ANTMINER S4 Discussion and Support Thread

[J4bberwock]

It's a shame the stock PSU was so useless.
With a correct one, you can get quite nice results.



The only limit was the gui not allowing over 268.75 frequency.
This is with a custom breakout board able to supply 2400w on 220/240v and 1800w on 110/120v

[P4ndoraBox7]

Shit....awesome speed.

[P4ndoraBox7]

So basically, even if the stock PSU was resent for a good one, you say I should reduce my frequency ?

Take a look at my stats, just in case...

[Prelude]

Yep, a proper PSU could definately unleash them. I used DPS-2000BB PSUs with mine when I had them, ran them at about 2.5TH/s each.

Only reduce if you don't care about the PSU dying.

[l8nit3]

So I was setting up an S4 for a buddy recently, and had an issue I've never seen before. I was hoping someone here could maybe shed some light for me.
Basically, I setup the S4 (as I have many times) and began running, at stock mhz. On boot all of the o's were actually ---- for about 2-3 mins, However it was showing a correct hashrate. Then after approx 5-20 mins (different depending on my troubleshooting) one board would slow down and eventually stop running, leaving me with half the hashrate.

During the slow down process a few of the o's would change to x's. Sometimes it would be one, sometimes up to 5.... but never the same o's. After approx 30-60 mins the x's would go away and hashing would resume as normal.... but only for 1-2 hours, then it would repeat the process.

For the first cycle of this strange occurence, It was always the 'bottom' board shown on the web interface. However subsequent cycles of the issue (after say 3 hours) the x's would be spread through both boards.... still no more then 5, and still always in different places.

My thoughts after messing with it for a few days, is that the controller board must be on its way out, and only getting intermittent connection to each board. Any thoughts?

edit
single s4 running on a corsair AX850 gold
/edit

2nd edit
My bad just realized, it had to be an s5 since it was 2 boards not 4.... but theory's should still apply the same i think
/2nd edit

[P4ndoraBox7]

Yep, a proper PSU could definately unleash them. I used DPS-2000BB PSUs with mine when I had them, ran them at about 2.5TH/s each.

Only reduce if you don't care about the PSU dying.

well this what i'm asking...if everything ok from the numbers I shared...could I try to overclock it some more too.

[notlist3d]

Yep, a proper PSU could definately unleash them. I used DPS-2000BB PSUs with mine when I had them, ran them at about 2.5TH/s each.

Only reduce if you don't care about the PSU dying.

well this what i'm asking...if everything ok from the numbers I shared...could I try to overclock it some more too.

It varies is the anwser some PSU's were better then others.  I ran 3 for a long time had 2 of PCIE one with the poles.  But I did OC at times and had no issues  I was lucky the PSU's worked great and never had a issue.

Some PSU's are not as lucky if you push it... yes more chance of pushing psu and it dying.   This is honestly a personal decision as it varies so much.

[P4ndoraBox7]

Yep, a proper PSU could definately unleash them. I used DPS-2000BB PSUs with mine when I had them, ran them at about 2.5TH/s each.

Only reduce if you don't care about the PSU dying.

well this what i'm asking...if everything ok from the numbers I shared...could I try to overclock it some more too.

It varies is the anwser some PSU's were better then others.  I ran 3 for a long time had 2 of PCIE one with the poles.  But I did OC at times and had no issues  I was lucky the PSU's worked great and never had a issue.

Some PSU's are not as lucky if you push it... yes more chance of pushing psu and it dying.   This is honestly a personal decision as it varies so much.

Allright thx.

[notlist3d]

Yep, a proper PSU could definately unleash them. I used DPS-2000BB PSUs with mine when I had them, ran them at about 2.5TH/s each.

Only reduce if you don't care about the PSU dying.

well this what i'm asking...if everything ok from the numbers I shared...could I try to overclock it some more too.

It varies is the anwser some PSU's were better then others.  I ran 3 for a long time had 2 of PCIE one with the poles.  But I did OC at times and had no issues  I was lucky the PSU's worked great and never had a issue.

Some PSU's are not as lucky if you push it... yes more chance of pushing psu and it dying.   This is honestly a personal decision as it varies so much.

Allright thx.

Your welcome!  In most cases it would not matter.  But some of the S4's PSU's were not up to spec or pushed so hard they died.

One thing I would be interested to know is how many still die.  One would think those PSU's that were not up to spec most have died.   So thinking about it a lot bigger percentage of S4's should have a good PSU as they have ran so long.... but I could be wrong.  

[Prelude]

I'm pretty sure it would be easier to tally how many are still working instead fo how many have died. Doubt there's more than 30%* still functional.





*Figure is, of course, pulled out of my ass

[notlist3d]

I'm pretty sure it would be easier to tally how many are still working instead fo how many have died. Doubt there's more than 30%* still functional.





*Figure is, of course, pulled out of my ass

I guess i was lucky I only sold mine a few months ago ran through a hot summer, and no problems.  3 out of 3 ran good and did not have bad PSU's.

And they were pushed in summer.  So not the best conditions compared to winter.   But I could have been lucky.

[P4ndoraBox7]

How can I check how much watt my second PSU has ?

And yep I know the ''sad'' story about the S4. Thats why I'm hesitating to push it more or whatsoever.

[Prelude]

I'd say you were lucky. I had 7 of them, 5 failed. I decided to pull the remaining 2 before they followed suit and replaced them with real power supplies.

[Prelude]

How can I check how much watt my second PSU has ?

And yep I know the ''sad'' story about the S4. Thats why I'm hesitating to push it more or whatsoever.

It's specifications are 1400w gold, which means pulling ~1555w measured at the wall when maxed out. I don't suggest pullsing more than 1300w DC from them if you want any sort of reliability. That would be approximately 1444w at the wall.

[notlist3d]

How can I check how much watt my second PSU has ?

And yep I know the ''sad'' story about the S4. Thats why I'm hesitating to push it more or whatsoever.

It's specifications are 1400w gold, which means pulling ~1555w measured at the wall when maxed out. I don't suggest pullsing more than 1300w DC from them if you want any sort of reliability. That would be approximately 1444w at the wall.

And if your wondering on watt's.... be careful on using a kill-a-watt on these.  One early mistake I did  when getting mine was using a kill-a-watt... it was not pretty it now instead of white has a brown mark between prongs.   It's not made for this kinda watt's.

But guess I was lucky to catch it and no real damage done.  Just was a little scare.

[Prelude]

How can I check how much watt my second PSU has ?

And yep I know the ''sad'' story about the S4. Thats why I'm hesitating to push it more or whatsoever.

It's specifications are 1400w gold, which means pulling ~1555w measured at the wall when maxed out. I don't suggest pullsing more than 1300w DC from them if you want any sort of reliability. That would be approximately 1444w at the wall.

And if your wondering on watt's.... be careful on using a kill-a-watt on these.  One early mistake I did  when getting mine was using a kill-a-watt... it was not pretty it now instead of white has a brown mark between prongs.   It's not made for this kinda watt's.

But guess I was lucky to catch it and no real damage done.  Just was a little scare.

That's weird, I've used my kill a Watt on 2000+ Watt loads on 240v,and maxed out on 120v plenty of times. Never any issues. They're rated for 15A 125V. (yes it works on 240v)

[notlist3d]

How can I check how much watt my second PSU has ?

And yep I know the ''sad'' story about the S4. Thats why I'm hesitating to push it more or whatsoever.

It's specifications are 1400w gold, which means pulling ~1555w measured at the wall when maxed out. I don't suggest pullsing more than 1300w DC from them if you want any sort of reliability. That would be approximately 1444w at the wall.

And if your wondering on watt's.... be careful on using a kill-a-watt on these.  One early mistake I did  when getting mine was using a kill-a-watt... it was not pretty it now instead of white has a brown mark between prongs.   It's not made for this kinda watt's.

But guess I was lucky to catch it and no real damage done.  Just was a little scare.

That's weird, I've used my kill a Watt on 2000+ Watt loads on 240v,and maxed out on 120v plenty of times. Never any issues. They're rated for 15A 125V. (yes it works on 240v)

Interesting maybe mine was a dud then.   I did leave it on for around a 2 week period.   I was testing all kinds of freq's and watched it slowly over the 2 weeks or so.

I never smelt any burning, or problem.  But when unplug it say the brown between prong's and was last time I used it on that much.  If rated for 240 mine must be a dud as was just 1 S4.

[Prelude]

No, it's rated for 125v. I'm just saying that it also works on 240v.

[notlist3d]

No, it's rated for 125v. I'm just saying that it also works on 240v.

Ahhh that is a big difference.  I still might have got a dud then.  it was 120v it was in with a single S4.   

So nothing over 125v.   But it sure made it ugly under plugs.

[Xeil]

How do I get after a power failure, he turned on? It included only the power supply.