Very odd and there's nothing about it in the log. Let's try this, run the machine with only the bad hashboard plugged in. Then let's see the logs.

There is some evidence that replacing only one fan can create a airflow imbalance, stressing the old fans because even with identical RPMs and current draw you can have different actual air flow depending on design or even age.

Many models of s19 are now supported by third-party firmware, these solutions can just ignore the RPMs.

To be completely honest very few texts except for ones at  bitmain certified centers have had their hands in the new + variants. If they are anything like the previous generation of s19 variants they are made with aluminum boards and have no mosfets. Meaning when one board throws an error the power supply is forced to turn off all three boards instead of just the one to protect the machine.

Your next step in diagnosing the issue would be to test the power rails using a multimeter when you see this issue occurring. With this info and more complete logs we likely can still help you.

I thought they did this because the heat sinks would just literally fall off while running sideways. With the heat sinks now spring loaded and clamped on I'm having difficulty figuring out why this is still done. I concluded is the old folly, " it's just how we've always done it". I've considered the convective benefits and see no appreciable difference, at least with such high pressure forced air. I suppose if you lay them horizontally you have to use more shelves which take up more space which don't let you put as many miners in one location which for obvious reasons will allow them all to perform better with the same cooling a denser configuration would in an identical pod. Although, I actually just learned that bitmain is going to attempt soldering heat sinks on again, this isn't one of those is it?

First you would use an S17+ control board because it's unlocked and available for modification of firmware.

Next you would become a Linux God who can compile code and create shell scripts.

Then, you would tear apart a test jig to understand how to send commands to the  hashboard and receive the nonce.

Once you have the nonce you have to know how to send that to your pool for credit. Somewhere around the second or third step you're going to figure out how to fiddle with little things like frequency and voltage too get your desired effects.


That's just a high level, back of the envelope, look at the top of the iceberg.

This means one of three things, you have an amlogic "cousin wangs covid ARM sale" control board(thumb microUSB drive replaced microSD), a beagle bone control board(SD card slot is inside the control board housing), or it is one of the new "cousin wang's 2023 ARM CPU sale" control board I've started hearing rumors about(idek).

The logs look fine also.

You can try these options.
1. Restore defaults in web interface and reload pool info.
2. Try configuring it with a remote tool like BTCtools.
3. Load/reload the latest firmware update from bitmain's website.

Number one worked for me last week when I had a similar seemingly fine log that simply didn't want to hadh but that machine was new to me, yours was running fine.

After it was clear there was no future and the monotony of warranty repair had worn me down, I quit ASIC Master. After the treatment I received for not spending more time on our new paying customer, there was simply no reason to stay. I decades of being an electronics technician I have never been told to stop speaking to a customer. This is when I learned what a hobby business was. Now I am a freelance ASIC repair instructor who does repair only on the side. I can be found on LinkedIn.

I am also now trying out an association with Scott Offord. He has a five seat Wisconsin training center, I'll be  teaching in(on site is my preference though). Also, if you have an S19 repair order with him, it will likely be me doing the work. And yes, his bots likely got him in trouble(if it was something else I'm all ears). Perhaps an admin could take a look at his account.

To answer your question, I don't know. Many may have been there since before I began Feb2022, and for a year I didn't pay any attention to the S&R desk.

Regards,
Maestro Juergen

That middle hash board is more than a little low, don’t be surprised if you come in and find “SM1 loss balance” one day. you should actually go into “miner API log” and take note of which chip is acting up. This will make the repair easier later on. I’m still here to help you whenever you need.
Regards,
Jay

This is driving me nuts. Where is the ADC that is reading the power entering the board aka “actural_vol = 0.00”? I’m looking at a S17 but I believe this applies to all bitmain boards. The ADC is short for analog-to-digital converter. It’s what is required to convert a signal like a voltage swing into binary, something usable by logic circuitry.
Thanks,
J

ASIC Master update from the front lines…

Recovering the previous regime is nearly done(I will however endorse his superb teaching ability without reservation). Having finished processing our largest orders, I’ve begun contacting and fixing our small S17 series customers. Most have been understanding, I’m even offering a backlog clearance price. Only one flamer so far, it really stings being called a scammer when you are trying to make things right (in this case having to inform him we don’t repair DR5 hashboards). With the clearing of the backlog we are moving our focus to S19, M2x, and M3x series miner repair only.

As of last week, Whatsminers are the machine we can repair in the largest volume because MicroBT sent factory support staff from their headquarters in Shenzhen to train our entire staff on their machines! (I was also able to carve out some time to learn about power supply and control board repair. coming soon.)

We had a rough start this year, not gonna lie.  I’m not gonna tell you to choose ASIC Master because we have a cool name and are centrally located in Chicago. I wont bullshit people with pieces of paper that prove we can spend money and are great on tests. I’ll just be as transparent as I’m able, let you know what’s going on, and be here when you need us.

With Gratitude,
Jay

Cost and convenience.
Cost: When time is money especially for our purpose designing something and getting it running on a chip with just a few clicks of a button is huge. When you decide to take the final design from development to production the cost is rather shocking, to reinforce the first point you then have to get in line at one of the world’s fabs and wait for that process to complete. In the meantime you may already be shipping thousands of FPGA‘s running your design to clients(Who made later ask why did you ever ship FPGAs, they are so less efficient, lol). And that’s exactly what we see, when hashing moved from the CPU it spent a little time in FPGA‘s before the ASICS came out.
Convenience: When designing a chip, simulation is really slow, pushing it out to an FPGA for debug is just the most practical thing to do. Later when it’s out in the field (where it gets its name) sending out a tweak is an extremely valuable feature to have. There is programmable logic that is not reprogrammable that preceded FPGAs. It should be noted though they’re not infinitely reprogrammable in the same way ram is infinitely rewriteable.

I find this thread highly entertaining. Wether people really understand the differences between the architectures, software, and silicone processes are used to create these entirely different technologies and try to dumb it way Way way way down with various degrees of hilarity, or people just guessing based on popular belief or outright hearsay; All while the actual simple truth is that there are no lines of demarcation, everything is in ASIC, a FPGA is a Programmable ASIC, a Pentium processor is general computing ASIC(the x86 granddad the 4004 was a calculator ASIC), etc.

FPGA‘s are only in efficient because miners are using them wrong. Hell even the ZYNQ FPGA on most bitmain control boards are being used wrong. As many know control board functions can be done by generic microprocessor like the TI Arm Processor on the begle bone and the A113D ARM processor, The programmable logic area is almost completely unused. But you know what, it’s still going to be powered on(it works because, you guessed it, there are two ARM CPU‘s built into every ZYNQ, The logs look familiar because it’s all ARM Linux), and you know what they call that, inefficiency. Do you know what would be a proper use of the FPGA, a miner that switched 10 different coins/algos in one day, because if you just tried to design an ASIC that could reconfigure itself for that many different algos, when you were done you’d realize you just designed an FPGA. They don’t have to be slower either, ZYNQ’s Goliath big brother Ultrascale is an absolute beast if you want to pay many thousand dollars for a single chip.

Any questions?

The number one major design flaw of these high density 17 Series is the fact that soldered on the heatsinks only sound good in theory. In reality you have to bond copper to Silicon in order for the heatsinks to have something to solder onto. Things really turn to shit though when the ASIC gets hot enough so that the copper coating actually dissolves into the solder! Of course when there’s no surface to cling to anymore the heat sink simply falls off, sometimes they call this black ASIC. Once this occurs, you either have to replace the chip or use a thermal heat sink epoxy. You cannot rely on any other adhesive at these temperatures, the old Bitmain vinyl thermal set adhesive can’t do it, silicone can’t do it, etc.

I noticed some heat warnings in the log, if you have a Heatsinks hanging on by a thread they are certainly not carrying away heat like they are supposed to.

It is not inconceivable for a bad fan to cause a connection error. Each control board has a Minimum of four voltage sources plus the fan power supply. Being so tightly interconnected it is possible for a short in the fan to cause a brown out or voltage drop somewhere else in the control board leading to a unexplained error. Try fans plugged in one at a time, maybe even different locations. Let us know how that goes.

Looking at the log and looking at the manual can be a little overwhelming. To be slightly more specific, if you look at the manual (linked above) in section VII #4 You will find examples of the few lines of code Which you’ll find in your log which tell you exactly which hashboard has gone bad.

Step one you need a SD card that has less than 32 GB, the perfect card is a 4GB or 8GB SanDisk brand one.
Step two you need a serial USB adapter, I recommend a CP210x based model.(this is the only way you’ll know what’s going on while you are flashing your control board).
When you have these two extremely cheap items you can complete all instructions with a little help from us. No need to mail anything anywhere.

Examples:
https://www.amazon.com/SanDisk-Industrial-MicroSD-SDSDQAF3-008G-I-Adapter/dp/B07BZ5SY18/ref=mp_s_a_1_16?crid=2H98SIOP2OIVY&keywords=8gb+sandisk&qid=1665503995&qu=eyJxc2MiOiIzLjMxIiwicXNhIjoiMy4zOCIsInFzcCI6IjIuOTEifQ%3D%3D&sprefix=8gb+sandisk%2Caps%2C175&sr=8-16


https://www.amazon.com/HiLetgo-CP2102-Converter-Adapter-Downloader/dp/B00LODGRV8/ref=mp_s_a_1_1_sspa?crid=3GQ4PFXZLG8G1&keywords=cp2102+usb+ttl&qid=1665503648&qu=eyJxc2MiOiIzLjMyIiwicXNhIjoiMi43NCIsInFzcCI6IjIuNTMifQ%3D%3D&sprefix=cp210%2Caps%2C236&sr=8-1-spons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUExOVFRSTlNODgzMEpIJmVuY3J5cHRlZElkPUEwNjA2NDg4M080MzVXSkRNT1Y5UCZlbmNyeXB0ZWRBZElkPUEwMjA5OTgyMjdaNURVMDdFTVFNRiZ3aWRnZXROYW1lPXNwX3Bob25lX3NlYXJjaF9hdGYmYWN0aW9uPWNsaWNrUmVkaXJlY3QmZG9Ob3RMb2dDbGljaz10cnVl#

There is only ONE way to remove braiins and that is with their downloadable toolbox, Which conveniently enough also restores the factory image in one step. I hope you don’t Disconnect dashboard three but remove it completely.

As a connection gets warm it will oxidize. Resistors are made out of metal oxides. As power continues to flow through the resistive material it only gets hotter and eventually compass or melts the receptacle. Heavy duty receptacles have greater holding force, more points of contact and higher temperature plastics/ceramics. A stronger holding force can reduce the amount of oxygen reaching the junction mitigating one factor. More points of contact distributes the flow over wider surface reducing the initial temperature increase. And the other factor just allows the receptacle to function much longer through the degradation. There is one Super cheap option that might just allow you to continue without upgrades, dielectric grease. The only purpose of dielectric grease is the cover surfaces and prevent oxygen interaction and oxide formation.
J

I’d go a step further and say if the metal capacitors by the power rails look exploded or leaky this is the only repair you can make without the electronics expert. They are the aluminum capped ones with the blue or red striped on the side.
J

I have already explained this on page 1 exhaustively with documentation. You’re  pretty close tho.

The answer is a lot more simple than that. Efficiency is a factor of two variables, the number of transistors needed to perform the task and the nanometer process used to create those transistors. FPGA‘s are wonderfully versatile meaning there are gobs more transistors to perform a function. It’s the classic trade-off Efficiency versus function. The other factor, process just means the smaller the transistor the less physical matter there is to consume power and thus give off waste heat. In other words, When you designate a function to a FPGA you’re basically turning off everything you don’t need but without physically unplugging everything unneeded that power to suppress those transistors is still being wasted, given the number of transistors in a modern FPGA this is not an insignificant number.