CFW for KSGER/Quicko STM32 Soldering Stations

[cosmin1]

Yes, i believe so.
From pictures it's the same board. But mine was without onboard battery. Must be attached through the white connector.

[DavidAlfa]

I don't see the utility of a RTC on the soldering station.
For what? Having a clock? Replacing the battery every few months?

[cosmin1]

Hahahaha... exactly.
My Quicko statipn doesn't have clock. And i don't miss it.

[DavidAlfa]

In fact, the I see the PTDreamer stock firmware too complicated.
Too much actions and submenus required for simple things.
I'm trying to improve that.
For example, entering the screen Brightness menu shows only a big number, rotate to adjust, and exit at  the first click.
without having to navigate up/down/ or having to go to exit/save. But its my personal liking.

The waiting for the C245 to arrive will be endless...

[DavidAlfa]

I decided to put a bit of humour when the thing crashes... 1KB less but worths it!

[dreamcat4]

Hey guys, Sorry for being absent here. As per usual had some other interruptions after fixing my pcb.

dismembered the dead tips  and what a deception. The rigid wires that go between the heating element and the connector had wayyy high impedance, in most it had ~0.7ohm( I cut 6 and set them in series, it measured ~4ohm).
So from 72W, 60W are at the tip. and 12W at the wires! I guess (I hope!) this is not the case of genuine Hakko tips.

Very interesting!

But can you also tell us which aliexpress store you bought these T12 tips from? Was it KSGER or the QUICKO store? It would be helpful to know the source.

Because I have this so-called "high quality" black tip K from the QUICKO store, and it cost $7. Instead of the normal $3.50, for the regular cheap T12 tips.

Measured it's total impedance from pins 2-3 between heater + and heater -. And the resistive impedance (at room temperature) measured about 9.0 ohms. After subtracting the 0.9 ohm impedance of my multimeter probe leads.

Not sure if that is good or bad. Have asked some people on Discord, will see if there is anybody who will reply. To compare others.

One of them looked better, with thicker wires, but no! It must be a different material, was 1.6ohm per wire! 3ohm in the wires!
I remember one of these bastards heated like hell, so much that the nylon in the connector had slighly melted. After a while of soldering you couldn't hold the handle anymore!
That must have been this tip. No wonder why,  27W in the wires, 45W in the tip...!

Good to know! 

[dreamcat4]

How do the fake JBC tips do?

Not sure, but might be pretty good? For $15 I purchased a C245-029 tip (=JS02  bent tip, smallest) from "KNOKOO Store" on aliexpress, here is nearest current items, they only sell in bulk now:

https://www.aliexpress.com/item/457967413.html

And it measures 3.6 ohm (lowest) across the heater (pins 2-3) on my multimeter. However my cheap multimeter is not very accurate! (subtract about -0.8? ohm in the leads?). So that is under 3 ohms.

I believe that looks like it's a good one, right?

The tip is marked "864858 JBC   |<|" and look of a high quality. It would not be surprising... if they were made in the same factory as the real JBC tips. The KNOKOO seller also seems to be selling many other high quality soldering tools / related products for manufacturing.

I've found this and it's very tempting! JBC handle + 2 tips for 30 bucks!
https://es.aliexpress.com/item/1005001530229488.html

Edit: I bit the bulled and bought the JBC ! Canceled the T12s

Absolutely fantasstic. Great! Thanks for getting the JBC handle! I have to say that is an excellent price, even disregarding the free tips, a bonus!

My handle cost +$10 than that, and it didn't come with any tips. So it's a great value. I think we can take photos of both our tips (when yours arrive). And compare the laser markings and the tool markings. To see if there is any problem. Hopefully they both source from the same factory. Then it's super value indeed. At that price. Great! We shall see.

Maybe there is room for both T12 and JBC tips in the world? Since T12 tips are "usually cheaper" than the T245. However the performance is less than JBC. The quality is also a big question.

Do not see the point of buying genuine Hakko tips. Because the price is the same as genuine JBC tips, who are better? JBC. So rather it's about the relative quality and price of these "not genuine" chinese tips.

[dreamcat4]

When this tip failed, the temperature rised instantly +50ºC when the power was on and the next cycle -50ºC. So it was something wrong in TC.
That tip had been done that randomly since last week. I knew it was the tip and not the software because all the rest worked well.

This is an interesting place to be. Because others (the Russians, in russian forums). Are blaming transient spikes on the power supply, not smoothed. The line noise on 3v3 which then makes it onto the op-amp high gain circuit (before reaching the ADC). Throwing off the temperature suddenly. Or they are blaming the op-amp itself being bad. As a chinese part which is either low quality or clone / fake. With some user reporting the problem goes away after changing the op-amp for different version (better alternative part).

IDK what is the truth, because I have not experienced this problem yet myself). Initially, it has been working OK for me, the temperature sensing. However on this Blue pcb at the moment I have 1.1MHz switching style 3v3 step down regulator ("should be good"). While the op-amp is not known. Unable to identify from markings. So-called "bad" 61LJV op-amp, which the Russian forums user explains here:

https://www.eevblog.com/forum/repair/brand-new-ksger-t12/msg3096285/#msg3096285

But what you are saying that for a while at the beginning everything was fine... then either something has slowly degraded. Or you have been running the tip for longer heating durations without break. And that caused the melting due to the high resistance in the leads.

Which is a different reason for the problem. IDK which is the reason. But it's really helpful to know why. Or perhaps more than 1 reason gives the same symptom....

Always watching to see if it happens to mine!

 

[dreamcat4]

Found out why my mosfet blew up today!

Ironically 

as you were buying a T245 handle, I bought 2nd T12 handle for myself. This part was also recommended by Russians. But it should be better this time! ,hopefully.

https://www.aliexpress.com/item/4000311213864.html

 

This is to replace my aluminium one, which was the problem.

Blown mosfet was in fact caused by dodgy handle shorting out. It's an aluminium handle see... and it's not insulated inside the 3 terminals.

The picture explains better. So what was happening? In brief: When strain or bend on the tip, it shorts the pin 3 heater + to the heater - & ground of the handle metal case.



This is such a poor handle, I do not recommend anybody else buy it. Multiple reasons:

* Pins not insulated to handle body. Could be fixed by wrapping electrical tape inside.
* Tip sticks out too far, so the tip to handle distance is too long
* There is no room inside the handle body, for the wiring to not get all squished and torn up / damaged and broken.

 

Bad. Avoid!

[cosmin1]

From what i remember, mine have some plastic insulator over the connections. Something like in GX12/16 connectors, so no shorting problem.
I really like those aluminium handles.

[dreamcat4]

I have been told that there are 2 different version, of this same aluminium handle. The "original version", and the "new version". The original version was better than the new version. At least in terms of the distance from the tip to the handle.

But they stopped making that now. It is no longer available. So I have got the new version here. This was discussed recently in the Russian forum (RadioKot). Except for my finding about how it was not properly insulated. Perhaps your example is, but not all from China are coming like that. So it could just as easily... be bad like mine. Who knows?

So the big problem is trusting it. But why cannot we just check it ourselves? To make sure? Because:

You cannot safely take apart this aluminium handle. Without damaging / destroying the fragile wiring in the process. Mine broke. Unless you use silicone grease lubricant on the strain relief. Because the wires are very thin, and there is no room inside.

Yet you need to dismantle the handle, in order to see / check if it was properly insulated. Or not safe, as mine was not. But in the process you damage the handle, and cannot safely fit it together, with a proper heatshrink on the wires. There is no room! Then you just get frustrated, and try to cram it all in there, (somehow). At which point the strain relief pops off the other end and will not fit back on anymore. Because it's being pushed out by the wires (no space). So you try to force the strain relief back on, really damaging the thin wires (yet again). That you cannot see / check they are ok anymore. Also they strain relief is damaged and will not fit anymore either.

 

But that was my problem. Not to worry, maybe other people to not have such troubles. Because they use silicone grease on the strain relief. As recommended by Russian forums.

But why would you buy one in the first place, if you already knew this? In fact there was yet another drawback of it, I forgot to mention earlier:

* The metal conducts heat from the tip into your hand.
* This problem only occur (too hot to hold) after a longer duration of continuus usage.
* Problem is a lot worse when the tip itself generating a lot of wasted heat up at the wrong end, near the contacts, because that is within the handle. This happens if the tip is faulty, or is poor quality. Such as DavidAlpha measured earlier. High resistance!

By comparison:

The carbon fibre handle is better because it:

* Has a shorter distance from the handle to the tip. So it has better control like a pencil.
* Does not conduct the heat, but insulates / protects your hand from high temperature. No worries. It will never get hot.
* Does not short itself out randomly by accident.
* About the same price. Perhaps +$3 more.

So this was the recommendation given by RadioKot forum user. I come back to tell my experience after receiving my Carbon Fibre handle. And if any problem with it. Good or bad (either way).

 

[DavidAlfa]

About the tip oscillation, for sure, that was the only tip doing that. Even having extremely measuring delays didn't help.

Omfg chinese quality strikes again. From the total value of the, what's the cost of a plastic isolator...?

So the c245 is ~3-4ohm?  I hope the Power supply handles that load.

Yes, there's no reason for ditching the T12s. The working method is the same, only the connections and the TC voltage are different.

I finally found the gremlin that crashed the mcu when flash usage was >9x%.
After a long afternoon seeing how the settings checksum was different on each booting, totally senseless, it suddenly came into my head!
I didn't took in count the system settings. It directly reads and writes to the last 1KB flash!
So the compiler is unaware, unless you touch the linker script.
Well, reduced the iron tips from 10 to 7 and that way it fits in 512bytes. As the flash page has to be erased whole, a single byte over the limit will need another 256bytes.
I think 7 tips are enought! Or are we T12 tip collectors?
I wanted the biggest font possible for the temperature screen, so to use the 22*36 font without eating the whole flash, I had to wildy strip and patch it to only use ~20 chars.
But that let me add more fonts, so now everything is nicely readable.
I think this is the limit for this fw in 16/64KB devices. Tried to add another menu to change the temperature units between Celsius, Kelvin, Farenheit... Errrrrr no way!
(Yes, I removed the trollface screen and neither, its the RAM). We'll see in the stm32F101/103.
I don't know why, but malloc did work last time I used it to find the biggest chuck of free ram.
Now it allocates everything, it will return a valid pointer allocating a bigger buffer than the RAM size itself!
That's a problem because now I can't measure the RAM usage by the screens. I keep adding stuff until it crashes....

I made a little video to show how It currently looks.
Keep in mind that I have no tips, so I'm faking the temperature, actually it's measuring 80°C with a pull-down resistor.
That's why the pwm doesn't fall when it enters sleep mode.
After flashing the settings are not yet initialized in flash, so it gives the error and resets everything.
That only happens the first time or if the memory gets corrupted.

[DavidAlfa]

However, I already targeted this fw. Looks very nice and I suspect is way more memory-efficient.

[dreamcat4]

So the c245 is ~3-4ohm?  I hope the Power supply handles that load.

Its less than 3 ohm. So for 24v drive... P=V^2/R = 24^2/3. This would be 192 watts or 8 amps full load at 100% PWM. However I have seen a video of somebody getting 10 amps on a C245 cartridge.

This is why the JBC tip is faster than others to bring the tip up to temperature. The high current is only for a few seconds during this time. It should not going to stress the PSU for too long. Unless there is a bug in the firmware etc, or it remains full on 100% PWM by mistake.

Then, once at temperature. What happens then? The current is limited by the rate of heat transfer, for the rate of dumping new energy into a heat load. It is limited quite a lot by the thermodynamics, across the thermal junction. Somebody else said that this upper limit can be as low as 30 watts. However I'm not sure myself. Maybe it can be more than 30w continuus at the worst case. But surely it must also depends on the size of the tip area. Which is quite small surface over which to transfer the heat. And the relative temperature differential between the 2 metal parts.

SO getting back to the start: the maximum "burst" current is up to 10 amps, but only for 3 seconds. Therefore the minimum spec for the PSU should be... enough to provide that amount. For a short time. In order to get the maximum performance. Otherwise the heating up time, that performance will be limited by your PSU.

The cheapest Chinese PSU are not likely to perform up to their given spec. Unless it is of high quality, like Meanwell brand, etc.

So what I did was to buy a cheap 20 amps PSU here from China. To make sure and be "double rated". And not stress the PSU too hard. But also not pay very much. The PSU is big and heavy. If you want to pay more, then you might be able to get something higher quality, that is also more compact. And not need such a high over-rating as 2x.

Perhaps a really high current bench lab PSU. Like the Riden RD6018 for example. Although it has a high ripple current. And only 1 channel. But you can then also use it to power other high current loads too, for other tasks. And it will be a variable voltage, and with OCP over current protection. To help limit any possible damage.

Well, reduced the tips from 10 to 7 and that way it fits in 512bytes.
As I added the biggest font possible for the temperature screen, I had to wildy strip and patch it to only hace around 20 chars.
But that let me add more fonts, so now everything is nicely readable.
....
(Yes, I removed the trollface screen and neither )

To speaking for myself, I would like to see one day, each of these options configurable with C macros. Then having different build targets for each MCU. With enabling a recommended set of options. So that higher end MCUs with more memory can progressively have better font, more tips, etc. as the space allows. And not be penalized to the lowest supported hardware.

That is not urgent or anything. It can only come after supporting multiple build targets. And unifying the code base all together.

Great video BTW. The screens look pretty good. It's very impressive! You should be proud of your work here. It is also of a high quality, like the JBC tips.

 

[Il_Marco]

I checked some original JBC C245 tips of different type/code I have:
The HEATER Resistance ranges from 2,26Ω to 2,64Ω.
Thermocouple resistance ranges from 33mΩ to 38mΩ.

My only (by now) T12 tip:
HEATER Resistance of 8,62Ω.

I received my unit and it's not as expected, just another v2.1S hardware version!!!
The unit is a full one with Power Supply from KSGER, suffering from the well known two "BUGS":
- High Voltage trace under D2 Heatsink
- Missing Earth connection to the Case, Encoder and GX12 Connector.
Some Mica Insulator, a drill, and a little work will solve that without too much troubles.

Here is my v2.1S controller unit. (SW Version2.12)
 My_V2.1S.jpg

Here is its schematics. I put in the schematic my idea on how to use both T12 and JBC handles by using a GX12 PIN as a jumper to select T12 (jumper connected) or JBC (non connected) handle. Do you think it could work or maybe it will simply add noise and more troubles? I will try it when I get some GX12 connectors I ordered from aliexpress. I plan to cut out the GX12 connection board and wire a new connector so that I can freely remove the controller from the panel.
 My_V2.1S.pdf

[DavidAlfa]

I don't see the problem using the jumper. The root of the noise problem is usually in the supply to the ADC and the op-amp.
The TC has very low impedance, so the noise has not so easy way to come it.
However, once it passes through the amp... and it's badly done in these ksgers.
At least  they should have added a small resistor in series with adc and the amp supply.
A 100 Ohm resistor with 10uF low esr ceramic cap after it would be already a big noise killer, the current is very low (1mA for the ADC) and won't cause more than few mV drop, but it's a very effective filtering option for low current devices.
 
The kser doesn't even have a cap on the amp feedback to prevent high frequency noise or oscillation.
And the effect gets worse with high gains.
I would solder a 10nF on top of R5. The bandwidth needed is really low, the temperature will not oscillate more than few Hz.

See this simulations. The input signal is 8mVpp, and the added noise is 2mV. A bit exaggerated but gives the idea.


Being more optimistic, with only 100uV noise, multiply it with 250 gain and you got 25mV noise!
That translates to 12.5ºC noise in Hakko language! And gets much worse with JBC tips.

[dreamcat4]

My only (by now) T12 tip:
HEATER Resistance of 8,62Ω.

Unless you actually tell us the name of the seller you purchased it from.... it does not help us find any better source better T12 tips. Or tell us if the clone tip is different from the hakko. Also including the tip type ot shape will help.

I received my unit and it's not as expected, just another v2.1S hardware version!!!

Looking at your PCB, it looks like the best v2.1 original version. I don't think you worry about the pcb. Thanks for including this. But again... which banggood seller was it?

[DavidAlfa]

And the seller... It was a year ago. I don't think that it'll any difference now.
They probably change their suppliers pretty often. So todays's tips might be totally different from what you buy in a week.

[dreamcat4]

@DavidAlpha please also see this issue, for the JBC T245 handle. Another shorting problem. If this happens it can damage the cable, which is moulded into the plastic handle. Therefore the cable is not so easy to replace.

https://youtu.be/BxvPT4n7p4E?t=592

[Il_Marco]

Here is the link where I bought my unit.
If you look at the reviews you can see that old lots were different.  So you never know what you will get In future (or maybe just now). I ordered it from CN store (and was on sale, about 10€ less)

 [41,45€ 19% OFF]V2.1S T12 Digital Temperature Controller Soldering Station Electric Soldering Iron Tips T12-K + 907 Handle Professional Tools from Tools on banggood
https://banggood.app.link/ELa95pSkObb

My only T12 tip came with the unit and is a K type one marked like this
T12-K 823151 31520

@DavidAlfa, I would add a 10nF cap parallel to R5 as you suggested.

I also ordered a Tip tester that I will get in a week or two, so that I can take some temperature measurements.
[12,42€ 27% OFF]DANIU FG-100 Soldering Iron Tip Thermometer Temperature Detector Tester 0-700℃ Professional Tools from Tools on banggood
https://banggood.app.link/7snIaDrmObb

I must say that, as far as I can see, T12 temperature seems quite stable.  I think that K type tip I have was shipped with some lead free timon it. I checked and it melted at about 220°C set on the unit.

About T245 handle,  the one a have was a wasted one... I disassembled it and the + heater contact was deep in the melted plastic! And that was an original handle. I don't know how that happened,  but happened, once in a long time. By the way, JBC units should be in standby when you change the tip connecting the tip to the base extraction tool.
One last thing. JBC handles doesn't have the shake switch (and NTC), so you should take care of putting it in standby (or maybe sleeping mode) when needed (also when you change tips? )

[DavidAlfa]

It seems strange to me. I worked almost 7 years in a shop using a C245, repairing phones/computers, everything...
I had a round tip and a "knife" tip for serious pcbs. I only replaced a tip in those years.
I didn't need anything else, had a cheap hot air station also. The C245 worked so well...Used almost everyday without issues.
If you maintain the tip clean and don't abuse it (I rarely went over 380ºC) it wil last a lot.
When I replaced it, wasn't because it had stopped working, the tip had worn and looked like one of these...



Maybe is the cheapo brand and its crappy internals. Being for hobby use, I think it'll be ok.

Today it was cloudy, cold as f** and of course, we are still quarantined. Like UK, 300 out of 365 days!
So played a bit making a Quicko pcb. The analog is powered from the TL431 and nicely filtered. Designed for 5mA, should be enough, but can take more if needed.
And using a high side mosfet driver (LTC4440) and a stm32F411.
I could panellize it and get 15 pcbs from jlcpcb for $6, but I don't think I'll do!

[illiac4]

Hi. I am attaching pictures of my two controllers. Both are the same model 2.12s but have different STM32 chips, different op-amp, crystal seems to be different also.
Are those controllers already compatible with your firmware?

DavidAlfa: Can you please share pictures of hardware mods done to the 3.3V rail and op-amp you have replaced. Maybee also add links to aliexpress, where those components can be sourced. Maybe best on your git so it can be all found on one place.

TNX

[illiac4]

2.nd controller

[DavidAlfa]

Huh? I didn't mod anything, worked pretty well as it is. Maybe you meant what dreamcat4 was doing, he's replacing the voltage regulator and the amplifier.
Well, and a blown mosfet haha...
I dodn't have these boards, so I can't try myself. If you have programmer we should get it working soon.
But I need someone for testing.
Most it's done, only the hardware layer needs to changed.

[illiac4]

I do have stlink v2 programmer. Are there any quick manual hot to flash this controller and compile your firmware?
Also i can sacrifice the one with 61jlv U3 op-amp to flash it.

Is there an option to flash it back with original firmware and then activate it with codes, if your firmware will not work? I do not have any spare stm32 chips so will not desolder and solder it, will just overwrite them.
I saw on the Russian google drive some dumped hex files, are those original dumps from KSGER and how do I identify which are suitable for which?

P.S.:
May I ask why is the original firmware so bad and need alternative firmware? From your video I can see that there are some things missing that are present in the original one. Like stepping, startup behavior.