Just realized my ground bond is bypassing the shunt. How to fix?

[zamboni]

Okay, i've got the system in my RV completely offline, battery pack is totally disconnected, and the bond from batt(-) to chassis ground is removed. But while checking things out, i realized i still have continuity between the chassis itself and batt(-) wiring. Since the chassis bond connects directly to the battery terminal, it's upstream of my shunt, so now i realize my battery monitor is useless because with secondary current paths through chassis, the shunt is partially bypassed. My rig is a frame-up rebuild and i know where every wire is: The batt(-) bond is the only tie that i made to chassis in the whole system. After reading SunKing's comments on an earlier thread, i realize i must have some internally-bonded equipment. If not the Inverter, then surely the slide and landing legs motors. UGH.

So, i have to relocate that bonding cable to the other side of the shunt...but that's going to be a big undertaking. The breaker box (MNDC250 with the shunt mounted inside) is already quite full and i'm not sure how to add another 4/0 cable in there. The only option i see involves taking pretty much the whole system apart and starting over. If that's what it has to be, then okay, my week is going to /really/ suck...but i'm hoping there might be some other options?

Based upon my catastrophic fuse rating (300A), i applied the 125% safety factor, and came up with needing 4/0 (@90*C) welding cable for that ground bond. But without that safety factor, 2/0 meets the ampacity requirement and would be a LOT less difficult to install! I don't intend to cut corners on safety, but am wondering if i overdid things by applying the 125% here, since a ground fault is not a continuous load. Which way is correct?


Thanks for any help y'all can offer.

- Jerud
------------------------------------------------------------
1220W array (4x 305W Astronergy panels @ 48V)
1000Ah LFP house bank (5P4S HiPower 200Ah cells)
MPPT solar charge controller (MidniteSolar Classic 150)
2800W PSW Inverter (Magnum MS2812)
ME-RC, Trimetric, and JLD404

2001 Fleetwood Prowler 5th wheel 25 foot, self-rebuilt

[Sunking]

You have a lot of things wrong. Example Bonding Jumpers are not required to carry full load current. Bonding Jumpers are based on the size of the fuse. It does no harm over sizing, but not required. Example a 300 amp fuse only requires a #4 AWG bonding jumper.

However this is not the source of your problem, and you cannot fix it if you use a Grounded System. The problem is in your equipment like Charge Controller, Inverters, and load equipment (gizmos). Most of your equipment has the Battery Return (negative polarity) bonded to the Chassis and Ground Terminal. Thus your vehicle chassis is in parallel with your battery Negative conductors.

You could fix it easily by moving the Shunt to the positive polarity, however that is very dangerous and most systems will not work or allow you to do that. The other way is to use a Floating System but that requires OCPD (over current protection devices) to be installed on both Polarities, and isolating every piece of equipment from the vehicle chassis and . Just one major problem doing that. Most likely most of your gizmos using battery power have negative polarity bonded to chassis.

So short story is you have to live with it and nothing you can do about it. FWIW your battery monitor would still be usless even if you did not have current flowing through the chassis. You cannot determine SOC from voltage on a working system.

[zamboni]

Why's there nothing i can do about it? It seems like you just gave me the solution: If the ground conductor does not have to be sized for continuous current and can therefore be 4awg, then i have enough space to move that ground bond connection to downstream of the shunt. As you explained, i can't change the internal ground bonding of my equipment, and i don't want to float the whole system, but i can make sure all the current which flows through the chassis must still pass through the shunt first. That will let my monitor count everything, and that solves my problem. Am i missing something?

It is a "real" monitor (Trimetric), not a voltage-based "fuel gauge". I assumed that discussing a shunt established my awareness of the distinction, but could have been clearer.

Aside from the two motors i mentioned, i can't think of anything that has the possibility of electrical connection to the chassis through the case or mounting bolts, but i'm going to disconnect everything one at a time and re-check continuity to find out where the internal bond is actually happening, just so i know in case it's helpful in the future.

Thanks for your feedback!

- Jerud
------------------------------------------------------------
1220W array (4x 305W Astronergy panels @ 48V)
1000Ah LFP house bank (5P4S HiPower 200Ah cells)
MPPT solar charge controller (MidniteSolar Classic 150)
2800W PSW Inverter (Magnum MS2812)
ME-RC, Trimetric, and JLD404

2001 Fleetwood Prowler 5th wheel 25 foot, self-rebuilt

[Sunking]

Am i missing something?

Yes Sir you are missing something. That #4 AWG to bond the system to the chassis to make it a Grounded System; MUST be connected directly to the Battery Negative Term Post. It cannot be connected anywhere down stream with intentional resistance inserted into the path. A Shunt is a RESISTOR.

To make your Trimetric work requires two shunts if I recall correctly. One to measure current going into the battery from the Controller, and another one to measure Current going to the load. If your Trmetric can do this, is to use the positive polarity. However I do not think you have that option because one of the shunt leads is referenced to the chassis and it would smoke and melt the Shunt wire as soon as you connected it.

What you are likely going to find is your charge controller and Inverter have the chassis bonded to the battery negative terminal. Now there might be, and that is a big maybe, a manufacture that makes a Controller and Inverter that isolates both polarities from the chassis allowing the user to use either Positive or Negative grounded system or non-grounded system. Good luck with finding both.

I am not going to tell you this, and I did not say it, is you can mount the controller and inverter on plywood or something non-conductive, and only run Battery Positive and Negative conductors. However that may not work because if you are powering any gizmos with battery, there is a very good chance the gizmo has negative bonded to its chassis.

Do as you wish. But the right tool for the job is not a SHUNT, It is a HALL EFFECT TRANSDUCER on the ungrounded Positive Polarities which I do not think your Trimetric is not equipped to handle. CLICK HERE for Hall Effect Transducer. Not sure what voltage your SHUNTS are but I suspect 20 or 50 millivolts. You would have to find a Hall Effect Transducer with the right scale (example 0 - 300 Amps) with the correct milli-volt output. You will need two of them, otherwise you are pissing in the wind. You have to know what current is going in/out of the battery, and the load current. That takes two sensors and from those two sensors you can measure current from Controller, Battery Current in/out, and load current. Otherwise your Trimetric is completely useless.

[zamboni]

Sorry for my late response, i forgot to manually check the thread and didn't get an email about your reply. Thanks again for so much information.

Yes Sir you are missing something. That #4 AWG to bond the system to the chassis to make it a Grounded System; MUST be connected directly to the Battery Negative Term Post. It cannot be connected anywhere down stream with intentional resistance inserted into the path. A Shunt is a RESISTOR.

Well $#!t.
Does the inclusion of the very-low resistance shunt in the ground current path completely undermine the function of the "catastrophic" fuse? Is this a question of strict adherence to code vs. real-world practical application? I have come across a couple sources now which explicitly advocate connecting the system ground bond downstream of the shunt. That includes the installation manual for the Trimetric battery monitor. Is this just more bad info on the internet?

the right tool for the job is not a SHUNT, It is a HALL EFFECT TRANSDUCER on the ungrounded Positive Polarities

A hall effect transducer is the same device used in an "amp clamp" meter, correct? That sounds great...if i put a hall sensor on the positive battery lead, that would capture all the incoming and outgoing current, but as you say i'm not sure i can get a sensor that will follow the right scale to work with a battery monitor that is expecting a shunt. As for moving the shunt to the positive side, you are of course correct that the trimetric would instantly smoke.

My system is too far disassembled for me to actually measure, but i'm goign to compare the resistance of my negative cables vs. the chassis ground-path and figure out how much current is flying under the radar. I might just have to accept that the monitor is going to lose a few tenths of a percent. So long as the batteries hit "full" (not actually 3.65VPC; i use 3.5VPC) every once in a while and reset the Ah count that inaccuracy should not wreck things totally.

[zamboni]

Okay, obviously some time has passed, but this thread has never fully made sense to me, and since I'm thinking about a new system I thought I should take the time to get it right. So I reread the above, and also reviewed Sunking's sticky on bonding and grounding, too (which, I note, is in the off-grid section but not the mobile / RV section). All the stuff above about the shunt and hall sensors...that's not my problem. My question is about the grounding bond size. In the sticky, SunKing makes it very clear that the GEC is the conductor that connects to dirt, and everything else is a bonding jumper. Well, in an RV, the "ground" conductor goes to chassis, which is not dirt. So that makes it a bonding jumper...right? And that's why earlier in this thread, SunKing told me it doesn't have to be sized for continuous current, but only needs to be 4awg. Sure enough, I got ahold of NEC 2020 and in Table 250.102(C), that's what I see... for my 4/0 ungrounded conductors it's actually 2awg, but probably SK just assumed I had 2/0 since that's common with RV solar. This all seemed to make sense.

But then, back to the off-grid grounding sticky. The diagram at the bottom shows three green conductors entering the "intersystem bonding bar", and one green conductor leaving it, going to earth. The one going to earth is labeled "GEC". So far, so good. I don't have a GEC but if I did, that's where it'd be. But now, the three conductors entering the bonding bar are labeled "same size as largest conductor used." And that's where I'm confused. Aren't these bonding jumpers? Shouldn't they be 2awg or 4awg -- sized per 250.102(C)?

If someone actually sees this, I'd be grateful if you can set me straight on this. Thanks!

[bob-n]

This may be a really misguided suggestion, but what about using a hall effect sensor rather than a shunt? For example:

http://www.kohshin-ele.com/en/products/cs_hallsensor.html

There are many makers of these devices. The only challenges would be sizing the sensor for the required current and interfacing the sensor output to your monitor. The good news is that you can put this in series with either terminal of the battery, of the array, or anything else to sense DC and AC current.

[bcroe]

Hall effect is used, but note in your clamp on meter, it must be zeroed for each use.
They drift, and so have limited resolution. They work if you can live with those
limitations. Bruce Roe

[zamboni]

Hall effect is used, but note in your clamp on meter, it must be zeroed for each use.
They drift, and so have limited resolution. They work if you can live with those
limitations. Bruce Roe

For the moment, I'm not trying to sort out the shunt / hall sensor question, although Bruce Roe did mention earlier that there are ways to put shunts on the positive supply lines, which intrigued me. But here's the thing: it's pretty easy and quick to buy a battery monitor like Trimetric / Victron, install a shunt in your system (one way or another), and have your SOC% (from Ah in minus Ah out) up and running with a nice user interface that you can teach the whole family to use. I get that less user-friendly and DIY options can be made that are superior, but that's not a very widespread skillset.

And anyway, it's the ground bonding I'm trying to get sorted out right now : )


P.S.
Something I love about this forum is that old posts don't seem to die like on so many others. I think it cuts down on a lot of duplicate content, actually : )

- Jerud
------------------------------------------------------------
1220W array / 1000Ah LFP house bank
MidniteSolar Classic, Magnum MS2812
ME-RC, Trimetric, and JLD404
Full-time 100% electric boondocking (no propane, no genny) since 2015
2001 Fleetwood Prowler 5th wheel 25 foot, self-rebuilt

https://www.livesmallridefree.com