No GFI at RV refrigerator outlets?

The friendliest place on the web for anyone with an RV or an interest in RVing!
If you have answers, please help by responding to the unanswered posts.

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
Today, I did some experimentation at my outlet that goes to my Dometic RM 2652 RV refrigerator.

The only outlet in my entire RV without GFI protection is in the refrigerator compartment. Sounds reasonable, because people are not expected to use this outlet for anything else. But read on.

As Gary mentioned in a different thread, a common problem is for the 120 VAC heater element to short to the metal grounded tube and cause one BIG ground fault that will NORMALLLY trip GFI. This recently happened to me while I was running my generator. While there was a dead short (0.0 ohms) from the outer shell of the 120 VAC heater element to the tube, my generator would still run for 15 seconds and shut down with an error code that didn't make a lot of sense. Also covered in the section.

But the 20 amp circuit breaker, in the generator did not trip. Obvious reason. The AC path between the K2 (propane/AC switching relay) and J7 (to the heater element) on the refrigerator control board is just a thin trace. No way is it going to handle above 20 amps for long and it will melt way before the generator CB will trip.

However, if you're at an RV park with GFI, the park GFI should trip in time to save the control board. But what if you're running your generator? I doubt many switch to gas when they run their generator. I leave my refrigerator on "AUTO" at all times. I assume most others do too, who have a model similar to my refrigerator.

So then I thought about this. While GFI is usually for people protection and not for equipment protection, it will trip VERY fast at the first sign of a short in an RV when the heater element shorts to the metal tube it is in. This should save the refrigerator control board for those who don't want to re-solder a trace. And that board is rather difficult to solder on (you may ask me how I know). I ended up soldering a wire wrapped around J7 and the other end at the K2 relay.

These can easily be added, see my yellow thingy below. No need to change the outlet. The KW meter will be removed, that was just part of my testing, but the GFI protection stays in.

I switched back and fourth many times between propane and 120 VAC (includes generator). There was no false tripping at all.

-Don- Auburn, CAGFI.JPG
 

NY_Dutch

Well-known member
Joined
Nov 22, 2010
Posts
6,488
Location
Where our wheels take us!
GFCI's do not sense a short to ground. They only sense differences in the current flow in the hot and neutral lines, tripping if the difference is more than about 6 ma. The GFCI sense circuitry uses no ground connection. A short to ground causes a brief high current load that's equal on both legs until the associated fuse or breaker trips. Inaccessible residential kitchen refrigerator outlets don't require GFCI protection per the NEC.
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
GFCI's do not sense a short to ground. They only sense differences in the current flow in the hot and neutral lines,
I know how they work. And I tested it, it works.

BTW, a path to ground certainly can make the difference in detected voltages and current. If not, please explain how else the current can be different in a simple series circuit. In fact, how else can you lose the current?

Say 49.353 ma flows through your body to EARTH ground from the hot black. Now EXACTLY 49.353 MA is missing on the white, so it trips open the black wire towards you as that is detected by the differential amplifier that makes the GFI open the black hot towards you.

BTW, do you know how a GFI tester works? It has a button to put current on the green wire from the black wire. Now the white has less current than the black wire. The missing current is in the green wire with path back to the source that causes the black to go open.

Try to make such a tester with only two leads when you're above ground. Cannot be done. No way to lose current in a ground that does not exist.

BTW, you probably want to read this:

"Suppose now that one of the turns within the “C” phase winding were to accidently contact the generator’s metal frame, such as what might happen as a result of insulation damage. This ground fault will cause a third path for current in the faulted winding. and will now be imbalanced by an amount equal to the fault current :"

-Don- Auburn, CA
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
Inaccessible residential kitchen refrigerator outlets don't require GFCI protection per the NEC.
Of course not. Why would the NEC care if your refrigerator control board craps out? They don't. I was talking about EQUIPMENT protection as I said in my first post. A very unusual use for GFI. But it works and that is all that counts.

-Don-
 

John From Detroit

Well-known member
Joined
Apr 12, 2005
Posts
24,506
Location
Davison Michigan
In that generator story a GFCI likely would not have tripped either. To understand why you need to understand both GFCI and The Generator.

The reason for a GFCI is simple.. When on Shore power if you say pick up an electric Grill or a hand held grinder (Skill's answer to Dremel) and there is a short to the case YOU become part of the return path (By the way that's why I bought my first Dremel. the boss' SKILL had such a short and I got tired of tossing it across the room, yes, that's not what normally happens, (you clench normally) I'm different.

The dremel was plastic

But with the fridge even if the element shorts there are other, easier, paths to ground so you won't be in danger.

On the Generator. Ground has no meaning nor connection so it will not become part of the path.
 

Gary RV_Wizard

Site Team
Joined
Feb 2, 2005
Posts
72,693
Location
At our Silver Springs FL home
GFCI's do not sense a short to ground.
Semantics, I think. A GFCI will trip if there is a short to ground (i.e. a ground fault), so it detects a short to ground and acts on it. How it does that is a sidebar discussion.

I could also argue that a circuit breaker doesn't "sense" a current overload. It has no clue how many amps are flowing - all it does is react to excessive heat in the breaker and snap open the circuit. That doesn't make it any less effective as an over-current protection device.
 

Mark_K5LXP

Well-known member
Joined
Nov 17, 2018
Posts
889
Location
Albuquerque, NM
a circuit breaker doesn't "sense" a current overload. It has no clue how many amps are flowing - all it does is react to excessive heat in the breaker and snap open the circuit. That doesn't make it any less effective as an over-current protection device.

They absolutely react to current, but there are two current profiles to monitor - "overloads" and "faults". Most common residential AC circuit breakers are "T-M" - Thermal-Magnetic. The thermal portion senses long term overloads and the magnetic portion trips it instantly with short circuits. The fallacy that many believe is circuit breakers protect equipment, which is false.

Mark B.
Albuquerque, NM
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
Don - You simply do not understand how a GFI works. They are not a breaker. It won't trip with a dead short to ground.
Not this again. :)

If you have a dead short to the green, it will trip every time. How do you think a GFI tester works? The button puts current on the green, which is then missing from the white and then trips.

A short from black hot to ground or to the green wire on one side will have current. Lots of it. That same current is missing from the white so it trips. When it trips, the black side is open. Then there is no hot to destroy equipment or people.

BTW, try to invent a GFI tester for the two wire GFIs where there is no green wire . . .

A GFI will NOT trip on an overload between hot black and the white return. Not even if a billion amps if nothing is lost to ground. That is why circuit breakers are needed. But a short to ground (or green wire ) is a new path.

I am surprised how many people do NOT really understand how GFI works. But I see at least Gary does, as I expected he would.

-Don- Auburn, CA
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
They are not a breaker. It won't trip with a dead short to ground.

It will not trip when hot black is shorted to white. Say that short is 50 amps on a 20 amp CB. The CB opens the circuit. But the GFI does NOT trip with overloads UNLESS there is a ground path to cause current missing to the white wire.

It's the path that doesn't belong that trips the GFI, not the overload.

-Don- Auburn, CA
 

SeilerBird

Well-known member
Joined
Feb 25, 2012
Posts
15,570
Location
St Cloud Florida USA
BTW, try to invent a GFI tester for the two wire GFIs where there is no green wire . . .

It has already been invented. Every hair dryer and every GFI breaker has no ground wire and it has a test button. And here is a schematic and it is plain to see the test button does not use the ground.
 

Attachments

  • GFI.jpg
    GFI.jpg
    15.1 KB · Views: 15
Last edited:

JayArr

Well-known member
Joined
Jun 13, 2020
Posts
153
Location
Mission British Columbia Canada
Hey Don

That trace on the circuit board is thin for a reason, it's supposed to blow. The traces on circuit boards are often calibrated for maximum current flow. The engineers know how thick the copper cladding is and they can calculate width of a copper trace for a determined current rating.

If you've soldered some big fat wire on there you may have defeated a safety feature of the refrigerator.

JayArr
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
It has already been invented. Every hair dryer and every GFI breaker has no ground wire and it has a test button. And here is a schematic and it is plain to see the test button does not use the ground.
And here is a schematic and it is plain to see the test button does not use the ground.
The most important part of that schematic is missing. What does the other side of the test switch connect to?

Notice your own schematic is showing two ground connections. Where are those grounds?

BTW, I own such a two wire hair dryer and that has me curious when I see your schematic.

While I realize you can have GFI on only two wires, how do they test it without a ground of any type? I need to know where that switch goes to, to understand how it works.

I notice your schematic does show two grounds, but it doesn't show where they are really grounded.

-Don- Auburn, CA
 
Last edited:

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
That trace on the circuit board is thin for a reason, it's supposed to blow. The traces on circuit boards are often calibrated for maximum current flow. The engineers know how thick the copper cladding is and they can calculate width of a copper trace for a determined current rating.

If you've soldered some big fat wire on there you may have defeated a safety feature of the refrigerator.
I can believe all of that! But with my added GFI, I still have the safety regardless of how thick my wire is. And it saves the board!

IOW, my way is MUCH better, at least in this case.

-Don- Auburn, CA
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
I am not trying to make any point here with this message, but I was curious so I just did this test. . .

I just now did a test with an ohmmeter on a GFI tester. What it does, when I press the "test" button, is to put a 18KΩ between the hot black (shorter slot in an outlet) and the green (round prong).

The exact measurement on mine is 17.85KΩ.

So it's obvious how these work, by putting just a little current to cause a few ma of current to the green (grounded wire).

E over R or 120 VAC over 17,850Ω=6.723 ma for the trip test. Looks like these will not trip the very old GFI's that can be as high as 20 ma. But modern ones are 5 ma.

BTW, IMO, they should have stuck with the 20 ma. The world isn't perfect and there are too many false trips with 5 ma, IMO, on some equipment. It's a common issue with Zero motorcycles when plugged into some GFI's, for one example, so what many do is carry an adapter with the ground prong broke off, which defeats the entire purpose of the GFI. I carry one myself with my electric motorcycles, as they do sometimes trip GFI when charging from 120 VAC on some newer GFI outlets.

-Don- Auburn, CA
 
Last edited:

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
I just did some more testing. But this time on my GFI adapter. Mine uses a relay. It has to be reset every time after AC is lost. In one way, I like this. But in one way I don't and many will consider it a flaw in my idea.

What I like, is when I run my generator with my refrigerator on "auto" my added adapter will keep my refrigerator working on propane. It prevents the unnecessary switching when I run the generator for a few minutes, as I sometimes do to heat up something (but with my new battery, I will probably just use the inverter from now on for the MW oven).

But the downside is, at least with my GFI adapter, I need to press the reset button at an RV park when I connect up the AC. Not a big deal to me, but it could be to some.

A true GFI outlet won't do this, so I don't know why my GFI adapter does. And I don't know if new GFI adapters also work this way. I will order a new one just to find out! Mine is very old. Old design. I don't see the reason for a mechanical relay in mine, which has to be reset every time used, but it is what it is.

Even if new ones do not do this, I like my old. It's no big deal to press the button one time as I set up at an RV park. Takes only a few seconds.

I don't even want it to switch when I run my generator, so I see that as a plus.

The flaw in my testing yesterday was to unplug at the refrigerator after the added GFI adapter. I should have unplugged at the RV AC cord instead.

But I like the way it works, and will keep this one installed and I simply have to remember to take the added ten seconds or so to press a button when I am setting up at an RV park. And even if I forget, it's not a big deal to run on a refrigerator on propane while at an RV park for a few days.

But I doubt if new GFI adapters work the same way. Mine is probably 20 years old. I will order a few new ones and I will post here the results. But I will continue to use the old one in my RV. I would rather switch to AC myself.

I realize the same thing can be done by pressing the "gas only " button on my refrigerator when I run my generator. But my old GFI adapter will do that for me.

The flaw in my testing yesterday was unplugging the refrigerator after the GFI adapter, as I was right there, made it easier, but . . .

-Don- Auburn, CA
 

JayArr

Well-known member
Joined
Jun 13, 2020
Posts
153
Location
Mission British Columbia Canada
Hi Don

Have you tested the new setup by shorting the AC element to ground while on generator with the GFCI adapter installed?

What about the same test while on shore power?



As to the adapter the relay style is cheaper, relays are less than $1 but shunt-trip contacts are a lot more.

Shunt-trip contacts are more reliable, that's why they use them in circuit breakers and GFICs.

I'm curious, does the "adapter" you are using have UL approval or is it some Chinese thing that was never certified.
 

DonTom

Well-known member
Joined
Apr 21, 2005
Posts
3,179
Location
Auburn, CA or Reno, NV or Cold Springs Valley, NV
Have you tested the new setup by shorting the AC element to ground while on generator with the GFCI adapter installed?
Kinda, the test button on the GFI tester is the exact same path, just a lot more resistance (17.85KΩ). Also, when I had the shorted element (before it was replaced) it would trip my GFI in my garage here. That is what led me to all this.
What about the same test while on shore power?
See above, and that was also what I used mostly yesterday. But also the generator a couple of times.
I'm curious, does the "adapter" you are using have UL approval or is it some Chinese thing that was never certified.
It's a "UL listed" LR68658.

Made in Honduras.

-Don- Auburn, CA
 
Top Bottom