Microwave tripping AFCI during the night

[Doc777]

The AFCI protecting the under cabinet microwave is tripping during the night while not in use. New home is 7 years old, 1100W microwave is also 7 years old. While unplugging the microwave from the dedicated 15A duplex receptacle (on a 20A breaker and 12/2 single run to the outlet) I discovered the receptacle was lightly scorched and the heavy rubber cord end right at the hot & Neutral prongs was slightly melted. Obviously there was/is heat buildup at the plug/receptacle junction. At first I attributed this to items in the cupboard where the outlet is located as pushing the cord end around in the receptacle. No other problems are present with the microwave.

After inspecting the circuit wires and finding no evidence of damage, I replaced the Duplex receptacle with a 20A dedicated single receptacle, and replaced the power cord on the microwave (it can not be hard wired), and made sure items in the cupboard did not effect the cord end. For extra protection I installed a new Siemens 20A AFCI breaker into the panel at location #4 with other 20A non AFCI loads around it. No other breaker tripping is occurring in the panel.

The AFCI will trip during the night (It sometimes will go 2 nights before tripping) while the family is asleep and the unit is not in use other than a clock display. It never trips during use. The LED trip lights on the breaker indicate an arc fault to ground. I'm tempted to uninstall the AFCI breaker in favor of a standard breaker but am concerned that there may be a an actual fire causing issue with this.

Any ideas on whats going on here?

 

[growler]

Any ideas on whats going on here?

Not really, sounds like a tough one. The only thing you know for sure is that there is a problem.

Is the problem in the microwave or is the problem in the branch circuit?

If this is a dedicated microwave circuit I think I would just meg the conductors of the branch circuit.


I saw a problem once where it had taken 20 years for a staple driven in to tight caused a section of romex to burn/open . It didn't cause a fire but I have a feeling that it would have tripped an AFCI years before. The owner of the house wanted the wall opened so we could see if the siding guys had driven a nail into the romex ( new siding) but it was a problem from original wiring.

 

[readydave8]

Sounds like the microwave is trying to burn down the house and the AFCI breaker is preventing it from doing so, should you consider changing out the microwave rather than the breaker?

 

[growler]

Sounds like the microwave is trying to burn down the house and the AFCI breaker is preventing it from doing so, should you consider changing out the microwave rather than the breaker?

Sure it sounds like a bad microwave but between the cost of buying and installing a new microwave is would be nice to know for sure.

If it were my house I would run a cord over to the microwave and leave it on a different circuit (AFCI protected) for a few days to see if it trips. It not under a load so a cord wouldn't be a problem.

 

[Ohms law]

The AFCI protecting the under cabinet microwave is tripping during the night while not in use. New home is 7 years old, 1100W microwave is also 7 years old. While unplugging the microwave from the dedicated 15A duplex receptacle (on a 20A breaker and 12/2 single run to the outlet) I discovered the receptacle was lightly scorched and the heavy rubber cord end right at the hot & Neutral prongs was slightly melted. Obviously there was/is heat buildup at the plug/receptacle junction. At first I attributed this to items in the cupboard where the outlet is located as pushing the cord end around in the receptacle. No other problems are present with the microwave.

After inspecting the circuit wires and finding no evidence of damage, I replaced the Duplex receptacle with a 20A dedicated single receptacle, and replaced the power cord on the microwave (it can not be hard wired), and made sure items in the cupboard did not effect the cord end. For extra protection I installed a new Siemens 20A AFCI breaker into the panel at location #4 with other 20A non AFCI loads around it. No other breaker tripping is occurring in the panel.

The AFCI will trip during the night (It sometimes will go 2 nights before tripping) while the family is asleep and the unit is not in use other than a clock display. It never trips during use. The LED trip lights on the breaker indicate an arc fault to ground. I'm tempted to uninstall the AFCI breaker in favor of a standard breaker but am concerned that there may be a an actual fire causing issue with this.

Any ideas on whats going on here?

Check to see if the ground wire is not touching the neutral terminal.

 

[Dennis Alwon]

Why is the mw on an afci? Is it in the kitchen?

 

[Dennis Alwon]

I would also unplug the mw and see if it still trips the afci.

 

[readydave8]

Sure it sounds like a bad microwave but between the cost of buying and installing a new microwave is would be nice to know for sure.

If it were my house I would run a cord over to the microwave and leave it on a different circuit (AFCI protected) for a few days to see if it trips. It not under a load so a cord wouldn't be a problem.

I agree, I was responding to mention of possibly changing out AFCI to regular breaker.

 

[readydave8]

I'm tempted to uninstall the AFCI breaker in favor of a standard breaker but am concerned that there may be a an actual fire causing issue with this.

that

 

[Doc777]

I put the AFCI in because I'm not sure what caused the MW cord end to melt around he prongs. Something caused it to get hot. I'm leaning towards the MW. What puzzles me is it happens at night when not in use. No other time. I think its time to try plugging it into another AFCI Cercuit for a few days to see what happens.

 

[Ohms law]

I put the AFCI in because I'm not sure what caused the MW cord end to melt around he prongs. Something caused it to get hot. I'm leaning towards the MW. What puzzles me is it happens at night when not in use. No other time. I think its time to try plugging it into another AFCI Cercuit for a few days to see what happens.

It is either a loose connection in your receptacle or you spilt some form of liquid on it. I went to a service call one time and the range receptacle was a surface mount type, someone spilt something behind the range and the whole plug was melted to the receptacle.

 

[Dennis Alwon]

Whenever I have seen a cord melt it was either a loose connection in either the cord cap or the receptacle. Are you sure the mw is on its own circuit?

 

[junkhound]

Having repaired a number of microwaves, there is a fairly common failure mode that would cause an AFCI to trip.

Totally unrelated to the bad plug/outlet, which may be due to loose high impedance plug outlet only without arcing there.


Most mircowave have at least one spot where a 12 AWG internal wire is soldered into a circuit board. This connections soetimes powers both the 7kV HV transformer and the timer/display circuits. This solder connection often becomes loose due to literally melting the solder during microwave operation. During normal operation there is a high enough current to locally melt solder spots and not arc. At night, at the very low loads just to keep the display clock lit, there is arcing at that loose solder joint, tripping your breaker.

Easy to find resolder - BUT, AND A BIG BUT, do not open a microwave, even unplugged, unless you are familiar with high voltage.

 

[jaggedben]

... What puzzles me is it happens at night when not in use. ...

Here's a hypothesis: Some piece of metal is contracting at lower temperatures, and/or there's condensation somewhere, or both, that is causing an intermittent high resistance short. When it happens might change with the seasons.

 

[curt swartz]

I would try the suggestion of leaving the micro unplugged for a few days to see if the breaker holds.

Siemens AFCI's have indicator lights to show what type of fault tripped the breaker. Have you look at these?

 

[kwired]

During normal operation there is a high enough current to locally melt solder spots and not arc. At night, at the very low loads just to keep the display clock lit, there is arcing at that loose solder joint, tripping your breaker.

But is is drawing enough current to even have the AFCI look for arc signature? The clock probably only has a load in the milliamp ranges.

 

[growler]

I installed a new Siemens 20A AFCI breaker into the panel.

The LED trip lights on the breaker indicate an arc fault to ground.

Siemens AFCI's have indicator lights to show what type of fault tripped the breaker. Have you look at these?

:slaphead:
It sounds like he took your advice before you even gave it.



That arc fault to ground makes me think the cable could be damaged or as Dennis says there may be something other than the microwave on this cicuit.

I had a customer with a similar problem years ago. I installed GFCI protected receptacles on his counter top ( had to, code ). A couple months later the guy calls and says his microwave will not work, keep tripping GFCI. His wife tells me he spilled a bowl of water in the microwave. I plug the refrigerator into the GFCI protected receptacle and leave it running for the rest of the day and come back that evening and tell him to get a new microwave or have that on repaired. This thing would trip when in use. I'm not sure what would cause and arc to ground when not in use.

I still think there is a possiblity of a bad cable discovered totally by accident. if you can eliminate the cable then it's probably best to replace the microwave ( 7 years old), there really shouldn't be that much current flow in clock mode. . They can always contact the manufacturer to see if this model would even work without tripping an arc fault.

 

[junkhound]

clock probably only has a load in the milliamp ranges

Very true from a minutes long RMS standpoint, probably under a mA RMS.

What often happens in the circuitry is that the solder joint is a high impendance at low loads and much lower impedance at a few hundred mA or more. The control circuitry is usually a full bridge rectifier on a small xfmer secondary or nowadays a rectifier bridge right on the line followed by a power factor correction circuit and a high frequency switching regulator for the low voltage controls. When the 120 Vac solder joint is high impedance (or open), the few hundred microfarad cap on the rectifier output drifts down to a lower level, the voltage drop now increases across the solder joint and it acts like an arc with a few hundred mA or even a few amps current inrush for just a few microseconds to recharge the capacitor. EMI signatures of this type low level arc tend to cluster around 0.5 to 20 MHz, just the type of thing the 'advanced electronic circuitry' in AFCIs look for. Ham radio stations in nearby homes are said to have tripped AFCIs.

Some of the solid state power electronics have very subtle failure modes. In high volume commercial electronics, transformers are being replaced by silicon switches and much smaller 50 kHz and above ferrite transformers - which in large volumes are much cheaper than the old standby small 60 Hz class II transformer. Even wall warts are going away from small 60 Hz transformers to switching regualtors.

 

[curt swartz]

:slaphead:
It sounds like he took your advice before you even gave it.

I guess I need to wear glasses!! Or Learn to read!!

 

[growler]

the breaker indicate an arc fault to ground.

What often happens in the circuitry is that the solder joint is a high impendance at low loads and much lower impedance at a few hundred mA or more. The control circuitry is usually a full bridge rectifier on a small xfmer secondary or nowadays a rectifier bridge right on the line followed by a power factor correction circuit and a high frequency switching regulator for the low voltage controls. When the 120 Vac solder joint is high impedance (or open), the few hundred microfarad cap on the rectifier output drifts down to a lower level, the voltage drop now increases across the solder joint and it acts like an arc with a few hundred mA or even a few amps current inrush for just a few microseconds to recharge the capacitor. EMI signatures of this type low level arc tend to cluster around 0.5 to 20 MHz, just the type of thing the 'advanced electronic circuitry' in AFCIs look for. Ham radio stations in nearby homes are said to have tripped AFCIs.

This type of breaker is supposed to be able to tell the difference between an arc fault and a fault to ground. This is displayed on the breaker useing LEDs.


Wouldn't this show up as an arc fault and not a fault to ground?

I would think it would be seen as a series fault.


What going to cause a fault to ground during non use period.

They could put the microwave on a GFCI breaker and see if that trips just to make sure.