ANTMINER S3+ Discussion and Support Thread

[Knoxx]

Not sure if posted yet, but this may help some of the fellow miners out there that don't like the paperclip trick.

http://www.frozencpu.com/products/5141/psu-218/FrozenCPU_24-pin_Power_Supply_Jump_Start_Connector.html

Might be US only- not sure.

I am waiting for my 30 units still. Held up in Hong Kong the last two days shipped via dhl


Knoxx/BlitzStix

worth ordering a jumper for those who might feel timid about it.
Side note-my EVGA1300 (from rakuten) came with a motherboard dummy WITH a jumper inserted. I guess they are aware who there customers are 

Nice to know. I just ordered 10 evga 1300 g2's for my setup. I won't order those from frozen.

How much did you pay per evga 1300?

I paid 195$ each.

[9600]

So! All units are hashing over 440gh (Two over 490), but only when using a cronjob to restart the cgminer service every 1/2 hour - 1hr depending on the miner.
Regardless of how long I leave any of these guys up to "stabilize", they always fall to 390-398 after 48 hours.

There's only one thing. I am noticing a rather high number of duplicate shares poolside since starting this restart routine. (surprise)
Is there a cgminer flag among the bitmain options to clear the asic of all work upon a cgminer restart? I know the A1 miners have a similar command with '--hwreset'

Restarting every 30 minutes is a bad practice.
Try replacing thermal paste grease and I think the issue will be solved.

I'm pretty aware how just how shitty it is, but unfortunately after re-pasting two units I saw a decrease in running temperatures but was still experiencing a steep drop off in hash after about 25-35 minutes. One of the units will spike back up to about 470 but never enough to average out to 440 without cgminer restarts.

I wish I could attribute this to poor labor skill on my part, but I'm just about 100% certain I don't suck at apply thermal paste. =/

what kind of thermal paste did u use ??

Ceramique

[Duce]

Ok, so I posted a few days ago about my batch 2 S3.  At stock clocks I get a huge hash drop at 2-3 hours and one chip is showing a "-" where the "O" should be.   I contacted bitmain and they said to re-paste, which I did.    The paste seems do have done absolutely nothing.   I would rather not run it at 212 but that what it is at now, has anyone found a reproducible solution for the underperforming miners? 

I have one in the same situation, it just does not like anything higher than 212. I know there are others in the same situation. On the positive side they consume less power. I did everything possible to remove thermal problems and none help. Mine is a B1 so I would just let it run. The amount of time working on them takes away from production. If you are like me you just want to get it up to speed but sometimes you just have to let one go.

[9600]

Ok, so I posted a few days ago about my batch 2 S3.  At stock clocks I get a huge hash drop at 2-3 hours and one chip is showing a "-" where the "O" should be.   I contacted bitmain and they said to re-paste, which I did.    The paste seems do have done absolutely nothing.   I would rather not run it at 212 but that what it is at now, has anyone found a reproducible solution for the underperforming miners? 

I have one in the same situation, it just does not like anything higher than 212. I know there are others in the same situation. On the positive side they consume less power. I did everything possible to remove thermal problems and none help. Mine is a B1 so I would just let it run. The amount of time working on them takes away from production. If you are like me you just want to get it up to speed but sometimes you just have to let one go.

Look at my previous posts on this thread, you can follow my saga. My personal solution has been to institute scheduled restarts. I am now between 460-490 for about 3 days now and have the accepted shares to show.

Dupes are a bit high, but that's to be expected with this config.

[Duce]

Can someone let me know what the Antminer S3 status showing -o-o-o-o -o-o-o-o -o-o-o-o -ooo-o-o  means?  Only one chain is showing this.  The other is oooooooo oooooooo .  Reason I ask is my Hash rate is 345 average.  Should be higher.  

that is a bad chain.  

fixes may be:
 1) lower freq setting 212.5 has helped
2) reboot via gui  may help
3) make sure you have a good psu

If the above are not working, go back to this thread and read last two days posts.
Your problem will get solved.

My Guess is power supply.  It's a cheap 520 watt from Best Buy.  Will try a Corsair 750w later today.  

How do I lower frequency?  Currently at 218.75.

Highly recommending to immediately replace your power supply with a new good quality one like you said.
I will recommend you to buy an 800Watts power supply unit for you instead of the 750Watts PSU. Keep and eye on the single +12V rail.
Why you want to lower your mining power? Some do for some particular reason like placing in office and so.
Underclocking is simple, uncomment lower frequency lines(3) to select lower frequency.

Ok got the new power supply.  Corsair 750M.  Still getting the same -o-o-o-o -o-o-o-o -o-o-o-o -ooo-o-o for Chain 2.  Any other tips?  Can the Antminer be a bad one?

Post a screenshot of your S3 Miner Stats page.

[IM]https://i.imgur.com/dPdNW5y.jpg[/img]

It seems like a thermal paste issue. But before doing anything do a firmware update or factory reset and check.
If its not working, open up the cover of your Antminer S3. Look for any loose cables.
Then swap your cables which run from control board to each blades.
After 5 minutes check the miner stats page, if its working, you are lucky.
Else if, there are any changes from your previous image post the new image.
This time if its not working next step is to try a thermal paste replacement.
I expect this will solve your issue.

allcoinminer is correct in that your GUI indicates 32 ASICs for the one blade. Someone early on had the same problem that I believe was resolved with step No.1, firmware update. Curious to know if that does correct your problem.

[Rmcdermott927]

Ok, so I posted a few days ago about my batch 2 S3.  At stock clocks I get a huge hash drop at 2-3 hours and one chip is showing a "-" where the "O" should be.   I contacted bitmain and they said to re-paste, which I did.    The paste seems do have done absolutely nothing.   I would rather not run it at 212 but that what it is at now, has anyone found a reproducible solution for the underperforming miners? 

I have one in the same situation, it just does not like anything higher than 212. I know there are others in the same situation. On the positive side they consume less power. I did everything possible to remove thermal problems and none help. Mine is a B1 so I would just let it run. The amount of time working on them takes away from production. If you are like me you just want to get it up to speed but sometimes you just have to let one go.

Look at my previous posts on this thread, you can follow my saga. My personal solution has been to institute scheduled restarts. I am now between 460-490 for about 3 days now and have the accepted shares to show.

Dupes are a bit high, but that's to be expected with this config.

What are you using for the restarts schedule?  It appears the problem with mine is that BOTH a hard reboot and GUI reboot are necessary to get my miner back on track.

[visdude]

Not sure if posted yet, but this may help some of the fellow miners out there that don't like the paperclip trick.

http://www.frozencpu.com/products/5141/psu-218/FrozenCPU_24-pin_Power_Supply_Jump_Start_Connector.html

Might be US only- not sure.

I am waiting for my 30 units still. Held up in Hong Kong the last two days shipped via dhl


Knoxx/BlitzStix

worth ordering a jumper for those who might feel timid about it.
Side note-my EVGA1300 (from rakuten) came with a motherboard dummy WITH a jumper inserted. I guess they are aware who there customers are 

Nice to know. I just ordered 10 evga 1300 g2's for my setup. I won't order those from frozen.

How much did you pay per evga 1300?

They're about ~$200 everywhere.

No way. pay with bitcoin at newegg and they are about $160. Awesome deal
http://www.newegg.com/Product/Product.aspx?Item=N82E16817438011&cm_re=evga_1300-_-17-438-011-_-Product

Jenna is in CA - that discount doesn't work.

Plus, you don't need to nitpic on a couple of dollars. I literally just said it's about ~$200, which is what the price will be after tax/discounts or whatever. 

Would 2 cx750 be good to over clock 2 AntMiner s3

If you meant one dedicated CX750 for each S3, then it's enough.  In fact, it's more than enough considering that I was able to clock the S3 at 250M with a CX500 without any issues whatsoever.  Some figures on my S3/CX500 test here.

[aztecminer]

So! All units are hashing over 440gh (Two over 490), but only when using a cronjob to restart the cgminer service every 1/2 hour - 1hr depending on the miner.
Regardless of how long I leave any of these guys up to "stabilize", they always fall to 390-398 after 48 hours.

There's only one thing. I am noticing a rather high number of duplicate shares poolside since starting this restart routine. (surprise)
Is there a cgminer flag among the bitmain options to clear the asic of all work upon a cgminer restart? I know the A1 miners have a similar command with '--hwreset'

Restarting every 30 minutes is a bad practice.
Try replacing thermal paste grease and I think the issue will be solved.

I'm pretty aware how just how shitty it is, but unfortunately after re-pasting two units I saw a decrease in running temperatures but was still experiencing a steep drop off in hash after about 25-35 minutes. One of the units will spike back up to about 470 but never enough to average out to 440 without cgminer restarts.

I wish I could attribute this to poor labor skill on my part, but I'm just about 100% certain I don't suck at apply thermal paste. =/

what kind of thermal paste did u use ??

Ceramique

ok.. do not use AS5 thermal paste and do not place
the plate upside down like this idiot tried to do below:


check your heat sink is placed right way:


with all those "-" it looks like you have a cooling issue on that blade.
look at your plate did you place it upside down by accident ??

[Fahlcor]

Quick question.  Some of you have showed using a monitor software to watch all the miners at one time.  how do I get that?

Is there a link somewhere?

Thanks

[Mudbankkeith]

Quick question.  Some of you have showed using a monitor software to watch all the miners at one time.  how do I get that?

Is there a link somewhere?

Thanks

You could try MultiMiner

[visdude]

So! All units are hashing over 440gh (Two over 490), but only when using a cronjob to restart the cgminer service every 1/2 hour - 1hr depending on the miner.
Regardless of how long I leave any of these guys up to "stabilize", they always fall to 390-398 after 48 hours.

There's only one thing. I am noticing a rather high number of duplicate shares poolside since starting this restart routine. (surprise)
Is there a cgminer flag among the bitmain options to clear the asic of all work upon a cgminer restart? I know the A1 miners have a similar command with '--hwreset'

Restarting every 30 minutes is a bad practice.
Try replacing thermal paste grease and I think the issue will be solved.

I'm pretty aware how just how shitty it is, but unfortunately after re-pasting two units I saw a decrease in running temperatures but was still experiencing a steep drop off in hash after about 25-35 minutes. One of the units will spike back up to about 470 but never enough to average out to 440 without cgminer restarts.

I wish I could attribute this to poor labor skill on my part, but I'm just about 100% certain I don't suck at apply thermal paste. =/

what kind of thermal paste did u use ??

Ceramique

ok.. do not use AS5 thermal paste and do not place
the plate upside down like this idiot tried to do below:
http://i62.tinypic.com/k9bapy.jpg

check your heat sink is placed right way:
http://i59.tinypic.com/2l9nubs.jpg

with all those "-" it looks like you have a cooling issue on that blade.
look at your plate did you place it upside down by accident ??

Ha!  Classic.

I wonder if re-torquing the heat sink screws first was ever attempted to make sure they are snug before going through with the elaborate heat sink reseating gymnastics (and the lost precious hashing time that goes along with it).  Most of the time, that's all what's needed (never have had to reseat heat sinks to date).  That's all I did with the S3s as it was with the other ASIC devices (BFL, AsicMiner, Blue Fury, AntMiner, etc.) in the past.  That's pretty much S.O.P. upon receiving a new (or not so new ) ASIC device.

[jpchrist]

Quick question.  Some of you have showed using a monitor software to watch all the miners at one time.  how do I get that?

Is there a link somewhere?

Thanks

Booya!  https://bitcointalk.org/index.php?topic=596178.0

[aztecminer]

So! All units are hashing over 440gh (Two over 490), but only when using a cronjob to restart the cgminer service every 1/2 hour - 1hr depending on the miner.
Regardless of how long I leave any of these guys up to "stabilize", they always fall to 390-398 after 48 hours.

There's only one thing. I am noticing a rather high number of duplicate shares poolside since starting this restart routine. (surprise)
Is there a cgminer flag among the bitmain options to clear the asic of all work upon a cgminer restart? I know the A1 miners have a similar command with '--hwreset'

Restarting every 30 minutes is a bad practice.
Try replacing thermal paste grease and I think the issue will be solved.

I'm pretty aware how just how shitty it is, but unfortunately after re-pasting two units I saw a decrease in running temperatures but was still experiencing a steep drop off in hash after about 25-35 minutes. One of the units will spike back up to about 470 but never enough to average out to 440 without cgminer restarts.

I wish I could attribute this to poor labor skill on my part, but I'm just about 100% certain I don't suck at apply thermal paste. =/

what kind of thermal paste did u use ??

Ceramique

ok.. do not use AS5 thermal paste and do not place
the plate upside down like this idiot tried to do below:
http://i62.tinypic.com/k9bapy.jpg

check your heat sink is placed right way:
http://i59.tinypic.com/2l9nubs.jpg

with all those "-" it looks like you have a cooling issue on that blade.
look at your plate did you place it upside down by accident ??

Ha!  Classic.

I wonder if re-torquing the heat sink screws first was ever attempted to make sure they are snug before going through with the elaborate heat sink reseating gymnastics (and the lost precious hashing time that goes along with it).  Most of the time, that's all what's needed (never have had to reseat heat sinks to date).  That's all I did with the S3s as it was with the other ASIC devices (BFL, AsicMiner, Blue Fury, AntMiner, etc.) in the past.  That's pretty much S.O.P. upon receiving a new (or not so new ) ASIC device.

that first screen show a heat sink placed upside down after replacing stock thermal paste ..
if the heat sink is placed upside down then it does not make contact w the asic chips..
of course that means the asic chips will not cool well and therefore not hash well .

[visdude]

Corsair CX500 = $29.99 (after rebate - expires 8/11/14) currently at the Egg + 10% BTC payment discount.  Also, when available, additional instant discounts with Promo Codes (appears just below the product listing title on a week to week basis):

http://www.newegg.com/Product/Product.aspx?Item=N82E16817139027

[Biodom]

Ok, so I posted a few days ago about my batch 2 S3.  At stock clocks I get a huge hash drop at 2-3 hours and one chip is showing a "-" where the "O" should be.   I contacted bitmain and they said to re-paste, which I did.    The paste seems do have done absolutely nothing.   I would rather not run it at 212 but that what it is at now, has anyone found a reproducible solution for the underperforming miners?  

I have one in the same situation, it just does not like anything higher than 212. I know there are others in the same situation. On the positive side they consume less power. I did everything possible to remove thermal problems and none help. Mine is a B1 so I would just let it run. The amount of time working on them takes away from production. If you are like me you just want to get it up to speed but sometimes you just have to let one go.

Look at my previous posts on this thread, you can follow my saga. My personal solution has been to institute scheduled restarts. I am now between 460-490 for about 3 days now and have the accepted shares to show.

Dupes are a bit high, but that's to be expected with this config.

What are you using for the restarts schedule?  It appears the problem with mine is that BOTH a hard reboot and GUI reboot are necessary to get my miner back on track.

Hey, I also have one miner that (only working at 212.5) does great on restart for 24-36 hr, averaging ~430Gh, then goes to 400 and stays there.
I would like to use the scheduled tasks trick, but there is already a string of code there (in scheduled tasks):

Code:

*/3 *  *  *  *    /usr/bin/cgminer-monitor

should a put the string like this

Code:

0 1 * * * /etc/init.d/cgminer restart

below it to do a once in 24hr (at 1am) restart or replace whatever is there already with a new string? what that does anyway: cgminer-monitor?

[MinersAnonymous]

Under my account I sent a message on bitmaintech.com and now my message options have disappeared, has this happened to anyone before? Please help, thank you!

[bluebottle73]

Under my account I sent a message on bitmaintech.com and now my message options have disappeared, has this happened to anyone before? Please help, thank you!

Must be a change on the site.  The message options are gone for me too.  I had attempted to communicate with them via that method before, but it wasn't until I sent an email instead that I got anything back.

[JennaK]

A cx600 would be useless to over clock a s3! I need 2 cx750 to have two overclocked s3s

[faetos]

Under my account I sent a message on bitmaintech.com and now my message options have disappeared, has this happened to anyone before? Please help, thank you!

Must be a change on the site.  The message options are gone for me too.  I had attempted to communicate with them via that method before, but it wasn't until I sent an email instead that I got anything back.

Gone from mine as well. Did you e-mail address did you get support from them? I need to ask them about a dead miner I have. Thanks!

[bluebottle73]

Under my account I sent a message on bitmaintech.com and now my message options have disappeared, has this happened to anyone before? Please help, thank you!

Must be a change on the site.  The message options are gone for me too.  I had attempted to communicate with them via that method before, but it wasn't until I sent an email instead that I got anything back.

Gone from mine as well. Did you e-mail address did you get support from them? I need to ask them about a dead miner I have. Thanks!

I sent my request for tech support to info@bitmaintech.com and got a reasonably fast reply, though it took several days before they finally sent me a shipping label for me to send back the faulty control board.