2nd Front board is no working fully too! Now I only gotta figure out what is wrong with 2nd backboard that shows heater open with both frontboards even when I short Q10 to Q2. Was a not properly soldered U3.
And the sensor board that requires me to put the handle into holder after tip change for the tipchange to go away, i guess some pull down resistor or sth like that is not working. But yeah actually thinking about leaving it like this.
The idea was that you have failed J4, J8 cable, soldering won't help you unless you replace it, moreover, the "SHUNT" signal is what you need and passes through it - pin5. My guess is the time you play with soldering, you move ( heated ) that cable and that was changes for.
Yeah sorry, should have mentioned that a 2nd back board was read fine by the same frontboard + same cable (changed them multiple times and actually tried a 2nd cable too) so I was fairly sure it is not the cable itself. That is why I redid the connector and Q2 and Q8 instead. I am fairly certain that one of the legs of Q2 was actually not properly connected (maybe with high resistance).
Thank you, will see were the SHUNT/SGND goes on the front board an see if I find any faults along the way there on the none working frontboard. (like U13 and PIN12 of U5)
Actually for a single tool we have 2 resistors by Q!0 and Q11 switching. You have clean oscillograms, which means Q2 and Q8 work correctly. Look at the pull-up R24 ( for 5.2C ) for good soldering and double the J8, J4/p3 with extra cable
Thank you, got it to properly read the ID resistors work by redoing both connectors and Q2 and Q8.
But sadly now it just permanently says heater open unless I short Vout1- and Vout1+ then it starts heating the wire used to short it and displays room temp (as one would expect when doing that or not? Idk maybe it should actually show sensor open xD since I disconnected the iron). Weird part through is, I can measure like 3.5Ohm from the iron (C245 tip) at were I short it and J2, so there should not be a connection problem from there to the iron.
Any chance you know through which components it goes to measure the resistance? Also could this be cause by maybe not 100% correct calibration? Also on my 2nd Frontboard it works properly with this backboard. Only my main (first assembled front board) does the heater open.
I have some problem with instrument detection. It shows no instrument and the IRON ID in Calibration it first showed 0505. Then I redid solder joints on Q11, Q10 and U2. Now it shows 0E05. (should be 1805 for T245)
The problem should be somewhere with Q10 right? Since the resistor between Vout1- and ID is not read properly? The resistor measures at 150ohm. (Actually while the controller is on it seems to measure at like 170ohm, when its off it's 150, idk how that happens)
I captured the id check with oscilloscope. Yellow = HSEL (at U2) Pink = ID (at the header J6) Blue = Gate of Q10 Green = Gate of Q11
I don't know something is wrong with my front board. I modified my backup fronboard now too and everything is fine with that one and using the same sensboard, power board, holder and iron. room temp and holder ADC readings are stable.
Once I get a new temp sensor and new ref voltage chip I am going to replace those and if it doesn't help reflow the PIC32 on it and after replace it if the reading is still broken. As I said I took a look at the voltages on the ADC pins with an oscilloscope and they don't fluctuate like the ADC readings would suggest, if I set the rising edge trigger even just 40mV above the expected voltage from Temp sensor it doesn't trigger at all (except some times on noise I am pretty sure is external, but not how it should trigger if the temp reading on screen is from the voltage)
The thing on the picture is intentional shorting of pin 53 and 54 or what ever it was for hardware version detection. Uh 1805 is not JBC T245? What iron id should I have?
Tip change is now working properly after I fixed broken ground connection on my sensor board. But in the calibration menu even the HLD changes between 0 and like 200+ (not sure if that is normal).
So I flashed old firmware version because then whole thing started with flashing the new firmware, and I checked reference voltage and the voltage coming from the temperature sensor and they are both fine, just the displayed reading done by the ADC of the PIC32 is totally weird. And under firmware contained in https://github.com/sparkybg/UniSolder-5.2/tree/5c84f029290762bfdbef9727b845d6a508c5c5f9 the temperature reading is stable and corresponds to the voltage it shows on my oscilloscope. Will try replacing temperature chip once I get one or find the spare one, might have destroyed/lost the spare one during inital assembly through.
So today I got a round to look at a problem I have had with my unisolder since a while. The room temperature always shoots up to 30-40°C and then drops down again to 22°C (actual room temperature) over a few seconds and then the cycle repeats itself. I took a look at the voltage applied on pin 29 (Vrt) and vout of U4 with my oszilloscope and the voltage looks stable, and it does not show any significant change during those temperature change cycles. ( I would have expected it to change by spec of U4 ~10mV/°C so at least like 80mV) Should I just replace the PIC?
Also an other question, I modified my unisolder for the holder/tip detection by shorting pin 8 to 28 and 53 to 54 and added the appropriate sensing board. And it now detects having the iron in the holder and goes into Tip change when I touch the tip removing plate, but is there any way to get out of tip change mode again, because my station never does so I need to power cycle it. I would have expected it would leave it after detecting a heater open and then no heater open again or something like that?
Yes, because you are powering the board from PicKit, and as I said, PicKit cannot provide enough power for everything.
But you must now forget about PicKit3 and update your UniSolder from PC software only, so it doesn't matter any more.
I did not power it from pickit later on, yes I did the one time where I posed logs, but honestly even there I was already switch between trying to power it from pickit and powering it from the power board. I had power disabled in mplab_ipe and power on on unisolder, powering it over power connector of the unisolder and even then it only flashes bootloader correcly when I disconnect the OLED and the CNTR cable. But yeah in the end it doesn't matter anymore and I am just happy it works now.
You are definitely doing something wrong. I don't know what exactly.
Both the bootloader and firmware are OK and tested.
Yes, as I said everything works perfectly if during flashing of the bootloader I disconnect the screen and the data cable (CNTR) from the front PCB. Apparently I did that for the first time when flashing the 02.12.2020 bootloader/firmware. It works now.
If I have the screen and CNTR connected it is a failure 100% of the time
After remembering what else I did different and trying it out again. It does not work if: I flash the bootloader while the unisolder is completly connected and powered by it's own supply. The bootloader seems to be fine, and flashing firmware in the software seems to work. (except after flashing it goes dark)
I does work if: I have the data cable (and screen) unplugged from front board during flash of the bootloader and only the power cable connected. So apparently I can't have the data cable or screen connected to my frontboard or it will destroy something during flashing.
Okay, i don't know what is going on, but it didn't work with windows either when I used the firmware stuff direly from git master. But now I used the bootloader/firmware from the 02.12.2020 release and flashed the bootloader under linux with pickit3 and then used the software to flash the withbootloader firmware from inside a windows vm. and it works
Also it still keeps working after I flashed the git master firmware using the software afterwards.
You need to flash the "with bootloader" firmware, not the "standalone" :-)
Tried that too (and again just now just to be sure) and it didn't make a difference
Is there any chance that I am maybe flashing the bootloader to a wrong memory region or something? So that then the address it trys to use to call the firmware is wrong?
When flashing Bootloader the output is as follows: Connecting to MPLAB PICkit 3...
Currently loaded firmware on PICkit 3 Firmware Suite Version.....01.56.07 Firmware type..............PIC32MX Programmer to target power is enabled - VDD = 3.375000 volts. Target device PIC32MX564F128H found. Device ID Revision = A2 2021-01-16 18:12:55 +0100 - Programming...
Device Erased...
Programming...
The following memory area(s) will be programmed: program memory: start address = 0x1d000000, end address = 0x1d002fff boot config memory configuration memory Programming/Verify complete 2021-01-16 18:13:00 +0100 - Programming complete
I gonna go install normal windows instead of using it from a VM and see if that makes any difference But yeah flashed bootloader with PICkit3 and then installed the withbootloader firmware, understood that now that withbootloader doesn't mean it is with bootloader but that is should be used with the bootloader.
1. Flash the bootloader only. 2. Using the USB and the PC software, upload the "with bootloader" firmware HEX.
This is what the bootloader is made for - uploading the firmware without PicKit, but just using the PC software and USB cable.
When I only flash bootloader and then on windows load the standalone firmware using the PC software, the screen goes blank after a few seconds. And then everything is the same as if I flash bootloader+firmware with the pickit3. The bootloader is there (if I press the middle button during boot it shows the moving squares) but some how the firmware does not get loaded or idk.
Guess I just gonna live with not having the bootloader.
Open Source Hardware
