Attempt to repair GW Instak GPS-3303C power supply --- succeed

[shadow.dark]

My GW Instek GPS-3303C power supply malfunctioned after I returned from vacation.The supply has a "Load ON/OFF" button , when turn off, the output is disable.when turn on, it enable the output.
Now, the button is completely ineffective, after PSU power on，settings are all displayed as zero. Press the button, nothing happened.
But interestingly, after a random period of time（At least ten or more minutes）, it will enable output for no reason, and at this point, the display of voltage and current will return to normal.But at this time, you still can no turn off the output. That "Load ON/OFF" button still doesn't work.
I try to find service manual or the schematic for GPS-3303,But it seems that developers have never published such a file.
If anyone has seen a similar problem, can you provide a clue?

If I use the wrong tense or tone with an offensive tone, I'm sorry that I'm just using poor online translation tool.

[shadow.dark]

It seems that we have encountered a tricky problem this time. It's not that the button itself has poor contact or other issues, and there are no visible fault clues inside the power supply. Another troublesome thing is that as this is a three way isolated power supply, I find it difficult to find a suitable reference GND point on the board. You can't even find large pieces of copper on the board.

There are rumors that the TEK PS280 is similar to this model, but TEK seems does not have this  "Load ON/OFF" switch. I am also looking for other service manual that could help.

[Swake]

It looks like the power supply is all analog. No micro controller.

Likely the 'control' part of this lab supply has its own power supply somewhat isolated from the power output circuits. This might be the reason why you find several traces that should all be ground but are in fact unconnected. You have to find that internal power supply and check it because there is a good chance it is  bad.

If that power supply is confirmed good then you have to find how the output button switches the output because something is wrong with that circuit. You checked the button itself, that is a good thing. If you are still in doubt about the button you can solder some wires on it and simulate the button by short-cutting the wires.

The first suspects are the relays on the board, are these used to switch the output? If yes, check how these are energized. I think the button might activate some transistor circuit or similar. Then check those transistor, diodes, capacitors or whatever in that energizing circuit.

If the relays have nothing to do with switching the output then trace the PCB from the button and see what components it is going to. You light find one or more transistors (FETs) in line with the output.

[RomDump]

I try to find service manual or the schematic for GPS-3303,But it seems that developers have never published such a file.

You can e-mail Instek for portion of the schematic. They wouldn't release the full schematic.

The Tenma 72-6905 Schematic is similar.

[shadow.dark]

Thank you very much. Compared with the parts I have already explored, the Tenma 72-6905 schematic is consistent with GPS-3303C. Thank you again for your help. By next weekend, I will continue troubleshooting and post the latest news here.

[shadow.dark]

Better late than never. I will provide a detailed explanation of the troubleshooting process as follows. Thank you again, RomDump. The drawings are in good match with the power supply.So I am attaching it here, although it may waste some storage resources.
The entire switch function is built around the MC14013 dual D trigger. However, not both triggers are used as triggers, and the first D trigger circuit is more like a buffer amplifier.
When I measur it using the nodes in the blue box as a reference GND, I found that the voltage of the nodes in the red box was at 13V，When the button is pressed, the voltage rises to 15V, which is clearly not a normal input voltage for CMOS integrated circuits.
I dismantled and soldered C505 Q501 U501, almost all of which were connected to the problem node to find the leakage point.
But finally it was discovered that the problem was still with the switch itself.
Yes, the switch can be turned on normally, but when it is turned off, there is still a conduction resistance of several thousand ohms, which is very different from the situation I often encountered before.
Perhaps something entered the switch, but I don't want to delve deeper. After replacing the switch, the power supply returned to normal.

[Swake]

Good catch !
A button is such a simple component and yet it can fail in so many different ways which are sometimes very difficult to find. This one was no exception to the rule.