Microwave Runs at Half Power [SOLVED]

[jpanhalt]

Kenmore Elite 1200 W microwave approximately 2003 vintage.  Inverter type. Came with house and worked fine until Friday morning, May 24.  It is an above the range, built-in model and a bit heavy for me alone to take off the wall.

Tore into it yesterday and checked what I could without opening the inverter section.  The magnetron is Panasonic, and I presume inverter board is too.  Magnetron is made in Korea.  Attached is a schematic from Panasonic.  It appears generic.  Magnetron filament tested OK with very low resistance and no short to ground. 

After being without power for more than an hour, it was reconnected to power and seemed to work better but quickly returned to half power based on time to warm cream in my coffee cup.

I have no prior experience repairing microwaves and read that magnetrons do age, but assumed that aging is gradual.  There are 2 diodes in the half-wave doubler.  What is the chance one of those failed open and is causing the reduced power?  There are some service tests in the manual from which the schematic came, but they require opening the inverter section.

Thanks, John

[Gyro]

We had a Panasonic inverter microwave. Similar symptoms, full power initially and then tailing off. It turned out to be cracked magnets on the Magnetron causing reduced field and overheating as it slowly melted its internal anode structure...  https://www.eevblog.com/forum/projects/will-a-magnetron-draw-480-watts-without-radiating-energy/msg3210292/#msg3210292

The-tell tale sign (apart from cracked magnets!) is loose contact between the cooling fins and the Anode. It's possible that it might be an inverter problem, but given the symptoms I doubt it - and you really don't want to be messing with it! A replacement magnetron is an option but from the sound of it, it doesn't owe you anything.

[soldar]

I find that schematic interesting. I am familiar with older microwaves which worked at mains frequency. This one seems to have a flyback transformer to provide the high voltage and I assume it varies the power by changing the working frequency.

I would start by checking the components on the high voltage side of the transformer, capacitors, diodes, resistor. Needless to say, be very careful and make sure capacitors are discharged although such low values are much less dangerous than the higher value, low frequency ones.

If everything checks out then I would try to check the frequency at the base of Q701 but it would be better if you knew what frequency to expect.

I will be following this thread. Good luck.

[jpanhalt]

@Gyro
Buying a replacement is not prohibitively expensive.  Finding one that is front vented with a vent fan for the electric range below is, which I didn't mention, is hard.  Black & Decker make a 1000W, sold by Home Depot for about $250.  That's the only one I Have found so far.  I don't know who really makes it.  I will check out the fins when I reopen it.

@soldar
Its circuit is similar to current ones.  According to Panasonic, line voltage is rectified, then an IGBT chops it to something between 20 kHz and 40 kHz followed by a small transformer and half-wave rectifier with voltage doubler.  An MCU on another board controls the IGBT for power control using PWM.

[soldar]

@soldar
Its circuit is similar to current ones.  According to Panasonic, line voltage is rectified, then an IGBT chops it to something between 20 kHz and 40 kHz followed by a small transformer and half-wave rectifier with voltage doubler.  An MCU on another board controls the IGBT for power control using PWM.

That has to be a flyback. You cannot put DC into a direct transformer.

So if it works between 20 and 40 KHz then the highest frequency would be the highest power. If the frequency is low then the power will be low. maybe something to check if you can do it easily.

[jpanhalt]

That has to be a flyback. You cannot put DC into a direct transformer.

So if it works between 20 and 40 KHz then the highest frequency would be the highest power. If the frequency is low then the power will be low. maybe something to check if you can do it easily.

Q701 provides oscillation of current with PWM by connecting and disconnecting the low-side of the coil to ground. 

[Gyro]

@jpanhalt: Ah yes, finding a suitable built-in front vented model is certainly going to limit your choices (and price options), it would here anyway. This could be a case where, after mechanical confirmation, it may be worth changing the Magnetron.

@both: Yes, the schematic does look like a simple flyback rather surprising considering the 1200W rating. The 'standard' (probably depending on age) configuration for the Panasonic inverter is shown and described on this page:  https://streampowers.blogspot.com/2012/09/panasonic-microwave-oven-inverter-hv.html and uses two IGBTs. Maybe some cost 'optimization' has taken place since. The one shown matches the one that I removed from our microwave. I'm still wondering what to do with it. I've already stripped some of the HV secondary parts and had some thoughts of rewinding the transformer secondary and trying to turn it into some form of induction heater but thoughts are as far as I've got!

[soldar]

Q701 provides oscillation of current with PWM by connecting and disconnecting the low-side of the coil to ground.

Do you understand the difference between a transformer in forward mode and flyback mode?

Do you understand that with that configuration it cannot function in forward mode? A forward transformer cannot function with current going only in one direction.

[soldar]

Yes, the schematic does look like a simple flyback rather surprising considering the 1200W rating.

My first guess would be that Q701 conducts for a fixed period of time every time and the off period is lengthened to throttle power.

So it could be something like 15 microseconds on, 10 microsecond off at full power and 15 microseconds on, 35 microseconds off at half power.

You cannot throttle more with this system because the power to the magnetron filament is also diminished.

First thing I would check is D701, D702, C703, C704 and R701 for good measure.

[SeanB]

Most likely magnetron dying, and needs replacing. the power supply is either on or off, it operates by varying the run time of the power stage, turning it on for a few milliseconds per mains cycle, so the power applied per mains half cycle is varied from nothing (off state) to power being applied for the full half cycle. IC 702 provides a line zero cross reference to the main board, which then allows power to be varied smoothly over a range, allowing enough time to have the heater fully hot, and the power being the amount of time on per half cycle of the mains as fine control, and coarse control, for things like defrost, by skipping cycles, relying on the heater staying hot over the few skipped cycles to run through. Power operation at anything under full is bad on the cathode, as it runs cool, so they are very high in thorium to keep emission up, at the expense of running very hot at full power. But a replacement magnetron, generics are available, no need to get the Matsushita branded one, as they really only have a few variants now for microwaves, almost all the variants are in the fin orientation, output probe depth, and the orientation of the heater connections.

[jpanhalt]

Thanks.  That confirms my impression.  In any event I need a friend to help remove it as my strength has gotten a lot less since I was just 60.  We will need to pull the range out for better handling, but that is not a big problem -- just a little inconvenient.

In the meantime, I am searching for a new, over-the-range, recirculating oven that has front exhaust, will fit the space, and is in stock.  I ordered the Black&Decker that seemed to fit those requirements from Home Depot last week, and HD notified me yesterday that it was now out of stock.  The closest thing that seems to be in stock is a Panasonic + mounting pieces (about $250 for just those pieces), and it ends up too tall and needs to me mounted on a shelf, which doesn't exist.  I don't want this magnetron to turn into a kitchen remodeling project.

[Hamelec]

https://dokumen.tips/documents/magnetron-y-sustitutos-no-2.html?page=1

[jdragoset]

Check if one of the GE over-the-stove microwaves at HD will fit.
It seems they all come with the ability to vent forward, into the kitchen or up or back to vent into exhaust ducting.

[jpanhalt]

[SOLVED]

I changed the magnetron for about $60 total as that was very easy to do and high voltage board had no obvious hot spots on top.  Power was still at about 50% (temperature rise of 1L of water for 60 seconds test), so I tore deeper and found the attached (HV board).  Both IGBT connections on board bottom were well darkened with soot.  End of story.  I got the Frigidaire Gallery 1000W at $299 vs. current "discount price" of $279.  That sale ended at Lowes yesterday, but a kind supervisor gave me the same price today. That's the cost of experience, which is several orders of magnitude less than my most expensive experience, so I am happy.  I will save the high voltage  (3kV) caps and bleeder resistors.

John