AFCI Nuisance tripping.

Status
Not open for further replies.

grasfulls

Senior Member
EATON Troubleshooting breakers

EATON Troubleshooting breakers

If there is something funky in the current waveform causing the tripping it will not matter where the device is in the circuit because this waveform is at every point in series with whatever is causing the funky to happen.
hhmmm

Well, Eaton sent me two breakers that they say may be purchased at the supply house. They will provide data as to why they trip....crud, it means I have to put them in and let them trip again.
 

grasfulls

Senior Member
what about a true online UPS

what about a true online UPS

If there is something funky in the current waveform causing the tripping it will not matter where the device is in the circuit because this waveform is at every point in series with whatever is causing the funky to happen.

Is it just current abnormalities that may cause a trip? Also, let's say you are powering an online UPS, will abnormalities on the Line side exist on the load side?
 

grasfulls

Senior Member
true online

true online

That will likely depend on the design of the UPS.

Well, Sort of knowing you now in here, I think you may know what I am trying got get at. I think (though eating crow MANY times has shown my thinking inaccurate) there may be devices that on the line side have different characteristics than the load side. So, if an abnormality causing the AFCI to trip only exists on the line side, well, in this case the electronics of the lighting module, then the AFCI tripping is just a nuisance and not real protection of a human being or equipment.

It was an apparent issue as soon as AFCI's were introduced into lighting systems, so they dropped the max dimming to 1000 watts. Now they claim we can fully load them (at least Lutron dimming modules with specific AFCI breakers). So they must have worked out what the weirdness was, at least with Lutron. So, if the modules exist in an enclosure that no normal householder is going to be into, why stick nuisance AFCI breakers on their line side? It would be neat if they could build the protection in one the load side is all I am saying.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
You are correct Grasfulls,
A simple series circuit is does not equate to a househouse circuit with several parallel connected devices and/or circuit branches.
The current in the entire circuit is likely not the same.

A circuit element that is easy to understand this for is one that includes an EMI filter. A filter with common mode filter elements.
The current is not the same on both sides of the device.

With these dimmer devices they may likely produce additional high frequency currents on the line side as well.

Over many AFCI problem threads I have suggested adding an EMI filter to the device in question causing the nuisance trip ( like treadmills).
Adding an EMI filter to an entire circuit may well eliminate nuisance tripping but the question then becomes will you then prevent the ability of the AFCI to also detect a true arc signature and thus render the AFCI install completely worthless?

One thing that I have not seen any manufactures make obvious is the fact that AFCI's may indeed be rendered worthless if a circuit has enough high frequency impedance between it and the load outlets it is intended to protect.

I can see where given time to evolve, the algorithms used to detect arcs, that manufacturers would get better at distinguishing between a true arc and a common dimmer current signature.
In particular a true arc signature will have some random elements associated with it whereas a dimmer signature will be constant ( given a constant dimmer brilliance setting).

When the dimmer setting is changed it would then take some finite amount of time for the algorithm to recognize the change and adapt. This finite amount of time is what makes the task so difficult. Attempting to interrupt the circuit fast enough to prevent a fire, yet delay long enough to recognize and prevent nuisance trips.

Do the Eaton breakers you received have the diagnostic LEDs on them or is there another method used to read out the failure diagnostic information?

No manufacturer of AFCI's should ever have been allowed to release a device without some sort of diagnostic readout. At bare minimum the ability to distinguish between a GF or an arc signature is must.
 

kwired

Electron manager
Location
NE Nebraska
No manufacturer of AFCI's should ever have been allowed to release a device without some sort of diagnostic readout. At bare minimum the ability to distinguish between a GF or an arc signature is must.
:thumbsup::thumbsup:

Even better would be to have a miniature oscilliscope that displays the current waveform and retains the waveform that caused it to trip, included with each device - then they could also increase the price by at least 500%:)
 

grasfulls

Senior Member
LED diagnostic

LED diagnostic

A simple series circuit is does not equate to a househouse circuit with several parallel connected devices and/or circuit branches.
...
Over many AFCI problem threads I have suggested adding an EMI filter to the device in question causing the nuisance trip ( like treadmills).
Does an EMI filter need to be designed for a specific range of filtering or are they generic?

One thing that I have not seen any manufactures make obvious is the fact that AFCI's may indeed be rendered worthless if a circuit has enough high frequency impedance between it and the load outlets it is intended to protect.
I would think if the manufacturer does not state what may NOT be connected to the circuit, then the rendering of worthless, if so, is their fault. One might argue the manufacturing of a device designed with the intent to render worthless does not need to be specifically listed, just the reference to any such device. But, I do not think a filter is being manufactured with that intent.

Do the Eaton breakers you received have the diagnostic LEDs on them or is there another method used to read out the failure diagnostic information?
It appears they will flash a pattern 30 times after following the procedure in its instructions. Interestingly, one is ZERO....I would think a failed diagnostic device may also render ZERO flashes, sheesh. Anyway, there are seven different flash patterns to watch for.

No manufacturer of AFCI's should ever have been allowed to release a device without some sort of diagnostic readout. At bare minimum the ability to distinguish between a GF or an arc signature is must.
Agreed
 

grasfulls

Senior Member
Scope

Scope

:thumbsup::thumbsup:

Even better would be to have a miniature oscilliscope that displays the current waveform and retains the waveform that caused it to trip, included with each device - then they could also increase the price by at least 500%:)

Though this must have been written with finger-in-cheek, I would imagine the only way to definitively diagnose what is causing a trip would be a device that does provide a scope type output ALONG with something monitoring events like power outage, a button press calling for a load at a specific percentage value, AND date/time stamps, etc. Oh, and along with a brain that can read scope (ok, interpret the waveforms), and this is not me.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
EMI filters come in many different configurations.
A quality manufacturer will list several frequency ranges and the attenuation figure in each range.


When I first heard about AFCI's many years back I thought they would have to be very intelligent devices. I envisioned them storing many "typical arc signature waveforms" and having a very fast processor that could do a near real time comparative analysis against these stored signatures. Of course such a device would be prohibitively expensive for general use.

I looked deep into the QO AFCI several years back in order to learn how they did things. It was nothing nearly so complex and as we have all learned relatively easily confused :happysad:
 

grasfulls

Senior Member
Ranges and Attenuation

Ranges and Attenuation

EMI filters come in many different configurations.
A quality manufacturer will list several frequency ranges and the attenuation figure in each range.


When I first heard about AFCI's many years back I thought they would have to be very intelligent devices. I envisioned them storing many "typical arc signature waveforms" and having a very fast processor that could do a near real time comparative analysis against these stored signatures. Of course such a device would be prohibitively expensive for general use.

I looked deep into the QO AFCI several years back in order to learn how they did things. It was nothing nearly so complex and as we have all learned relatively easily confused :happysad:

Well, I suppose we get back to a scope analysis to determine what type EMI is needed, so for me this is back to square one. I think the manufacturer of the item causing the trip or the breaker manufacturers should do testing to determine what will work best with items in general use, such as a Lutron lighting system module(or in my case a SAVANT system module). How can we (or ME - simple electrician) perform or afford to perform any comprehensive testing, especially if it involves simulating what causes the trip. I can turn a main breaker off and on, but I cannot take the utility down and bring it back up...oh...wait, I think I can take it down.
 

grasfulls

Senior Member
TDR's?

TDR's?

What about this?
Get a mechanically held lighting contactor or contactor(s) with enough poles to cover all the AFCI breakers that are tripping and attach the coil to a TDR that is a DELAY ON when Energized. Then, after the initial surge, say 4 or 5 seconds to be rather safe, it can engage the contactor? I have to think about how to disengage a mechanically held contactor upon power loss...
 

grasfulls

Senior Member
UPS with dry contacts

UPS with dry contacts

What about this?
Get a mechanically held lighting contactor or contactor(s) with enough poles to cover all the AFCI breakers that are tripping and attach the coil to a TDR that is a DELAY ON when Energized. Then, after the initial surge, say 4 or 5 seconds to be rather safe, it can engage the contactor? I have to think about how to disengage a mechanically held contactor upon power loss...

Ok... A UPS with dry contacts. On power out the dry contacts change state. I use one to send power to a relay through which 110 volts goes to the mechanically held contactor to disengage it. On power restore that relay disengages and a NO contact that now closes when the UPS is being fed by utility power energizes a TDR that closes the lighting contactor 4 seconds later. I can probably use the UPS to feed the lighting controller the whole time as well.

One bummer is those contactors really BANG when engaged or disengaged.
 
Status
Not open for further replies.
Top