Topic: ANTMINER S5: 1155GH(+OverClock Potential), In Stock $0.319/GH & 0.51W/GH - page 258. (Read 451266 times)

The C1 only doubled the hashrate because they doubled the boards. 4 boards to the C1 vs 2 boards for the s3. They could do this because of the ability to fit 4 boards into the existing cases of the s3's with only minor alterations. Now if they were to double the boards(which would double the chips) then it could be a 2TH/s machine, however that is a lot of heat in a very small case, I don't think it would be possible to dissipate the heat fast enough with 4 boards so close together. The chips might be fine but the rest of the PCB would probably be toast quite quickly. If one were to mount the boards into something like an S4 case, that would allow enough room to create in theory a 3-4 TH/s unit with a watercooling block sandwiched between 2 boards. I would say 6 boards total for approx. 3.3TH/s keeping in mind the need for 2 separate PSU's to power the boards, and enough room for cooling lines to be run, fans to cool the PCB's and an extra controller (unless they develop one that can carry more then 4 boards) At max I could see potential for 3.8-4.5TH/s machine setup as the chips can be chained up to 256 chips, but I think that would take a lot more R&D. Also I don't know about pricing. It would be really cool to see something like this, however the cost would likely skyrocket with this type of setup. One can dream though.

If bitmain cared to donate enough boards to me, I could see what I could come up with......... 

it likely would be 1.6TH/800W as 800W was about the amount of heat the C1 could handle. 2.2TH/1.1kW would likely be too hot

please make it double the s5 hashrate, as c1 double the s3!

Hi sk8erskid420,
Partner received the PCBs and control boards already, is developing the liquid cooling solution. Wait for update...

Hi Gents,

What is your average ambient temperatures?
Here in South Africa we have been experiencing temperatures from 27-39 degrees C.
My S1 was running at 51-54 degrees. I read that the units should not reach 60 degrees.

I was looking at the S3+ units here in ZA. They are selling for $478 (R5200).

So that made me think I should start looking at the S5.
What temperatures can be expected from this beast?
What is the average and max temperatures for this new unit?

I had a thought, now please remember this is just a thought and probably not to practical(cost efficient) for the home miner, but what about a waterblock cooler, something similar to the C1. If the heat is dissipated by the top of the chips then this should be an easy application, perhaps not by us, but perhaps bitmain in future releases. By that I mean it wouldn't be worth the cost for a home miner to buy all the required parts to watercool. if they could machine a waterblock that could intake/exhaust the coolant at the top of the block, perhaps intake at the top back and exhaust at the top front or something like that, then one could still stack sideways, cool extremely efficiently, and run a full loop system that could cool perhaps up to 3-5 units per radiator(haven't done the math, just spitballing). In theory I would guess that at stock speeds, this would keep the chips running under temp compared to a fan setup. Also as a bonus I would think a single fan used as a booster cooler to cool the interior of the unit would be an idea. Just remember this is only an idea that popped in my head so please do comment, I'd love to hear what others think of this, Good or bad. I'll prob post this in a few S5 threads to get some extra feedback.

Philip1957 wasn't too keen so far either.

other than Pat has anyone else gotten any antduds?

Bitmain always seems to have a poor start with new models and then improves with age. Hope the S5 goes ok for you Xian

I've got a bad feeling about S5 B1's... Fuck that's bad. Surely it was merely "damaged in shipping"

ಠ_ಠ

My S5's just landed in Cincinnati, so with any luck, I should have my own review up by Wednesday.

Don't turn the BBB around.. lol..

I agree with klodike, everything he said.. take a breather.. then I would try swapping the ribbon cables, and pay close attention to the orientation.  move the cable from the good board to the bad one..


Rich

take a break. sometimes ideas come to you. My post was just to make sure you eliminate a bad ribbon cable or bad connector on the controller from the possibilities. Otherwise its likely that on the bad board the fault is in a component that drives both chip chains, as it seems strange that both chains are dead (ie: less likely that at least two hashing chips failed in a short timespan)

2015 will present new opportunities, but may have to wait until Q2 before they get released. Tipple return? Not likely.

$.09 kwh western Wa.  But even then I within a month was making nearly half the btc then the prior month.  I'm not concerned with the current price of btc but more over acumulating it in a cost effective manor.  Curse these huge farms.  Antminers are rock solid in my opinion just can't see much value in pud cost of mining unless new tech comes out and doubles or tripples returns.

Ug. Don't say that - I've disconnected it again now......

I'm pretty sure it's the hash board, but I don't have a multimeter to do what you ask I'm afraid. I've disconnected & reconnected the cables so many times now I'm sure they're worn out..... Seriously though, I've been messing with this for hours - I think we should either try the replacement hash board (with new cables?), or I can box it up for you if you send me a label.

Edit: Just read your post. Right, well if you think it could be a shared component fault - that's either the BBB or the controller board? I didn't try turning the BBB around - just made sure it was properly seated & there were no dry solders anywhere - it & the controller board seemed to visibly check out though. I can go over it all again if you want me to, but it won't be tonight I'm afraid, up early tomorrow & it's getting late. I'm also getting tired of looking at it tbh.......

Have a think & let me know, I gotta eat.....

Best share = 0?

so to check; have you tried flipping the BBB around and using the working ribbon cable from board one to attach board two AS board #1? this would eliminate any issues with the controller connections or the ribbon cable to the miner.

beyond that, if there are no visible faults on the bad board, a multimeter and some poking around for 0.8V (& 0.8v multiples) and 12V test points may identify whether the DC voltage isnt getting beyond or to a particular component. Seems strange that both chains on the board went together, i am wondering if there is a shared component fault

hummmm  Thank you for trying out!

I'll be back if I think other options...

This was all done previously, but thanks anyway 

OK - here it is connected to the RM1000 using four separate pcie connections - no difference. I then swapped the data cables over & tried again - the same board was still dead, so it looks like that's the problem eh?