I don't know why all this talk of AC * Mains systems, the OP has already declared an automotive application back in post #10!
Max.

Because it is essential to understand the existing tech and how noise has affected the existing solutions before proceeding. That was clearly explained in my post. Noise does not care if it is AC or DC, and even in DC systems, the current never "steady". There is always a time varying element and often, the time varying current is way larger than the DC current (driving a single phase inverter, etc.).

If they are trying to solve for existing standards that trip at 5mA but there is 2Arms of noise, it is not going to be easy to make a GF solution that works. I do not understand how the OP sufficiently explained what the spec is that is requiring GF for a DC application. As I see it, a cost effective design is not going to be possible. The OP needs to describe better what they are attempting. I was just trying to give them guidance and history to help their learning path. Which is almost 100% AC power systems.

 

GFCI is synonymous with AC. I've never heard of GFCI with a DC system. I doubt a GFCI outlet would be effective as a ground path problem with low voltage DC. May not even be effective with low voltage AC. When we hear of GFCI we think of 110 AC (or higher). But even I picked up on the fact that the TS was talking about automotive electrical systems and not AC.

[edit] After reviewing the last time the TS visited, he's been absent from this thread since last Thursday. Here it is now Tuesday. I'd guess either he got his answer or got the answer that has moved him away from this project. I DO know people go absent sometimes. I'd like to hear what the thinking is? As @MaxHeadRoom mentioned, the (and I paraphrase) antithesis of the neutral line is (in an automobile) the negative side of the circuit, the body, the frame. In that sense, there is no ground for the GFCI to sense current on. If there's a way to do it - I'd like to see it. Not that I can think of a need for it, just more so out of curiosity.
[end edit]

 

A slightly less typical description for AC wiring--

'Ground Fault' means that current is found to be flowing on the ground lead. This normally only occurs if power from the hot or neutral lead is bridged somehow to Ground (Indicating that power is taking an incorrect path from hot or neutral through something (like a human) to the outside case/shell of a device which is tied to ground (like the outside of a toaster or piece of machinery chassis). Ground is used as an alternate 'least resistance path' for current to follow if it escapes the engineered domain between hot and neutral.

The idea is that if too high a current is discovered flowing from hot or neutral to ground- someone could be caught, so a breaker is tripped to save their life. All devices that come into human use should have some means to disconnect from power, or cause a breaker to be thrown in the case of over-current for safety reasons for the user, and liability reasons for the vendor.

If you're creating a DC ECU, then use opto-isolation or some other method of isolation to separate your control circuit from MAINS.

NO NO NO!!!, a ground fault detector senses the difference between the current in the line wire (black) and the current in the neutral (white) wire. I have appliances with GFCI devices on the end of the cord with a 2 wire plug and a 2 wire cord. the ground wire plays no part in the detection process.
Would you really accept a device that depended on that third pin integrity?? I certainly would NOT. Besides, if the fault current were flowing through a person and not back through the green wire noting would be detected. While the instructions always tell you to connect the green wire ground, that is to provide an adequate ground connection to the ground pin of the outlet.

 

One consideration is the vehicls that have 120 volt AC outlets as part of the OEM electrical system. My Gand Caravan is a good example of that. What I have not discovered is if that 120 volts AC is ground referenced to the body ground of the car, or not. And looking in the owners manual is a joke, as it is on a CD without much help in how to find things. Hundreds of pages of assorted information without any particular arrangement.

 

One consideration is the vehicls that have 120 volt AC outlets as part of the OEM electrical system. My Gand Caravan is a good example of that. What I have not discovered is if that 120 volts AC is ground referenced to the body ground of the car, or not. And looking in the owners manual is a joke, as it is on a CD without much help in how to find things. Hundreds of pages of assorted information without any particular arrangement.

If the ground of the 120 volts in the vehicle is not referenced to the body, you have a problem. In fact, the "neutral" should be tied to it at one point where the power originates, same as in homes.

 

If the ground of the 120 volts in the vehicle is not referenced to the body, you have a problem. In fact, the "neutral" should be tied to it at one point where the power originates, same as in homes.

Please explain the benefit of having one side of a completely isolated power source connected to the vehicle frame. What benefit does that offer? Certainly connecting one side to the frame assures that the other side is a shock hazard relative to every conductive part of the vehicle. How can that be a benefit? Please explain your thinking.

 

NO NO NO!!!, a ground fault detector senses the difference between the current in the line wire (black) and the current in the neutral (white) wire.the ground wire plays no part in the detection process.

So if they are unbalanced, where is the current difference flowing?

 

So if they are unbalanced, where is the current difference flowing?

The unbalance portion of the current is flowing out where ever the fault connection has been made. THAT is what the whole system is intended to detect, which is that some of the current is leaving the loop beyond the GFCI device. It might be that it is flowing back through the appliance green-wire ground, but the hazard is that it would be flowing through the appliance user. And the main purpose of a GFCI device is to protect the users.

 

I am aware how it works, per OSHA https://www.osha.gov/SLTC/etools/construction/electrical_incidents/gfci.html
The time to detect a fault current is less than 1/40 of a second.and detects a current miss-match as little as 5 ma in order to trip the GFCI/RCD.
Max.

 

@MisterBill2 brings up a good point about devices such as my wife's hair dryer. When plugged in, they're ONLY connected to Line and Neutral. Since the hair dryer is often used in a bathroom there's the chance it could be dropped into a sink or tub. Perhaps when there is a path to ground other than the neutral, maybe that's when it trips. Many older homes do not have GFCI outlets in their bathrooms and kitchens. Having such a device on a hair dryer or maybe a curling iron might be looking for the difference Bill mentioned.

Me? What do I know?! I only know that a GFCI is required within either 3 or 6 feet of a source of water. Not sure which. Which brings up a different topic within this automotive thread - water piping. For years I've known ground to be either a galvanized pipe or copper pipe. But with the introduction of PEX, you can't just simply touch a ground wire to the plastic pipe. So what if we put a short length of copper in the WATER circuit? I doubt that would suffice. Even though the water can act as a conductor, does it conduct well enough to function as a ground path? My thoughts, though I've not looked this up, is that it's not sufficient or safe enough. Maybe it is. I'll have to wait to hear.

But this thread isn't about proper application of a GFCI, it's about putting such on a car. I just don't see any other path back other than the chassis of the vehicle. And that's already at ground, a.k.a. negative potential.

 

@MisterBill2 brings up a good point about devices such as my wife's hair dryer. When plugged in, they're ONLY connected to Line and Neutral. Since the hair dryer is often used in a bathroom there's the chance it could be dropped into a sink or tub. Perhaps when there is a path to ground other than the neutral, maybe that's when it trips. Many older homes do not have GFCI outlets in their bathrooms and kitchens. Having such a device on a hair dryer or maybe a curling iron might be looking for the difference Bill mentioned.

But this thread isn't about proper application of a GFCI, it's about putting such on a car. I just don't see any other path back other than the chassis of the vehicle. And that's already at ground, a.k.a. negative potential.

All outside outlets should be CFGI versions, where many portable power tools are double insulated now, (two pin) But if used outdoors in a wet environment, it is possible for the tool to become immersed in water etc.
In this case the GFCI will still operate off, due to the moisture conduction to ground.
Only 5ma current required to operate!
I agree that in an automotive area, the frame is usually the return conductor.
Max.

 

Simply put they're mostly used around sinks and where water runs.
Brzrkr

 

Curiosity got the better of me. Here is what I found regarding those boxes at the end of the hair dryer:

[edit] those boxes are called "Appliance Leakage Current Interrupters" or "ALCI's".

 

Curiosity got the better of me. Here is what I found regarding those boxes at the end of the hair dryer:

[edit] those boxes are called "Appliance Leakage Current Interrupters" or "ALCI's".

That ALCI is a totally correct name because there is no ground within the appliance for there to be any leakage to. So it would have to be to the outside of the appliance.

 

And still no reply to my question as to any benefit of adding a ground-to-the-frame connection to that isolated power source in the car. Most cars do not have water plumbing for the passengers, nor sinks and drains. AND none of the cars while driving provide connections to the AC mains common.
So I am still waiting for that explanation as to the mechanism of any benefit of grounding an AC supply in a car.

 

still no reply to my question as to any benefit of adding a ground-to-the-frame connection to that isolated power source in the car.

Probably because there isn't an answer.

 

If you have a portable generator, you can:
a) use it isolated (neutral-ground) bonded at the generator.
b) Provide an external ground - ground rod

Now if the generator is connected to a house the ground-neutral bond should be broken and the house neutral-ground bond should be used.

If the car is the generator, I think the only type of equipment that ight care if ground is connected to the body is a transmitter or RF receiver.

 

If you have a portable generator, you can:
a) use it isolated (neutral-ground) bonded at the generator.
b) Provide an external ground - ground rod

Now if the generator is connected to a house the ground-neutral bond should be broken and the house neutral-ground bond should be used.

If the car is the generator, I think the only type of equipment that ight care if ground is connected to the body is a transmitter or RF receiver.

The TS wants to put GFCI on an automobile. So far we've gotten no further information as to how or why.

 

Probably because there isn't an answer.

I was hoping that while trying to explain it the reality would dawn on the person.

 

Please explain the benefit of having one side of a completely isolated power source connected to the vehicle frame. What benefit does that offer? Certainly connecting one side to the frame assures that the other side is a shock hazard relative to every conductive part of the vehicle. How can that be a benefit? Please explain your thinking.

Same reasons that we do this in our homes. First, a fault from hot to ground will trip a breaker. Otherwise, you would never know until a second fault appears. Second, although less likely in a vehicle, issues from lightening to other static charging of the system, if floating, can result in a shock or static damage to electronics or ignite gas or gasoline in the case of cars. Actually, the European system where the center tap of the 240 volts is tied to ground and no current-carrying neutral is used, is better, and possibly the center of the 120 volts in a 120 volt only vehicle system could be grounded instead.