waiting for this recurring cost to be nailed down. >snip< Bullet #1 starts uncertain, ends strongly with them signing up with ASIC Hosting Service. Looks like they are detailing the service contract. Don't want to clutter the thread, just to keep it current. |
|
|
|
|
