GATO Soldering Controller, for: C245, C210, T12 tips.

[.rpv]

Hi!

So I'm using my soldering station controllers based on stm32duino for years now, I started with the 907-types and later with the T12-type tips and now I'm playing with the JBC C245 type tips:

Test on a ground plane:
https://www.youtube.com/watch?v=4Jc7nBBVmdQ
*I've used this test a bit so there's no flux on the thin.

Upgrade to 1.3" and soldering on a TO-3
https://www.youtube.com/watch?v=wl1LVijpSf4

So it's a work in progress thing, there are a few things I need to solve:

• Giant initial overshoot
• PID values need a touch
• Performance overall can be improved

*I'm not using original tips so performance can be considerable better with original tips.

C245, C210 and T12/T15 tips are pretty similar:

• C245 and T12/T15 use almost the same gain (around 300x) on the amplifiers
• C210 and T12/T15 use the same construction (thermocouple and heater in series)
• C210 needs like twice the gain (500x-600x)

so use this controller on those tips will require just changing a few resistors and the way the handle it's connected to the controller (so not simultaneous with the current hardware).

I've tested this controller up to ~170watts without problems.

The current setup uses:

• A bluepill F103C8
• ST core v1.9
• STLink for uploads.

*Edit current schematic: https://github.com/r-pv/GATO_CONTROLLER ... O_rev0.pdf
I'll modify a few things before making proper PCBs.

What do you think?

[dannyf]

two quick comments, based on my experience building a few of those things:
1. not much difference between on/off type of controls vs. pid - the large thermal mass in my view negates the advantages of a pid loop;
2. overshoots are due to high gain / loop delays. you may want to use two or more algorithms based on how far off you are to the target temperature. aka high gain when you are cold starting and dial it back as you get closer to the target temperature...

hope that helps.

[.rpv]

two quick comments, based on my experience building a few of those things:
1. not much difference between on/off type of controls vs. pid - the large thermal mass in my view negates the advantages of a pid loop;
2. overshoots are due to high gain / loop delays. you may want to use two or more algorithms based on how far off you are to the target temperature. aka high gain when you are cold starting and dial it back as you get closer to the target temperature...

hope that helps.

Hi, sure it helps, thanks!

1. You're probably right, but I like the PID performance overall it just the turn-on overshoot also I don't see a downside to use PID control, maybe some little unnecessary computing time?.
2. You're on the money, I'll try to use 2 set of PID values, one shy on the start and the current values later if still doesn't work I'll try to do it manually and then use the PID with the current values.

[dannyf]

I'll try to use 2 set of PID values,

or a hybrid:

1) if you are too far off the target, full on / off;
2) otherwise, pid of your choosing.