ANTMINER S5: 1155GH(+OverClock Potential), In Stock $0.319/GH & 0.51W/GH

[Digitalmocking]

Looks like its working!  

What are the big update ?? My fans seems to run lower. Anything else??

Give me your btc address for tip

After upgrading all the S5's and scoping out the temp charts, I picked a miner with the highest fan speed and temperature and took it offline. (Note to Bitmain: add a cool-down cycle that is triggered by unplugging the net cable).  After an air blast in the garage to get rid of the dust, proceeded to take off the fan and end brackets using a folded piece of cardboard to separate the fins on the heat sink.  Next, I removed one of the two ASIC boards with the dozen or so spring screws (clever!) and noticed that the thermal compound on the 1384 chips was mostly missing !  Maybe an unfortunate over-heat in a prior life, who knows.  After cleaning up the heatsink and board with contact cleaner followed by iso alcohol, I applied some GELID Solutions GC-Extreme thermal compound to the chips, installed the first board, ditto for the second one.

Bottom line, new thermal compound resulted in 600 RPM fan reduction and 8-10 Deg C in temperature !  Took about an hour for the first one and $5.00 in Gelid compound.  Happy camper here !

mine runs about 50c and the fan is around 3800 rpm..

is this what you are getting? i dont have another one to compare.. should i take apart and repaste?

That S5 was running in the low 60's at 3960 RPM - After the thermal compound it dropped to 50 at 3240 with the same room temperature.  Lower HW errors too.  I'll pull the next worst one tonight and repaste it for comparison.   The Gelid stuff is really stiff and gluey compared to the Arctic Silver compound.   That's probably a good thing  If I were completely ADHD I'd sort the boards across units and get matched sets lol.

Of the 6 I just ordered one is running noticeably hotter than the rest.  I think I know what I'm doing next weekend.

[1715019734]

1715019734

[1715019734]

1715019734

[aarons6]

Looks like its working!  

What are the big update ?? My fans seems to run lower. Anything else??

Give me your btc address for tip

After upgrading all the S5's and scoping out the temp charts, I picked a miner with the highest fan speed and temperature and took it offline. (Note to Bitmain: add a cool-down cycle that is triggered by unplugging the net cable).  After an air blast in the garage to get rid of the dust, proceeded to take off the fan and end brackets using a folded piece of cardboard to separate the fins on the heat sink.  Next, I removed one of the two ASIC boards with the dozen or so spring screws (clever!) and noticed that the thermal compound on the 1384 chips was mostly missing !  Maybe an unfortunate over-heat in a prior life, who knows.  After cleaning up the heatsink and board with contact cleaner followed by iso alcohol, I applied some GELID Solutions GC-Extreme thermal compound to the chips, installed the first board, ditto for the second one.

Bottom line, new thermal compound resulted in 600 RPM fan reduction and 8-10 Deg C in temperature !  Took about an hour for the first one and $5.00 in Gelid compound.  Happy camper here !

mine runs about 50c and the fan is around 3800 rpm..

is this what you are getting? i dont have another one to compare.. should i take apart and repaste?

That S5 was running in the low 60's at 3960 RPM - After the thermal compound it dropped to 50 at 3240 with the same room temperature.  Lower HW errors too.  I'll pull the next worst one tonight and repaste it for comparison.   The Gelid stuff is really stiff and gluey compared to the Arctic Silver compound.   That's probably a good thing  If I were completely ADHD I'd sort the boards across units and get matched sets lol.

Of the 6 I just ordered one is running noticeably hotter than the rest.  I think I know what I'm doing next weekend.

that is a pretty large gap..
board #1 is hotter on mine then #2 but only by 6c.

id wait till the diff increase to take it down and repaste it.. its going to be a big increase and id mine as much as you can before.

[aarons6]

has anyone ran any tests on wattage vs speed?

after the diff increase i am thinking about turning my s5 down to 300..
hopefully it will run cooler, the fan wont be so loud and it might use a little less watts..

[MyRig]

@Digitalmocking

Temperature difference between the 2 hashing PCB was reported to the engineers and it was reviewed.  So far, it is not the hardware caused issues, but rather where and how the miners are being deployed and managed per the engineers.

[Digitalmocking]

@Digitalmocking

Temperature difference between the 2 hashing PCB was reported to the engineers and it was reviewed.  So far, it is not the hardware caused issues, but rather where and how the miners are being deployed and managed per the engineers.

All of my miners are in a datacenter with 21c input temps, of the 6 miners all boards are running between 45 - 51 except for the one board that runs 10c hotter than all the others.

I'm going to pull the system apart this weekend and check the thermal compound.

[ilpirata79]

I have a 1200W Corsair PSU... how much can I safely overclock my C1 miner?

Thanks,
ilpirata79

The PSU is able to provide 1200watts, the C1 is drawing ~900 watts.
You can safely OC the C1 till max frequency, the PSU will survive only the C1 can get a bit stressed

Cheeers

Sorry to bother you with this request... what are the suggested voltages for the different frequencies?


Thanks,
ilpirata79

[klondike_bar]

Thinking of two S5, outside of footprint, is there a real compatibility advantage to buying one EVGA Supernova 1300 Gold Vs two Corsair CX750M Bronze?  I understand the CX750M measures as an 825W supply (per Dogie), so it seems to me the reason to choose the 1300 is less space and fewer power cords.  Right or wrong?

Overall you'll have less loss using a single, larger PSU versus two individual ones.  Granted, its not much loss, but hey, electricity is electricity.

The CX750M is bronze rated, which means that using a pair of them vs a single 1300W GOLD is about 2-4% more electricity usage. its not a whole lot, but its still $2-3/month savings with a more efficient PSU. over the span of a year, you'll have saved about $30 by using a gold-rated PSu over a bronze-rated one.

[ilpirata79]

Anyone willing to sell its C1?
I can pay around 100$ for it (without hoping to roi neverthless).

Best regards,
ilpirata79

[toptekk]

so far all my S5 run about the same as this one so im very happy with all mine .



Next difficulty retarget occurs at block 350783.0 (eta 5.3 days): 50426253807.9 / +7.9% [est.] that still not to bad id be worried it went to 56 or higher .

[sinewave]

Is it possible to power up and use one of the two hashing boards on the S5? I currently only have 2 PCI-e connectors on my PSU and wonder if I can use my s5 on "half rate" while I wait for my DPS-8000GB adapter boards to arrive.

[scyth33]

Is it possible to power up and use one of the two hashing boards on the S5? I currently only have 2 PCI-e connectors on my PSU and wonder if I can use my s5 on "half rate" while I wait for my DPS-8000GB adapter boards to arrive.

yes possible. just remove the data cable from the board that you do not want to use.

[Lexis77]

On my question - why i recieved secondhand shit different revisions? Bitmain answer - miner work well? If yes - go to ass hole!

[aarons6]

Wrote in support of an angry e-mail. We'll wait for that answer.

On my question - why i recieved secondhand shit different revisions? Bitmain answer - miner work well? If yes - go to ass hole!

if you were rude to them why would they be nice to you??

also, does it work?? even if it had some dust on it, doesnt mean it was sold to someone and returned.. it could have been sitting around the warehouse or one of their control test models. (which would probably still qualify as new if it passed their tests)

[smit1237]

 Why ibm does not sell servers covered in dust? Servers need qc too. So ibm can't and bitmain can? This is not right i think

[Xian01]

Wrote in support of an angry e-mail. We'll wait for that answer.

On my question - why i recieved secondhand shit different revisions? Bitmain answer - miner work well? If yes - go to ass hole!

if you were rude to them why would they be nice to you??
also, does it work?? even if it had some dust on it, doesnt mean it was sold to someone and returned.. it could have been sitting around the warehouse or one of their control test models. (which would probably still qualify as new if it passed their tests)

new ↮ dust

[Mikestang]

Anyone willing to sell its C1?
I can pay around 100$ for it (without hoping to roi neverthless).

Best regards,
ilpirata79

This is a thread for the S5, not C1, but regardless C1s resell for like $500, I don't think you'll find one for $100.

[Mikestang]

if you were rude to them why would they be nice to you??

also, does it work?? even if it had some dust on it, doesnt mean it was sold to someone and returned.. it could have been sitting around the warehouse or one of their control test models. (which would probably still qualify as new if it passed their tests)

Business have to be nice to customers, that's how it works if they want more customers.

Electronics do not get dusty internals from sitting around a warehouse, and no test procedure should run long enough to get the device dusty.  A dusty device = a used device, plain and simple.

[Lexis77]

Wrote in support of an angry e-mail. We'll wait for that answer.

On my question - why i recieved secondhand shit different revisions? Bitmain answer - miner work well? If yes - go to ass hole!

if you were rude to them why would they be nice to you??

also, does it work?? even if it had some dust on it, doesnt mean it was sold to someone and returned.. it could have been sitting around the warehouse or one of their control test models. (which would probably still qualify as new if it passed their tests)

I'm not rude! I asked a normal question. In response, I hear - rejoice that works. I know how to distinguish brand new from used. I am more than 20 years doing repairs at the component level microelectronics.
I have always been loyal to bitmain. Even the theme started with the history and support on our forum. But after such a deception, their reputation will suffer. I promise!

[dogie]

Of the 6 I just ordered one is running noticeably hotter than the rest.  I think I know what I'm doing next weekend.

Temperature difference between the 2 hashing PCB was reported to the engineers and it was reviewed.  So far, it is not the hardware caused issues, but rather where and how the miners are being deployed and managed per the engineers.

The temperature sensors are also rather crude, and on a large scale there will be plenty of 'miscalibrated' ones. I'd be more suspicious of the sensor reading incorrectly than one board running significantly hotter.

[dogie]

Thinking of two S5, outside of footprint, is there a real compatibility advantage to buying one EVGA Supernova 1300 Gold Vs two Corsair CX750M Bronze?  I understand the CX750M measures as an 825W supply (per Dogie), so it seems to me the reason to choose the 1300 is less space and fewer power cords.  Right or wrong?

[For reference, almost all of the Corsair PSUs have an additional 10% 'peak' rating and so have additional capacity.]