CFW for KSGER/Quicko STM32 Soldering Stations

[sevgiduvari_7]

counterclockwise, standby and sleep works fine. But when you turn it clockwise, the BC3 icon appears on the screen only and the boost does not work. Please one last adjustment.

[sevgiduvari_7]

When you press it and turn it clockwise, the boost will increase to +50 and it will return to normal operation to -50 at the end of the set time, please, a final touch. Also, although I choose sleep at the beginning in the settings, there is no sleep at the beginning, it starts to work normally.

[DavidAlfa]

Do not push while rotating. Nowhere is said that to set boost mode.
Rotate normally to make the setpoint screen appear, then click.

Rotate anticlockwise while pressed: Set sleep and standby modes.
Rotate clockwise while pressed: Enable tip selection.
Rotate ONLY, no press: Shows setpoint screen. Click before 1 second to trigger boost mode. If clicked later, it will be ignored.

For the sleep issue, go to Iron screen, below "Err Time" you'll see "Resume". Set it to "LAST".
I have to fix the issue but that's the workaround for now.

[sevgiduvari_7]

Unfortunately, boost doesn't work. When you don't click, the setting goes up. If possible, reverse the boost setting with the tip scimi so that the tip does not select but the boost works.

[sevgiduvari_7]

For the sleep issue, go to Iron screen, below "Err Time" you'll see "Resume". Set it to "LAST".
thanks sleep solved please get boost.

[DavidAlfa]

Sorry, I don't know what you're doing. If settings are showing, you're clicking for a long time.
Mobile users: Warning, link to 2MB gif!
It's really this simple

When booting, there's always some error detected before the voltages and sensor readings are initialized.
Thanks for reporting the booting issue, I never had it, as I always had the resume mode in "Last".
I quickly figured out that it was the Resume option overriding the Boot mode.

Fixed the issue by adding a boot flag, set when exiting the boot screen.
If any error is cleared before the boot flag is set, force the configured boot mode. Otherwise, set the Resume mode.

I also increased the shake filter timing window: Accept changes happening between 100-500ms, instead 200-400ms
Still, it's good to ignore very slow or very fast changes, usually it's just vibration or noise.

Hotfix builds released. It's so much easier now I automated it...


It there's something confusing in the operation manual, let me know.
I want to avoid the same questions being asked everytime.

[Polf]

David, the new firmware is working great! Thanks!

[wickated]

main problem with these DIY T12 stations is simply they are manufactured not by companies but by chinese amateurs on CNC thats why 99,9999% of these soldering stations have funny potentiometer (rotary encoder) instead of three simply buttons to set temperature !!! it seems they have big problem to drill 3 smallest holes for buttons to make these stations managment much more better just take a look on aliexpress most or (whole) of oldest design 936 soldering stations to use with older tips technology have buttons to setup temperature like they should because they manufactured by chinese companies !

1) u always have throughholes for three buttons on pcb. since u need them, just buy pcb only and add buttons instead
2) encoder is not pot
3) angle of screen does nothing with job. its not a clock

[wickated]

When you press it and turn it clockwise, the boost will increase to +50 and it will return to normal operation to -50 at the end of the set time, please, a final touch. Also, although I choose sleep at the beginning in the settings, there is no sleep at the beginning, it starts to work normally.

how the holy f// did u manage to install custon firmware?   

[sevgiduvari_7]

it was very nice with the new software, thanks it was boost

[M-Reimer]

I've added a socket to my station which allows to plug in an external stand switch.

I want to directly connect an original Hakko stand to this so I followed their pinout. I guess if copying a feature from the original Hakko station it may be a good idea to follow their pinout so original Hakko equipment can be connected. I even took some photos to document how a socket can be added in a way that the shake switch reconnects if no plug is inserted.

But now there is another problem where I want to ask first before doing any mods to anything.

For some time I was wondering why my switch does not work at all or only very unreliable. Until I realized that, if the switch is open, the pin, it is connected to, only rises to about 1.6 V and not all the way up to 3.3 V. So I finally took apart the Hakko stand and I found that they have an 4.7k resistor in parallel to the switch. So basically this switch toggles between 0 Ohms and 4.7k Ohms in this setup. My guess is that Hakko does this to detect if a stand is connected at all.

0 Ohms: Stand connected, Iron in place
4.7k Ohms: Stand connected, Iron not in holder
open connection: No Stand switch connected

I guess that the processor pin which is used for the shake switch is not an analog input port and so it is not possible to detect voltages below the "logical high" level? Would it be safe to add a bit of additional pull up to raise the voltage up to a region where the station properly detects it? If possible I would really prefer to make my new "switch socket" compliant in a way that original Hakko stuff works on without modifying it.

EDIT: Actually I have found an hacky solution. If I add a diode in series to my socket, then this adds half a diode drop (around 0.6 V) to my 1.6 V which results in around 1.9V and this is slightly above the lower "minimum" that is described in the STM32 datasheet for "logical high". Seems to work but I still think, if possible with not too much effort, reading the pin in "analog mode" would actually be nicer.

[DavidAlfa]

Nope, I already thought that ages ago... The shake input doesn't have analog connection in most boards.
Remove the 4k7 resistor, easy.

[ricktendo]

Erm... 65 euro for that, must be a joke... 

Agreed, but its kinda interesting to see the new irons and the different menu item names and layout. Here is a video with teardown at 12:45 (it appears to use Atmel micro)

[mastershake]

well they also copied the malectrics welder and are charging like 250$ for it

[lfldp]

main problem with these DIY T12 stations is simply they are manufactured not by companies but by chinese amateurs on CNC thats why 99,9999% of these soldering stations have funny potentiometer (rotary encoder) instead of three simply buttons to set temperature !!! it seems they have big problem to drill 3 smallest holes for buttons to make these stations managment much more better just take a look on aliexpress most or (whole) of oldest design 936 soldering stations to use with older tips technology have buttons to setup temperature like they should because they manufactured by chinese companies !

1) u always have throughholes for three buttons on pcb. since u need them, just buy pcb only and add buttons instead
2) encoder is not pot
3) angle of screen does nothing with job. its not a clock

ok but point 1 need to modiffy panel case too

[M-Reimer]

I decided to keep the diode in. Seems to be pretty stable this way without modifying anything at the stand at all. And worst case the voltage level is still too low and the iron does not properly leave standby mode. Not that big of a deal. I can still fix this problem if I really have issues in future.

Maybe I now have found the first small issue. I have a "Standby time" set in the "IRON" menu. I also would like to keep this in case I unplug the holder and want to use the small station "on the go" with a holder without switch. In this case I only want to change one setting (stand to vibration switch). The way it is programmed currently, a configured "Standby time" combined with "stand switch mode" causes the iron to go into standby while working with it if it is not placed regularly to the stand. In my opinion in "Stand mode" the "Standby time" setting should be ignored. Only the "Sleep time" should be still active but the iron should only go to sleep if it is kept into its stand for the full "sleep time" value.

[DavidAlfa]

The timeouts are always active, even in stand mode.
However they won't be refreshed without the shake sensor.

[wickated]

I decided to keep the diode in. Seems to be pretty stable this way without modifying anything at the stand at all. And worst case the voltage level is still too low and the iron does not properly leave standby mode. Not that big of a deal. I can still fix this problem if I really have issues in future.

Maybe I now have found the first small issue. I have a "Standby time" set in the "IRON" menu. I also would like to keep this in case I unplug the holder and want to use the small station "on the go" with a holder without switch. In this case I only want to change one setting (stand to vibration switch). The way it is programmed currently, a configured "Standby time" combined with "stand switch mode" causes the iron to go into standby while working with it if it is not placed regularly to the stand. In my opinion in "Stand mode" the "Standby time" setting should be ignored. Only the "Sleep time" should be still active but the iron should only go to sleep if it is kept into its stand for the full "sleep time" value.

fire safety reasons. just set to 10 mins.

[M-Reimer]

The timeouts are always active, even in stand mode.
However they won't be refreshed without the shake sensor.

And as the pin is shared between either shake sensor or stand switch there can't be a shake sensor once the stand switch is connected.

If there is no way to add a new GPIO just for the stand switch (I guess it would be possible to repurpose one of the programming pins for example) and you don't want to disable the timeouts, then maybe just allow to configure which kind of "sensor" is connected. And if this is set to "stand" then 0 minutes for "standby time" means "go to standby immediately if iron placed to stand" and "sleep time" on 5 minutes means "go to sleep if the iron was in stand for 5 minutes".

Just thinking. Stuff that only works properly with shake switch is pretty useless to keep running once the firmware was told that there is no shake switch connected.

[DavidAlfa]

What are you talking about? You can adjust all that 
0 standby time will skip that mode and check sleep timeout instead.
You can adjust the desired the mode when the handle is put in the stand.
The timeouts are there and will always be, otherwise the station would run forever if the handle is not correctly placed or the stand sensor fails, adjust them as required.

[M-Reimer]

The problem for my usecase is that my plan was to use the station "on the go" sometimes. That's why I designed it the way that unplugging my stand switch reconnects the vibration sensor so the station works in the "original configuration". I have the station without power supply (as I don't trust the Chinese power supplys and prefer to place the bulky power supply below my workbench). So on the go it would be possible to run the station from a power bank.

It would be not a big issue to change one setting in the menu (even though I would prefer if the station would detect if the plug is pulled and automatically switches mode. There is actually a switch contact left that could be used for that. But actually that would need another GPIO to connect this to  ). But the way it is currently designed I would have to change two settings. One setting in "SYSTEM" to reenable the shake sensor and one setting in "IRON" to reenable the standby time. And the same in reverse as soon as I place the station back to my workbench.

Maybe I'm the only one who uses the stand switch feature at all, so the target group is too small, but I would really love to have some simple way to switch from "workshop mode" (stand switch, no standby time) to "on the go mode" (shake sensor, standby time set). Maybe configuration profiles that can be easily switched or something like this.

[wickated]

u can cut wire to shake sensor and place switch on station case and add jack for stand switch/ i still see absolutely no problem anywhere.  im using stand and feel nothing wrong with standby timer.

[SomeUsername]

thanks for the response, sorry for missing it, I just found this, which seems to have the stm32 chip, and seems to be cheaper than the ones you linked, is it good? or should I get a different one? it's a quecoo from the quicko store

https://www.aliexpress.com/item/4000045522658.html?

[DavidAlfa]

Why are you disabling the standby time in stand mode?
In stand, the timeouts are secondary, basically used as a protection method, the running state will be set instantly when you insert or remove the handle from the stand.

When you unplug the stand, you said the wiring is connected back to the shake sensor.
The timeout will be refreshed with the handle movement, so what's the problem?

From what you're saying, I think you only need to switch between stand and shake modes.

In any case, I can't make a custom firmwares for everyone, I would be making mods for all people on the forum...
The problem isn't just making the mod, I would also need to make them on each new update!

Adding a new "mode" which reads an external gpio... I'll need checking the schematics of all controllers, finding available gpios that are easy to reach, modifying the profiles and testing, explaining 1000 times what they're doing wrong in the mod...
Huh, not currently on my plans.

[mastershake]

i did this for someone and as david said just added a switch was the easiest way to fix it and it works great. he runs that station off a battery board i made for it. works great