KSGER T12 Clock Backup Battery Problem Fixed

[@Rchelicopterfun]

Since recording this back in 2019, this issue has long since been resolved and was only a problem when these 2.1S T12 stations first came out. April 2023 Update: My Ksger T12 2.1S Soldering station is still working like a champ and I've used it alot. Such great little direct drive technology stations for such a low cost.

[@matrix9987]

Soldering battery directly is not a solution.

[@Rchelicopterfun]

@Carl Jito It's now being reported (late August) that the latest batch of Ksger T12 stations are coming without the R10 resistor mounted on the board so it would seem there is no reason to have it there, and the manufacturer has addressed and corrected the issue.

[@stevensweeting7191]

@@Rchelicopterfun receive mine today, no R10 resistor...battery still dead, and the buzzer works once or twice then it makes this very faint high pitch noise??

[@MM-hx3ss]

some chips have a battery backup feature where if VCC is lower than Vbatt, it switches to Vbatt power. A pull down resistor to ground is used so as to prevent emi noise generating a voltage greater than VCC on the floating Vbatt pin and making the chip switch to Vbatt, but there actually isnt a battery there, so the chip turns off leading to data loss. But the battery in this design isnt used to keep the chip powered when VCC is removed. So yeah, the resistor is pointless.

[@squee222]

My battery cable wasn't even connected, and the tips kept resetting with the clock (maybe different software). One of the tips they sent me was shorting the thing out too.... so quality control is obviously an issue for these things, but the price can't be beat. Thanks for doing these videos, you saved me some headache.

[@Rchelicopterfun]

What brand (Ksger or other) and what version of firmware (V2.1, V2.01, V2.1S, V3) if you don't mind me asking?

[@squee222]

​@@Rchelicopterfun KSGER. Hardware 2.00 Software 2.09 is what it says It's a hardware problem with the tips. They agreed to send me a new tip for free but only if I bought something else to cover the shipping so I bought a few more tips and am still waiting for them in the mail. Very happy with the product after I took the time to fix it.... haha - common with these cheap chinese electronics.

[@retro_boy_advance]

Thanks John. I had this battery problem and removing R10 seems to have solved it. UNFORTUNATELY, due to my stupid clumsy hands, I also accidentally removed the Y1 component next to R10. Thankfully, my T12 still seems to be working fine... but I'm worried for how long? Does anyone know what adverse effect removing Y1 and leaving it that way would have on my T12 in the long run?

[@DmitryKiktenko]

Might be xtal for microcontroller? Well, it seems controller works somehow on internal clock. You might experience pid drift over heating of controller. Or maybe you'll have problems with time counting. I think nothing more up to surprise you.

[@DmitryKiktenko]

Thank you for sharing your experience of the problem. In my case there's probally a battery drains to vcc, cause screen is zoomed, but still keeps going even with power plug off. But after time the battery drains out and screen isn't show anything until battery is changed. I desoldered encoder and found no problem, but left a stick of tape under it just for a case. Overall desoldering connector to get board free is pretty inconvenient, and oleds are much burned out by constant showing time, so i do think to buy new board, and then try to bugfix an old one. I think i have to find where it drains.

[@DriverWay8848]

I have this exact soldering station from six months. It is the best one I've try. The errorr it was showing when u change the new tip, i think is about the different resistance on the first one ho is not the same from the new one. When is hot enough, u just need to shutdown the device and leave it for a couple of seconds to cool down. Then u start it again and the resistance of the two tips will have about the same value. After that everything should be ok.

[@cmajorh9769]

Good catch👍🏾

[@greengohm]

I think I have really early model, because the PCB layout is different and, lo and behold, there is not path from the pin on the IC to the battery, even though I have the battery (and it's full after those couple of years). Still, very interesting video! Thank you for sharing this, John!

[@Rchelicopterfun]

Yep, either earlier or later model as they fixed this problem soon afterward (only was an issue for a few months). I of course attract issues like this; me and Murphy's Laws are good companions, so it's only fitting I got one out of the bad run of boards. 🙂

[@knortn]

I have two of those soldering stations. Both had drained the battery within a few weeks. However on mine it's not the switch that shorts the resistor to ground. The trace is directly connected. I don't think the resistor is supposed to be connecting the battery to ground, it doesn't make sense.

[@Rchelicopterfun]

That grounded resistor can be removed by all accounts talking to others. I really don't know why it's even in there? As I said to another commenter, the schematics I have found for STM32 Vbatt connectivity usually only show a capacitor between that trace and ground to smooth out possible ripple voltage. There may be a resistor between the cap and the Vbatt but not direct between the Vbatt and ground. Who knows, my shorted switch housing may have taken out something that is grounding that resistor and that's why mine worked fine afterward?

[@knortn]

Yep, i removed the resistor on both of my T12 stations and they're perfectly fine. One of them also had a broken EEPROM which had me reconfigure it each time i turned it on. The EEPROM had some rows that couldn't be written to. Replaced it with a new AT24C08 and it asked me for a code. I found the code in the contents of the old EEPROM by comparing it with the one from the working station. It might have also worked by simply copying the contents of the broken EEPROM onto the new one but by that time I had already figured the code out from all the swapping between both stations...

[@pr0xZen]

I think the designer went by the first example/diagram in the STM AN4718 application design notes (page 4) on STM32 Vbat design. And either _originally intended_ to use a supercap instead of battery, or (by inexperience?) didn't realize that the resistor should *only* be there if using a supercap. When going just battery, the res has to go too.

[@fr3k513]

What about the other settings (for example tip calibration settings), are they also reset to defaults without battery?

[@Rchelicopterfun]

Nope, only RTC. Regardless, moot point since this issue was addressed by Ksger over a year ago now.

[@fr3k513]

@@Rchelicopterfun Oh, so they store main settings in the flash memory. Nice to know. Thanks.

[@Rchelicopterfun]

Exactly 🙂

[@MinhPhamGators]

I'm thinking about getting one of these. Would the unit function without the backup battery? It is only for keeping RTC or something else like user settings?

[@Rchelicopterfun]

Works fine without the battery as I show in the unboxing/review video. The 2032 cell is only used as the clock/date backup; all user settings and tip calibration PIDs are saved in the non volatile memory.

[@MinhPhamGators]

@@Rchelicopterfun That is strange to have a soldering station with time and date. Why couldn't they have made it wifi-able? Then we all could hook it up to Home Assistant and watch the temperature rise and fall in there, and the time keeping would be free through NTP. 😁

[@pr0xZen]

@@MinhPhamGators Just another utterly crap secured IoT device for hackers to fiddle with, burning down your house - or even screwing with your standby settings and messing up your tip calibrations! Those evil bastards! _You will never touch my calibrations! _*_NEVER I SAY!_*

[@MinhPhamGators]

@@pr0xZen Don't blame you. So many bad actors out there. I had my ssh port open and watched auth.log. I saw a bunch of these guys trying to brute-force their way into my machine. I shut that port down in nanoseconds. With the scammers, at least you can detect and play them, with these hackers, they are in the shadow, and are skilled - don't want to mess with them..

[@discedoce1827]

hi, did you reinsert the 10k resistor after sorting out the shorted switch? tqvm.

[@Rchelicopterfun]

Yes, but others have not had the same luck. I'm almost thinking that switch short took out whatever is grounding the resistor trace on my particular board. Others are reporting leaving out the resistor and not having any issues. As I have mentioned several times now, any schematic I can find of the STM32 vbatt connectivity only shows a capacitor between vbatt and ground to smooth out any possible ripple voltage/current. A few show a low value resistor, but it's placed between the cap and ground, not direct to ground.

[@briancatsup5661]

I cannot decide which version should i buy, either 2.01 or 2.1s? what's the difference?

[@Rchelicopterfun]

Sorry, don't know other than the 2.1 is more recent.

[@AlanDike]

Did you do anything to remask the trace where the switch was shorting to ground?

[@Rchelicopterfun]

I stuck a square of Kapton tape on the back of the switch. Shouldn't be needed once the metal tabs were bent back into position, I just figured it was a good precaution.

[@AlanDike]

@@Rchelicopterfun I agree it shouldn't be needed... But they mask off ground plane to prevent incidental contact.. so I completely agree with your decision to repair the masking (I'd probably have used something like a paint on coating.. bit high heat tape or just about anything would work.. I applaud your thoroughness

[@NordicDan]

Got my T12 from Banggood a few days ago. Looks like a list of things to check and a couple small modifications but it seems the consensus is once the issues are fixed it lands you with a pretty good soldering station for $50 plus some easy work. I'll add this to my list of things to check lol

[@Rchelicopterfun]

The clock backup battery grounding issue was fixed on the T12 last year shortly after I made this review. You should see the resistor removed on yours.

[@alexstone691]

Mine has battery with 3V but it forgets everything every time I reset

[@SevenDeMagnus]

Thanks. Mine seems to the be the newly designed one, even though the firmware is not 3.1S. I truly love the KSGER T12 and KSGER brand (Quicko and Quecoo are rebranded KSGER. Sometimes I wish I got the Quicko T12 Mini with the aluminum and black iron combo and bought the power supply and power plug separately, for just US$39 on our local online shop but then the MINI might be harder to stack:-)

[@kodoyama]

I thought the battery was saving other user settings, not just time and date. Or is that not the case? If it's only for time and date I'll probably just remove mine.

[@Rchelicopterfun]

All user defined settings are stored in the STM32's NVM (non-volatile memory). Only the RTC within the STM32 requires voltage at VBAT to keep the RTC register counting.

[@kodoyama]

That makes sense, thanks for clarifying.

[@JohnCena-iw2vk]

my KSGER keeps resetting its tip selection.

[@andre137]

I own the KSGER T12 since today. My battery is empty too. So i checked the resistor and it is grounded. But my switch is fine. How can i fix this? Just remove the resistor? Is it safe?

[@Rchelicopterfun]

I don't know nearly enough about the circuit to advise what will/might happen when running it without that specific 10K resistor. It seems to work fine however without it in place. Looking at several STM32 backup battery schematics would also suggest it's not needed. Most show a capacitor in series from VBAT to ground to absorb ripple. Another person also commented having two units with the exact same problem. This IMO, is more than coincidence. Hopefully as more people see this video and comment, we might have a better scope of the problem and if it's only only this specific 2.1S version or others as well. It's obviously not only due to the rotary switch shorting to ground.

[@stanimir4197]

@@Rchelicopterfun 10k resistor to ground feels dumb, a capacitor makes sense. I guess it's a design flaw or the very least the resistor should be 1M5+. Edit: is C1 parallel to R10, perhaps they wanted them in series (which also makes sense).

[@momfiethehunter504]

Interesting. I have the same issue with mine but I have a silver unit but maybe the board in mine is from the same batch as yours

[@momfiethehunter504]

Never mind that. I have a different board but maybe it can still have a similar issue

[@antonjansenvanrensburg4145]

i have 2 with the same problem, both got a short to ground on that resistor but not the same problem on the pot :( guess i will leave it on forever...

[@seyul]

it came today from banggood and same problem.:(

[@SevenDeMagnus]

This might apply to other gadgets the Dreamcast's VMU which eat up batteries:-)

[@michaelo2l]

Just posted on another T12 video related to battery drain... I mentioned that on my version of the board (purchased Sept. 2019), that R10 had been removed (this was true for at least two other posters)... It's entirely possibility the switch was the culprit but as no one had spotted it, they simply removed R10... Will test the switch for a short and correct if necessary and mention your findings on the other video...... in the mean time, nice catch ;)

[@zack12aaaa]

mine remove as well..thanks

[@alexstone691]

Mine had died in 6 days, suddenly It was v3.1s mini idk about software

[@DarthMaul41]

Good design, but bad execution on these chinese things....

[@Rchelicopterfun]

As others have commented, the R10 resistor grounding issue was fixed on subsequent product runs shortly after I put this video up.