AFCI

hbiss said:
The high frequency noise from the inverter just rides on the 60Hz and has no effect on the body.
Click to expand...
If harmonics cause problems on 60Hz systems resulting in additional current, and it's resultant heating, then it is reasonable to assume there is additional leakage current, to ground, also. It is possible that zero sequence CTs used in GFCI devices need to saturate at a frequency level which means the harmonic leakage current does not cause nuisance tripping.
 
jim dungar said:
RCD is not safer than GFCI. They are fundamentally the same technology.
The big difference is between their pickup points and their time-current curves.

So the argument should be, why does the US remain at 6mA when the rest of the world doesn't?
Click to expand...
Yeah thats the question, or put another way why are we so obsessed with mA and not the time aspect?
My understanding is interference of fault current with the hearts T-phase occurs at around 400 ms of shock exposure.
If you study the time / current curve of UL 943 between 5ma and 30mA Per the UL formula as the resistance goes up and ma goes down to 5 the trip time stretches out to something like 7200ms or 7.2 seconds at 30mA its around 560ms or around 33 cycles of AC.
For thoes 33 cycles your hit with the full available fault current.
By contrast a Residual Current Device (RCD) take for example the spec used in Germany, they are required (DIN VDE 0100-410) to trip in 300ms when tested at its rated residual operating current, 30mA.
So at a 30mA test level RCD standards are safer for your heart than the GFCI UL 943,
Not to mention major US manufacturers like Schneider (the parent company to square D), as well as Siemens happen to be industry leaders in RCD's world wide, and if there was some UL way to allow a RCD it would take them less than 6 months to have one for the US market, mean while UL wants to re-invent the wheel and develop things that don't exist that we wait decades for.
We'll probably be running off Fusion energy before this is sorted out.
 
Last edited:
jim dungar said:
It is not RCD versus GFCI. The technology is all but identical


It is pickup point versus clearing time.
Until you get UL and the NEC, maybe even the CDC, away from GFCI type A performance levels it makes no difference what you call the device.
Click to expand...
Agreed, I am not a UL person but perhaps what the NEC can do is clarify what its goal of protection is in the 5-30mA gfci dead band anyways.
 
tortuga said:
Agreed, I am not a UL person but perhaps what the NEC can do is clarify what its goal of protection is in the 5-30mA gfci dead band anyways.
Click to expand...
The performance is not defined by the NEC.
It may have been CPSC or CDC that pushed the "5mA" value back in the late 60s and early 70s.
You would need to prove that they have been wrong for some 60 years.
Remember European electrical systems, and components, are different and applied differently than those in North America.
 
jim dungar said:
The performance is not defined by the NEC.
It may have been CPSC or CDC that pushed the "5mA" value back in the late 60s and early 70s.
You would need to prove that they have been wrong for some 60 years.
Click to expand...
Its not about right or wrong, more like we have better information now. But yeah we agree.
jim dungar said:
Remember European electrical systems, and components, are different and applied differently than those in North America.
Click to expand...
Their systems are not that much different
The RCD could be very cost effective with our 480/277 systems which resemble their 400/230
 
tortuga said:
Their systems are not that much different
The RCD could be very cost effective with our 480/277 systems which resemble their 400/230
Click to expand...
You are only looking at the source voltage. They have loop circuits which we don't have. This has forced them to have a single 'house' leakage protection while we are able to have one per circuit.

There is nothing in the NEC that would prevent you from using 30mA protection on systems greater than 150V to ground.
 
jim dungar said:
loop circuits
Click to expand...
Yeah valid point 'ring circuits' as they call them, though I think they were only in the UK, or countries that followed the British standards and are not in new installations.
If the newer research data showed that dead band between 5-30ma was meaningful their standards would reflect that, as we all know the EU is not a place known to be light on government standards, but what the reserch shows is the response time is more important.
There were a few members on here in years past from Nordic countries, Sweden or Norway i think, I have herd the old system described as what we call corner grounded 240 then branch circuits are line - line 230, but thats being phased out in favor of 400/230 wye, however in rural areas with no three phase its the same as here open delta, but they don't ground mid-phase, or simply end grounded 230.
The other major thing they had that drove the RCD requirement is a TT grounding system,
while everything new there is TN-S (floating neutral) or TN-C-S (MGN then floating neutral like we have here) a TT system you can have a ground rod at the pole, grounded neutral, then neutral floats and is not bonded to the buildings local grounding electrode system, also common in Japan if I am not mistaken. With A TT system you really need a RCD/GFCI.
I need to get over there and do some in person research on what the current standards are.
 
kwired said:
So they apparently will know when HF load is being supplied and when it is not. How about when there is mix of HF and non HF load? Or maybe it is only looking at frequency of any differential current on the protected conductors when it comes to how should it respond?
Click to expand...
It is not a high frequency load, but a load that creates high frequency leakage current. It looks at the frequency of the leakage current and changes the trip point based on the frequency. The must trip for 60 hertz remains 6mA, at 2500 hertz, it is 12.9mA and at 10,000 hertz, it is 35.82.
 
infinity said:
Since we're all complaining why do we require hard wired sump pumps to be GFCI protected? There two feet down in a hole where no one can touch them. :mad:
Click to expand...
They can require it all they want.
It's not going to happen in my basement on my plugged in sump pump.
Nor on my freezers.
 
don_resqcapt19 said:
It is not a high frequency load, but a load that creates high frequency leakage current. It looks at the frequency of the leakage current and changes the trip point based on the frequency. The must trip for 60 hertz remains 6mA, at 2500 hertz, it is 12.9mA and at 10,000
Click to expand...
I submit that the high frequency has no current or voltage. It's only noise or interference much like RF from a transmitter in the same location as the GFCI that causes them to trip.

I can see what they are doing- fudging the trip point upwards when it sees HF noise. Ain't fooling anybody.

-Hal
 
qcroanoke said:
They can require it all they want.
It's not going to happen in my basement on my plugged in sump pump.
Nor on my freezers.
Click to expand...
Fair enough, I am sure many think and do the same, I happen to have some old equipment and appliances I like to tinker with so personally I want know if something starts dumping more than 30mA down the equipment ground, I don't want to wait for a fault to trip a inverse time breaker so in my own home I have a few GFPE circuits for various things.
 
hbiss said:
I submit that the high frequency has no current or voltage. It's only noise or interference much like RF from a transmitter in the same location as the GFCI that causes them to trip.
Click to expand...
Do you have any scientific support for your opinion?
How have you ruled out harmonics?
 
I'm waiting to hear from the manufacturer how it's done and the basis of their research, but you can bet that it will be proprietary and we will be kept guessing just like with AFCIs.

By bet is on my theory because I don't have to BS anybody.

-Hal
 
hbiss said:
And a fine job they did with AFCIs. Remember the neon sign transformer and razor blade test setup for lamp cord?
Click to expand...
They did a good job with GFCIs, their parameters lasted some 50 years, before they needed tweaking.

You may continue to focus only on AFCIs, but that is not an area I have ever had to work with. The NEC and UL are more than the residential world.
 
You must log in or register to reply here.
Top