Cheap T12 clone soldering stations

[sn4k3]

@CapnBry
my metal handle arrives today.
Where you insert the tip there are one plastic spacer + one rubber ring + one plastic spacer. When i remove the tip sometimes the rubber ring and the top plastic spacer come out attached to tip. This happen to you?

Note: Now they sell a new version, it come with a rubber grip in green, similar to JBC. But my verison is same as yours

EDIT: The 3d colar works for me after drill with 12mm bit  Printed mine in PETG RED

[CapnBry]

Nope, I've never had anything but the tip/cartridge come out when I remove it. I've never seen the rubber ring and spacer you're talking about (and now I feel lucky!).

Good news about the collar too! The inside of it had small ridges (0.2mm) on it to lock it into the grooves on the iron so drilling it out probably removed those too. If it ends up falling off you might try to rebuild the STL with i_dia increased a tiny bit from 12.0 to 12.1 or 12.2mm. I just mucked with the value until I got a good fit from my printer. Of course, I bet a drop of CA glue or something would have worked just as well.

I got an (unsolicited) email from the KSGER seller informing me they have many different metal handles now and that the one we got is EOL. They have the one with the green foam, one that's all black, and a stainless shiny one, all with different shapes. Hopefully I am not on their mailing list now for every minor thing they do because there's no way to unsubscribe. 

[sn4k3]

Nope, I've never had anything but the tip/cartridge come out when I remove it. I've never seen the rubber ring and spacer you're talking about (and now I feel lucky!).

Good news about the collar too! The inside of it had small ridges (0.2mm) on it to lock it into the grooves on the iron so drilling it out probably removed those too. If it ends up falling off you might try to rebuild the STL with i_dia increased a tiny bit from 12.0 to 12.1 or 12.2mm. I just mucked with the value until I got a good fit from my printer. Of course, I bet a drop of CA glue or something would have worked just as well.

I got an (unsolicited) email from the KSGER seller informing me they have many different metal handles now and that the one we got is EOL. They have the one with the green foam, one that's all black, and a stainless shiny one, all with different shapes. Hopefully I am not on their mailing list now for every minor thing they do because there's no way to unsubscribe. 

After contacting the seller he told me that pieces are glued and maybe i need to glue them. When you insert or remove the tip you dont fell a bit of rubber friction?
Yes with drill the ridges are lost but even drilled it stay on place without even move.
The new handle with the green foarm is the same as our but with that foarm that can be removed, seller told me he can send me one for free on a next order. And he send me the links of the handpieces, they suck/ugly xD

[GregDunn]

I've read all the T12-type soldering station threads thoroughly, and there seems to be only a vague consensus about which actual brand/model is a worthwhile purchase.

What I'm looking for mainly is a secondary station to make use of my T12/T15 tips: one I can stick in a box, take to a friend's house and make repairs without worrying about packing up my Hakko FX-951 and accessories.  If it gets dropped or stolen, no big loss.  I'd like a built-in power supply w/IEC socket and fuse, and the ability to plug a FM-2028 compatible handle in.  If I have to fix the internal grounding and stick X- and Y-caps in it, that's fine.  It just needs to be cheap and not complete junk.  Bonus if it has the microcontroller / OLED display as well.  It won't be taking the place of my FX-951 so accuracy is not even a big concern.

It seems like there are several KSGER and Quicko kits which fit my needs, but it's unclear which ones of them (for example) have all the necessary connections to the handle in order for the T-12 tips to operate correctly.  Re-reading the threads didn't provide a clear recommendation in this regard.

Is this one, for example, any good?

https://www.amazon.com/KSGER-Soldering-Electric-Temperature-Controller/dp/B07DKZ6PDT/

Or this one, which looks like a metal case?

https://www.amazon.com/KSGER-Soldering-Electric-Temperature-Controller/dp/B07DSCH4CJ/

[exe]

which ones of them (for example) have all the necessary connections to the handle in order for the T-12 tips to operate correctly.

I don't understand what you mean. I just bought a station, handpiece and tips from ksger and it "just works". Never heard of incorrect operation (except when using a tip for the very first time temperature jumps like crazy for 2-5mins, but it settles). I don't know if ksger stations work with original tips, handpieces etc.

[KL27x]

and the ability to plug a FM-2028 compatible handle in.

I don't know of any clones which are compatible with a genuine 2028 handpiece.

There are like 8 pins on the real 2028 handpiece. All the clones I have seen only have either a 3 or 5 pin plug. If you want something compatible with your real 2028, you will have to do some modification. Probably better off modding a 3 pin station. The 5 pin might have a different component in the handle from the 2028. (I am guessing that it's for a thermistor for cold junction compensation? Or maybe for a sensor for sleep?  ).   

You might as well just buy a fake 2028 handpiece for your clone station and keep it in your to-go kit. But you can buy 8 pin DIN connectors that are the same style as used on the 951, if you are determined. I haven't found them in any component supply store. They are not a common connector. But they sell them out of China on Ebay. Probably Ali/Banggood, too.

It should take only 5 pins for the thing to work completely properly..... power, ground, earth, and 2 (or 1 and ground) for the cold junction sensor. But I don't know which clones/handpieces actually utilize one. Someone else may know. I haven't done any testing for drift, but in 2 months of using a 3 pin T12 clone for some pretty heavy lead soldering, I didn't notice any issue with lack of this sensor. The cartridges seem to be so long and thermally insulated and efficient enough that this isn't a big deal at moderate temps and output.

In case it's not clear, the cold junction sensor is supposed to measure the temp at the back of the cartridge. This is where the thermocouple ends, and it is the "cold junction." If the iron gets hot enough, the temp at the end of the cartridge can rise. And then the thermocouple will send back a falsely low reading. Theoretically, the station will pump more power, the cold junction gets hotter, and this could result in a positive feedback that could make the iron go sorta nuclear. The sensor compensates for this. IIRC, i have seen people design clone station with the cold junction sensor on the station side. This is not going to do any good; it will just make the iron temp fluctuate based on how hot the station gets.

Come to think of it, every T18 clone I have ever seen comes with a reverse plug from the genuine Hakko. Maybe Hakko looks the other way as long as the plug is different?

[GregDunn]

and the ability to plug a FM-2028 compatible handle in.

It should take only 5 pins for the thing to work completely properly..... power, ground, earth, and 2 (or 1 and ground) for the cold junction sensor. But I don't know which clones/handpieces actually utilize one. Someone else may know. I haven't done any testing for drift, but in 2 months of using a 3 pin T12 clone for some pretty heavy lead soldering, I didn't notice any issue with lack of this sensor. The cartridges seem to be so long and thermally insulated and efficient enough that this isn't a big deal at moderate temps and output.

[/quote]

This appears to be the concern.  By "compatible" I mean that a Hakko tip plugs into the clone handpiece and works correctly, just like it would on a Hakko station; station pin compatibility is not essential.  Obviously if it supplies power only, it would behave more or less normally except that you won't be able to measure the actual tip temperature.  Assuming the station is a closed feedback loop, it has to have a tip temperature in order to know when it's reached the setting and provide some level of accuracy - not to mention the temperature display.  Incidentally, the tips only have 3 contacts, so as I understand it, the temperature measurement has to be made in between applications of power to the heater.  Any other pins brought back to the station have to be for added functions, such as a sleep switch or the cold junction.

But the cold junction is only supposed to provide a temperature reference; that is, it doesn't matter where the cold junction is, as long it's constructed properly and you know its temperature.  Thermocouples act as a differential measurement device, so as long as the reference temp is known (or compensated for), you can calculate the tip temp. 

I worked for a while designing and installing process control systems; when we used industrial thermocouples to measure heat treating furnaces or other devices, the cold junction was frequently mounted at the recording device where the operator resided, far away from the hot junction; it was calibrated against a reference standard (the company made one just for this sort of use) and the offset dialed in before it was used to measure the system.  All that mattered was that the offset was set up right initially, not the cold junction actual temperature.  The measurement potentiometer was designed so that any subsequent temperature drift at the cold junction was nulled out.  I would argue that putting the reference junction in the handle or in the station doesn't matter IF it's correctly compensated. And that's what we don't seem to know - does Hakko have a special thermistor with a known temperature coefficient at the junction in the handle?  And do the clones also use this?  They might just use a known resistor composition and set a "typical" value in the controller.  It would probably be OK for a known style of tip.

It would be useful info to know which of the clones have 3, 4, 5 wires connected at the handle end.  Lacking that, practical experience with drift/accuracy is worth hearing, because that's really the concern.

[stj]

3pin like the Bakon have a ceramic cap across the element,
5pin like quiko / ksger have a tilt-switch.

the way i would do it is make an adapter with a 3/5pin plug and an inline hakko-compatable socket with a short bit of flex between them.
that way you wont trash any warranty's or render the base no-longer compatable with the original handpiece.

[exe]

Assuming the station is a closed feedback loop, it has to have a tip temperature in order to know when it's reached the setting and provide some level of accuracy - not to mention the temperature display.

It does have a closed-loop control, it displays temperature and output power and has different profiles for different tips (but you have to dial-in the tip type, it doesn't have automatic tip type recognition, nor I think tips support it).

Honestly, I feel like we discuss how to buy a Ferrari for the price of a bicycle. If hakko-like performance is needed then buy... hakko or jbc . But these stations are about the best China can supply. I don't really see alternatives (other than trying to find something secondhand). So, to buy or not to buy is sort of silly to me because I don't see many alternatives (but there are some, like bakon, metcal clones, ts100, etc).

[GregDunn]

Assuming the station is a closed feedback loop, it has to have a tip temperature in order to know when it's reached the setting and provide some level of accuracy - not to mention the temperature display.

It does have a closed-loop control, it displays temperature and output power and has different profiles for different tips (but you have to dial-in the tip type, it doesn't have automatic tip type recognition, nor I think tips support it).

Honestly, I feel like we discuss how to buy a Ferrari for the price of a bicycle. If hakko-like performance is needed then buy... hakko or jbc . But these stations are about the best China can supply. I don't really see alternatives (other than trying to find something secondhand). So, to buy or not to buy is sort of silly to me because I don't see many alternatives (but there are some, like bakon, metcal clones, ts100, etc).

Yes, obviously it's closed-loop; I was just being specific to the method of control being discussed.

Not looking for a Ferrari at a bicycle price - just a T12 iron which has predictable behavior and won't break the bank if it gets broken or lost.  It looks like all but the cheapest clones meet this expectation, and more.  I won't be using it much on the bench; that's where my Hakko or Ungar units get used.    It seems to be safe to take a chance on one of the KSGER units.

[KL27x]

They are all compatible with T12 tips. They are all temp-controlled/thermostatic. They all pulse the power and do thermocouple readings between pulses. Most of them use a microcontroller to do this. The dirt cheap ones use a dual opamp to make a comparator and one-shot circuit that reads the thermocouple and (re)fires when it's below the setpoint.

If you goggle, you can probably find a pinout of the real 2028 somewhere. Google is failing me right now, but I have seen it. IIRC, 2 pins are used for a cold junction thermistor (which doesn't seem to be necessary; by all accounts, clone stations have no problem holding repeatable temp; it might serve as safety feature more than for accuracy in the real 951?), and i think 2 were used for a bicolor LED (which was used in the original 2027; omitted in the 2028). But my memory isn't very reliable.

If you want to see for yourself, you have a 2028. The little yellow ring is a separate piece. It unscrews from the blue parts on either side.   

There are no clones that use a dumb variable power. Even the stations that cost less than 1 genuine tip are using at least the "616" op amp circuit to hold to an adjustable set temp. And they work, well. It only takes a dollar in parts.

I am pretty sure the 951 uses a 24V trafo and TRIAC. You could theorize that the DC/FET clones can cause more electrolytic migration of the heater over time. But the life of the plating is more than likely the limiting factor in tip life for 99.9% of soldering scenarios.

The clones with integral PSU, I've seen DC 19.5-24.5V. At higher than 24V, the heater life is supposed to noticeably degrade. I have found my Bakon @ 19.5V, warms up in roughly 17 seconds from cold, depending on the tip. @ 24.5V my Suhan 616 clone takes roughly 10 seconds, which is pretty similar to genuine 951 in this specific aspect. The genuine is doing it (I think) with the voltage drop of a TRIAC, so I would bet the 616 circuit is not optimal, but it works pretty good.

BTW, per what you wanted in a travel station, the Suhan 616 station has an IEC socket, is dirty cheap (I got it for $26 or $27), and it's probably the smallest thing going. I don't see any physical fuse in there, but the 24.5V PSU is obviously not cheap chinese barebones switcher; I rather think it's a PSU originally designed by Apple. It's super complex and painstakingly compacted into a rectangle. The case is form fitted to just barely fit the PSU with the socket and front panel sqeezed on the ends. Essentially a PSU dunked in black plasti-dip. (Warning: the cord it ships with has live and neutral swapped). But it is analog and the handpiece it comes with is a turd.

The Bakon is only 19.5V, but it has an AVR running things and a clunky 2 button interface and 3 digit display. The power cord is integral. The PSU is a very simple switcher. I don't know much about switcher design, but it's all air in there. It has 1/10th the components in the Suhan PSU. The handpiece it ships with is very good; it's better in some ways than the 2028 styles, IMO. It has a thinner footprint, but it's longer and taller and probably twice the volume.

I didn't do any testing of temp drift over time/temp. I did warm up speed testing, as already stated. And I did one test for thermal droop/gradient (how much higher the set temp must be to get 200C at the tip surface on a specific heatsink). In a sample of one tip (3mm bevel) tested at one temperature, both clones worked roughly as well as my 888. I haven't really used the Suhan much in real world soldering, though. The Bakon is heavily tested and approved.

The reason I gravitated to the Bakon is the thinner and longer form factor makes it the perfect footprint to integrate a DIY soldering iron holder right on top of it. And I rather like the thinner, more flexible, integrated power cord compared to a thick IEC cord on a tiny device. I modified the Bakon so the cord exits through the "front end" next to the iron connector, it being an auxiliary/second iron for me rather than a permanent bench fixture.

[GregDunn]

They are all compatible with T12 tips. They are all temp-controlled/thermostatic. They all pulse the power and do thermocouple readings between pulses. Most of them use a microcontroller to do this. The dirt cheap ones use a dual opamp to make a comparator and one-shot circuit that reads the thermocouple and (re)fires when it's below the setpoint.

If you goggle, you can probably find a pinout of the real 2028 somewhere. Google is failing me right now, but I have seen it. IIRC, 2 pins are used for a cold junction thermistor (which doesn't seem to be necessary; by all accounts, clone stations have no problem holding repeatable temp; it might serve as safety feature more than for accuracy in the real 951?), and i think 2 were used for a bicolor LED (which was used in the original 2027; omitted in the 2028). But my memory isn't very reliable.

So far, my Google-fu has failed on finding a 2028 pinout, but I'll keep looking.

If you want to see for yourself, you have a 2028. The little yellow ring is a separate piece. It unscrews from the blue parts on either side.   

There are no clones that use a dumb variable power. Even the stations that cost less than 1 genuine tip are using at least the "616" op amp circuit to hold to an adjustable set temp. And they work, well. It only takes a dollar in parts.

Sometimes, though, the builder cheaps out on that extra dollar - like dropping X/Y caps, hard mounting the IEC connector (or not using one), poor plastic in the handpiece that breaks easily.  That's one reason I wanted to see if anyone had a definite yes or no on any of the cheap clones.  Just so I was aware of what I was buying; again, I don't expect perfection, I just want to be prepared to make any necessary changes in the equipment.  It does sound like the ones I was considering are all fine in the important design aspects.  I'll probably take my 2028 apart when the new unit arrives, for comparison - and hopefully to share with the community.

I am pretty sure the 951 uses a 24V trafo and TRIAC. You could theorize that the DC/FET clones can cause more electrolytic migration of the heater over time. But the life of the plating is more than likely the limiting factor in tip life for 99.9% of soldering scenarios.

The clones with integral PSU, I've seen DC 19.5-24.5V. At higher than 24V, the heater life is supposed to noticeably degrade. I have found my Bakon @ 19.5V, warms up in roughly 17 seconds from cold, depending on the tip. @ 24.5V my Suhan 616 clone takes roughly 10 seconds, which is pretty similar to genuine 951 in this specific aspect. The genuine is doing it (I think) with the voltage drop of a TRIAC, so I would bet the 616 circuit is not optimal, but it works pretty good.

BTW, per what you wanted in a travel station, the Suhan 616 station has an IEC socket, is dirty cheap (I got it for $26 or $27), and it's probably the smallest thing going. I don't see any physical fuse in there, but the 24.5V PSU is obviously not cheap chinese barebones switcher; I rather think it's a PSU originally designed by Apple. It's super complex and painstakingly compacted into a rectangle. The case is form fitted to just barely fit the PSU with the socket and front panel sqeezed on the ends. Essentially a PSU dunked in black plasti-dip. (Warning: the cord it ships with has live and neutral swapped). But it is analog and the handpiece it comes with is a turd.

Not coming up with any hits on that either; it keeps redirecting me to KSGER and Quicko units..

The Bakon is only 19.5V, but it has an AVR running things and a clunky 2 button interface and 3 digit display. The power cord is integral. The PSU is a very simple switcher. I don't know much about switcher design, but it's all air in there. It has 1/10th the components in the Suhan PSU. The handpiece it ships with is very good; it's better in some ways than the 2028 styles, IMO. It has a thinner footprint, but it's longer and taller and probably twice the volume.

I didn't do any testing of temp drift over time/temp. I did warm up speed testing, as already stated. And I did one test for thermal droop/gradient (how much higher the set temp must be to get 200C at the tip surface on a specific heatsink). In a sample of one tip (3mm bevel) tested at one temperature, both clones worked roughly as well as my 888. I haven't really used the Suhan much in real world soldering, though. The Bakon is heavily tested and approved.

The reason I gravitated to the Bakon is the thinner and longer form factor makes it the perfect footprint to integrate a DIY soldering iron holder right on top of it. And I rather like the thinner, more flexible, integrated power cord compared to a thick IEC cord on a tiny device. I modified the Bakon so the cord exits through the "front end" next to the iron connector, it being an auxiliary/second iron for me rather than a permanent bench fixture.

This is all good info, and appreciated.  If I feel the need to make any ergonomic changes, I'll definitely refer back to this.

[KL27x]

https://www.ebay.com/itm/110-220V-Soldering-Iron-Station-Temperature-Control-Handle-T12-Tip-for-Hakko-/222098597348?hash=item33b61ba5e4

Had to look back through my history. I actually paid under 25.00 delivered, lol, in 2017. I have no idea what search term to find it, but hitting similar items in my history made it pop up. This is a different seller, and it is still under 25.00. I also found one on auction that starts at 13.50 with free shipping.

If you look at the pic, you will see "Suhan" on the front plate on the left. It is the same thing I have, down to the 936 franken-handle and the assuredly backwards wired IEC cord. I think mine has "616" printed on it, somewhere. But I can't remember where I put it.

"616" appears to be a model number for the circuitry/board to some extent. You can buy a tiny pcb that is labelled "616" for like $3.00 to make a T12 station with your own power supply. The circuit in these little Suhan station is a larger pcb with some extra parts in it and works at the 24.5V of the really neat PSU. The little 616 boards only work to 20V.  The important bits on either boards are a dual 358 and a FET.

It has a single R/G LED that turns red when the heater is on and green when it switches off and a nice smooth pot. As stated, the PSU looks impressive (from someone who doesn't know what they're looking at and didn't do any testing!). You could barely fit an extra quarter in the case. The KSGR PSU looks similar to what's going on inside the Bakon. I can't find it, but it's probably a lot smaller than the KSGR... in probably every dimension. There was no way I could remove the IEC connector and even run a permanent power cord through the front panel.

*Edit: It was on my bench the whole time, lol.
2.97" wide
4.09" long (4.53" with the knob)
1.48" tall, not including the rubber feet.
And it says "616-A1" on it.

I wonder if this digital station by Suhan uses the same PSU and case.
https://www.ebay.com/itm/2017-Quick-Heating-T12-STC-Digital-Soldering-station-Electronic-welding-Iron/142430838684?hash=item21298a1f9c%3Ag%3AW18AAOSwjvJZVftD&_sacat=0&_nkw=t12+station&_from=R40&rt=nc&_trksid=p2047675.m570.l1313.TR1.TRC0.A0.H0.Xt12+station.TRS0.TSS0

Heh, yeah. in one picture it says "Apple 24.5V 2.65A." I probably saw this, before, planting the "apple" idea. It sure looks the part. I may never use this as a soldering station, but it may come in handy one day for the nicely enclosed PSU.

[GregDunn]

OK!  First, thanks to all for their input.  I found a KSGER iron which looked to have all the necessary features and ordered it.  It arrived and I proceeded to run it through its paces.  It's a STM32-based unit with internal switcher, IEC power connector, etc.

I checked the continuity from earth ground to tip - grounded.  Also measured tip and case/connector voltages when powered up - only a few tenths of a VAC to any of the grounds on my bench.  That's probably inductive pickup, actually; I have a lot of AC stuff on the bench.

It's a little disconcerting to have a piece of equipment like this exhibit a power-up delay where nothing appears to happen for several seconds.     I was able to adjust the cold junction temp and watched the tip temp gradually settle in as the element warmed up.  The manual says there will be a period of inconsistency for the first few minutes of powering up a new tip, and so there is.  After about 5 minutes it was rock solid.  The UI is a little fiddly but I was able to set everything up just fine eventually.

Except.  There's nothing in the manual or online (that I've been able to find) explaining how to set C or F on the display.  What's the trick?

[exe]

There's nothing in the manual or online (that I've been able to find) explaining how to set C or F on the display.  What's the trick?

(insert here a typical joke about US and SI)

Did you check setup menu? If it's not here, then, probably, it doesn't support this (I don't have my unit at hand to check).

[GregDunn]

   As an engineering student, I got used to thinking in C and K all the time so the units don't bother me, per se.  What bugs me is that I can't get all of my temperature measuring devices to use the same units.  It's annoying to have a calibrator and one iron permanently set to C, another iron and thermocouple which only read F, and only one iron and an IR thermometer which can be switched.  I waste too much time converting units, and it shouldn't be necessary.

Edit: the setup menu only seems to allow switching between English and Chinese for the menu text - which is amusing because the manual is bilingual in English and Russian...

[nanofrog]

  As an engineering student, I got used to thinking in C and K all the time so the units don't bother me, per se.  What bugs me is that I can't get all of my temperature measuring devices to use the same units.  It's annoying to have a calibrator and one iron permanently set to C, another iron and thermocouple which only read F, and only one iron and an IR thermometer which can be switched.  I waste too much time converting units, and it shouldn't be necessary.

Edit: the setup menu only seems to allow switching between English and Chinese for the menu text - which is amusing because the manual is bilingual in English and Russian...

I seem to recall that what you want to do, and more, is possible from at least a couple of YT vids. So I'd suggest checking around and see if there's better firmware available. 

Good luck. 

[GregDunn]

I'm sure better firmware would be handy as well!  The problem is that I don't know how to verify that it would run on my version of the board (HW2.0, FW2.1) - I'll see if I can scrounge up some details on that.

Oddly, I had trouble getting the case open on my unit - the back comes off, as well as the nut holding the rotary encoder, but the case seems to be either glued to the front panel or there's a hidden fastener.  I've found all the externally accessible screws (even looked under the rubber feet).

Edit: unplugging the unit seems to make it forget all the settings too.  That doesn't seem right.

[stj]

that's interesting, some of the kits/pcb modules on ali clearly show a coincell in heatshrink on flyleads.

[exe]

that's interesting, some of the kits/pcb modules on ali clearly show a coincell in heatshrink on flyleads.

That's for RTC (real time clock). Unfortunately, no alarm function and it needs to be powered to display time (I'm kidding).

[GregDunn]

that's interesting, some of the kits/pcb modules on ali clearly show a coincell in heatshrink on flyleads.

That's for RTC (real time clock). Unfortunately, no alarm function and it needs to be powered to display time (I'm kidding).

I also find it somewhat amusing that they went to the effort to allow set and display time/date through the UI but not C/F.     Still haven't found an option for that.  Not a real issue, just something to remember if I have to switch irons for whatever reason.

I left the unit unplugged overnight and when turning it back on today it did remember the tip type and cold junction settings.  Wonder if there's something not making good contact inside...

[exe]

that's interesting, some of the kits/pcb modules on ali clearly show a coincell in heatshrink on flyleads.

they went to the effort to allow set and display time/date through the UI but not C/F.

They are in China . Honestly, adding C/F option wouldn't come to my mind as well unless asked...

BTW, if I'm not mistaken, firmware for 2.1S was written by a Russian guy. I think you can find him in the Internet and ask to add this feature. Oh, quick googling showed this thread about another alternative firmware: https://www.eevblog.com/forum/reviews/stm32-oled-digital-soldering-station-for-t12-handle/ . Please let us know if an alternative firmware any better than the stock one. But, honestly, I like the station the way it is. It's a bit hard to remember how to control all the features from the single encode (long press, short press, press and rotate...), but I think it's a good hobby-grade station (may even on par with some more branded devices).

I left the unit unplugged overnight and when turning it back on today it did remember the tip type and cold junction settings.  Wonder if there's something not making good contact inside...

Mine just works . But I don't know if it stores settings immediately, or with timeout (to reduce write wear). When I change settings and want to switch off the station I wait a few seconds (so not to interrupt power while it's writing to flash). May be I'm too paranoid.

[GregDunn]

BTW, if I'm not mistaken, firmware for 2.1S was written by a Russian guy. I think you can find him in the Internet and ask to add this feature. Oh, quick googling showed this thread about another alternative firmware: https://www.eevblog.com/forum/reviews/stm32-oled-digital-soldering-station-for-t12-handle/ . Please let us know if an alternative firmware any better than the stock one. But, honestly, I like the station the way it is. It's a bit hard to remember how to control all the features from the single encode (long press, short press, press and rotate...), but I think it's a good hobby-grade station (may even on par with some more branded devices).

I left the unit unplugged overnight and when turning it back on today it did remember the tip type and cold junction settings.  Wonder if there's something not making good contact inside...

Mine just works . But I don't know if it stores settings immediately, or with timeout (to reduce write wear). When I change settings and want to switch off the station I wait a few seconds (so not to interrupt power while it's writing to flash). May be I'm too paranoid.

I really don't have a problem with the single encoder control (telling it to change tips with the press-rotate is a bit weird, but OK).  I've read that alternative firmware thread and don't even want to think about it till I'm sure my STM32 revision is compatible, because bricking the unit seems inevitable.  Since this is not my primary iron, I'll probably just let it slide for now.  Especially since I still can't figure out how to get the case apart despite much peeking and poking.  The fact that it's so different from other cheap T12 controllers in the minor details doesn't fill me with confidence that it will work unmodified with the CFW.

I'll keep an eye on the presets.  It may well be that I did something hurriedly and trashed the NVRAM entries during one of my tests.

Oh, one very nice thing I noticed while trying to get the case apart.  The screws holding the rear panel on are driven into brass inserts rather than into the bare plastic; one of several things which pointed to some care in mechanical design.

[GregDunn]

Aaaaand... dumb.  There is a fascia over the front panel hiding the extra screws.  I should have known there would be one, but it was so carefully applied that I didn't see it even while looking for it.   

So here's the controller board.  Note STM32F102C8T6, whereas the CFW is for a F103 chip and the author says the code would have to be rewritten to support that chip.  Apologies, but I really have a lot on my plate at the moment, so I'm not going to get into writing new firmware yet.     There doesn't appear to be any obvious jumper to change the C/F display so I'll just deal with it on the occasions where I need to use the iron.

[stj]

aha! i see the 2pin battery connector.