Defective Siemens AFCI's

[Bob Knauff]

In the last two weeks I have come across two service panels having Siemens type AFCI's like those in the photo. In the first box, 2 of the 4 breakers would not trip when the test button was pushed and one tripped but would not re-set. One was fine. In the second box, 2 of 4 would not trip when the button was pushed. 2 were fine. I tried rapidly turning the bad ones off and on a few times and then re-testing with no difference. A search of the Internet turned up nothing. I have to believe there is some sort of recall or service bulletin about this issue. The failure rate is just too high here.

Anyone?

Similar Threads:

• Hot AFCI'S?
• OPEN GROUND Defective or Maintenance
• Any recent recall on defective shingles?
• Possible defective shingles
• AFCI's in condos

[Aaron Miller]

Siemens - AFCI Test Button Failure

[H.G. Watson, Sr.]

So the big question is....was the ambient temperture BELOW 32F (in August?? during an inspection?? presume panel wasn't installed in a refrigerator!) I would call Siemens with the date codes and panel info.

[Bob Knauff]

Thanks for the link Aaron. It at least provides a path to pursue.

You're correct H.G. not much danger of getting to or below zero here. The ambient temp. at the time of testing was near 100 degrees outside. Perhaps it's an issue of being too hot?!

[H.G. Watson, Sr.]

Assuming that the "problem" is a "defective" or "damaged" AFCI is not logical.

You haven't pictured or indicated if the Circuit Breaker/branch feeder type AFCIs were installed to the panel correctly, if the panel is a service panel, or if not service equipment if the N & G are isolated.

When the test doesn't occur or the combo/branch feeder AFCI will not reset following same is an indication of a problem with the installation, damage, something causing leakage, etc. and as per instructions, time to call in a qualified electrician to diagnose.

Occupants/Owners rarely (if ever) actually "TEST" combination breakers or gfci receptacles, etc.

Did you see ANY "night light" type (illuninated) switches installed anywhere in these homes? Any retrofit ceiling fan/light controls? Retrofit dimmers or controllers for retrofit/upgraded lighting systems? X-10 or remote controller interfaces?

My first suspicion in both cases and your area, (assuming AFCIs appear properly installed in panel) if sellers/occupants claim new problem, AND sellers/occupants claim they HAVE tested and reset without problem previously AND claim no changes to electrical system inside and no new devices in place, would be to check if both properties have a POCO controlled peak time controller for the HP or AC - or other "smart metering" or DEMAND/PEAK/controller system retrofit (WHs, etc.). Next any remote controll systems interfaces, next alarm systems, next wireless or semi-wireless routers (those that transmit signals via electrical system not hardwired communications cable).

Usually unplug everything from receptacles, remove any CFLs, LEDs, etc., turn off all hardwired switches/controllers, just like any other CB or combo CB (GFCI, GFE, AFCI, etc.) that won't reset or test cycle) is the HOs first step prior to calling elect. and disabling/removing/turning off/unpluging all RF producing devices anywhere in the home.

That vintage most always defects in installation system or problematic devices. I wonder if the homes in question may have been using communications equipment that singals on the branch circuits (certain semi-wireless router systems) or some other "wireless" controll system (X-10, "wireless" ceiling fan remotes, "wireless" socket type receiver controll switches for lamp holders esp. containing CFLs, timer-containing appliances, "remote control" cord sets, "power supplies" inverters, switching, battery backups, etc.) or any combination controllers for ceiling fan/lights transmitting signal on switch loop rather than separately wired lighting loop vs. fan supply loop. All are pretty common.

First, its an older "branch feeder type" not a modern type. Appears to be of the second generation of same (indicator) yet indicator does not appear "lit" in your photos.

Second, the symptoms you describe are indicative of a wiring error, a system problem, or leakage from something powered producing either RF or leakage. Third, you offer no pictures or details/specifications/date code of the panel itself, nor the age of home.

Forth, the symptoms you describe indicate need for a qualified party examine and diagnose.Fifth, you do not indicate what is in place and connected to the branch circuits in question. Leakage, LG/LN MOV, RF, crossed Ns, N-G, damage, etc. lighting systems, etc.Notorious issues with different generations of ceiling fan & controllers, dimmers, electronics, switching power supplies, defective surge suppressor strips, etc. transformer based lighting/systems,

Early installations and later unqualified modifications possible. Filters or removal of certain devices that might be on the circuit that leak.With photos, details on the breakers, the panel itself, etc. contacting seimens, would likely result in referral to a qualified party as below.Refer to qualifed electrician to diagnose & remediate - siemens offers specialized diagnostic tool(s) for the qualified person to diagnose/test their AFCIs in the field.

Remember branch-type AFCIs have been phased out for good reason, and that one has to start at the beginning for tracking down an issue. That begins with unplugging all devices at receptacles, etc.

It's worth noting that the test button on an AFCI does not simply force the mechanical internal switch of the AFCI to trip. Rather, the test button on an AFCI tests the arc fault detection circuitry to be sure that it is working properly, that it will respond to an arc fault, and that the circuitry will in turn cause the mechanical internal switch to open.

Following links you might find interesting:

(copy also attached for the first as its small enough, and when I the link here it displays "page not found?!?):

http://www.sea.siemens.com/us/intern...S0001-0903.pdf


Next is a scanned file of a Seimens booklet document (too big to attach) entitled "A guide to Electrical Interference and AFCIs" and I couldn't find a direct link on Siemens' site that would work when I posted it here, so I'm refering to a copiy on the Oregon (state gov bcd) site link here:

http://www.bcd.oregon.gov/boards/ele...JohnPowell.pdf

The Literature that came with that Branch Feeder type AFCI should include specific information for steps to take when the circumstances you described occur in the "monthly testing" section of the instructions.

HTH.

[Garry Blankenship]

The arc fault technology is improving with time, but long from perfected. I'm not a fan of more electronics to safe gaurd more electronics. I believe one has little choice but to defer any performance concerns to the appropriate licensed professional. A pro should at least be on top of what mfg. or mfg. production series may have some known defects. There are too many potential problems for a Home Inspector to attempt any specific identification, as pointed out in H.G.s post. Even non-electronic circuit breakers will trip from electronic circuitry induced problems like poor power factor and others already mentioned.

[H.G. Watson, Sr.]

Bob Knauf,

Follow-up question to your latest post: Are these installed in exterior/ panels? Combo Meter/panels at lateral risers?

Other thoughts pondered...

Was there a "downstream" dead-front, or combination GFCI present on the non-testing and/or non-resetting AFCI branch feeder type circuit breaker? What "vintage" was same? was it in a reset postion? did you check to see if it too was tripped or "locked out"?

Were these circuits wired with 12/2/2 NMB? (two neutrals, one for each circuit, note you have 20A AFCI circuit breaker, branch feeder type,pictured)? If so, did you re-trip or open the companion circuit and then was the non-reseting breaker able to reset & test with the other circuit open?

Any photo-cell activated or motion-detector lighting/fan controls?

See link below to the NEMA 2011 white paper, followed by some contractors in oregon documenting some "issues" in application & the feed-back they received from factory reps - specifically pdf doc page 15 of 25 from Bear Elec.

http://www.cbs.state.or.us/bcd/commi...r_Electric.pdf

[Bob Knauff]

A lot to ponder HG, thanks. I'll respond with details as time permits. Suddenly I am very busy here. Good deal!

[Lon Henderson]

H. G., that is some great info, but it seems to me, that all an HI should do is just write up something like "AFCI breakers failed to respond properly to the test button and further evaluation by a licensed electrician is recommended." And move on.........

But I am wondering how many of you guys test AFCI breakers in occupied houses?

[Richard D. Fornataro]

As an electrical inspector it has been my experience more often than not that the AFCI circuit breaker is NOT the fault. (Pardon the pun.)

Many times, electricians, when discovering that the circuit breaker will not "hold" on the existing circuit, they simply do not connect the neutral or cut the neutral conductor so that there is no "stripped" conductor making electrical contact with the terminal.

Without a neutral reference, the test button does not function and neither does the AFCI for the purpose intended.

NYS has a requirement that all habitable rooms have at least two receptacles. (Except a bathroom.)

In the past, when cited for such deficiency in the number of receptacles, electricians were allowed to simply extend the circuit by use of Wiremold, etc. to facilitate the installation of additional receptacles.

If called to such a situation today, they are required to AFCI protect any branch circuit wiring that is extended or modified per NYS.

2011 NEC will require AFCI protection when merely replacing a receptacle.

When they encounter a circuit that will not "hold" an AFCI in the operating position due to faulty BX wiring, reverse polarity, shared neutral, etc., some less confident or legitimate contractors will result to the trickery previously reiterated to attempt to get a final inspection for such an installation.

I'm not indicating that this is the case here but I am putting it out there for consideration since I have encountered such in the course of my daily endeavor.

[Robert Mattison]

Good post, with useful information, Richard D. F.

[H.G. Watson, Sr.]

You do realize you need to disconnect all loads to reset and/or test if it won't otherwise? High inrush such as a bank of lights/lamps, etc. can be enough to cause the difficulties described with testing or resetting those branch-type AFCIs - as they DO have a GFPE feature (GFI @ 30 mA).

[Ken Harbeck]

But I am wondering how many of you guys test AFCI breakers in occupied houses?

We do not. Too many complaints from HO's about reset alarm clocks. Weak I know, but what'ya going to do? We are a guest in their home during the inspection.

[Bob Knauff]

First, HG stated that "Assuming that the "problem" is a "defective" or "damaged" AFCI is not logical". I disagree. Consider the Square D AFCI recall of a few years back.

I do however find it very odd that that many Siemens AFCI's can be "defective" and no one has issued some sort of recall notice. That's what prompted my research and post here.

In an attempt to answer some of HG's earlier questions and concerns regarding the AFCI issues:

The AFCI's are mounted in exterior combo Meter/Service Panels; all appeared to be installed properly with grounded conductor attached to device and not buss bar; no lighted switches but not sure if homes had wall mounted fan controls; both homes vacant; AFCI indicators did not light (if they have them); both homes just under 10 years old; no GFCI's in AFCI circuits; cables typical 12/2 NM w/ground. That's all I can add.

For safety, I report recommendation that client immediately have a licensed electrician investigate further as a RR issue, of course.

As has been brought up, I agree with the "never test AFCI's in occupied homes". Too big a chance that a computer or timer may be disrupted and the owner get very unhappy.

Also, I have heard that many inspectors do not test these devices at all. I found that curious since they are as important as GFCI's, IMHO and those are tested by them regularly.

I also agree with the idea that if found to be not operating properly the issue, no matter what it is, should be directed to the appropriately licensed professional. However I also believe that one needs to know how things work and try to understand WHY and WHAT CAUSES them to fail in order to be able to better understand and communicate the issue to clients or even contractors or perhaps even to defend ones position in making the call.

[Bob Knauff]

An AFCI update.

It was difficult to photograph but the hot conductors from the cable bundle are attached to the AFCIs; the neutrals from the bundle are pigtailed and attached to the panel neutral bus bar (2 rear wire nuts); the pigtails from the AFCI are capped with a wire nut (2 forward wire nuts) and left dangling.

The seller complained that "It must be OK because it was like that when the house was new about 3 years ago so it HAD to have been inspected and passed by the city!". They questioned my recommendation to correct the situation.

[Lon Henderson]

Been there, heard that.........

It does make you wonder how they ever got wired up that way?

Ran into Siemens AFCIs in three panels on a tri-plex two days ago. I didn't test the breakers on the two occupied units. P****oed tenants do not make for smooth inspections.

But the alert on Siemens is that below zero degrees is when they have problems responding to the test button. I explained all this to the buyer who asked if he had to pay me to come back out on some cold evening this January to check them.

I said "Yep"

[H.G. Watson, Sr.]

An AFCI update.

It was difficult to photograph but the hot conductors from the cable bundle are attached to the AFCIs; the neutrals from the bundle are pigtailed and attached to the panel neutral bus bar (2 rear wire nuts); the pigtails from the AFCI are capped with a wire nut (2 forward wire nuts) and left dangling.

The seller complained that "It must be OK because it was like that when the house was new about 3 years ago so it HAD to have been inspected and passed by the city!". They questioned my recommendation to correct the situation.

This of course is an incorrect installation, and would not allow an AFCI breaker to successfully complete the test and reset cycle, and would furthermore defeat the correct functioning of the device.

[Bob Knauff]

In my previous post the reader was given the benefit of the doubt that they knew how to properly install an AFCI device and the ramifications if it was not done properly. Perhaps there are still those out there who did not understand the problem so, in that case, the attached page from a Square D AFCI install manual will help clarify. And, as HG points out, the AFCI protection is not present if the device is not wired correctly even if it would act as a standard breaker anyway.