Topic: [ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs - page 123. (Read 1260226 times)

Post your image link between the square brackets "][":

[img]"here"["/"img]

thats about ideal. if you see more than 2-3 running at 125C you might need to tune back slightly, but temps seem fine. *slightly lower temps are slightly better for the system

Don't worry, the one at 125 will get throttle down to lower the temp.

Hmm most of my asics are around 85 to 100 Celcius, got some at 115 Celcius and one is running at 125!!

we were talking about asic temp readings (visible on ASIC stats pane), not exhaust, although exhaust 79 is high.

How about the SP30?
I have it running in an air-conditioned but small room, at "slow fans, medium rate" setting (Fan Speed 60 Start Voltage  0.65 / 0.65 Max Voltage  0.73 Max Watts  1100 / 1100), my temp readings are Temp Front / Back T,B 32 °C / 68,79 °C. I had it running on "medium fans, high rate" for a few hours, but the last temperature number was around 85, and for some reason the miner restarted by itself (maybe reached cutoff temperature?)
And what do "T" and "B" on the back temperature mean?

Indeed !

jtoomim - Most of what you just said is stuff I've either looked at, am currently designing on or am currently prototyping.

Cool. Thank You all.

So if I keep it constant at about 80-90C it's still more than adequate.
Very nice. 

I've been thinking about ways of doing string based miners but with a adjustable SMPS-like bypass, such that over 90% of the current would go through each of the ASICs in series but a small proportion gets shunted through a P-MOSFET and then an inductor before continuing to the next ASIC. The alternate input path for the inductor would be from Gnd, probably with a diode to prevent current flowing backwards when the inductor isn't charged. Opening the P-MOSFET would drop the voltage across the ASIC in question while increasing the voltage across the rest of the ASiCs in the string (both above and below). That gives proportionality control for different ASICs, and gives you an option to bypass ASICs that are causing problems without bollucksing the whole string. Total power going through the bypass path should be small in typical cases, making it impact efficiency little. However, peak power can be high, so the parts cost would pretty close to doing a standard VRM design. An alternate approach would be to have a binary 0%/100% shunt approach to bypass bad ASICs, and that could probably be made pretty cheaply.

If you want to change the voltage applied to the whole string (i.e. the average ASIC voltage), the place to do it is on the PSU/AC2DC converter. Unfortunately, most PSUs do not make that functionality readily available. I know at least some models have a trimming pot inside that lets you change the voltage. I'd guess that this is true of many, and the ones that don't have a pot could probably be trimmed with a pencil mod on a resistor. What's really needed is a PSU with voltage that's adjustable by i2c. Probably exists and isn't cheap, for no good reason other than weak demand.

sp20 shows real chip temp while real temp in s5 is actually at least 20-25C higher than what you see on the interface
if you choose auto and high enough power and voltage max limit, miner will bump the temp and drop the fan until the highest temp reaches 115C, suggesting that it is the highest steady state temp spt is comfortable with.

Not off question, most of the previous page has been off topic really

Asics will self regulate and throttle back when they hit 125°C so you are safe to run them up to 110 - 115 really but the cooler the better.

The once furthest from the intake will allways be hotter in my experience.

Off-question, but what are the 'safe' temperatures of the SP20?

When looking at the Asic stats page, LOOP3 is always the hottest, even if I set the max power supply watts lower than the others. (At 150 now).

Seems to like to climb to 80C. The others are in the 50's and 60s.

Thanks

Yeah I was right there in the first 20-odd pages of the S5 thread questioning the particulars of the design and reliability, how they handled certain likely issues and stuff. Prismas made a mediocre showing for string design, but the BE200 wasn't built from the ground up to be used in that topology like the BM1384 was (or the BE300 would be).

We've needed an economical home miner for two years. There's never really been much in that market sector that really suited every need (dB volume, price, efficiency, etc) exceptin' maybe AM Tubes, the S1, S3 and a couple others. Not enough, anyway. SP20 was really good gear, especially when it stopped costing $1000, so it's sad SPTech probably won't be doing any more machines like that but honestly it's never been their priority.

By many I meant quite a few other people who were regular on this board posting that S5 string design would never work, will be constantly throwing whole boards off, etc, etc. Largely, it did not happen.

I think that we need an economical new home miner before 21e6 will introduce their product, whatever it is (I strongly suspect that it is a router).

What I mean is, his revolutionary design was a string miner. I was being sarcastic about it being revolutionary, since folks had made BitFury string miners a year before. Strings are good because of fewer parts count and increased electrical efficiency, but the lack of voltage regulation and adjustability sucks. So there's still room for improvement.

sure, better design would be beneficial, although many were skeptical re strings and they seem to work OK (at last in case of S5).

No I didn't. You may notice from reading that thread that I was cynical about his design, given that when he posted his revolutionary idea there had been several other string-based miners already come to market. Between Novak and I we have enough knowhow on the hardware and software to do what we want to do. Bitmain hasn't shipped chips yet because I forgot to send them payment yesterday.

Regarding PSUs, if he's running 1400W from a PSU and then adding one fourth of an SP20, he's looking at adding probably about another 250W so that puts the supply at 83% rated capacity (or about 69% what it's actually capable of). Might be alright.

the SP20 has two internal boards, each with 2 PCIe connectors (vertically above eachother). AFAIK you can only use a maximum of two seperate PSUs to run it, 1 per board (like with an antminer)


I'm always cautious about this. If for any reason one PSU trips or fails, the other one will take the load and fail within seconds (or instantly). Generally, load-sharing is better for applications where power draw ranges from 20%-100%, rather than being 100% all the time like in bitcoin mining.

Did you try to contact that legkodumov fella who designed the string board Bitmain used (they mentioned it themselves on some obscure board here).
EDIT: found it
https://bitcointalksearch.org/topic/m.10435170
I think that this guy was saying that he can give the design to anyone with chips (maybe for large volume, though, but it wouldn't hurt to ask).