I have no reason to have any faith at all in AFCI technology, as I have not yet seen any empirical evidence to show that they work in the manner that they are supposed to.
However, that doesn't take these guys off the hook. I would like to know what is causing them to trip. Any theories ?
EMF from the inverter or slightly abnormal sign wave out put. My guess. There is no way a $35 breaker is sophisticated enough or well thought out enough to differentiate between a normal arc, abnormal arc, EMI interference and distorted sign wave. My understanding is these breakers detect arcs by simply looking for current ripples, and there is a lot out of the abnormal arc fault that can cause that.
If AFCI breakers really could detect abnormal faults well and not trip on anything outside of that they would literally have the computing power of an Iphone at the very least or a laptop.
I am being literal about that. MV Recloser and MV/HV substation breakers equipped with arc fault logic relays have that type of computing power. Most have on board oscilloscopes/sine wave analyzers. And considering that these are applications where if an arc signature is detected near 100% of the time its a downed wire or tree branch contact. No customer will have a brush motor or flat screen TV that will load a 13.8kv 1,200amp circuit half way or a 60MW sub transmission line. Yet despite that that the added logic to prevent nuisance tripping and make fault finding easy.
AFCIs are a full blown gimmick at the 120 volt level.