KSGER T12 keeps resetting

[ronnybkk]

Hello, I just received a KSGER T12 and I'm having some issues. I looked through the threads but nothing came up that is quite similar.
When I turn the unit on, it stays on, I get a solid 25V DC, but once I insert the tip and the temperature ramps up, the blue LED on the PSU dims, turns off and the unit resets.
I noticed the bigger the tip, the faster the reset so it seems correlated to the current drawn.
I also noticed a short coil whine when it resets.

I first thought the handle had some shorts but upon investigation everything seems normal.
I'm suspecting the PSU is defective and I'm being sent a new one by KSGER, but it's gonna take months to arrive so if anyone could help me troubleshooting the unit it would be greatly appreciated.

I guess I could start by using a laptop power supply and see if it's indeed a PSU issue?

The PCB is black and the version "Soldering Power v2.05". I can provide pics if necessary.

[sam_sam_sam]

Please post a picture of the switching power supply  board because there are some issues with one of the version of this power supply

[ronnybkk]

Thanks for your prompt reply, here are a few pics and a video of the issue:
https://photos.app.goo.gl/nZ2a7H1tb9xhBBaa8

[DavidAlfa]

Don't wait, claim to the seller so it sends you a new psu.

[ronnybkk]

Don't wait, claim to the seller so it sends you a new psu.

Yes he already sent one but shipping from AliExpress takes up to 4 months here so if I could fix it....

[DavidAlfa]

Does the power supply voltage drop? Ensure that's not the controller itself.
Whines and other noises can be normal as the load changes.
Try loading the power supply with few power resistors, with the controller disconnected.
If it handles 70w without problem, start searching the problem in the controller.
A badly soldered capacitor, or a stm32 pin...

[floobydust]

I drew a schematic of the PSU Soldering Power V2.04 here and have to see more pics to see what they changed in your V2.05

A weak PSU could be due to many things.

There is a risk the controller board's power mosfet will get damaged if the PSU is sagging, so I would use a dummy load for troubleshooting. It's a bit dangerous to work with a SMPS so make sure you have the safety aspect covered. I would measure primary DC voltage and see if it is sagging or not due to say a bridge rectifier has open diode, PTC is bad.  Inspect the transformer core see if it got cracked in shipping.
The JST connector is not good for the 3A, it's not usually used, they direct wire to the controller board.

[ronnybkk]

I checked the PSU's PCB but didn't notice any crack or suspicious component.
Thanks for your v2.04 schematics, I didn't spot any difference with my v2.05 but I added a pic of the back of the PCB in the gallery if you wanna have a look.

I put a 1A load on the PSU, it worked for 20 seconds then it went dead. Whatever was faulty and triggered the protection from the controller is now toast.
Troubleshooting this tiny PSU live is a bit too risky indeed so I just checked all the passive components offline and noticed D5, a RS1M was open circuit.
Is it the result of my manipulation or the cause of my issue? I guess I'll know when I receive the new part.

[floobydust]

In the pictures your transformer has a massive blob of solder at the bobbin, at the end pin by R3.
This is not good because anything that could cause shorts is a safety issue. Maybe take a closeup or if it looks wonky I would pull the transformer out and inspect it. There are some really crap quality, unsafe SMPS transformers coming out of china. KSGER brand was never that bad, usually better quality but even they get copied by their countrymen. edit: it looks like glare instead?
It doesn't make sense D5 failed unless it got broken somehow or was defective, or there is a short at the transformer. D5 powers the SMPS IC, if it was overloaded R2 might burn up too.

V2.05 fixed the HV PCB trace under the heatsink, upsized the primary cap from 68uF to 82uF, and changed the earth grounding. I'm not sure if the 24VDC(-) is floating or connected to it. V2.04 had the two connected together so the iron was grounded and you just needed a wire from the metal case to the protective earth-ground pad for safety. Can you check if 24VDC(-) is connected to PE GND? and I will draw a new schematic for it.

*Caution there are no bleeder resistors for the big primary cap, so always ensure it is discharged (check with a voltmeter) and safe to poke around before going in.

[ronnybkk]

edit: it looks like glare instead?

I believe so, nothing stands out to me.

It doesn't make sense D5 failed unless it got broken somehow or was defective, or there is a short at the transformer. D5 powers the SMPS IC, if it was overloaded R2 might burn up too.

R2 in circuit measures around 5ohm so it looks ok. C7 R3 D4 all measure 0ohm as they're connected to both sides of the transformer so I guess they would need to be tested off circuit to make sure they're ok. Only D5 gives an infinite value, hence my assumption this might be the culprit.

V2.05 fixed the HV PCB trace under the heatsink, upsized the primary cap from 68uF to 82uF, and changed the earth grounding. I'm not sure if the 24VDC(-) is floating or connected to it. V2.04 had the two connected together so the iron was grounded and you just needed a wire from the metal case to the protective earth-ground pad for safety. Can you check if 24VDC(-) is connected to PE GND? and I will draw a new schematic for it.

PE and 24VDC(-) are not connected, PE seems to be floating indeed. The track goes all the way to a solder pad next to the negative terminal of the JST connector. I guess it's just a matter of bridging both but for what reason they haven't done it?

*Caution there are no bleeder resistors for the big primary cap, so always ensure it is discharged (check with a voltmeter) and safe to poke around before going in.

Yes, thanks. This is the first thing I check when working on audio amplifiers. The primary cap usually has a residual charge that I drain with a 10W resistor.

[floobydust]

PE and 24VDC(-) are not connected, PE seems to be floating indeed. The track goes all the way to a solder pad next to the negative terminal of the JST connector. I guess it's just a matter of bridging both but for what reason they haven't done it?

Are you sure about the PE not connected to 24VDC(-)?  In the previous version power supply V2.04 it was, which thus grounded the soldering iron.
V2.05 I see the trace to the JST but it goes to something else maybe the rectifier heatsink pin and I thought another trace.

R2 in circuit measures around 5ohm so it looks ok. C7 R3 D4 all measure 0ohm as they're connected to both sides of the transformer so I guess they would need to be tested off circuit to make sure they're ok. Only D5 gives an infinite value, hence my assumption this might be the culprit.

Across C7 or D4 there should not be a dead short, as they are arranged. Snubber cap C7 is unfortunately a 100V part so it works too hard.  I put in a 400V part.
Maybe try replace D5, it's the rectifier for the aux winding and low stress. Unless it acted as a fuse for some weird other problem.
Worst case scenario the mosfet shorted and the IC as well, but that always makes smoke. The fuses are huge and do not blow until most of the parts have exploded.
I would use a series light-bulb on mains to limit current when repairing it.

[ronnybkk]

Are you sure about the PE not connected to 24VDC(-)?  In the previous version power supply V2.04 it was, which thus grounded the soldering iron.
V2.05 I see the trace to the JST but it goes to something else maybe the rectifier heatsink pin and I thought another trace.

Yup, I checked several time. I'll try to take more pics when I'm back to my place as you can't see from that angle.

Across C7 or D4 there should not be a dead short, as they are arranged. Snubber cap C7 is unfortunately a 100V part so it works too hard.  I put in a 400V part.
Maybe try replace D5, it's the rectifier for the aux winding and low stress. Unless it acted as a fuse for some weird other problem.
Worst case scenario the mosfet shorted and the IC as well, but that always makes smoke. The fuses are huge and do not blow until most of the parts have exploded.
I would use a series light-bulb on mains to limit current when repairing it.

I'll wait to get the new D5 and will tests C7 and D4 off circuit to see if they're fine.
Thanks for your guidance, the replacement PSU might suffer from the same issues so it's all usable information.
I understand fuses are currently 5A, what would you recommend?