Dangerous Solder Station KSGER T12

[DangerZone]

Hello EEVBLOG Community

Got something Interesting:
Today I had unpacked my KSGER T12 Solder Station and went right into Soldering.
My ESD matt and strap was also brand new.
As I was Soldering my strap felt extemly uncomfortable because it bit me very hard.
First I thought that the Material was not good, but while I was soldering LEDs, they began to produce light.
Then I measured the Voltage from the Solder Iron tip to my ESD matt and measured 73V AC and directly to Ground 80V AC 
So yeah the weird bitting was because it always shocked me a little bit.
All my effort to protect my Circuit from ESD was for the Cats 

Then I sent those Results to my Friend (which has a Brand ersa solder station at his company) and he said he also noticed that when he was Soldering that the LEDs sometimes light up!
He measured 15V AC to the matt and 40V AC directly to Ground.

Without further investigation (maybe I will do it when I have time), how is this possible (how could the Solder Iron have 80V AC on the Tip)?
Doesn't that mean, that my Solder Iron isn't galvanically Isolated?
Does anybody also have noticed such a problem (because also one of the Brand ersa seemed to have no galvanic isolation)?
Should it not be a standart to Ground the Soldering Tip, or are there applications when you don't wanna have this?

[AVGresponding]

It could be galvanically isolated, but AC-coupled via suppression capacitors. If the earth connection from the unit to the wall outlet is faulty, then the path to earth for this leakage current becomes you and your ESD strap and mat...

Looking at the internal construction of this unit, it looks like the mains connections could easily become open, due to mechanical stress from inserting and removing the power lead.

[BrokenYugo]

It is indeed standard to use a 3 wire cord and tie the tip directly to the earth wire, both to shunt any mains leakage current (what you're feeling, hopefully) and prevent any static buildup.

Open the unit up and have a look around, this cheap Chinese stuff should really be treated as a kit and inspected before power up, sloppy work with the grounding is typical, along with the occasional more hazardous mistakes in design or assembly.

[DangerZone]

yes indeed, I opened it up, and it hasn't even the aluminium case grounded, also main powerlines running underneath the heat sink with just the solder paste as isolation.
I also broken the thing trying to fix all issues, now the heat regulation isn't working anymore and it gets glowing hot.
What a waste of time and money.
Never buy something again from china 

[floobydust]

KSGER did change their power supply's PC board grounding.

Soldering Power V2.04 (green pcb) has mains PE ground from IEC connecting to the 24VDC secondary side GND, so the controller board and heating-element is grounded. The enclosure is not.
Safety mod is add a ground wire to the case, that also covers ESD to the knob/encoder. The heatsink over the HV PCB trace, add some insulating tape or lift the heatsink.

Soldering Power V2.05 (black pcb) has mains PE ground from IEC connecting to a large pad and open jumper/pin3 DC output connector, to the 24VDC secondary side GND, so the controller board and heating-element is NOT grounded and floats up to HV due to the Y-cap    The enclosure is not grounded either. They fixed the HV trace spacing bug in older V2.04
Safety mod is to connect jumper, and add a ground wire to the case, it also covers ESD to the knob/encoder.

If you want your soldering iron isolated/floating and trust the "double insulation" of the power transformer etc. instead of the hard grounding, use a 1MEG resistor from PE to DC GND. Just measure continuity, and then ACV when it's running to confirm the tip is not floating up to something that can damage semiconductors or people.

edit: updated the KSGER PSU schematics and posting here:
https://www.eevblog.com/forum/reviews/t12-stm32-v2-1s-soldering-station-controller-schematic-etc/msg2467236/#msg2467236

[gmc]

Very interesting post. I bought one of these as well and am having exactly the same issue as you described. I opened it up and also have the black PCB v2.05

To fix this is it just a case of adding a 1M between PE and DC GND?

Update: Guess, not. Tried that and it made no difference. Still 30-40VAC between the tip and ground.  I don't feel safe using this and worse I don't want it to damage any semiconductors I'm busy soldering.

Lesson learnt - don't but cheap crap from China. Looks like I'm on the hunt now for a more expensive decent brand.

[Slartibartfast]

It is indeed standard to use a 3 wire cord

No, it's not. I have a professional grade ERSA soldering station and a Weller soldering station on standby. Both have the euro plug installed factorywise (I'm in Germany), which does not have three wires. That is good because it does not make sense to use three wires.

https://en.wikipedia.org/wiki/Europlug

and tie the tip directly to the earth wire, both to shunt any mains leakage current (what you're feeling, hopefully) and prevent any static buildup.

Not a good idea. If you do it that way, significant spurious voltage drops can still occur.

The only proper way is to shunt those leakage currents through your ESD installation. To this end, and this is really standard, every soldering station features a ground socket, usually the common 4mm type socket (a.k.a. banana socket), to accept a connection directly going to the ESD safety gear.

Cheers  Peter

[floobydust]

Very interesting post. I bought one of these as well and am having exactly the same issue as you described. I opened it up and also have the black PCB v2.05

To fix this is it just a case of adding a 1M between PE and DC GND?

Update: Guess, not. Tried that and it made no difference. Still 30-40VAC between the tip and ground.  I don't feel safe using this and worse I don't want it to damage any semiconductors I'm busy soldering.

Lesson learnt - don't but cheap crap from China. Looks like I'm on the hunt now for a more expensive decent brand.

KSGER V2.05 power supply, if adding a 1MEG between PE and 24VDC GND (iron tip) only reduced the tip voltage to ~30-40VAC, then I would say the transformer is poorly wound or the Y-cap has a problem, leakage current is high.

and tie the tip directly to the earth wire, both to shunt any mains leakage current (what you're feeling, hopefully) and prevent any static buildup.

Not a good idea. If you do it that way, significant spurious voltage drops can still occur.

The only proper way is to shunt those leakage currents through your ESD installation. To this end, and this is really standard, every soldering station features a ground socket, usually the common 4mm type socket (a.k.a. banana socket), to accept a connection directly going to the ESD safety gear.

Cheers  Peter

You're comparing a German-made soldering station with a chinese one?    I looked at an Ersa RDS 80 soldering station, mains cord is "C - Euro plug CEE 7/16" and is ungrounded.
"Since type C sockets are unearthed, they have become illegal almost everywhere..."
Front has a banana jack for "potential equalization through 220k resistor" to connect to workbench and your wrist-strap for ESD protection.
It looks like no PE connection at all- meeting safety standards it would need double-insulation and no SMPS Y-capacitor for leakage current to mains.

Expecting a chinese power supply to have a decent (insulation) power transformer, wound properly for low capacitance (leakage current) and to have any safety approvals whatsoever with a hi-pot test - well, no.
These products are cheap for a reason, walk into the casino and see how your luck is.

[Slartibartfast]

"Since type C sockets are unearthed, they have become illegal almost everywhere..."

World would have a lot less BS if people would stop the habit of ignoring context.

Type C sockets are not illegal anywhere. They are illegal in many countries for wall mounting. That is a huge difference. There are still plenty around legally, attached to extension cords, adapters (Example), power stripes, etc.

Cheers  Peter

[makken]

I did some measuring on my KSGER V2.05 as well, and I also had 85 VAC between tip and ground.
And around 4-5 VAC when the power switch was set to off.

For a beginner in the field, me, this seems very off. Is it? Is the "problem" solved by connecting DC- from the tip to PE on the main board?

[gmc]

Unfortunately that didn't work for me. I've just dumped it and purchased a hakko.

[floobydust]

For V2.05 power supply, I would add a ground wire that connects PE ground from the IEC connector to the 24VDC(-), and it should connect to the metal case (I use the encoder mounting tab to solder a wire there) as well.

[makken]

For V2.05 power supply, I would add a ground wire that connects PE ground from the IEC connector to the 24VDC(-), and it should connect to the metal case (I use the encoder mounting tab to solder a wire there) as well.

Thanks for the clarification! I did just that, and now i have continuity from the tip to the ground pin. Before I made that solder i measured current between these two points. The measurement read 0.1mA, so I guess the leakage current was just that.

[floobydust]

The leakage current is small but the voltage high, you don't want to damage semiconductors when soldering them. If your work surface (i.e. ESD mat) is not at the same potential as the soldering iron tip, then the parts get a zap.
I'm not sure what china's problem is grounding things, they seem to prefer floating soldering stations as if PE ground in the country is unsafe.

[McClane]

Hi everyone!
I just received my KSGER T12 with Power Board v2.05 and already grounded the enclosure.
I’m now wondering if I should also connect the tip to ground, and if so, how.
I measure around 15V AC between the tip and ground, so much lower than what others measured. With a wire between 24V DC GND and PE this drops to around 0V. With a 1MOhm resistor instead of the wire, it only drops to around 5V AC.
What would be your advice - leave as is, wire or 1 MOhm resistor?
Thanks!

[floobydust]

Place your bets at the casino.
Some people prefer ungrounded tips (not directly connected to PE) to avoid sparks when soldering equipment that has power present. Very rare to have that happen I think. With 1MEG you got 5VAC and another post got 30-40VAC- which means the quality is joy luck.
The majority of soldering stations have PE grounded tip for best safety and no stray voltage damaging parts. This is what I would do.

If the power transformer is wound properly with correct amount of UL-approved insulating tape, and spacings between primary-secondary windings and at the bobbin are correct, as well as the Y-cap having safety-agency ratings, and the PC board spacings properly done, then the risk of electrocution is low.
But chinese transformer winding quality can be deadly. They never seem to hi-pot test them, unskilled labour winding them incorrectly. KSGER power supply quality varies widely, especially the transformer construction. I have not taken one apart but they look not the greatest. The PSU's are made by someone else.

Youtube Diode Gone Wild disassembles phone charger transformers and some are very unsafe designs, the primary and secondary wire crossing over each other, then it's just a thin layer of varnish between you and mains - which is deadly 

[McClane]

Thanks!
I think I’ll go for wire without resistor then.

I’m usually quite cautious about these things and don’t buy cheap Chinese stuff, but when I read about this soldering station I could not resist (and knew up front that I had to at least ground the enclosure to make it safer)
Otherwise, I think I’ll be fine using it. I’m in Germany and we have RCDs protecting the whole house, so even if there is a fault it should not be too dangerous.

[floobydust]

Post #6 Slartibartfast  mentioned ERSA and Weller having 2-prong power cord, in Germany. I assume the leakage current to the tip is very low, they use linear power supply not SMPS.
The KSGER danger is electric shock if the insulation fails, or damaging semiconductors with the stray voltage causing tip potential. If you have a bench ESD mat or wrist-strap grounded to PE but the tip is at 30VAC- this is trouble.

What's weird is power supply V2.04 the tip was PE grounded, and V2.05 is not.

[McClane]

The quality of ERSA and Weller is probably much better…

But I assume by connecting the enclosure and DC GND to PE it should be okayish
If the insulation fails and mains voltage gets connected to exposed parts, the RCD should pop.

[Kadano]

For V2.05 power supply, I would add a ground wire that connects PE ground from the IEC connector to the 24VDC(-), and it should connect to the metal case (I use the encoder mounting tab to solder a wire there) as well.

To add on to this, I have some information I gathered today by testing my recently acquired KSGER T12 soldering station, V2.05 power supply board (black color).

A major challenge was that there's really nothing you can assume to be correct on these Chinese boards. My mistake was assuming that if there's a black wire and a red wire going to the DC board, black is negative and red is positive.

Turned out that the crimp connector was more of an afterthought and the wiring is completely inconsistent. While usually you'd wire connector cables so that it doesn't matter which end you plug in where, with this board revision (or at least my unit), they wired the connectors in mirrored fashion, so that if you plug them in the other way (power board end where DC board end should go), the wire colors are wrong (red for negative and black for positive).



I painted marks of red and blue color for which wire is supposed to be in the respective spot.

Earlier today, when I assumed the spot where the black wire was originally (red spot marked in the picture above), I soldered Protective Earth to that solder point and thus to DC+. I was puzzled as to why my voltmeter showed -24V on the hot soldering iron tip, but with the confusing wiring it now makes sense.



Of course, soldering protective ground to DC+ is not a good idea at all, so I figured I'd post about this to hopefully prevent other people from encountering the same issue.


Since the question of direct ground vs. 1M resistor was talked about here, I decided to install a toggle switch (shown on the left in the photo above) to switch between infinite, zero and 1M resistance and check the noise levels at the tip with my oscilloscope for each.

With the tip floating, as it does by default, there is some severe noise when measured with Protective Earth as a reference:

Both pink and yellow here are connected to the hot soldering iron tip, yellow is set to much higher sensitivity. Noise is so bad by default that the high sensitivity yellow line exceeds the display area by a lot. Even on the pink line, we still have extreme noise with up to 180V.

Toggling on the 1M resistor helped decrease noise levels by a bit:

However, we still have a maximum of 122V, personally I don't feel very safe soldering on any components this way.

With direct ground connection, things get much, much better:

With the default vertical scaling, noise level was now too low to give accurate values for the 50V max display. Since I didn't want to mess around with my oscilloscope measure settings from what I usually use them for too much, I adjusted the pink signal line to 1V to get precise noise level values:

Now, noise level is at around 1.2-1.5V, a range that I feel okay with.

For reference, I measured my Ersa Analog 60 station as well, where the tip is connected to ground by default:


I'm not an expert with electronics at all, so I'm not sure how much the measured ~170V AC at the tip matters, since it's measured respective to Protective Earth that it's floating against. I'd appreciate an expert's estimation for sure. Perhaps the voltage looks worse on the scope than it really is effectively.

What I did notice is that, when measuring the voltage with a multimeter set to DC 200V, again with the COM wire going to internal Protective Earth, it only fluctuated between about -2 and +4V. Not sure whether that just means that the voltage was fluctuating too quickly for the multimeter to pick up accurately. In that case, probably I should have set it to AC instead, something I forgot to do.
After having spent 8 hours today researching about and working on this multimeter, I don't really want to put any more time into it right now though.

Related to that, my next soldering station purchase, whenever that will be, will definitely be a JBC or similar. When I take into account the time spent on fixing all the minor issues of these Chinese KSGER / Quicko stations and their handles, 300€ for a JBC doesn't seem like much any longer.

[Slartibartfast]

Post #6 Slartibartfast  mentioned ERSA and Weller having 2-prong power cord, in Germany. I assume the leakage current to the tip is very low, they use linear power supply not SMPS.

I should know about the Weller as someone gave it to me as scrap. When I refurbished it I saw the inside but I do not remember it. Like the ERSA, it is quite heavy, so I'd assume their transformer is big enough to work at 50Hz (linear power supply). Both stations are a couple decades old, when I look into new offers from, e.g. CONRAD I find most stations nowadays seem to use IEC-60320 C14 (i.e. with PE connected). Maybe indeed that is related to them using SMPS.

[floobydust]

With a double-insulated (linear) power transformer there is very little leakage current and capacitance from pri-sec. They do weight a lot though, heavy.
Kadano's scope traces show the situation with new KSGER V2.05 SMPS is terrible, leakage much worse than just the 2,200pF Y-cap. I would expect IC's to get damaged, never mind the transformer winding quality is hit or miss.

[sofakng]

I posted in another thread (about the CFW for KSGER) but I'll ask the question again in here for completeness...

What should be modified on the V2.05 (black) PCB to increase safety ?

I also checked continuity between the solder tip and earth ground (3-prong USA plug) and it's not connected.

[nikitasius]

Ksger T12 on 2.04: heating 30mV AC, standby 13mV AC.
From tip to ground (tip is grounded).

[nikitasius]

Here are the photos