What's new
What's new

How do AFCI breakers behave with electric motors?

52 Ford

Stainless
Joined
May 20, 2021
I plugged a treadmill into an outlet that was on an AFCI breaker. Tripped within 1 second of start up (I think the machine has some sort of soft start, so maybe it'd trip quicker with a conventional startup). The same treadmill works fine after I swapped the 15A AFCI for a regular 15A breaker. The machine has a PMDC motor, not sure about this exact model, but probably 90 or 120 VDC, around 2HP.

Do AFCIs just not work well with the high start up current? Or does it have to do with the motor being brushed? Or....

Sent from my SAMSUNG-SM-G891A using Tapatalk
 
I'm the last guy you'd want to listen to on this, but I suspect its the arcing brushes. Open-air arcs emit a boatload of crap that could be detected at the breaker. On the other hand, I would think a lot of that would be buffered by the speed controller. Just thinking out loud, and not very clearly to boot!
 
I'm the last guy you'd want to listen to on this, but I suspect its the arcing brushes. Open-air arcs emit a boatload of crap that could be detected at the breaker. On the other hand, I would think a lot of that would be buffered by the speed controller. Just thinking out loud, and not very clearly to boot!

Arcing brushes would be my guess as well. It's unlikely that the controller filters enough of the RF hash that the breaker detects and trips on to prevent tripping.
 
I'm not an Electrical Engineer, but if you're talking motor brushes then you're taking DC. I don't think it's likely that that arcing is seen on the AC line.

AFCIs and GFCIs are sensitive components. Most treadmills for home use probably function fine on 15A circuits as they are designed/ intended. I've had situations where the above mentioned devices have tripped out on normal tool usage such as power tools, drills and such. Weaker older devices exacerbate the situation even more; it also depends what else is on the circuit or is it's dedicated.

The higher load, harmonics on the line, older faulty equipment, all stack the odds against you. That would just be my initial assessment. A standard breaker is a lot more forgiving to all of those conditions but you also risk the fact that now your entire circuit (i.e. every other receptacle) is no longer protected from arc faults - for what it's worth.
 
Not to mention the fact that there are such things as rectifiers, commonly used to develop 90V and 180V for DC motors off AC..... And treadmills have very commonly used that technique.

AFCI breakers look for a certain type of noise. Maybe some are not very discriminating. But arcs generate noise in the radio frequency region. At one employer, we had a detector for arcs that we developed. It looked in a specific range of frequencies. I think my NDA is still active, so I cannot discuss what frequencies, but they are probably not what you think.

Anyhow, the AFCI looks for HF noise, and trips if it finds sufficient such. My guess is that it would have to avoid the common AM broadcast band, and related frequencies, or they would trip often. I'll have to feed one various frequencies and see what it is looking for.

In any case, rectifiers can make "damped oscillations" which are similar to what arcs produce, and rectifiers can make a unit fail FCC or CE testing unless some filtering is added.

What is worse is that many treadmills have used SCR controls. SCRs create really bad radio "hash", and would probably cause an AFCI to nearly jump out of the wall, if not sufficiently filtered. Triacs have also been used, and do similar, but being often slower devices, they are not quite as bad.

More recent appear to use PWM techniques in more of a "real motor controller". The PWM also makes a lot of noise, which with insufficient filtering could trip an AFCI as well.
 
Not the brushes. It would be the same with a "brushless' PM DC motor.

It's the power supply and the way it provides variable speed & torque. Fast-switching artifacts.

Not THAT hard to filter. As I do.

But I also eschew AFCI/GFCI goods anywhere in the shop.
That works even better!

Cheaper, too!

:)
Treadmill is at the house. Shop is all standard breakers and a couple GFCI outlets. I've honestly never had an issue with the GFCI outlets. Only time they pop is when I do something stupid and I'm glad they did.

The last AFCI I bought for the house was $50ish. That was for a single pole 15A breaker. Makes me thing... a little spark here and there isn't THAT big a deal.

Sent from my SAMSUNG-SM-G891A using Tapatalk
 
In an arcing circuit the voltage rises to the breakdown voltage and suddenly draws a large spike of current until the current drops to basically zero on every half wave.

In an SCR motor drive the voltage rises to the threshhold voltage and the motor suddenly draws a large spike of current until it drops to basically zero on every half wave.

I quite like GFI breakers. I quite despise AFCI breakers. It's just not possible to reliably determine whether an unknown load is working correctly or not by listening in on the wires.

Edit: I should mention this is an oversimplification and that arcing can have many behaviors, but this is one of them.
 
The AFCI devices are really designed to work on circuits that have reasonably well-behaved loads; lamps, etc, even computer power supplies don't seem to faze them.

Intended to be used in bedrooms etc, where it would be bad if the lamp or electric blanket started arcing in the middle of the night, and kept it up until something was lit. Keeps people from being directly barbecued by electrical faults.

Unless you have machines in the bedroom, maybe you do not need the AFCI?

GFCI is required in residential use, at least, for garages and basements, etc. Places with a conductive floor. Even many machines want GFCI, although not the residential 5 mA nominal type. More usually 30 mA, or 100mA.
 
If you think AFCI and GFCI nuisance tripping are bad, take a guess at what happens when the GFPE controller for the main beaker in a high rise substation accidentally gets left at the factory setting of 20 amps. All it takes is one temp light covered in 277V sharpie scrawls that some jackass inconsiderately rewired for 120V.

"Oh shit!"

*queue every sparky in the building frantically converging in the basement to pull off an emergency re-enactment of that one Jurassic Park scene*
:D
 
If you think AFCI and GFCI nuisance tripping are bad, take a guess at what happens when the GFPE controller for the main beaker in a high rise substation accidentally gets left at the factory setting of 20 amps. All it takes is one temp light covered in 277V sharpie scrawls that some jackass inconsiderately rewired for 120V.

"Oh shit!"

*queue every sparky in the building frantically converging in the basement to pull off an emergency re-enactment of that one Jurassic Park scene*
:D
LOL!


Haven't seen that movie in a long time. Thems fancy fuse boxes is a pain. Can't just put duct tape on them to add a few amps. Do that on a normal breaker and you can get a WAY more juice out the wire. Wont trip till the duct tape melts. [emoji16]


I do have a question.... ok, so, 240V... 2 120V legs. I had to replace my 60A (edit: 50 amp) two pole with two 30A single pole breakers (its what I had). I had a 50A 240V transformer stick welder plugged in, cranked it way up. Welded for a while and it popped ONE breaker - not both. I know that if they were tied together, it'd take more force to trip the pair. Anyway... why'd the 30A breaker pop? I'd expect 25A on each leg with the welder turned up. The machine welds GREAT... so I doubt it's something wrong with the machine. It isn't a problem, as I almost always burn under 150ish Amps - some 5/32, lots of 1/8 and 3/32 rods. I'm just wondering if there is some obvious reason behind it. I should add that it's always the same leg, the same breaker, that trips. I have DMMs, a fairly nice scope, and all that, but I keep it in the house, not at the shop, so I haven't tried measuring anything.
 
Last edited:
OK, how about: "The welder DOES pull more than 30, since it pulls from the two 120V in SERIES, so it does pull up to 50A, and the one breaker is just the one that is set a tad lower than the other one." ?
 
A 12kVA load on a 7.2kVA circuit plus a handle tie violation to boot.

Commonly referred to as a 'homeowner special'.

Granted it is a welding load so you could maybe cheese it by citing Article 630, but even so...
 
Solution would be an EMI filter between the device and ac line. Sometimes even clamp-on ferrite sleeves will attenuate the noise enough to prevent nuisance tripping. Might as well try the cheaper solution first.
 
A 12kVA load on a 7.2kVA circuit plus a handle tie violation to boot.

Commonly referred to as a 'homeowner special'.

Granted it is a welding load so you could maybe cheese it by citing Article 630, but even so...
30A breakers, but it's wired for 50A.

Like I said - it's what I had on hand. Rather be welding than waiting on parts - think it was a Saturday or Sunday evening when I was working on that. Holiday? Either way, I had a reason I couldn't get to a store.

FWIW, the welder I was using is the one that had two 14 AWG extension cords ran in parallel as a power cord. [emoji16] it was either that or use Romex.

Again, got the job done.

If it was someone else's welder or at someone elses shop, I wouldnt do any of that. Risk/reward... I'm fine with the amount of risk.
 
Solution would be an EMI filter between the device and ac line. Sometimes even clamp-on ferrite sleeves will attenuate the noise enough to prevent nuisance tripping. Might as well try the cheaper solution first.
Interesting. I'll have to try that. I haven't opened up this treadmill. However, all of them that I've worked on in the past had a big filter on the AC going to the main PCB.
 
30A breakers, but it's wired for 50A.

Let's see...

240V welder being powered from L1 and L2 of a 120/240V single phase, 3-wire system...

Two 30A single pole circuit breakers. That means you either have:

  • Breaker 'A' on L1 and breaker 'B' on L2 to arrive at a 240V, 30A circuit...
  • Both breakers 'A' and 'B' on L1 and your remaining leg landed on the neutral for a sketchy 120V, 60A circuit...
  • Or both breakers 'A' and 'B' on L1 and your remaining leg landed directly on the L2 bus for an even sketchier 240V, 60A circuit.

I'm betting that it is in fact a 30A circuit, not a 50 or a 60.
 
Ah. So it's a 30A 240V circuit. I was thinking about it wrong.

I get it now. Thank you.
 
You *could* put two 30 amp breakers in parallel on each leg.

Safety hazards. Gotta catch 'em all.
Right. What I did, though, was a SINGLE 30A 1 pole breaker on leg 1 and a SINGLE 30A 1 pole breaker on leg 2. I *THOUGHT* that would add up to 60 amperes. It don't. Oops. I know now, though.

Again, I don't mind taking risks, as long as I have a good understanding of what those risks are and what the benefit of taking said risk would be. And AGAIN again, I wouldn't cobble shit together like that for someone else - unless they NEEDED it and understand the risks involved and we had a discussion about that.
 








 
Back
Top