diy solar

diy solar

The "Meg" Build

That, and connect ground to a rod driven into the earth.
(depending on moisture content. The first U.S. soldier death in Iraq was electrocution while taking a shower.)
Not so fun fact, lol.

What about when boondocking in the middle of nowhere? Im not gonna drive a ground rod everywhere i go, but at home base, I might.
 
Try to track down the problem and fix it.
Use GFCI where possible.
Carry a voltmeter where ever you go, and test "step potential" before walking or climbing down from RV.
Jump with both feet when getting on and off steps. Shuffle with small steps across ground without picking up your feet.
 
For the hell of it, I connected back to the grid with 50a supply, and low and behold, only 216mV between rv and earth, and no shocking when I touch it.
This is not a surprise, The cord to shore power does the same thing as the screwdriver that was pounded into the ground and wired to the RV. The symptom went away, but the underlying problem still exists.

If I have followed correctly, the problem goes away when the PV is disconnected at the inverter but it does not go away when the PV is disconnected at the pannels.
Is that correct? If so that implies there is something shorting one of the PV lines to ground someplace along the run. However, there is still not an explanation of where the stray voltage is coming from.

BTW: Did you try to see if there is a DC voltage between Earth and the RV at the same time you are measuring the AC voltage? (Even though it is registering as an AC voltage, it could be a pulsed DC)
 
So does this mean I need to bond neutral/ground when not connected to grid power?
You should definitely make sure there is an N-G ground when not on shore power. However, it is not at all clear to me that the lack of the bond is part of the problem you are having.
 
If I have followed correctly, the problem goes away when the PV is disconnected at the inverter but it does not go away when the PV is disconnected at the pannels.
Is that correct? If so that implies there is something shorting one of the PV lines to ground someplace along the run. However, there is still not an explanation of where the stray voltage is coming from.
You are correct. But upon inspecting the wires, I have found nothing obviously wrong with them
BTW: Did you try to see if there is a DC voltage between Earth and the RV at the same time you are measuring the AC voltage? (Even though it is registering as an AC voltage, it could be a pulsed DC)
I did try this and forgot to mention it, when set to DC, the DMM shows 0.000V DC.
 
You are correct. But upon inspecting the wires, I have found nothing obviously wrong with them
It seems unlikely, but it could be a capacitive coupling between the wires and earth.


I did try this and forgot to mention it, when set to DC, the DMM shows 0.000V DC.
Interesting.

Is there a 'main breaker' in the AC breaker box? If so, turn it off while the inverter is still on. If the problem persists, it is an indication the problem is before the breaker box.
 
Last edited:
Not so fun fact, lol.

What about when boondocking in the middle of nowhere? Im not gonna drive a ground rod everywhere i go, but at home base, I might.

Definitely need NG bond when disconnected from shore. My Victron's automatically do this for me; not sure if your setup has any equipment that can be configured to do this when it's operating as an island.

Physical earth ground isn't necessary in a mobile rig.
 
but SOMETHING is earth grounding a hot lead, so chassis is carrying AC relative to earth.

Yes, just saying that a standalone RV when boondocking isn't necessary. Maybe I misunderstood but I thought the problem was being injected from something outside the RV?
 
Yes, just saying that a standalone RV when boondocking isn't necessary. Maybe I misunderstood but I thought the problem was being injected from something outside the RV?

I think problem occurs when disconnected from all external AC sources. But something connected to a shipping container? PV wires to a PV array, nothing else?

You guys nailed it, I flipped the PV disconnect switch and there is now 0V from the dirt to aluminum. This is even with absolutely no sun out too.

It should also be noted, that leaving the PV array that is screwed to the roof of the trailer on, has no effect. It only has voltage potential when the panels on the shipping container are connected.

So PV panels on shipping container, their frame has path to ground. Seems inverter is driving AC onto the PV wires leading to shed, and those panels leak that AC to their frame?

Maybe unplug all MC cables of that array and test for DC from each +/- lead to frame.
Panels have capacitance from leads to frame ... check PV leads for AC. Both MPPT operation and not.

Maybe just need to run a ground wire from shipping container to trailer, along with PV leads. I think that's something missing.
I always expected that to be a DC shock hazard in case of damaged panel.
Didn't expect AC leakage (but that would happen with my transformerless Sunny Boy if used on 3-phase.)
 
So PV panels on shipping container, their frame has path to ground. Seems inverter is driving AC onto the PV wires leading to shed, and those panels leak that AC to their frame?
But upon inspecting the wires, I have found nothing obviously wrong with them
I disconnect the PV wires from the Array (200ft away) and still have 54V AC between rv and earth.

The OP says that the problem occurs even if the PV wires are disconnected at the shipping container, but not when disconnected at the array.
That implies the wires are somehow involved, but the panels are not.

The OP also says the is no obvious problem with the wires. If the wires have no short to ground, the only thing I can think of is capacitive coupling to earth.

Other interesting points:
  • When hooked to shore power the symptoms go away, but that is probably because shore power ties the AC ground to earth Ground.
  • The OP reported that when the trailer has a wire from chassis to ground the current on the connection was very low (near zero). (If it is a capacitive coupling between earth and the PV wire, I would expect the current to be low.)
  • The OP reported that the voltage between the trailer and earth is AC and there was no DC reading.

I am a visual person so I drew up my understanding of the layout.
1635149380726.png
The only thing that came to mind in looking at the diagram is that there are probably some ground loops....but I don't think that could cause the issue.
 
OK.... this is a long shot but.....

With the way the xfer switch is put in, is the neutral of the AC input tied to the neutral of the AC output? I seem to remember that this is a no-no on some inverters.
 
no-no on some inverters.
My small brainstorming here running with a 120V “leak” idea. Maybe filterguy or hedges can make sense of it…

If it’s this ^^^ ATS issue we can call that “equipment related” and is fixable from my (not an) electrician’s perspective, or rewiring properly whatever that is will fix it.

From a engineering perspective (I’m not an EE either) if it’s not ATS/inverter “equipment related” I’m leaning on earlier discussions in this thread.
Two possibilities exist that would suggest
- A very small hot/black “leak” to something grounded (frame, bare 120 ‘ground,’ a box, or equipment chassis) OR
- There’s a bad/poor neutral/white connection somewhere and it’s finding the lowest potential to make the circuit. In this case it would be ‘nothing’ until one is on earth and touches the camper- completed circuit- and since the circuit’s only apparent path to earth is through the panel Neg(-) to the container….

…would suggest that disconnecting the ATS and wire-nutting inverter out to breaker panel feed so that ATS is eliminated as a possibility (true island, no switch) and retesting would prove out if there’s a facility-based leak or it’s the ATS issue suggested. (I can see how one might test using a meter without disconnecting but AC and tiny ‘leaks’ can sometimes lie to you, it seems removing a possible suspect is the most efficient means of obtaining proof.)

Wildcard: filterguy’s review of disconnecting at panels VS at facility equipment debunks this train of thought
OP says that the problem occurs even if the PV wires are disconnected at the shipping container, but not when disconnected at the array.
That implies the wires are somehow involved, but the panels are not.
By array…meaning RV mounted panels I assumed. Need clarification from OP if this was an A/B one at a time or an either/or with the alternate panels still connected maybe?

Does this make any sense? Or am I confusing things?
 
I think problem occurs when disconnected from all external AC sources. But something connected to a shipping container? PV wires to a PV array, nothing else?



So PV panels on shipping container, their frame has path to ground. Seems inverter is driving AC onto the PV wires leading to shed, and those panels leak that AC to their frame?

Maybe unplug all MC cables of that array and test for DC from each +/- lead to frame.
Panels have capacitance from leads to frame ... check PV leads for AC. Both MPPT operation and not.

Maybe just need to run a ground wire from shipping container to trailer, along with PV leads. I think that's something missing.
I always expected that to be a DC shock hazard in case of damaged panel.
Didn't expect AC leakage (but that would happen with my transformerless Sunny Boy if used on 3-phase.)
FilterGuy is correct as I disconnected the PV wires at the array, and I was still experiencing the same voltage potential between earth and RV.

I will add. When doing some random testing, I had disconnected all wires DIRECTLY at the disconnect switch, and while reconnecting the wires one by one, (there is 4 wires going into the Disconnect switch), I saw 27v with the first wire, then 38v with the 2nd wire, then 49v with the third, and 62v with the last. Very strange that each wire I connected added an incremental amount of voltage.

The OP says that the problem occurs even if the PV wires are disconnected at the shipping container, but not when disconnected at the array.
That implies the wires are somehow involved, but the panels are not.

The OP also says the is no obvious problem with the wires. If the wires have no short to ground, the only thing I can think of is capacitive coupling to earth.

Other interesting points:
  • When hooked to shore power the symptoms go away, but that is probably because shore power ties the AC ground to earth Ground.
  • The OP reported that when the trailer has a wire from chassis to ground the current on the connection was very low (near zero). (If it is a capacitive coupling between earth and the PV wire, I would expect the current to be low.)
  • The OP reported that the voltage between the trailer and earth is AC and there was no DC reading.

I am a visual person so I drew up my understanding of the layout.
View attachment 70079
The only thing that came to mind in looking at the diagram is that there are probably some ground loops....but I don't think that could cause the issue.
Everything in this post is correct. Nice picture too! I'm just not capable of doing something like that! :ROFLMAO:

OK.... this is a long shot but.....

With the way the xfer switch is put in, is the neutral of the AC input tied to the neutral of the AC output? I seem to remember that this is a no-no on some inverters.
It shouldn't be. I can check this by removing the AC Input directly at the inverters, this should pinpoint if anywhere on the input side is causing problems. While I am in there, I will check and see if we have a N-G bond inside the inverter. I vaguely remember playing with my DMM when the inverter was on the table and found continuity between the N and G, but it has been atleast 6 months since then. As I tell my boss when he asks me to remember something from yesterday "Yeah, I've slept since then, so...."

By array…meaning RV mounted panels I assumed. Need clarification from OP if this was an A/B one at a time or an either/or with the alternate panels still connected maybe?

Does this make any sense? Or am I confusing things?
By array, he means disconnecting the PV wires at the array on the container, (see his pic above with the 200ft between the two).

The array that is on the roof of my RV has no change to the voltage potential between earth and ground.

I thank you guys for trying to help me with all of this!
 
The OP says that the problem occurs even if the PV wires are disconnected at the shipping container, but not when disconnected at the array.
That implies the wires are somehow involved, but the panels are not.
By array, he means disconnecting the PV wires at the array on the container, (see his pic above with the 200ft between the two).
Then I’m confused at the difference between “disconnected at the shipping container” VS “disconnected at the array”
@FilterGuy?
The array that is on the roof of my RV has no change to the voltage potential between earth and ground
Q?: confirm:
-With the container connected and RV array connected you get ACV RV-to-earth y/n
-With the container disconnected and RV array connected you get zero ACV RV-to-earth y/n
-With the container connected and RV array not connected you get ACV RV-to-earth y/n
-With the container disconnected and RV array connected, if you disconnect the container wires in the RV at SCC input do you have ACV RV-to-earth?

The point is discovering if it’s a screw or chafing somewhere is minutely feeding true AC to bare/earth or if it’s actually equipment like filterguy theorized. No matter what there should be no AC in your chassis/frame.
(I also wonder if there’s voltage running around in the camper but without the shipping container or wires to/from making earth connectivity it’s not showing up?)

So if you prove out those situations now that it’s narrowed down you can eliminate possibilities and they’ll likely only be one thing left and THAT will be the actual problem. (Hedges’ capacitance theory is a thing but maybe I’m just not smart enough to figure out how that happens without a source or induction?)
 
Then I’m confused at the difference between “disconnected at the shipping container” VS “disconnected at the array”

Me too.
Wires from trailer unplugged at edge of shipping container and lying on the ground?
Vs. hanging in the air from trailer to shipping container, but panels on shipping container unplugged?
That would suggest to me wires on ground couple AC voltage to ground. Also PV panels couple voltage to container & ground.

What if you disconnect at the trailer?

(Hedges’ capacitance theory is a thing but maybe I’m just not smart enough to figure out how that happens without a source or induction?)

It requires an AC voltage imposed, either from 60 Hz inverter or higher frequency MPPT switcher. Then current flows through capacitance to ground.

The things we're doing at work (we happen to be looking at current rather than voltage), use a clamp ammeter or microammeter and it displays RMS current. Use a clamp probe with scope, it displays a time-domain waveform and FFT frequency peaks. Then we can see 60 Hz and its harmonics, 120 Hz from rectified 60 Hz, other frequencies from other sources.

Another way to isolate is turn off sources. In your case, run inverter to produce 60 Hz but turn off (isolate positive lead) charge controller. Then do the opposite, run charge controller but not inverter.

There should be a ground wire from frames of PV panels on shipping container back to system on trailer.
I'm confused by the results of where exactly you disconnect it, though, considering which continued to cause the problem.
 
I will add. When doing some random testing, I had disconnected all wires DIRECTLY at the disconnect switch, and while reconnecting the wires one by one, (there is 4 wires going into the Disconnect switch), I saw 27v with the first wire, then 38v with the 2nd wire, then 49v with the third, and 62v with the last. Very strange that each wire I connected added an incremental amount of voltage.
That kinda reinforces the idea that there is capacitive coupling between the wire and earth.
 
Is there an AC voltage between batteries and chassis?
That would be the case if inverter was +/-60V and transformerless, to produce 120V.
Alternatively, if isolated from battery and battery not grounded, could float and couple to have AC voltage on battery. In that case, grounding battery negative to chassis would fix it.

If battery carries an AC voltage, it would also be seen on all PV wires.
 
Then I’m confused at the difference between “disconnected at the shipping container” VS “disconnected at the array”
Same thing, array on the shipping container.
Me too.
Wires from trailer unplugged at edge of shipping container and lying on the ground?
Correct, I walked out to the shipping container, and unplugged the array on the shipping container from the branch connector, and laid them on the ground.
What if you disconnect at the trailer?
If I disconnect at the trailer, then I only get 1.XXXV AC.

-With the container connected and RV array connected you get ACV RV-to-earth y/n
Yes.
-With the container disconnected and RV array connected you get zero ACV RV-to-earth y/n
Yes.
-With the container connected and RV array not connected you get ACV RV-to-earth y/n
Yes.
-With the container disconnected and RV array connected, if you disconnect the container wires in the RV at SCC input do you have ACV RV-to-earth?
No. No ACV RV to earth when the container wires in the RV at SCC Input are disconnected.
Is there an AC voltage between batteries and chassis?
I will check for AC voltage between battery and chassis. Should I check between positive or negative? Or both?
That would be the case if inverter was +/-60V and transformerless, to produce 120V.
That would be interesting. How would you find out?
 
Back
Top