Sharp KC-D40EUB

[fisz]

I am devastated. How did that happened? Well, I touched LED1 anode with a soldering iron. While the power supply was plugged in.
RCD tripped and there was a spark.
What I tested then:
Power supply still supplies good voltages (+5V and +15V). There is no short on the power supply and the circuit breakers are ok. So I disconnected it.
There is only 300R resistance between +5V and GND on the board -- seems very low to me. But there was only one transistor between the point I touched and +5V.
I removed it and it tested, it seems fine. I 've also removed LED1. Still no luck. And the board doesn't work -- only one LED is on, and it doesn't react to buttons.
Please tell me that this is fixable...

And a bonus question: how did that happen? I know the soldering tip is mains earth referenced, but the power supply is isolated (only L and N, no earth connection). What closed the circuit?

[wraper]

And a bonus question: how did that happen? I know the soldering tip is mains earth referenced, but the power suppy is isolated (only L and N, no earth connection). What closed the circuit?

You have Y safety capacitor between mains voltage and GND in PSU. You discharged this cap into soldering iron tip through the circuit.

[Manul]

There is main chip there. Put multimeter into diode test mode, go and test every pin for any abnormalities (in general, or by comparing to other pins of similar function). You can go first negative to ground, then positive to ground (reverse). Some pins will be grounded, or go to power supply, you can trace them, they do not matter, focus on pins which are inputs and outputs. It is possible that you will find something. Usually pins without pullups or pulldowns will show 0.5 - 0.8v forward, and a bit lower backwards.

There is some other IC's. Test them similary.

Test similar looking transistors with same technique.

There is a lot of LEDs, so also test their anode and cathode points. With some skill and luck faults may often be found this way.

[wraper]

On second look, none of the circuit seems to be insulated from mains voltage. At least when PSU board is viewed from top. Which means you passed mains voltage through low voltage circuit into the iron tip. That's why RCD tripped. Frankly I don't think this is repairable, likely you killed MCU.

[wraper]

I know the soldering tip is mains earth referenced, but the power supply is isolated (only L and N, no earth connection). What closed the circuit?

This assumption is quite strange. How lack of earth wire as such suddenly makes PSU isolated? Isolation has nothing to do with that. You actually mislead me when I gave my first answer based on your claim it being isolated.
EDIT: I actually fear you may kill yourself by acting stupid. Only presence of RCD saved you from more dire consequences.

[Manul]

On second look, none of the circuit seems to be insulated from mains voltage. At least when PSU board is viewed from top. Which means you passed mains voltage through low voltage circuit into the iron tip. That's why RCD tripped. Frankly I don't think this is repairable, likely you killed MCU.

Yeah, but I highly doubt that it is actually true. Such a design should not exist because of safety standards. This would be crazy.

[wraper]

On second look, none of the circuit seems to be insulated from mains voltage. At least when PSU board is viewed from top. Which means you passed mains voltage through low voltage circuit into the iron tip. That's why RCD tripped. Frankly I don't think this is repairable, likely you killed MCU.

Yeah, but I highly doubt that it is actually true. Such a design should not exist because of safety standards. This would be crazy.

Look on PSU PCB, then say where isolation is supposed to be. There is nothing crazy about it. Everything is isolated from the outside.

[Manul]

Look on PSU PCB, then say where isolation is supposed to be. There is nothing crazy about it. Everything is isolated from the outside.

No, I agree with you. I'm just saying that it is hard for me to believe (and it is crazy by my standards). Isolated flybacks are so jelly bean, that I never actually saw a non isolated one (well, but I'm not so much into consumer electronics either).

[wraper]

Look on PSU PCB, then say where isolation is supposed to be. There is nothing crazy about it. Everything is isolated from the outside.

No, I agree with you. I'm just saying that it is hard for me to believe (and it is crazy by my standards). Isolated flybacks are so jelly bean, that I never actually saw a non isolated one (well, but I'm not so much into consumer electronics either).

There are reasons to not isolate. Like you need to somehow power a fan motor. If you isolate, then you also need to have isolated motor control.

[Manul]

If you isolate, then you also need to have isolated motor control.

Do you think they used AC fan? Well, ok I don't know how these devices are made. I'm just so happy, that no one asks me to save money when I design things

[fisz]

Here is the other side of the power supply board. Are you telling me that the 5V marking actually means +5V referenced to mains? And the only thing that stops people from electrocution is a little plastic cover of any of the buttons?

[fisz]

Which means you passed mains voltage through low voltage circuit into the iron tip. That's why RCD tripped.

Sorry, I don't understand. I didn't supply any voltage. Either it was low voltage circuit, or I've touched a point which was already at mains (so no low voltage).

[Manul]

Here is the other side of the power supply board. Are you telling me that the 5V marking actually means +5V referenced to mains? And the only thing that stops people from electrocution is a little plastic cover of any of the buttons?

Sadly and shortly, yes.

[wraper]

Which means you passed mains voltage through low voltage circuit into the iron tip. That's why RCD tripped.

Sorry, I don't understand. I didn't supply any voltage. Either it was low voltage circuit, or I've touched a point which was already at mains (so no low voltage).

It's low voltage circuit which is connected to mains. Means it floats at mains voltage but by itself it's low voltage. With what you have done, this circuit was exposed to over 200V across it, where only a few volts should be present normally.

[wraper]