Wonder if anybody else has run into this problem?
I have a cutler hammer ch panel, when the elevator goes up or down the afi trips(keep in mind this afi has nothing to do with elevator). I talked to cutler hammer and the engineer says that the freq. drive motor is putting noise on the line and when 3 amps are present on the circuit, that noise will trip the circuit (I know this is true because when all lights in sewing room and office are off...no problems)anyway Cutler Hammer says to filter the motor to eliminate the problem. The elevator guy is trying to say he's never had this problem before and of course doesn't want to pay for the filter and labor, his solution is to put in a siemans breaker(yea right). Have any of you guys had problems with residential elevators tripping AFI's? Any thoughts about responsibility for additional $$ cost's appreciated too.

Is this in a single family dwelling? or A complex type setting?
On the same phase- or shared ckt?

1st I've heard of it but am watching closely.

it is a single family home, the afi is a separate ckt. but the elevator is causing it to trip with the noise online.

Are you really sure you don't just have a ground fault?

Could you try installing a TVSS on that panel supplying the AFCI branch circuit?
It's possible that some TVSS's may offer basic noise filtering capabilities.
They aren’t too expensive and if it works, you could both invoice it as an extra and come out smelling like a rose.

The TVSS clamps spikes at 300v or so (for a 120/240 service). I can't see that helping and if something does help it would be on that circuit, not the whole panel. There may be some kind of noise suppression that would work but I would think it would need to be be right at the motor or VFD/starter.
I have always been skeptical about engineers saying "noise". I know it can happen but that is also a catch all for "I don't have a clue". Do you have access to a scope and a current probe?.
I would still start by swapping in a GFCI as a test.
If that trips I would be looking for an intermittent ground fault.

I believe in noise. In a 3 phase system, it might be harmonics. It appears that those spikes, in a single phase system, will trip an AFI. Would a line side reactor on the drive filter the spikes?

If I understand the way the snake renders it's oil in an AFCI they are looking for current spikes, not voltage spikes, that is why I would start with a scope and a current probe.

Triplen harmonics are not really noise, they are very regular beat frequency waves within the base wave that get superimposed on it.

You can certainly start easter egging around throwing anti-noise hardware at the problem but at a certain point you may need to take a look at what is really happening.
If this is not one of the new breakers that has indicators of whether they tripped on an arc or a ground fault, I would eliminate the possibility of this being a ground fault problem first.

no, its not a ground fault Gfret.

here is there solution, this filter:

http://www.mouser.com/ProductDetail/Tyco-Electronics-Corcom/36FC10/?qs=sGAEpiMZZMsSuP4UBDzYO6M5j8iiHFgDkEQeiwHgMeo%3D

The FC filters provide filtering for frequency inverters and
variable speed motor drives. Designed for very noisy
applications, they attenuate conducted interference at low
and high frequency ranges and protect programmable
logic controllers from RF noise on the AC power line. These
filters perform best when installed on the line side of the
motor drive. The FC filters better prevent noise from
returning to the line, enabling equipment to meet strict
European regulations on RFI.
Suited for field wiring applications, these filters are ideal for
EMC troubleshooting and field refurbishing. The FC filters
feature side flanges for easy mounting and DIN type
terminals which eliminate live metal parts, allowing safer
and easier connection. The filters could also be sold as an
accessory to motor drives marketed into Europe.

I do have a gfi breaker in the panel, but also 2 other afi's which don't trip...but neither has 3 amps on it at time elevator is called. When lights are off when elevator is called it doesn't trip either...only after 3 amps.

The AFCI that trips is not in the circuit for the elevator?

I guess I missed that. Have you swapped it with one that doesn't trip? If it stays with that circuit, what is special about that circuit? Does the wiring run through the elevator shaft or something?
Does it run parallel or share a box with the elevator circuit? (some place where they could be capacitively coupled)

This article from Eaton on hybrid TVSS and noise supression basically describes what I was talking about.

TVSS/Noise Suppression

I recommend you put a power quality analyzer on this circuit and watch the response when the elevator operates. That will help you narrow down the problems. If it's just a matter of the elevator frequency drive creating so much ITHD that VTHD is tripping this breaker, you're going to have other problems that go far beyond a nuissance trip and really need to bring the THD down. You can't know any of this, though, until you get some instrumentation on there.