Did Sense know my microwave was dead?

I’ve been eating the same breakfast for a few weeks now: 2 over medium fried eggs, 1 slice of toast, and meat - typically 2 frozen sausage links. I hear the frozen links in the microwave until hot and then brown them in the pan I used for the eggs. Yum.

This morning, my sausage was still frozen after the usual one minute in the microwave :disappointed_relieved:. Maybe I did something wrong, so I popped another minute on the timer. No luck. I unplugged it to see if I could reset something (works for my tv :smiley:) but still no dice. I ended up cooking the sausage completely in the pan.

Then, I wondered what Sense knew about my microwave, sense it’s detection has been pretty reliable. Turns out, the Sense device microwave showed it has been off for 2 days. It did not detect when I successfully heated the sausage yesterday, but it did know about the day before.

Could this last use of the microwave yesterday, before it gave up the ghost, have had a different enough signature that Sense’s algorithm didn’t notice? Could this give any clue to what may be wrong with it? This seemed like an interesting coincidence.


Interesting! Can you grab some power meter screenshots over the past few days?

Here are the last 2 uses Sense identified. Pretty boring.

Here’s the time before. More interesting.

And finally, the few times before that. (Check the date, I’m not sure if the post in order)

Theory (I’m not an electrically minded person): maybe the last successful use that sense didn’t detect used the last of the energy stored in the capacitor (is that the right word)? Then, with it being dead the next use didn’t work. Could this point me to which component to replace first?

The screenshot with the choppiness at 11:40 looks like a microwave being run at less than 100% power. By default, all microwaves I’ve seen run at 100% unless you make a change at startup. That waveform looks like 80%-90% power. Because how they produce less power is to have cycling on and off of the magnetron. Think of a light bulb being on 80% of the time. It would be all on or all off. This is controlled in your microwave control board.
This only applies to a microwave that is not so new that it’s inverter based, yours doesn’t appear to be that type.
We’d the 27th is the one I’m referring to, went and checked. Looks like it was set to less than 100% power for a period then set to finish at 100%. Like I would microwave a frozen sandwich. Defrost 1:30 high power for 1:00

Interesting insight. As far as I know, we never used anything less than 100%. So, could that be telling of a possible issue that eventually resulted in kaputs?

Whelp… the tech came out today to fix our fridge (another story for another time). While he was here, my wife asked if he also does microwaves.

Mind you, we’ve tried it a few times since noticing it was dead on Sunday. Last night, my wife ran it for 20 minutes by mistake instead of cooking the squash in the temporary microwace we borrowed to put on our counter until ours is fixed. Of course, the tech tested it and heated the water just fine. :man_facepalming:t2: My wife called me at work to let me know the results.

My first thought was to wonder if Sense detected it this time since it seemed to work correctly. Nope, the device in Sense shows last on 4+ days ago. When I got home, I had to test it myself. The power meter graph shows it comes on, but I don’t see it turn off (at the big drop).

The device page still shows 4+ days since last on. Is this because it didn’t recognize the off signature (only the on)?

I would t give up on it yet. If Sense could see the on signature then it may just need to refine the off. Of course, that’s if it continues working.
Do you notice a difference in wattage between four days ago and earlier compared to now?

We used it for dinner tonight. This time, Sense saw both on and off, and it registered in the device power meter. My wife thinks it didn’t heat as well as before, but maybe that’s a biased observation, knowing it was having trouble.

Here is the closeup for the 2 uses at dinner, 1 minute each.

Although the timeline shows each use for those two was different wattage, the device power meter clearly shows the same. I’m thinking Sense is just missing it sometimes which is common with a lot of detected devices.

Maybe, though the signature for this device has always been pretty reliable. Others have missed or had conflation, but not this one.

Still doesn’t explain why the microwave wasn’t working for a few days, suddenly started working again, and whether the Sense had any insight into that (causing the signature to miss, or if the different plots it did detect are telling in any way).

I thought the one waveform you had shown where the microwave was on then off a few times in a row was odd. That would only happen in reduced power. Say I cook something for 2 minutes on 50% power. It will power the magnatron for 30 seconds then off for 30, it will repeat this to complete the cycle. A total of 1 minute on and 1 minute off alternating back and forth. You may be right and this was your wanting of this device giving up the ghost.
You’ll find out for sure when you need it to work and it doesn’t
Or the wife is playing tricks to get that fancy new microwave she’s always wanted.


I’ve had the same experience of a device not showing on for a couple of days before it goes out. I find this to be an exciting and extremely useful perk and am hopeful the Sense team will be able to figure out how to send us warnings/alerts along these lines.

The microwave is not working again, and Sense is not detecting the off signature. I ran it for lunch, and Sense detected the on, but not the off. This doesn’t show in the device power meter, but it says the device has been off for less than an hour (when we were trying to reheat leftovers).

Here are the most recent uses that the Sense device power graph shows.

How old is your microwave. It could be that your magnetron tube has degraded. The on signature would be the same - the high voltage transformer feeding the characteristic load of the magnetron. But the off would likely be different - resonant energy wouldn’t build up in the magnetron so you wouldn’t see the results that built up resonant energy on the inbound current when the oven goes off.

1 Like

So it’s still showing on?
I’m not understanding because the wattage would also show?

Sense detects the on signature but no wattage shows in the device power meter. I’m not sure how to explain why that happens, only that’s what I observe.

We bought it maybe 6-7 years ago. It can’t be older than June 2011 because that’s when we bought the house.

I don’t know much about how microwaves work, but what I’ve learned on the internet is that the magnetron is the one component you probably don’t want to replace on your own — better to just buy a new one at that point.

@kevin1 do you think it’s the magnetron that is bad?

I didn’t see @kevin1 answer so thought I’d give you my experience with magnetrons. I believe they somehow deteriorate and while the input current stays the same, the output will go down. When you ha r a microwave that heats something at say 2:00 minutes and eventually that time increases to heat the same thing, the magnatron is usually to blame.
There is also a transformer in there to power everything than go bad

That sounds like it may be my problem. When the microwave started working again, it seemed like it maybe wasn’t heating things as quickly as it was before. I didn’t know if that made sense or if I was just looking for something wrong.

Mine started to take longer to heat and I looked into replacing parts like the magnatron. It was about the same price to buy another microwave as I just use a countertop model, so that’s the route I went.
Wish I would have had Sense at the time so I could have played around with it more