CFW for KSGER/Quicko STM32 Soldering Stations

[StephenR0]

I installed the latest firmware and started revisiting calibration.  I did get some thermocouples that I have more faith in.  Here's the calibration report.  I've changed the format slightly to reflect all of the numbers used in the calibration.

Code: [Select]

Model:  Quicko STM32/CKS T12
Tip:    T12-BC2

Calibration > Adjust >
                        ADC Value   Temperature
  Cal. Step 250°C       1050        256°C
  Cal. Step 350°C       1500        353°C
  Cal. Step 450°C       1900        445°C

Calibration > Start
                        Temperature ADC Value
  Cal250                224°C       1160
  Cal350                324°C       1602
  Cal450                419°C       2012

I started to calibrate a D24 tip, but I couldn't get the 450°C adjustment done without it going out on an error.  I'm burning it in some more assuming that's the problem.  Or should I be adjusting something else?

Edit:  Here's a little more detail about what happens.  I'm in the adjust portion of calibration.  I get through the 250°C and 350°C steps.  While on the 450°C step, the temperature is about 430°C.  Before I can adjust the ADC value, the temperature starts climbing.  At about 500°C, it throws an error and quits the adjustment.  I've burned this tip in for at least an hour at 300°C.  Do you have a suggestion about how to go forward?  I can certainly burn it in some more.

[DavidAlfa]

So the BC2 calibrated correctly, but not the D24? Select the BC2 profile, plug the D24 and carefully start rising the setpoint.
I guess weird things will start to happen around 400ºC. Latest fw version? Stock PWM settings?
When that happens I set an extreme delay, maybe 100mS, if the temperature stops jumping, then I decrease the delay in 10mS steps.
If the issue persists, then either the tip is defective or need a crazy amount of burn-in.

I've had some tips that didn't like going over ~420ºC. My last ones (Now RIP) for example.
In gave up and used them normally at 370-390, testing the 450ºC setting from time to time but no way.
Maybe 2 weeks later, I rememberedthis issue, and they were working much better!
It's hard to tell why it happens. It woul be nice if you can hook a scope to the handle and see the difference in the waveforms, because clearly there's something going on.
Do you have older tips that were working correctly, to compare with the new tips?

[StephenR0]

Actually, I don't expect to use this tip at anything close to 450ºC.  I'd be fine with only going to say 400ºC.  I'm sorry, I'm not as electronics knowledgeable as you are.  I have an old analog scope, but I don't really know how to use it.    And all of my tips are from my original purchase of the Quicko.   You've given me some ideas.  And I'm sure you're right that time will affect this.  I'll keep working with my other tips.  I bought five with the Quiko. 

Edit:  Yes, it's the latest firmware version and stock PWM settings.  If you'd like to test some changes, just let me know.  Thanks.

[DavidAlfa]

So the D24 did fine before?
The important thing is to figure out if it's only that tip, or there's something going on.

[StephenR0]

These are all new tips.  I'm only now trying to calibrate them.  I will be on the look out for any other issues.

[RobHon]

Hi David,

Hey, I suspect the CH32 reacted a little different to the ADC bug.
Give it another try!

I gave up of the CH32F. Returned it back to the seller.
I installed the CKS in the board and managed to mass erase it.
Just installed the last firmware. Tomorrow I play with the calibration.

Anyways, thank you for the remembrance.

[RobHon]

Updating

Code: [Select]

Model:  Quicko STM32/CKS T12
Tip:    T12-D24

Calibration > Adjust >
                        ADC Value   Temperature
  Cal. Step 250°C       1400        261°C
  Cal. Step 350°C       1800        349°C
  Cal. Step 450°C       2200        444°C

Calibration > Start
                        Temperature ADC Value
  Cal250                232°C       1495
  Cal350                321°C       1939
  Cal450                414°C       2354
now I'm using a FG-100 clone.

[wickated]

Code: [Select]

Model:  ksger STM32 2.1s/3.3 ad8605 x250
Tip:    jbc c245-944

Calibration > Adjust >
                        ADC Value   Temperature
  Cal. Step 250°C       1950        ~240
  Cal. Step 350°C       2650        ~340
  Cal. Step 450°C       3450        ~450

Calibration > Start
                        Temperature ADC Value
  Cal250                240°C       1960
  Cal350                340°C       2660
  Cal450                450°C       3450

Code: [Select]

Model:  ksger STM32 2.1s/3.3 ad8605 x250
Tip:    china sheet jbc

Calibration > Adjust >
                        ADC Value   Temperature
  Cal. Step 250°C       1400        ~250
  Cal. Step 350°C       2050        ~350
  Cal. Step 450°C       2650        ~450


[DavidAlfa]

Hmmm....   I don't trust your calibration wickated.
Tip calibration results are always a tad higher than Cal adjustment values, because they're temperature compensated.
But yours are almost the same, like if you did something yourself.
Or are you soldering at 0°C ambient?

[StephenR0]

I wondered about that.  I was confused that the measurements that I had just made were now lower during the calibration step.  You have now made that clear.  The problem is that we are always questioning our measurement hardware.  I'm beginning to have some confidence in my current thermocouples, but I still question everything.

Edit:  Maybe this is an item for the FAQ.

[DavidAlfa]

It's already there, sort of.
Cal adjustment is just a coarse calibration.
It didn't exist before, made just to avoid burning tips between different boards with huge measuring differences.
Why question everything? For you is just a number.

What you should take care of is the real temperature and avoiding burning/making holes in your boards 

[wickated]

Hmmm....   I don't trust your calibration wickated.
Tip calibration results are always a tad higher than Cal adjustment values, because they're temperature compensated.
But yours are almost the same, like if you did something yourself.
Or are you soldering at 0°C ambient?

ahh that explains everything. for example i set adj1000 for 250c, start cal, on 250 step i see 230, set 230c reading and after that see ~1000adc in tip menu. i thought that its temp compensation not used while setting adc but didnt mention should it be so or is a bug.
anyway instead of calibrating mulitple tips to tip end temp i prefer to set heater outer shell temp(place where TC is), so i use only 1 setup for all tips and its guaranteed that temp wont exceed onscreen value. tip end temp delta is written on small sheet of paper in ma brains.

[DavidAlfa]

When you set Cal_Adjust manually, it just sets the desired adc value as the setpoint, it isn't not ambient temperature-compensated.
That's why after that you get -20ºC or so. It doesnt matter, doesn't need to be spot-on, only needs to avoid burning your tip during calibration.
When you set ex. 235ºC as result, all the work is done by the calibration process, extrapolating the correct ADC value given the current ambient temp.
That's the last time I explain this.

And that's why I certainly know your calibration results were not done by the automatic wizzard, making it useless.
CAL_250 adjustment=1950, CAL_250 result=1960? No way!

Why you must thing everything is a bug? Talking like if you knew how the firmweare works? That really bugs me!
The calibration gives different values, what's the problem? If it calibrates accurately, that's the only thing you should take care of.
If you don't have enough knowledge, don't pretend me to explain you every single detail of the firmware. I'm a developer, not a teacher! .

[StephenR0]

So, I've been burning in my tips a lot more.  I just set the standby to an hour and let them run at 300ºC and walk away.  After an hour I come back and see how they run at 450ºC.  I've got five tips and in every case burn in seems to settle things down.  But sometimes it takes two or three hours of burn in to do it.  Not a problem, of course.  Here's the D24 calibration report.  I'll post the others as I get them.

Code: [Select]

Model:  Quicko STM32/CKS T12
Tip:    T12-D24

Calibration > Adjust >
                        ADC Value   Temperature
  Cal. Step 250°C       1050        251°C
  Cal. Step 350°C       1500        354°C
  Cal. Step 450°C       2000        445°C

Calibration > Start
                        Temperature ADC Value
  Cal250                220°C       1190
  Cal350                320°C       1645
  Cal450                423°C       2109


[wickated]

Why you must thing everything is a bug? Talking like if you knew how the firmweare works? That really bugs me!
The calibration gives different values, what's the problem? If it calibrates accurately, that's the only thing you should take care of.
If you don't have enough knowledge, don't pretend me to explain you every single detail of the firmware. I'm a developer, not a teacher! .

what u see wrong ? i can set cal_adjust in steps of 50/100 only. i have values written down on a paper. why should i use thermometer every single time i wanna reroll fw ? and u can see very clearly symbol "~" near preadjusted temp.
since i have paper with digits written and fast encoder rotation added to adc setup if i use 1 tip preset after flash i just directly go to tip menu and set 3 values directly. if i wanna use multiple tips i go and set nearest preadj, which are only 10 steps tight. why i should even bother measuring real temp while preadjust? i made if multiple times before and nothing in controller opamp part had changed. resulting error is within 50c, not 150c+. jbc tips are pretty dense, on my favourite 741 tip end is only ~5c lower then core

[DavidAlfa]

It seems you still don't understand that  pre adjust is done only once, when the settings have been erased?
Then you only run the normal calibration.
I usually don't bother much and use the same profile for all, the temp between tips doesn't change more than 5°C.

Of course you can go and restore the values, that's why the tip cal section was made.

[wickated]

that  pre adjust is done only once, when the settings have been erased?
Then you only run the normal calibration.

i literally wrote it

[DavidAlfa]

Well, I didn't understood that:

if i wanna use multiple tips i go and set nearest preadj, which are only 10 steps tight. why i should even bother measuring real temp while preadjust?

If you already know your typical adjust values, ok. If not, of course you should measure when adjusting.
Did you ever thought that you might be confusing others? There's a lot  more people reading the thread, and they don't know how it works first-hand like I do.

Test the attached firmwares. There are 3 tests.
As their names suggest,  the tests are SW only mode, HW only mode, and detection mode.
The tests show a bouncing ball and the FPS (The screen effective update rate)
HW I2C speed is set low for now., so don't expect more speed than SW mode.
In any case, it will show if it's working in HW or SW mode.

Althought it will never reach the crazy speeds of SPI  :

[wickated]

Did you ever thought that you might be confusing others? There's a lot  more people reading the thread, and they don't know how it works first-hand like I do.

i have analog devices 8605 opamp, which is not pretty default on those boards, its output is not default, so numbers i have shouldnt confuse anyone who have same modification.

57fps in sw_detect mode
58fps in sw mode
doesnt start in hw mode with mod from prev page

[DavidAlfa]

Post picture of your mod.
I limited the drawing rate, but not the OLED update rate.
Drawing eats much less cpu power now, so the fps will be limited by the i2c speed.
Also changed few other things, download and try again.


In detect mode, the speed will be the same as in a forced SW/HW builds. It only probes the screen at initialization, and sets a flag.
That flags is read by the graphics driver before sending data to the screen, using sw or hw method.
The purpose of SW and HW builds is just to ensure the detection works correctly.
And it seems to, as forced HW doesn't work either.

Overclocking the spi bus (16->32MHz)

[wickated]

lifted standard pins also

in newer build auto also 58 fps sw now

[DavidAlfa]

Remove these resistors for now.
Why lifted the pins?

[wickated]

Remove these resistors for now.
Why lifted the pins?

what resistors?
for test

[DavidAlfa]

What is those soldered pins that can be seen in the uper part of the image?

[wickated]

What is those soldered pins that can be seen in the uper part of the image?

sda scl display wires