Quecoo T12-958 soldering station - INVER error?

[redwire]

There must be a hardware change/mod to fix the INVER problem. If you can post a pic so we can go over it- I know the thermocouple op-amps some are crap quality, and that heater current thing seemed to trip way too high so the resistors I would guess got changed.

The schematic looks pretty good now

[spleenharvester]

Cheers redwire. Think I have fixed it all, attached V04 schematic. I've checked and rechecked the ADC section many times and can't find any errors, maybe someone else with this board can confirm.

As an aside, I've received my replacement T12-958v2. Interestingly although this one still shows as software V3.2S and has identical PSU and controller boards it does not suffer the INVERT error.

[macklij]

if it's STlink 2.0 or 2.1 you dont need programming software,
the stlink will show up like a usb flashdrive and you just copy the BIN or HEX file into it.
thats it - it will program the chip for you.

Thanks - I haven't played with the STLink yet

[redwire]

Hi all

My board hasn't actually got a version number printed on it at all, but software reports HW revision 3.2. I've fixed the BOOT0 and RTC issues in my schematic (I had BOOT0 connected like that as the traces are physically connected to it, but as someone said on EEVBlog it just seems to be grounded out anyway).


R6 is definitely 5 milliohm, has R005 printed on it. I have rechecked and am also convinced the other resistors are correct in that part of the schematic. I've added a missing 4K7 to ground on V+ but I don't think that makes a difference as it's not on +in or -in?

With that V03 schematic, the "ADC" section U7 trips above 26A if it's R6 0.005Ω sense-resistor on heater return. I calc 0.130V at the op-amp with 2.5V/20k/3k.
It could only be for a fast circuit breaker with a shorted cartridge? You're not getting 26A out for long, just a blip.
Hot-cartridge swaps can short at the handle/socket connector and blow a power mosfet, maybe Quicko is trying to cover for that.

The Temp sense sch is still wrong - op-amp is in non-inverting config, so the wires to R10 are flipped. It should be op-amp output to R10 with R23 to GND and the junction of R10/R23 only goes to (+) pin 3. op-amp out also goes to R18.
op-amp (-) input pin 1 is the one that goes to heater signal. So you have pin 1, 3 flipped as well.
Ferrite bead FB1 is in the wrong spot or it is more likely a NTC thermistor for cold-junction temp.

[stj]

if it's STlink 2.0 or 2.1 you dont need programming software,
the stlink will show up like a usb flashdrive and you just copy the BIN or HEX file into it.
thats it - it will program the chip for you.

[macklij]

Thanks both. Yes, I intend to use the STLink with the header pins at some point. I added them when I did the grounding mods for the tip and case (There was already a ground lug on the back panel, but with the case anodising it wasn't being conducted around the case). The ground pin of the 5 pin iron connector wasn't connected to anything either, meaning the soldering iron tip was floating.

[spleenharvester]

macklij The board already has header pins to use with an STLink which makes life much easier - just solder a header on and connect any STLink adapter, making sure the station isn't powered up at the same time. The T12 CFW github page has a really good guide on how to use the software.

[stj]

STlink is a debugging interface that uses the SWIM pin on the mcu - it's a serial jtag interface.
to program with the serial adapter you just use the boot0 jumper/pin and TXD / RXD on the mcu
and this software
https://sourceforge.net/projects/stm32flash/

[macklij]

yiou dont need STlink to program the chip, a usb to serial convertor with a 5v or 3.3v output to match the mcu voltage will be fine.

Thanks. I have both, but have never used the STLink. I don't even know if it works.

[spleenharvester]

Hi all

My board hasn't actually got a version number printed on it at all, but software reports HW revision 3.2. I've fixed the BOOT0 and RTC issues in my schematic (I had BOOT0 connected like that as the traces are physically connected to it, but as someone said on EEVBlog it just seems to be grounded out anyway).


R6 is definitely 5 milliohm, has R005 printed on it. I have rechecked and am also convinced the other resistors are correct in that part of the schematic. I've added a missing 4K7 to ground on V+ but I don't think that makes a difference as it's not on +in or -in?

[stj]

yiou dont need STlink to program the chip, a usb to serial convertor with a 5v or 3.3v output to match the mcu voltage will be fine.

[macklij]

And of course, there is no saying if the original 3.2 firmware will figure it out... or the custom one. I think I'll calibrate some tips and see how it goes with the original firmware. If I hate the menu structure then think again.

Thanks

[redwire]

I'm not sure which firmware load to use.
If a heater cartridge goes open-circuit, and these T12 clones are notorious for crappy carts/sockets and connections, there are two ways to tell.
Heater current will not be there when the mosfet is on, or the thermocouple voltage will peg when the heater mosfet is off. Good firmware can figure it out. But it is a shortcoming for some people to think outside the box.

[macklij]

Thanks both. So presumably their firmware is making use of this extra hardware, at a minimum to determine if the cartridge is fitted and working. Does the custom firmware determine the cartridge state in a different (perhaps inferior) way?

I'm basically just trying to decide if I should try the custom firmware or not. There is always the risk of bricking the unit or not being able to go back to Quecoo firmware. But I do have an ST-Link that says 'use me' on the box (well I think that's what the Chinese says)!

Thanks again.

[redwire]

Using the Post #24 schematic, U7 LMV321 is wired as a comparator but values with R6 5mΩ sense resistor and 130mV at pin3, it trips at >26A which does not make sense.
If it was 50mΩ then >2.6A trip seems reasonable? It does look like it only checks if there is heater current above a certain threshold.

Quicko put in the extra hardware to monitor this. I kinda like their controllers better.

[stj]

it's probably monitoring current just to detect if the cartridge is fitted and working

[macklij]

...
What's interesting is Quicko is also monitoring heater current, this is not done anywhere else.

I've just (today) received a Quecoo T12-958 that looks to be this version (3.2 + 3.2S) and am reading this thread with interest. Thanks for the interesting posts. 🙂

I had planned on installing the custom firmware. But if this Quecoo/Quicko is monitoring heater current, does it mean that the custom firmware should be avoided because the stock firmware is better?

[redwire]

What version # is written on your controller board?
When I get time, I'll review the schematic and mention what I think needs to be fixed. Like the thermocouple op-amp circuit does not make sense. BOOT0 is just connected to GND.
I forgot I was drawing this schematic for Quicko STM32 Ver3.3 controller last year, it has a different power supply than yours but some basics should be similar, including I/O assignments.

[spleenharvester]

I can see a few mistakes in your posted schematic. For alternate firmware like David's you need the I/O pin assignments.
What's interesting is Quicko is also monitoring heater current, this is not done anywhere else.
I've drawn a bunch of detailed KSGER T12 controller schematics. I quit after people took my schematics made a few edits and plastered their name all over it, for fame and glory.

Yeah more than likely there are some errors, I have no electrical engineering background so don't really have any idea what I'm tracing. Which bits are wrong as I'll correct them? Have found the RTC missing 3.3V assignment since,


All the connected I/O pins are there, David was able to produce CFW (well not from this one but someone else's which got the same pin assignments as mine). I mostly finished it for fun really!

[stj]

monitoring current is good for a cartridge system, dont want to blow a fet because a cartridge isnt fully plugged in!