Why do AFCI breakers still have thermal trip elements?

Status
Not open for further replies.
W

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
An AFCI breaker has a microprocessor to track current vs time to identify arc signatures and trip. Given that, why does it still have a conventional thermal trip element? The microprocessor could do the same time averaging of current that the thermal element physically does, and implement the trip curve electronically. The trip curve would not be dependent on ambient temperature like the behavior of a thermal element is. (Unless the electronic current measurement method is temperature dependent, not sure how that happens internally).

And then, given that the electronics could implement any trip curve desired, it should be a simple matter to make every AFCI breaker a 100% rated breaker. My understanding of the 125% factor for continuous currents is that it is a reflection of a mismatch between the time vs current shape of the trip curve of a thermal element, and the time vs current shape of installed conductor damage curves (or some other limit on allowable short duration overloads). Namely that when the thermal element is tuned to provide the desired behavior at short time periods, it ends up overly sensitive for long time periods, and may trip at long durations for current under the breaker's rating. But if the trip curve is implemented electronically, it can be reshaped to no longer have this mismatch between short and long durations.

So am I missing something here?

Cheers, Wayne
 
Just a few thoughts on this. I believe the AFCI function can utilize relatively low precision and to a large extent nonlinear current measurement techniques to perform its target function. If the current transformer, analog interface circuitry, A/D converter, etc. in the AFCI also had to accurately measure the RMS current level including harmonics, high frequencies from drivers, etc. then it would likely be more costly than it is now.

Also, the thermal element directly emulates in a physical way the heating from RMS current flowing through the conductors that need to be protected.
It has virtually no limitation on the bandwidth of its response for this application, it's immune to interference, etc.
Perhaps more importantly, the thermal elements have a very well proven reliability while the solid state components used in AFCIs, GFICs, etc. are more susceptible to damage due to surges, etc. Of course this could be improved but at a cost that might not be acceptable for residential use.
I think just the fact that AFCI and GFCI functionality needs to have a test button indicates that the reliability of their components may not be sufficient for the overcurrent protection application.

I suspect the cost of the thermal element itself is a very small part of the total cost of an AFCI. I think UL certification testing would also have to be more extensive, because the electronic detection approach has many more degrees of freedom in which it might not meet the requirements over various test conditions.

Still, it's an interesting thought and an "all electronic" breaker for residential use will probably be developed and deployed eventually.
 
Last edited:
Right now the cost parity for microprocessor breakers is around 100-400 A in industrial 3 pole breakers. As in around 100-400 A it can go either way. Above that current usually microprocessor trip units are cheaper and below that they aren’t. As with anything there are of course exceptions both ways.

Even if it was microprocessor controlled they all have one fundamental big issue. It takes some time for the CPU to boot up. So most microprocessor trip units still have a hardware magnetic trip or pseudo-magnetic.
 
So am I missing something here?
Click to expand...

What you are missing is they have to have the thermal trip to actually be a breaker. All the other stuff is just fluff that we don't know if it even works! Since there is less "nuisance" trips with later versions, how do we know that there is even anything in there that works detecting arcs? It could be just a regular breaker painted up to look like an AFCI!
 
retirede said:
Which in and of itself would not be an issue (manufacturers would just sell more breakers if they had a higher fail rate). The rub comes in by the fact that the failure mode has to result in a tripped breaker - not one that will not trip.
Click to expand...

If the electronics are protecting the wire from over heating I'd much rather the failure mode be trip and lockout.

However electronics are prone to fail, where as mechanical is not likely to fail (to begin with) tripping reliably when called to do so. Simpler is better.
 
mbrooke said:
If the electronics are protecting the wire from over heating I'd much rather the failure mode be trip and lockout.

However electronics are prone to fail, where as mechanical is not likely to fail (to begin with) tripping reliably when called to do so. Simpler is better.
Click to expand...

I think we agree...
 
Given that we now have to have surge protective devices on dwelling units, partly because of the sensitivity of AFCIs to surges, I would not want to see an all electronic device...at least when the electronics fail the thermal magnetic part of the AFCI is still functional.
 
don_resqcapt19 said:
Given that we now have to have surge protective devices on dwelling units, partly because of the sensitivity of AFCIs to surges, I would not want to see an all electronic device...at least when the electronics fail the thermal magnetic part of the AFCI is still functional.
Click to expand...



The thing is SPDs can not deal with sustained surges like open neutrals and 34.5kv failing into 12.47kv lines.

My understanding is that once the MOVs start conducting the heat up. If they heat up enough a thermal fuse opens taking them out of the circuit.
 
Status
Not open for further replies.
Top