Microwave oven random restarts of the panel

[junkpete]

My Samsung CE1185UB Microwave from 2008 is starting to die. Everything works fine (well, the main timer knob which is a rotary encoder does a few skips and jumps but a few days ago I saw that the display showed a blinking "12:00", which means that there was a power outage or similar. Such a power outage happens extremely rarely in our area but I ignored it and set the time. A few hours later it again showed a blinking "12:00". I put it some stuff and turned it on and there I saw what happens: the software restarts here and there and resets everything.

I disassembled the oven and examined the board where the microcontroller is sitting but did not see anything weird. I measured however 3.6V on the 5V rail (buffer capacitor) and 9.7V on the 12V rail (12V buffer capacitor). Right after plugging the oven into the power outlet, both voltages are a bit higher (4V and 11V) and then both slowly decrease. I guess that's the reason why at some point the microcontroller shuts down.

I replaced both buffer caps since their ESR was a bit high but that did not change anything. I found the repair manual including the schematics online but I am struggling a little bit to figure out what happens there, especially with that TNY266 which "integrates a 700 V power MOSFET, oscillator, high voltage switched current source, current limit and thermal shutdown circuitry onto a monolithic device.".

I don't really grasp how that TNY266 works so I have not idea how to debug it

Anyways, I measured a few other voltages and the rectifier BD101 creates from 233V AC a fairly stable 317V DC.

I attached the circuit with my measurements.

Any help is really appreciated...

[fzabkar]

317VDC seems low. It should be more like 329V.

I would replace the electrolytic capacitors (C102 and C108).

[junkpete]

OK thanks, I'll do. I already replaced C108 but that did not change anything. I had no 450V cap stocked... I'll order a few.

[richnormand]

initially:
I would check c102.
Make sure the bridge rectifier is all ok.
Check the zener zd102 in the feedback loop and see if it is leaky.

[junkpete]

The C102 has 3.9 Ohm ESR, I think that's a bit too high (see attachment).

The rectifier is fine, I just checked it.

How do I check if the Zener ZD102 is "leaky"?

[shakalnokturn]

Most of the time the TNY series die by spewing their guts and killing the primary inrush current limiting fusible resistor. (So I'd assume you're fine so far...)
A thermal camera would save you time here.

To check your Zeners you want the reverse current to be very low when the diode is reverse biased below the Zener voltage.

At this point I'd try isolating the 5V and 12V from the rest of the PCB to be sure something downstream isn't dragging it low, then check supply voltages again.
Make sure to keep the regulation loop closed, do not run it with anything less than what figures in your attached schematic.

[junkpete]

I measured 20.70 MOhm with my Keathley 197 and the value very slowly decreases (about 0.01 MOhm every 5 sec).

I quickly made a test by soldering a 1K resistor in series with the ZD102 and increasing the voltage across both of them. The voltage drop over the ZD102 is definitely not staying around 4.3V when it's higher. I went up to 10V and got 9.7V or so.

Thanks shakalnokturn for the info about the TNY, glad to hear I can probably rule it out.

Yes, isolating the 5V and 12V would be nice but I am a bit struggling with that.

[junkpete]

I replaced the Zener diode with a new one, but the 5V and 12V still read around 3.6V and 9.7V  I'll try to isolate them and also ordered a new 450V 10uF cap.

[shakalnokturn]

Vz (ZD102)+ opto's LED Vf set the regulation voltage on the +5V rail, the +12V follows this with a little bit of margin that is clamped to a safe value by ZD103.

[Jwillis]

The C102 has 3.9 Ohm ESR, I think that's a bit too high (see attachment).

The rectifier is fine, I just checked it.

How do I check if the Zener ZD102 is "leaky"?

Looks like a lot of smokey residue around that capacitor originating from the rectifier area . Check all connections for scorching . Microwave ovens are notorious for burned wires and connectors. Burned connections can cause power disconnection and weird voltages across the board.

[MarkT]

Lots of possilibities to go through - but the voltage drift suggests the feedback path which includes the opto-isolator.  The slow drift indicates thermal sensitivity (cool spray for diagnoses?).  But be careful in there, microwave ovens contain _extremely_ lethal high voltages.

Or give up on the old unit... A modern microwave oven with an inverter PSU would be my recommendation, more efficient, much lighter.

I keep mine switched off at the wall when not in use - probably going to extend its life as well as avoid all the standby current consumption.

[p.larner]

mine started doing random things,it was the tactile buttons making random contact,i replaced them all,its been fine since,that was probrably 3yrs ago,maybe worth a look.

[bdunham7]

You need to measure at both sides of the PC102 optoisolator to see whether the low voltage is due to feedback or if the system just can't get enough power across the transformer.  Then you'll know which path to follow.

[junkpete]

Thanks all for the comments. Today the 450V cap arrived but it did not change anything, the 5V and 12V rails are still below 4V respectively 10V and slowly decrease after powering the microwave on.

Don't worry, I disconnected the high-voltage parts.

I'll check the PC102 optoisolators next. At least now I can rule out all electrolytic caps.

[junkpete]

PC102 is fine. I desoldered it and the forward voltage from pin 1 to 2 is about 1.0V. The other side is also fine (works as expected in a test circuit) but I still replaced it with a fresh new one, just in case.

Not sure how to proceed. Replace the TNY266? Is there  a good way to test it?