Best BST-863 hot air station

[ant17]

hi guys i thought i would contribute. I just brought a best bst-863 and have had trouble removing components. I was trying to remove a fpc connector off a old mobile logic board at first i tried 350 degrees at 50 airflow for about 2 to 3 minutes. but still would not come off. eventualy i tried 400 degrees then after about 2 minutes it came off i am thinking its out of calibration but i don't think you can calibrate this unit like the quick

[gnavigator1007]

hi guys i thought i would contribute. I just brought a best bst-863 and have had trouble removing components. I was trying to remove a fpc connector off a old mobile logic board at first i tried 350 degrees at 50 airflow for about 2 to 3 minutes. but still would not come off. eventualy i tried 400 degrees then after about 2 minutes it came off i am thinking its out of calibration but i don't think you can calibrate this unit like the quick

What flux are you using? Got a thermocouple for a multimeter? How much distance? How many layers is the board?

Haven't checked my own for accurate temps in awhile, but I'm usually set at 400

[ant17]

interflux i have thermocouple setup thats partly how i new something was not right i am not sure howmany layers

[SteveyG]

hi guys i thought i would contribute. I just brought a best bst-863 and have had trouble removing components. I was trying to remove a fpc connector off a old mobile logic board at first i tried 350 degrees at 50 airflow for about 2 to 3 minutes. but still would not come off. eventualy i tried 400 degrees then after about 2 minutes it came off i am thinking its out of calibration but i don't think you can calibrate this unit like the quick

There's a lot of factors that dictate the suitable temperature and airflow. If you're concerned, check the temperature with a thermocouple.

1 year on, my BST-863 is still working excellently despite heavy use. I also reviewed several other hot air stations recently.

[dreamcat4]

Whilst i very much like the look of the BST 863 being discussed here, i also feel that only 3 preset slots is simply not really enough given how fiddly and how many button presses it takes to make any frequent adjustment and in only 1c increments.

Actually the point was not originally my own but brought to my attention during a discussion with Paul Daniels. And I agree because when you add heatshrinking plus the ability to cope with both very thick and heavy pcbs and also very thin and small boards. Then with a preheat mode also taking up a slot that is at least 4 minimum. However Paul says he actually needs as many as 6. For not needing to go back and keep making a manual adjustment.

Now I did email BEST company already (many months ago) asking them to make an update to the BST-863 just with better manual controls. Rotary encoders. However I myself am not high profile enough to bring attention to this. My request has fallen upon deaf ears.

So what about the possibility of modifying the firmware of this unit, to support a greater number of presets instead? The LCD has these 3 fixed symbols one for each preset. So I was just wondering if you can press or hold down 2 of the preset buttons simultaneously. And have than registered by the microcontroller. Then have a software mod that illuminated a combination of 2 preset symbols together as a pair. This would then let the capability to add an additional layer of 3 extra presets. Allongside the original 3 presets.

However if 2 button presses (or held down) cannot be registered simultaneously on the MCU. Then that would not be possible or elegant. At least not without some type of an additional hardware modification. For example to access unused GPIO pins and rig up exttra external buttons. Which of course is not really so nice when the front panel is already nicely mooulded etc. Which is also why I requested that BEST (the manufacturer) do a new front panel / revisions themselves in the first place.

I know making a custom firmware may sound a bit crazy however you'd be surprised its already been done to a much greater extend for other products such as the TS100 soldering iron, the STM32 T12 OLED soldering controller, and also the really cheap fan handle 858D+ hot air station.

Wheras i'm talking about is the ability to store some extra presets. Where those preset management routines already exist. And just need a bit of finessing. Rather than completely rewrite an entire firmware from scratch.

So what do you guys think? Wouldn't those be the next logical improvements to this unit? Improving the controls.... Considering that it already performs in almost all its other aspects so well? Who's with me here eh?

 

[ercapoccia]

Would the Quick nozzle fit the Best? Under the microscope bent nozzle are a must.
I

[kubino]

I ordered nozzles for QUICK and they fit nicely. No problem here ;-)

[SpecialK]

Just in case anyone is interested, here are the insides of the BST-863. Seems pretty good to me, I've used the Quick 856AD extensively and it pretty much does the same job (without the footpedal). Regarding the touch screen, it seems to work through gloves, but once you've set the presets you'd not need to touch the screen anyway.

Product link is here: https://bit.ly/2XRCkAQ

I've been looking at hot air stations for some time.  Hobby use, so I wasn't going to spend big money.  After seeing your videos and Voltlog videos, I was impressed.  I was able to pick this unit up for about half the price of a Quick.

I must say that this unit is quiet.  The 120V version I have seems to have a subdued fan compared to the 240V versions.  I use mine at 50% to 58% airflow, and the power up/power down squeal is very slight compared.  It really seems to remove components like butter at 400 to 410 celcius.  Is that too hot?

 

[wraper]

It really seems to remove components like butter at 400 to 410 celcius.  Is that too hot?

Yes, it is too much. I do not suggest going over 360^oC unless it's very hard to do the job otherwise, and preferably stay at 350^oC or lower. 400^oC as minimum will burn a component surface, especially noticeable on yellow tantalum capacitors which become brown. Also many Chinese stations actually output higher temperature than on the display to appear working better than they actually are. In that case you would need to set temperature even lower.
EDIT: this is for lead free. With leaded solder you should drop the temperature by about 30^oC

[SpecialK]

Okay. That's good to know.  So far I've just been trying it out on stripping components on some practice boards.  I would imagine there should be a certain dwell time and if it takes longer then you need to up the temp or airflow?  Any rule of thumb?

[luwii]

Thanks for photo of the board.

I took R12 off, instead of going through the hassle of taking too much apart. That stopped the buzzer, and the rest seems to work fine.

On mine, R12 is 10 . I then played around with a variable resistor in it's place and settled on changing to 330 . (A little higher value might be nice, but I did not have anything suitable in stock.)
It is not nearly as loud, but the up/down sound turns into an annoying chirping sound. Decided I do want some buzzer feedback, and hope that the up / down is not used too much. Time will tell if it take it out again.

[MrBean]

Hi All

First post from a newbie to electronics who's bought a Best BST-863 to play with 

I got it from Banggood, and after opening it up I first sorted the grounding issue out by filing the paint away below the earthing points on the case, however I'm perplexed by the internal wiring. It seems (photos attached) to my not very knowledgeable eyes that the AC plug is wired up incorrectly internally, with the live female spade connector going to the neutral male and vice-versa (this is the same on SDG Electronics' unit but different on VoltLog's unit - both reviews/teardowns on YouTube). Then where the wiring runs into the EMI filter, the live and neutral wires have been switched round on in/out (does placement matter for this?). Then on the screw block connector for the front panel, where on SDG Electronics' unit there are live wires, I have neutral wires and vice-versa. I had to tighten the screw connectors up - One was nearly falling out! The front board is quite messy too, lots of flux everywhere, unlike all the other internals I've seen. Mine didn't have the Best holographic sticker on the case, but I imagined they'd just ran out that day or decided to stop using them...

Would you knowledgeable folks please be able to take a look and tell me if I'm being silly or if it might be an issue? I haven't plugged it in yet as I'd like to check it's all OK first...

Thanks in advance! 

MrB

[fatmir]

Hi,

i tried both Best BST-863 and quick-861DW. I think there is no difference in the work. Just the touch panel at Best BST-863 worries me a little?
I would also like to try the ATTEN ST-862D, if the opportunity arises to test.

best regards
Fatmir

[RandomHacks]

I got this station April last year, the price is good for the quality but the firmware/interface... 
I got so annoyed by it that I designed a board from scratch. I'll leave the links to the videos here in case anybody is interested.
No more screaming buzzer

Presentation:


Installation:

[dreamcat4]

 

Awesome man!! You did it!  Pretty darn awesome... Looking forwards to it!

 

Very good work there. Can't wait

[tatel]

I got this station April last year, the price is good for the quality but the firmware/interface... 
I got so annoyed by it that I designed a board from scratch. I'll leave the links to the videos here in case anybody is interested.
No more screaming buzzer

Man, I have been looking for a hot air station for some months now. This had gone under my radar.  Then your post made me aware of it. It's now on Banggood for some €140, shipped from Spain. I got it and also ordered one of your kits from Tindie.

You are right, your board isn't the cheapest thing in the world  But, hey, I love open source things. I find there is value in being able to  fiddle with the FW, should the need arise. Moreover, it already seems to overcome the drawbacks exposed on this thread. And I'm willing to support people that makes us free of vendor-locking. So I'm going to happily give it a try. 

I think this combination can be good, better than getting the Quick 861DW for about €260 plus VAT, shipped from EU, without open source firmware. You have a good selling point.

Just two details: 1) I think you should really post a link to the code and 2) Although this is a minor consideration, the knobs should be included in the kit.

Best wishes

[SpecialK]

This looks pretty good to me.  I'm impressed.  I might just order one.

[dreamcat4]

OK i have a question about the inner workings of this device. So ones of the PCBs, is square shaped and it sits on the bottom next to the fan. And it is dual purpose. Both the PWM fan controller, but also it supply +15v DC power to the front panel control board (for MCU display etc).

And there is a simple 4 pin header in the top left corner of the photo here (see below). Markings are:

[+15, GND, A1, PWM].

Clearly when you set a fan speed on the front panel touch display, and unhook the handle. Then the MCU controller should be sending a pwm signal on the pin #4 PWM. Between 0 and 100% pwm. Is that correct?

Because on my machine it never happen. It has aproblem. So when I unhook the handle there is no PWM signal. At least not at any frequency that is low enough for me to detect here on my poor DSO150 toy.

And the A1 pin always remains at +5v.

Clearly I would like to debug this issue further. So I was just hoping to get some help for confirming that I should drive the board with +5v pwm signal? And maybe at a low frequency < 30khz ?

Because if the PWM freq is much higher then I cannot see it on my test equipment here. The DSO150 is just a toy. It does not go much higher than about 40khz without aliasing.

Also when I trace the A1 pin on this fan control board... it seems like an output signal from the fan controller, to tell the control board it's ok and ready?

Because this remains at +5v, maybe this pin must be dragged low, before the front panel MCU will actually emit any PWM fan signal? (or if not, then was just not capable to detect it, because PWM frequency is much higher... hehe)

Any help appreciated. Many thanks.

I was really also just wondering where the issue occur. Because since ordering the new upgrade kit. Then it will replace the old front panel main control board. So depending on the location of the problem, whether fix or not. Many thanks.

[dreamcat4]

Ah ok, to answer some of my own questions:

After disconnecting A1 and PWM pins from the front panel. Then I was able to see a +5v pwm signal being generated by the front panel, and it's at 10khz freq.

At least during power on cycle, for 2-3 sec. It does not seem to matter if A1 pin is tied to GND or not. It is the fan control board, pluggin in the PWM pin, it is dragging down the pwm signal to shallow bumps.

The other thing is after removing the handle. To begin with I was seeing a PWM fan signal on presets 2 and 3 (but not on preset #1). Which was reliable at first (and at that time the A1 was initially tied to GND). But then after some few times of power cycles. Then it stopped doing this. And now there is 0v pwm after lifting the handle. Despite the fan icon appear and then temp reads 0c.

So it leads me to believe that perhaps the front panel PCB is working. But not the fan control PCB, that one perhaps might have some type of an issue.

[RandomHacks]

I got this station April last year, the price is good for the quality but the firmware/interface... 
I got so annoyed by it that I designed a board from scratch. I'll leave the links to the videos here in case anybody is interested.
No more screaming buzzer

Just two details: 1) I think you should really post a link to the code and 2) Although this is a minor consideration, the knobs should be included in the kit.

Thanks for the feedback. This is the github repository for the code but I still have to organize some things before I post it, soon. https://github.com/RandomHacks-Git/Best-Improvement
As for the knobs I did order them ages ago but they never arrived and that is why they aren't included. I will order some more on july 1st as hopefully aliexress/ebay will start to collect vat at checkout because I don't feel like paying a service fee to the post now with the new customs rules.

Ah ok, to answer some of my own questions:

After disconnecting A1 and PWM pins from the front panel. Then I was able to see a +5v pwm signal being generated by the front panel, and it's at 10khz freq.

At least during power on cycle, for 2-3 sec. It does not seem to matter if A1 pin is tied to GND or not. It is the fan control board, pluggin in the PWM pin, it is dragging down the pwm signal to shallow bumps.

The other thing is after removing the handle. To begin with I was seeing a PWM fan signal on presets 2 and 3 (but not on preset #1). Which was reliable at first (and at that time the A1 was initially tied to GND). But then after some few times of power cycles. Then it stopped doing this. And now there is 0v pwm after lifting the handle. Despite the fan icon appear and then temp reads 0c.

So it leads me to believe that perhaps the front panel PCB is working. But not the fan control PCB, that one perhaps might have some type of an issue.

The A1 pin isn't used at all by the station (not on my stations atleast), it is supposed to be a feedback pin to check if the motor is running but they messed up the schematic of the ESC board so you don't get any signal from that pin (I looked at this ages ago so I don't exactly remember, but there is a current limiting resistor (in series of course) and immediatley after there is a pull down (or pull up? don't remember) resistor with a lower value so it drags the signal down (or up?) I'm sure I wrote the exact issue somewhere on my notes I would have to look it up, but long story short, the pin isn't used and can be unplugged with no effect. Swapping the two resistors around would solve this but it isn't implemented in the firmware on either the original board nor mine although I left the pin connected to the mcu in case someone would want that (I don't want to get into much detail as I don't quite remember the circtuit they used on the esc but I believe they tied the feedback pins of the chip together so if the motor is not running but twitching between two hall sensors there still wouldn't be a way to detect a fault)

So your blower does not work at all? You should have a PWM signal of around 10kHz on the PWM pin as you said, to check where the problem lies I would suggest injecting a signal on the ESC's pwm pin with a signal generator or arduino (don't forget to tie the GNDs together).
 

[dreamcat4]

Ah very interesting...

So then as you say if the true purpose of the A1 seems to be to return a feedback signal to the MCU, to say the motor is in fact spinning around. Because if it is not blowing any air (with a valid PWM signal)... then the MCU should probably turn off the heater. To prevent damage or overheating in that error condition. Since the heater requires an airflow to keep it cool. So this might also help to explain why there is no heating ATM. And the screen reads "0c". If it did not receive fan spin return signal.

OK so looking at what you are calling the ESC board (sorry I do not know these terminology so well). It is in fact an Allegro A4931 brushless DC fan controller chip. And so this is a great start already, since we know BLDC motors require pulsing at the right times and although I know almost nothing about bldc. Clearly there is a great potential for some high current upset or pulses / backfeeding voltage spikes etc. All on that BLDC board.

So while I still did not begin to debug the real issue (yet). I can see maybe a clue in my microscope, which is between the pin header contacts for the motor. Because between the different pins, and phases of the motor wires are very thin and long tin whiskers.

Actually I already broke them up mostly, in trying to wipe them away too soon. But clearly this is due to the lead free solder. And also the high current pulses due to BLDC motor driving. Which in combination with no protective coating etc... there was probably some sort of bridging occured. Due to these tin whiskers.

So maybe an event occured during operation, with a current from one coil got dumped to the pin for another coil. Or one of the hall sense wires. And then some large current or induced voltage spike on the wrong connection. Or at the wrong relative timings. Whatever happened. This seems the most likely reason for the failure as far as my limited knowledge permits. Because since this BLDC fan seems reasonably somewhat powerful output device, it might be enough to cause some sort of a damage.

Although these tin whiskers are really thin filament. So maybe if i just clean them up and no damage was done here. Then it might start to work again. If it is just a safety cut out feature engaged by the BLDC controller chip IDK.

But for everybody else out there with these stations.... I think you guys all should be unplugging and removing this board. And inspecting it for the tin whiskers. Then cleaning it up. Then maybe just apply some silastic, or other physical barrier underneath the long pin header pins. (that the soldered pins sticks out on the under side).

For the physical barrier to create IDK what is best. But maybe just some silastic, or glue gun. Or conformal coating IDK. Should be enough for only +15v DC. Its a low voltage.

Please take this advice because they really will grow over time. The more hours of operation for the fan and get worse. So it is worth the hassle to protect against this.

[RandomHacks]

Ah very interesting...

So then as you say if the true purpose of the A1 seems to be to return a feedback signal to the MCU, to say the motor is in fact spinning around. Because if it is not blowing any air (with a valid PWM signal)... then the MCU should probably turn off the heater. To prevent damage or overheating in that error condition. Since the heater requires an airflow to keep it cool. So this might also help to explain why there is no heating ATM. And the screen reads "0c". If it did not receive fan spin return signal.

OK so looking at what you are calling the ESC board (sorry I do not know these terminology so well). It is in fact an Allegro A4931 brushless DC fan controller chip. And so this is a great start already, since we know BLDC motors require pulsing at the right times and although I know almost nothing about bldc. Clearly there is a great potential for some high current upset or pulses / backfeeding voltage spikes etc. All on that BLDC board.

So while I still did not begin to debug the real issue (yet). I can see maybe a clue in my microscope, which is between the pin header contacts for the motor. Because between the different pins, and phases of the motor wires are very thin and long tin whiskers.

Actually I already broke them up mostly, in trying to wipe them away too soon. But clearly this is due to the lead free solder. And also the high current pulses due to BLDC motor driving. Which in combination with no protective coating etc... there was probably some sort of bridging occured. Due to these tin whiskers.

So maybe an event occured during operation, with a current from one coil got dumped to the pin for another coil. Or one of the hall sense wires. And then some large current or induced voltage spike on the wrong connection. Or at the wrong relative timings. Whatever happened. This seems the most likely reason for the failure as far as my limited knowledge permits. Because since this BLDC fan seems reasonably somewhat powerful output device, it might be enough to cause some sort of a damage.

Although these tin whiskers are really thin filament. So maybe if i just clean them up and no damage was done here. Then it might start to work again. If it is just a safety cut out feature engaged by the BLDC controller chip IDK.

But for everybody else out there with these stations.... I think you guys all should be unplugging and removing this board. And inspecting it for the tin whiskers. Then cleaning it up. Then maybe just apply some silastic, or other physical barrier underneath the long pin header pins. (that the soldered pins sticks out on the under side).

For the physical barrier to create IDK what is best. But maybe just some silastic, or glue gun. Or conformal coating IDK. Should be enough for only +15v DC. Its a low voltage.

Please take this advice because they really will grow over time. The more hours of operation for the fan and get worse. So it is worth the hassle to protect against this.

Does your blower spin up at all when turning the station on?
I just installed the original board back on one of my stations to test this. From what I understood you get a PWM signal only for a few seconds after turning the station on and you get 000ºC displayed on the screen right? this is the behaviour of turning the station on without having the handle in the support which makes me believe you have a faulty reed switch (magnetic switch) inside the handle. Check for continuity between the black and white wires on the first JST connector from the left, it should be close to 0 ohm if the handle is in the support and open if not.

After checking it turns out that if you tie A1 to GND the station also shows 0ºC and does not heat up but the blower works so I guess it is either the reed switch or something is pulling A1 down on your station.

[dreamcat4]

Does your blower spin up at all when turning the station on?

No, the fan never spins. However I can detect the PWM signal on my DSO150 toy oscilloscope. It a clean PWM signal during power on when the link is disconnected from the esc board.

However the PWM signal is very much smushed down (over loaded) when it is connected as factory. So it seems like the esc board is loading it. However when I measure to GND the pwm pin on the esc board, i get 10kohm. So it seems like some kind of an active circuitry doing that loading. Like going through a little transistor. Sorry I have not had time to check the data sheet yet for the BLDC controller.

I just installed the original board back on one of my stations to test this. From what I understood you get a PWM signal only for a few seconds after turning the station on and you get 000ºC displayed on the screen right?

Correct.

this is the behaviour of turning the station on without having the handle in the support which makes me believe you have a faulty reed switch (magnetic switch) inside the handle.

Yeah they are kindda dodgy. I added some extra magnet in the handle today. What still happens (after adding missing magnet) is if the handle gets stright vertical. Then the reed switch works well (no matter the rotation). However if the handle becomes tilted a little bit backwards in the cradle. Then there is some roatations positions that the reed switch does not close. And the station thinks handle is out of cradle (when it's just tilted in cradle).

Check for continuity between the black and white wires on the first JST connector from the left, it should be close to 0 ohm if the handle is in the support and open if not.

So yes, I just tested (thank you to telling me the wires!). And my reed switch does work as it should. The reed switch is not stuck. It will open and close depending on the magnet.

The only improvement left there is to help stop the handle tilting backwards so easily. Maybe with some extra added 'half' strap or loop going around behind, or other modification to the cradle. Like a way to jam it vertical.

Thanks though! Is helpful to debug the reed switch better. To make sure it works better. Hehe 

=====

The FAN issue seems the ESC board at this point.

Anyway, here is photos of the tin whiskers. This was after breaking them off (not realizing before). So there was more than 1 whiskers per solder joint. And clearly the whiskers a plenty long enough to bridge and reach the next contact. Before I messed them up sorry!

Just wanted to make sure there was at least some kind of photo. Before cleaning up the contacts and/or looking somewhere else.

I am not sure if as a current pathway, it was solid enough to cause a permanent damage. Or if there was any bigger whisker before. Sorry i didn't pay enough attention. But at least even a thin whisker could be enough to confuse the BLDC motor controller driver. Which needs to know the rotation position etc. To stop it from working properly.

Clearly no real experience before with any BLDC circuit. But maybe it can help for others who might get a similar issue in future. If you have any other comments to add please say so! 

[dreamcat4]

The A1 pin isn't used at all by the station (not on my stations atleast), it is supposed to be a feedback pin to check if the motor is running but they messed up the schematic of the ESC board so you don't get any signal from that pin (I looked at this ages ago so I don't exactly remember, but there is a current limiting resistor (in series of course) and immediatley after there is a pull down (or pull up? don't remember) resistor with a lower value so it drags the signal down (or up?) I'm sure I wrote the exact issue somewhere on my notes I would have to look it up

OK so (this is all on the ESC board) for the A1 pin i can see now, what they did. The have tapped off the HA wire from the motor itself, which gives 1/3 of the hall sensor click for the motor rotation. This HA pin on the motor header appears as R7 next to the bldc controller IC.

Then R7 is a 25kohm resistor, which connects up a ladder, to an NPN transistor on the left side of the board. But also with a smoothing capacitor. So if the motor is rotating, the base of the NPN transistor is supposed to become active. And this should then conduct the CE and ground the A1 pin.

But if the motor stops rotating. Then there will be no trigger signal repeating from the HA --> R7. So the capacitor cannot keep the transistor fed with current. It will decay, and the NPN transistor will turn off. This will then mean there is no path to GND anymore, and the A1 pin will be pulled HIGH (=+5v).

So it is not matter though, if the controller ignored this signal. Or I can just drag it to GND myself on the controller side. So long as I do not also end up on the other side of the transistor (before the Base), because dragging the HA line low would confuse the BLDC.

I think for debugging I should (power off) manually rotate the motor by hand. And check that all 3 of the hall sensors inside the motor are triggering (HA, HB, HC). So to know the motor is OK.

Hehe such a noob with these BLDC stuff. Please ignore me while I continue to learn more. Must read the datasheet too would help 

[RandomHacks]

The A1 pin isn't used at all by the station (not on my stations atleast), it is supposed to be a feedback pin to check if the motor is running but they messed up the schematic of the ESC board so you don't get any signal from that pin (I looked at this ages ago so I don't exactly remember, but there is a current limiting resistor (in series of course) and immediatley after there is a pull down (or pull up? don't remember) resistor with a lower value so it drags the signal down (or up?) I'm sure I wrote the exact issue somewhere on my notes I would have to look it up

OK so (this is all on the ESC board) for the A1 pin i can see now, what they did. The have tapped off the HA wire from the motor itself, which gives 1/3 of the hall sensor click for the motor rotation. This HA pin on the motor header appears as R7 next to the bldc controller IC.

Then R7 is a 25kohm resistor, which connects up a ladder, to an NPN transistor on the left side of the board. But also with a smoothing capacitor. So if the motor is rotating, the base of the NPN transistor is supposed to become active. And this should then conduct the CE and ground the A1 pin.

But if the motor stops rotating. Then there will be no trigger signal repeating from the HA --> R7. So the capacitor cannot keep the transistor fed with current. It will decay, and the NPN transistor will turn off. This will then mean there is no path to GND anymore, and the A1 pin will be pulled HIGH (=+5v).

So it is not matter though, if the controller ignored this signal. Or I can just drag it to GND myself on the controller side. So long as I do not also end up on the other side of the transistor (before the Base), because dragging the HA line low would confuse the BLDC.

I think for debugging I should (power off) manually rotate the motor by hand. And check that all 3 of the hall sensors inside the motor are triggering (HA, HB, HC). So to know the motor is OK.

Hehe such a noob with these BLDC stuff. Please ignore me while I continue to learn more. Must read the datasheet too would help 

I might be looking at this wrong but if the resistors on your ESC are the same as mine that little transistor "T7" will never turn on because R20 which is in series is 100k (+ the resistance of R7, R8 or R9) and R19 which is a pull down immediatley after is 10k so I don't think whatever tiny voltage that reaches the base of T7 will ever be high enough to turn it on and A1 will stay floating, if you swap R19 with R20 you will be able to see the signal on A1, I think they messed this up in the design.

The 5v you see on A1 come from the main board (pullup), if there are still 5v on that pin while it is connected to the ESC, this should not be what is causing the problem, as far as I can tell the station only shows 0ºC if this pin is grounded (or the handle isn't in the cradle/faulty reed switch as I mentioned earlier but we already ruled that out).

My R7 is actually 10k so maybe they changed the values. Did you try do disconnect A1 only and check if the blower works that way?

Those are some impressive tin wiskers, never saw anything like that in real life