Yes and no.
Yellow is the tachometer feedback into the miner and tells the controller how fast the fan runs. Blue is the speed control signal. Red & black are DC power in.
4 wire fans use a speed controller inside of the fan itself to chop the 12v power as needed to set speed so just using a (high power, remember the fans take several amps of current) pot in series with the power MIGHT not work.

That brings up the fact that there are 2 types of 4-wire fans: The most common will run at full speed in the absence of a signal on the blue wire - signal on the blue slows it down.To test, cut the blue wire. If it goes full speed only then can you fiddle with an external speed control to reduce the red wire +12v.

If the fan runs at a very low speed it is the 2nd type that uses a signal on the blue wire to speed it up and that is what you have to modulate.

^^ Correct
If one bothers to look it up, all ASIC means is Application Specific Integrated Circuit. That can be anything: a watch or alarm clock chip, sound chip in a toy, crypto calculator, etc. Just means it is a custom chip designed to do 1 thing (or at most a very limited number of things) which cannot be changed. It is the actual circuit design and layout of the chip  - not its function - that can be copyrighted.

No, as it says in the title of this thread, there is a 1% fee.

Which would imply that it is a dual PSU.
If not, barring using 2 power relays wired to monitor each input and open/close the circuits (pricey & bulky) the only other way to isolate the power inputs so you don't (usually) have a live cord end when unplugged would be to use 2 bridge rectifiers, 1 for each power inlet, to give the raw DC that is fed to the switching portions of the supply. While it would mostly satisfy that safety aspect it also raises a few questionable design points regarding the current ratings of the bridge rectifiers and what happens if a portion of one shorts.

Either way, as a designer I'd go with the single high current power inlet even if it means breaking the mold of how it was done in the past. It's just the right way to do it. In my book since large farms already have to deal with the very different cases, power inputs just become another part of what they need to change.

Or they just prefer to follow some semblance of proper electrical design (tho yes they *are* pushing the plug rating) vs winging it and letting the chips fall where they may.  

A single connector makes no difference to me. My pdu's have outlets for both c14 and c20 male ends so is piece of cake to use it with the correct c20 to c19 cordset. Of course can only power 2 per pdu though...

As I mentioned earlier in one of the BM threads, what I'd like to know is if the BM miners have 2 psu's in the 1 case as that is the only way using 2 feeds to 1 device would be permitted by electrical codes. IF those power inlets are simply in parallel with each other there are several very bad possible situations that come to mind, 1 being the male end of a cord being live when unplugged from the wall...

The miner needs time to do its auto-tuning, that can be anywhere from several minutes and by some reports up to an hour. During tuning you get the "Connection refused" message and it goes away when tuning is done.

How long are you waiting?
Any whitespace at the end of your pool and/or user info? If so delete the space because it makes those entries invalid.

Has anybody had a chance to look inside the PSU to see how BM is powering them?
They damn well better actually have 2x PSU's in the 1 case as just doubling the power inlets/cords to feed a single supply violates regional electrical codes everywhere on the planet.

That said, running multiple PSU's with 1 per-hash board or some other grouping would make sense as it gives the ability to have the PSU voltage be better optimized per-board.

Ditto.
For me (running 460-480TH) the risk is what happens when a 73TH miner goes down hard: That would be waaay too high of a percentage of my hash rate offline for at a minimum of a few weeks. Unless you are running a farm with 10's-100's of PH losing that much from a single miner failure is unacceptable.

Yes. Read https://bitcointalk.org/index.php?topic=2959630.msg30384065#msg30384065 point-3
Even on this pool which is targeted at very small miners, to be practical even for lottery mining these days the slowest BTC miner MUST be running at least several GHs. Running at 120MHs means that it will be several minutes or more for each share to be processed and that is a far cry from the approx 18 shares/min that most pools want to work with.

Wrong.
So far Canaan has been on sched with their A1047 miners. Got my latest 'Sept delivery' one last week with no reason (so far) to think their Nov and Dec deliveries would be delayed. Their 50THs beast - who knows but historically Canaan has been good with dates.

Back to the Whatsminers, I know they mentioned "Foundry troubles" which I rather take as (pure speculation) they got a large batch of off-spec chips that won't hit normal speeds. MicroBT does like to push their chips to the max and like BM does, will populate a miner with a combo of 2 great boards and 1 'meh' one. Miner still meets specs but I've love to see a M10 with all 3 boards running 523MHz vs for example, one of mine doing 523, 512, & 435MHz... Point is they need enough top-grade chips to make the 'great' boards.

I'd agree with you if it wasn't for them being so hungry. 27-30THS and drawing over 100-200W more than an A1047 (in "normal" mode) makes it a no-go for me. Ja price is good but that power draw would null out the price difference in less than a year.

Well I just finished a few tests with the A1047 I rcvd on Monday:

Date: Oct. 1, 2019
Miner: Canaan Avalon 1047
Rcvd: Sept. 30, 2019
Power source: Triplite 6kva dual-conversion UPS
Voltage out: 230V +/- 0.5%
Ambient temp: 85F (29.4C)
Note: UPS is also powering a M10 pulling 2.17-2.18kw, this baseline load is subtracted from UPS readings to give test miner load.

- Miner in Normal mode for 30 min -
Power: 2.06kw
GUI speed: 31.31THs
Efficiency: 65.79W/THs
After 24hrs speed/power are the same.

Date: Oct. 2, 2019
- Miner in High Performance mode for 1.5hrs -
Power: 2.65kw
GUI speed: 37.6 THs
Efficiency: 70.48W/THs

Do note that the miner is running in a fairly warm room and that definitely affects power usage.

The A1047 uses a C19 plug - NOT a c13. As noted above, it comes with a cord but the wall end is a 'crow foot' style.
I use www.stayonline.com for all my cables

Mostly the same and yes. Only difference I see on the one I got Monday is that the PSU is attached like the 1041 was *supposed* to use. (all 3 of my 1041's have the loose PSU) Speeds are >31THs in Normal mode and >36THs in High Performance

edit: Oh, and the A1047 uses a C19 plug for power - not a C13. It comes with a cord but the wall end is a 'crow foot' style.

That is, it would be nice for the person that finds it. Aside from that warm fuzzy feeling from knowing the pool hit a block, regarding income it would be meaningless to the others here...

Ja ^^
Most of the time when I've seen this what actually happened is that the AC input rectifiers shorted and/or one of the big raw DC filter caps shorted which directly overloaded the inrush limiter. In that case, think of it as a last-resort 'fuse' to keep the fireworks to a minimum   And yes, high line voltage surges will take out rectifiers/caps in a heartbeat if the mfgr did not properly de-rate them.

3rd suspect would be the actual switching MOSFET(s) but in even a half way decently built supply that should be rare.

1st read the 1st time/small miner reference. That will answer a lot of basic questions.

re: "free power"
Do be aware that all large miners are both VERY HOT and LOUD. For all intents and purposes they are very large heaters that convert all power fed into them into heat out. Only difference is that yes, depending your exact circumstances they will be very loud and hot heaters that pay you more that the electric costs. (in your case free). Considering that most of the new big miners take in over 2kw and many over 3kw think of them as giant space heaters.  Can you deal with the heat?

As for sound: Earsplitting is a good way to put it. Some do have a lower power mode that cuts it down but still are loud. Think a loud blow dryer or vacuum cleaner loud.

Thermistors are not only used for temperature sensing but also as current surge protectors. PTC (Positive Temperature Coefficient) thermistors are temp sensors, their resistance rises as temp increases. NTC (Negative Temperature Coefficient) ones are generally surge protectors, specifically, inrush current limiters. They start at certain (higher) resistance when cold to limit startup currents and as they get slightly warm from power flowing through them their resistance goes down to nearly zero to supply full power to a circuit.

In short - where is the bad thermistor located, near the AC power inlet and large bulk capacitors? If so it's a surge limiter.

In their infinite wisdom I don't think Bitmain ever came close to a standard fan assignment. What fan plugs into which controller fan socket varies between the batches. All that matters is that whatever sockets that batch look at get feedback from the fans.