I also moved from your previous firmware version to your new firmware version on two S9s.

I just went through all my logs and did not find the problem you are seeing.

The temperature change is due to a fix they applied to correct for hash boards that have no center temperature sensor chip which was overlooked in the previous version.

It's that whole free enterprise thing. If you think the buyers are paying too much, you should be a seller. People have various views on the value of things and that is what makes a "market".

Yes, the Dallas area.

Please let me know if you have one available.

lots of fans to choose from on digikey.com or mouser.com

Please let us know if you find something better

Jinx3825, you are correct, almost.

When it's raining or above 90% humidity, the intake air is nearly saturated with moisture which will likely still condense in the first part of the miner before the air temperature rises above the dew point enough to be non-condensing. It only needs a surface to condense upon, not a colder surface.

The evidence for this is empirically demonstrated by the many miners which have suffered destruction due to corrosion caused by repetitive condensation formation. Counter intuitively, it is unusual for electronics to short due to condensation. Shorts are more often due to dirt and dust accumulation, although sometimes triggered by condensation.

Please refer to the Dew Point description here.
https://en.wikipedia.org/wiki/Dew_point

If your intake air is 100% relative humidity, you only need to raise the air supply temperature about 5C to eliminate condensation formation. If you have the intake and exhaust of all your miners fully isolated, you can add a small fan to return a little of the exhaust air back into the intake plenum to provide the needed intake temperature rise. Really, you should heat the intake air whenever the humidity reaches 90% or whenever the Dew Point / Temperature spread is <5 degrees.

Please refer to the Dew Point description here.
https://en.wikipedia.org/wiki/Dew_point

If your intake air is 100% relative humidity, you only need to raise the air supply temperature about 5C to eliminate condensation formation. If you have the intake and exhaust of all your miners fully isolated, you can add a small fan to return a little of the exhaust air back into the intake plenum to provide the needed intake temperature rise. Really, you should heat the intake air whenever the humidity reaches 90% or whenever the Dew Point / Temperature spread is <5 degrees.

Filters are good. They must either be relatively huge or fan powered. I prefer to use a heavy metal grid to keep bad people and animals out, then a smaller gauge metal grid to stop the big stuff then finally the filter.

That's true in autofan mode. If we have the firmware source code it should be simple to disable the fan.

This is awesome.

Please let me know if you can also disable the fan so it can be unequipped.

Can one control board handle more than 3 hash boards. It looks like there's space for many more.

Can you just set the fan speed manually to 0? If the firmware expects 0 then no fan should do exactly what it was commanded to do.

The conservative answer offered here is certainly done with the best intention.

I did however run two new S9s on "110V" circuits. I put that in quotes because my "110V" circuit is actually 125V, as measured under load at the power supply input.

I took the approach to actually measure the performance of the system to determine exactly why we cannot run these on our existing home wiring. I did run them on these circuits for a couple weeks, until I installed some "220V" circuits, measured as 245V under load at the power supply input.

When I first connect one miner, I connected the unit soon found my SJT aka power cord cable was HOT. It looked like a heavy cord but alas it was only 21AWG. I dug out a 14AWG SJT which fixed that problem. I measured the 12VDC output as soon as I turned it on and found 12.08V at the inputs of the hash boards. I monitored the voltage for days to ensure the power supply could maintain this voltage regardless of temperature or hash load. It was 12.08 +- .01V for weeks.

I measured 12.5A on the power supply input. This was maintained +-0.2A over the period of test.

I used these measurements to double check my assumptions for my circuit design, capital and expense budgets and such.

I keep a small repair bench at my house where I still use my 125V circuit, only because it was already there.

So I do understand that many folks have had trouble with the implementation. So why did mine work but theirs didn't. Why does Bitmain take the position that you simply cannot?

I hope this information will help someone.

I measured 12.5A at 125V on my circuit. My circuit had only this one PSU on it as its load. So this is a dedicated circuit. It is also a 20 Amp circuit. Bitmain published specs on the APW++ as 1200W at 110V. If this is true then the power capacity at 125V is as follows;
1200W X 125/110 = 1364W.

I suspect they generalized 110V  but really used the minimum operation voltage spec of 100V. If that's true then there may be a bit more power available. No matter, we only need 1350W.

NFPA rules will not allow continuous load on a circuit to exceed 80% of the circuit rating. They provide an exception for 100% rated circuits, but you'll not find one of those in your house. I sometimes design these in data centers, as in certain cases I can allow 90% loading on a continuous load circuit.

So, bottom line is that you must have at least a 20A circuit at 125V (measured at full load). Because the miner would only present 12.5A on this circuit, you could use a branch off this circuit to supply an additional 3.5A of continuous or intermittent load per NEC definition. It doesn't sound like a lot but your light, computer, network gear should be fine but certainly not a coffee pot.

Please do not think you can get away with running on a 15A circuit. Smaller wire, higher voltage drop and against the rules which were written for good reasons.

Oh, if you are not qualified to operate a voltmeter, do not attempt. You could be seriously injured or killed.

Please be careful. If you need assistance operation the voltmeter, feel free to contact me I'll be happy to talk you through it on a phone call.

That would answer a lot of questions. I would connect the controller and one hash board to one 1200W supply to see if that will run.

The power supply may shutdown if it senses a short circuit.

I'd probably try this;

Measure power supply voltage with nothing connected. If no output voltage, measure input voltage. If input good and output bad, power supply is your problem.

If power supply is good, then connect controller and measure again. If output voltage fails, measure resistance between power input terminals of controller board.

If controller runs then add one hash board. Measure again.

If bad, swap hash board and try again. If good add one hash board.

The idea is to find the disruptive hash board.

Obviously power down between connection changes.

Halong has proven absolutely nothing. No amount of bold text can change that.
The did do an excellent job of mining, or maybe collecting, a lot of BTC. All from speculative, trusting, hopeful folks who likely earned that money legitimately. So from that standpoint, they got the money.

Low voltage is very destructive. I like to ensure the voltage is correct before I do anything else.

Yes, if the power supply looks like it's running but is not able to supply the load, the miner may not be able to start up.

I always start by measuring the DC voltage to each connector. I connect my meter and watch the voltage during the first power up. It should build up to 12VDC very quickly and stay there forever. If it dips below 11.9 there is a power restriction somewhere.

I'm referring to S9s assuming the M3 is about the same.

You won't need a datasheet. Just copy or build an electrically equivalent board to mount a stack of chips on. The only think tricky about this ASIC is the secret recipe inside.

All the circuitry of miners is very, very simple.

If you get ASICs to sample, we can just replace one ASIC on a working board to test. easy, easy.

Head's up. Common Chinesium scam method:
- send working samples, aka actual ASICs pulled from S9 miner, to build trust
- continue to send pulled chips to get the pump primed
- receive money for bulk order
- send 20% real chips, 80% fakes
- Oh, so sorry for your trouble, please return failed units so we can analyze
- replace the 80% fakes with another 20/80 mix
- Oh, so sorry for your trouble, our analysis indicates you have installation method problem. We will offer you another replacement. Please return the chips that failed.
- return a 30/70 mix
Oh, so sorry. It looks like you did not fully correct you installation techniques. Perhaps we should build these boards for you here. Send us more money so we can start.

You see how this works? Many people there get very, very rich with this scam.

Nope. All the current flows through both lines.

Measure current of one line X Voltage = Watts

The current on the other line should be exactly equal. If not, some current is going to neutral or ground from one of the lines.

I measured 6.25 amps per S9 @245V.

NFPA rules limit continuous loads to 80% of the circuit rating. So a 15A circuit is limited to 12A.
I use 20A circuits to feed 2 S9s.