Author Options:

GE Microwave Makes Humming Sound with Smell Answered

My GE microwave (JE1860SB 002) made a cracking sound and stopped working. After some research, I removed the cover and checked the fuse and it was good. I then checked the thermal flame/sensor and discovered it was burned out. I replaced the thermal flame/sensor and now the unit comes on but makes a loud humming sound and has a burnt smell. The next step (according to my research) is checking the diode (cable) which is attached to the capacitor. It appears I need to discharge the capacitor before proceeding? --- BEFORE replacing the sensor, the unit had been unplugged for over one week….AFTER replacing the sensor, the unit was plugged in for about 5 minutes on Jan 25, 2018.

- What is the best/safest way to discharge the capacitor? Is an insulated screw driver or pliers needed or will a rubber handle tool work considering the time it has been unplugged? (Capacitor pictured below)

- Should I remove the two connecting wires and diode before attempting to discharge the capacitor? - If the diode is bad should I still check the Magnetron?

- If the diode is good, what is the next item to check/test?



5 months ago

Then it never hurts to short the capacitor TWICE..(heavy plastic Screwdriver)..
Sounds Like a Shorted coil in the transformer and yes it could be a shorted diode..

Click pic 2see whole image


Thank you for your response. The unit has been unplugged for four days. Confirming that a heavy plastic screwdriver will be ok to use to discharge the capacitor? Also, the Thermostat flame/sensor is called "thermal cut-off" or "thermal fuse" on many other brands. If the fact that it blew gives you a better idea of the source of the problem, please advise. Thank you.

GE Thermal Sensor.jpgIMG_8904.JPG

I am familiar with that klixon thermal circuit interrupter..
They are self resetting when the thermal event goes back to room temperature..
As far as what thermal event set it off that is hard to tell...
A) the metal where it is attached could have tripped it off.
B) surrounding air got hot but there would be discoloration on the paint..
C) Very rare, a power line voltage spike could cause a internal arc and that heat tripping it..

I thought I would add to this thread since this subject is similar and I haven't found much online regarding this subject. Hopefully my reply will help someone else. My counter top GE microwave started to emit a burning smell while doing normal cooking (nothing burned inside). Then it lost all power - nothing worked on it. It also did this about a month ago (but no burning smell) and then after a day started working again. But this time I couldn't get it to restart after 2 days. I took the case off and saw that one of the two white plastic connectors to the "Thermostat Flame Sensor" (TFS)(the one on the black wire) that Bayoudude showed in his picture was blackened and melted to the TFS. That accounted for the smell. I can easily replace the TFS but I need a new white, plastic connector shown in his picture. Anyone know what they are called and where to find those? I've looked everywhere online and the local radio shack has closed. I don't know why those connectors are used in certain connections inside the microwave and other connections use the normal quick disconnect connectors that have the metal exposed. Thanks!

Usually a spade lug female crimp to wire connector available in small packs in hardware stores.


Thank you! I have those and I figured they would work but I wasn't going to use them since I thought GE obviously used the long, rectangular, white ones for a reason (couldn't theorize that reason, though!)

As a follow-up to my microwave fix: I did a continuity test on the TFS that had partially melted at the one wire connection and it surprisingly tested GOOD. I didn't trust it so I removed a TFS from an older GE microwave I hadn't thrown out yet that my son destroyed by burning popcorn. I also cut off one of those white connectors from the old microwave and soldered that to the cut wire on the newer microwave I'm fixing. The replacement (used) TFS and my solder job got the microwave running. Bottom line for my situation was the wire connection to the TFS shorted/burned and the TFS was OK even though I swapped it.

I suspect those insulators are formulated to be acceptable as Food Grade Material over the electrical properties of insulation... In fact, personally I would use unshielded bare connectors.

See the flame sensor is norm line volts, why worry insulation except arc which you already had !


Yeah, you are right. When I'm working with things I'm not educated with I simply don't like to assume parts are swappable unless I know for sure. I'm just surprised the fuse and the TFS were OK. Seems to me at least the fuse should've tripped but I guess the connection shorted/burned open before the fuse could trip.

You cannot beat fuse timing, but if the temp gets higher then 212 the TFS opens the circuit and re-starts the uP.

I suggest looking at the filament connection, on the magnetron tube, to see if that is visibly melted or charred.

If you want you can check those with an ohmmeter too, but that test is not definitive, unless there is a short there that persists, even when no high voltage is applied to the tube.

Ideally the DC resistance seen between the filament and the body of the magnetron tube will be infinite, or in a practical sense, too high for your meter to measure. E.g. my HarborFreight(r) multimeter cannot measure resistance greater than 2 megaohm = 2.0e6 ohm

I mean what I suspect, is in that little part that feeds the filament wires through the body of the magnetron tube (it might be called "filament feed-through"? "high voltage feed-through"?) there is a short circuit, or place were the high-voltage insulation has failed.

Thus when you energize the transformer and its little rectifier circuit, it is actually driving a short, or an arcing spark gap, and that might explain the louder buzzing noise; i.e. this is the sound of the transformer being overloaded.

If it is a failed high-voltage feed-through thing, on the magnetron tube, that is the part where most people give up, or go shopping for a new magnetron tube.

It is possible to replace the high-voltage feed-through thing, although this requires surgery on the magnetron tube. It is not that hard, but most people don't do it, and I guess the reason why is probably because there is fear of doing it wrong, or belief that it cannot be done.

I am trying to think if I have any pictures of an obviously burned, or melted, filament feed-through connector, just to show what that looks like, in the cases where it is visibly obvious.

I decided to up a picture of a magnetron tube, with an obvious problem with its filament feed-through connector. The thing to look for in the close-up picture of just the connector, is that black spot on the left side. Essentially the terminal on that side is trying to burn its way through its plastic housing.

Also measuring the (cold) resistance, in ohms, between the either one of the filament pins, and the metal body of the magnetron tube, reveals this resistance is much smaller than it should be.

For this particular broken magnetron tube, the meter sees about 30 ohms between either of those filament connector pins, and the metal body of the magnetron tube. I am guessing this resistance, or rather conductance, is due to charred, melted, gunk.

The same measurement for a solid, working, connector, should be too large for my meter too measure; i.e. the filament terminals not touch, and are well insulated from, the metal body of the magnetron tube.

By the way, the resistance of the filament itself, seems to be too small for my meter to measure; i.e. the two filament terminals look to be connected to each other by zero resistance, as if they were connected to each other by solid metal.


Vertical 500 ma / div Horizontal 0.05 volts / div = 150mv/200ma = 0.75 ohm filament
Click pic to see whole image !

C Plot.jpgC scale.jpgCurveTrace Setup2.jpg

Correct That Please !


Horz 200 mv Vert 1.5 A Filament Res = 0.20v / 1.5 A = 0.133 ohms

CurveTrace Setup2.jpgC scale.jpgC Plot.jpg

Could be cold resistance compared to hot..


5 months ago

Oh yea the acrid burnt smell is a transformer coil so hot that it smokes the insulating varnish.
A shorted diode could cause that transformer to heat up like that but it should also blow the fuse ?

A plastic handle on a big metal screwdriver may spark up to an hour after using the microwave..