Looking for Feedback on 400W PV System Design

[JonSchroeder]

Thanks for the feedback daleyocum. Your comment about the BMV-712 reminded me that my original schematic was missing a few items, and I hadn't posted an updated one. See below for latest design.

Regarding the need for a controller. I have the bluetooth dongle for the multiplus, and the VictronConnect app on my phone does allow me to switch the inverter on/off, and change the AC Input limits. I also have the switch on the front of the unit, which will be easily accessed. That said, I'm curious why you find yourself frequently switching on/off the inverter. I guess I have it in my head that would just leave it on most of the time.

Regarding the BMV-712, I do have one, but I'm not finding that it communicates with the MultiPlus (something about it being it being VE.Direct, not VE.Bus). Do you find that the battery info shown by the MP is aligned with that of your BMV? I wonder if your control planel allows these to 'talk'.

Regarding the DC ground heating up, I'm going to start with a 4AWG ground wire from DC Neg bus to the chassis (at the new battery location, not the panel). If that doesn't help, I'll be looking for more ideas...

Thanks again,
Jon

[daleyocum]

Originally Posted by JonSchroeder

Thanks for the feedback daleyocum. Your comment about the BMV-712 reminded me that my original schematic was missing a few items, and I hadn't posted an updated one. See below for latest design.

Regarding the need for a controller. I have the bluetooth dongle for the multiplus, and the VictronConnect app on my phone does allow me to switch the inverter on/off, and change the AC Input limits. I also have the switch on the front of the unit, which will be easily accessed. That said, I'm curious why you find yourself frequently switching on/off the inverter. I guess I have it in my head that would just leave it on most of the time.

Regarding the BMV-712, I do have one, but I'm not finding that it communicates with the MultiPlus (something about it being it being VE.Direct, not VE.Bus). Do you find that the battery info shown by the MP is aligned with that of your BMV? I wonder if your control planel allows these to 'talk'.

Regarding the DC ground heating up, I'm going to start with a 4AWG ground wire from DC Neg bus to the chassis (at the new battery location, not the panel). If that doesn't help, I'll be looking for more ideas...

Thanks again,
Jon

Because all these different Victron pieces were developed over the course of many years they haven’t done a great job of making them talk directly to one another. Ideally they’d have one communication bus for everything to talk over. Instead they have VE Direct, VE bus, ethernet, USB, and Bluetooth. Each piece is left having to make due with partial information and make guesses when it comes to Lithium with its flat discharge curve.

If you spend another $300 you can solve all this with a Victron Venus GX. This, along with big brother the Color GX, acts as a central controller. It integrates the BMV712 with the Multiplus and the Solar controller to do what you really want.

You’ll find that without it the solar controller will wake up in the morning, charge the battery, and then turn off for the rest of the day (guessing about when it should do that.) With the GX controller it will kick back on if you draw down the battery during the day because it’s watching the state of charge from the BMV712. In my case that happens when we turn on the AC during the day or make lunch with the microwave.

The GX controllers have a million plugs on them to connect to all your different Victron devices. For me it was annoying to run all those wires up to a Color GX in a public place so I saved a few hundred dollars and hours of work by using a the display-less Venus GX. I then just leave an old iPhone always connected to its hotspot to watch its pretty pictures.

With regards to needing to turn the inverter on and off, if you can easily get to the switch on the front of the unit then perhaps you don’t need the Multicontrol panel. You will probably want to leave the inverter off when you aren’t using 120VAC because the inverter will draw 3-5 amps just on standby. It would get really annoying to ask everyone in the trailer to do that with their phones.

[uncle_bob]

Hi

Any time you find a wire that gets hot to the touch, that's a bad sign. Either there is too much current (due to a load problem) or to small a wire (due to a wiring problem). Yes, it's the same thing, just looked at in terms of two different solutions.

250W at 12V is roughly 20A. Anything less than 12AWG will indeed get quite warm with that sort of current. A loose nut at the ground bus will get warm, even if it *is* 12AWG.

Unless you spend a lot of time playing with bare copper wire, the only way to be *sure* of the wire gauge is to measure it. A sub $10 vernier caliper is plenty good enough to do this. It also is not a big loss if you set something on it and destroy it (unlike the $100 version).

At 180W into lights, I think I'd schedule a "let's go LED" party somewhere in the foreseeable future

======

With modern Lithium's, the BMS does the fiddly charge control stuff. As far as the charger is concerned there are really only two states:

1) Full voltage to run up to equalization (likely 14.x V)

2) Cutback voltage to maintain charge (likely 13.5 to 13.9 V).

The charge system runs up to the point (that you set) and decides the batteries are charged. It then cuts back. The reason you get a Multi is that all those settings are where you can get at them.

Indeed a lot of the "Lithium" chargers on the market don't have two states. They simply run at the 14.x V point all the time. The batteries (apparently) are happy with this. Some of us suspect that they will last longer if you cut back the voltage. So far the manufacturers seem to be happy to recommend the constant voltage chargers.

Bob

[JonSchroeder]

Originally Posted by daleyocum

...If you spend another $300 you can solve all this with a Victron Venus GX. This, along with big brother the Color GX, acts as a central controller. It integrates the BMV712 with the Multiplus and the Solar controller to do what you really want.

I'm starting to realize that I should have factored in a 30% up-charge on all of my main Victron component costs to account for all the 'accessories' I'd be needing. Already purchased a MK3-USB adaptor to program the MultiPlus ($68) and a Bluetooth dongle to monitor it ($78), now it seems I need a 'Venus GX' (~$300) in order for my devices to talk to each other. At the very least, the USB adaptor should have come with the unit.

Regarding the draw of the inverter, I'll have to do some testing on that. Was hoping that without a load, it wouldn't be pulling anything close to 3-5 amps.

Thanks for the feedback daleyocum!
Jon

[JonSchroeder]

Originally Posted by uncle_bob

Any time you find a wire that gets hot to the touch, that's a bad sign. Either there is too much current (due to a load problem) or to small a wire (due to a wiring problem). Yes, it's the same thing, just looked at in terms of two different solutions.

Hi Bob,

Does it surprise you that its the chassis ground wire (i'm assuming that's what the uninsulated wire is) and not one of the 'Neg Load' wires that's getting hot? If all DC loads have a neg wire leading back to the panel, there really shouldn't be much current running through the grounding wire, right? Seems like this may indicate that my grounding wire (possibly from the fridge) is shorted to the frame somewhere. Or am I misunderstanding the wiring? Is there a way to test for a short on the negative side?

Thanks,
Jon

[daleyocum]

Originally Posted by JonSchroeder

Hi Bob,

Does it surprise you that its the chassis ground wire (i'm assuming that's what the uninsulated wire is) and not one of the 'Neg Load' wires that's getting hot? If all DC loads have a neg wire leading back to the panel, there really shouldn't be much current running through the grounding wire, right? Seems like this may indicate that my grounding wire (possibly from the fridge) is shorted to the frame somewhere. Or am I misunderstanding the wiring? Is there a way to test for a short on the negative side?

Thanks,
Jon

I’d suggest using a multimeter on the Multiplus to see if the chassis is floating or is internally connected to the negative 12v input terminal. I did not know off the top of my head. (Obviously disconnect the negative wire from the multiplus before doing this.) I suspect you’ll find the Multiplus chassis is internally connected to negative. If that’s the case something substantial in the trailer is getting it’s negative from the chassis and not from a negative wire running to it.

[uncle_bob]

Originally Posted by daleyocum

I’d suggest using a multimeter on the Multiplus to see if the chassis is floating or is internally connected to the negative 12v input terminal. I did not know off the top of my head. (Obviously disconnect the negative wire from the multiplus before doing this.) I suspect you’ll find the Multiplus chassis is internally connected to negative. If that’s the case something substantial in the trailer is getting it’s negative from the chassis and not from a negative wire running to it.

Hi

Since the Multi has an independent ground wire that attaches to the chassis / outer metal shell, my guess is that the internal electronics are isolated from the shell.

Bob

[daleyocum]

Originally Posted by uncle_bob

Hi

Since the Multi has an independent ground wire that attaches to the chassis / outer metal shell, my guess is that the internal electronics are isolated from the shell.

Bob

Bob, I hope you’re wrong because that would mean there’s something very mysterious (and very bad) going on in order for that ground wire to get hot!

[rickst29]

Originally Posted by JonSchroeder

.... On further inspection, it seemed to be the non-insulated grounding wire that was getting hot (so hot I couldn't keep my finger on it for long).

Does this make sense? Shouldn't the fridge have a neg. wire running back to the panel? Or could this just be chassis grounded?

If the only connection to the "battery -" terminals is through your new #4 AWG wire on the Distribution Panel grounding bus, then something with a large draw (or multiple smaller things) are using the frame as part of a common Grounding Plane, and the sum of those current amounts must go through that wire (in order to reach the batteries).

I think that you have 3 options, and they can be combined:

1. You can search for 'major appliances' which are grounded to the frame, and run individual wires back to the Grounding Bus. (This is the way that modern AS Trailers are built.) Your biggest suspects are the Fridge (in DC mode), the water heater (in DC mode), and the Propane Heater. The main heater, when running, uses a lot of 12V power to run it's fan. The Air Conditioner may or may not use 12V for the same purpose, it's also worth a check.
2. You should upgrade the size of that wire. You should also check that its frame lug is very tight, and free of rust. Your 1984 trailer has probably never had that lug and connecting wire segment cleaned. (It's a connection between copper wire, brass lug, and steel frame: Galvanic Corrosion occurs among those metals.) It's actually possible that the heat is being generated at the lug connection - with very high resistance, and then transmitted along the copper wire because it has the best heat distribution.
3. You might also consider another frame lug and grounding wire near the batteries. Just be sure that it connects into the "Trailer return current side" of your battery monitor shunt, rather than going directly to the batteries.

My own "biggest appliance" is the Inverter. It is adjacent to the battery (a big LiFePO4 within the coach), and wired into "Battery +"and "Battery Monitor shunt -" power distribution blocks with short segments of #2. My power Distribution Panel (with its "12V Grounding Bus") is a long distance from the battery, so I do use the frame to reach between the Battery "-" Distribution Block and the 12V "-" grounding bus at the Distribution Panel.

BTW, I have no idea why your panel includes fuses for "Battery -" connections. I do not see that in more modern designs.

[rickst29]

Originally Posted by rickst29

I think that you have 3 options, and they can be combined:

1. You can search for 'major appliances' which are grounded to the frame, and run individual wires back to the Grounding Bus. (This is the way that modern AS Trailers are built.) Your biggest suspects are the Fridge (in DC mode), the water heater (in DC mode), and the Propane Heater. The main heater, when running, uses a lot of 12V power to run it's fan. The Air Conditioner may or may not use 12V for the same purpose, it's also worth a check.
2. You should upgrade the size of that wire....

If you have a suitable shunt (allowing metered current below the 10A limit of most multi-meters), then you could look at the current while turning various 12V electrical things on and off.

To install the shunt (or to do anything else with the "12V negative ground bus"), I suggest that you use your big battery "+" shutoff switch (or circuit breaker) to remove battery power from the 12V Distribution Center. Be sure to also pull the main Trailer power cord, to prevent 12V power from arriving through the Power Converter.

If you don't have such a breaker at the batteries, I kinda strongly recommend that you get one.

Anyway - if you can't measure current, you have another alternative (but I'd wear a pair of non-conducting gloves for this one): After disconnecting and shutting down the three +12V power sources, disconnect that frame "ground" wire from the bus.

I'd actually put in a new replacement right then: stranded insulated white, 95c temp rated, size #4 AWG. If the frame lug can't accept #4, #6 might also be OK - because we will next identify the biggest power users, and later give them individual wires. Don't connect the new wire to the bus yet, leave it wrapped in electric tape. Now start turning on your 12V appliances. Anything which doesn't become active is using the frame.

Pull new stranded white "12v ground" wires for the biggest appliances which didn't turn on successfully.

[rickst29]

Your old panel, with separate battery fuses for individual batteries on BOTH "+" and "-", exhibits a very bad idea (illegal in 120V house wiring, and really bad in 12V trailer wiring as well).


You should never, ever have a fuse or breaker which is placed to interrupt "grounded neutral" wires, green (or bare) grounding wires, or 12V "negative ground" in a negative grounded Travel Trailer. Fuses and Breakers should only be installed on the "hot" wires.

[JonSchroeder]

So, I had a chance to do some work on the trailer this past weekend, and I've made some progress on resolving the hot grounding wire issue.

First, as recommended by all, I've added a heavy (4awg) grounding wire from the DC Neg bus (at the battery location, not the DC panel). I did this by cutting a hole in the subfloor above a frame rail and drilling/tapping a hole for a 5/16" bolt (see pic below). This seems to have resolved the hot wire issue at the DC panel, as I can now turn on all lights and the fridge without any warming of the neg wires at the panel. This is great news, but it doesn't solve the mystery of why that chassis ground wire was getting hot...

Rickst29, your point about the fused Neg wires is a good one. I have no idea why those would be there. Clearly, it came from the factor like that, but it that doesn't mean its the right way to do it. I'm going to replace the current DC fuse panel with a modern one from BlueSea. I'll replace the replace the chassis ground wire with something heavier at that time. I'll also add a small non-lugged blade fuse block for the 7-pin connections to get rid of all the old style fuses.

I plan to do some additional diagnosing regarding the DC neg wiring to identify exactly which appliance(s) are using the chassis ground. At this point, I'm pretty sure the fridge is, but I don't know if that's by design, or due to a short. Regardless, If I confirm that the fridge is in fact running its neg through the chassis, do you think it's worth trying to run new DC wires to it?

Thanks for all the helpful feedback!
Jon

[daleyocum]

I think the reason modern Airstreams don’t use the frame as ground is to avoid ground loops as well as the chance of corrosion at all those various connections between dissimilar metals. It also means every metal joint doesn’t have to be electrically sound. Is it worth chasing all those down and running new ground wires in a trailer that wasn’t designed that way? That’s up to you. I personally wouldn’t bother. If it truly is just one appliance then it is probably, as you suspect, an open wire or bad connection somewhere.

[rickst29]

Good job! Just one thing I'd like to check with you: the kind of bolt and washer which you installed. It has two places for "galvanic corrosion" to occur: #1, between the copper wire lug and the washer (or bolt head); and #2, between the bolt and the steel frame (which you can't see).

Here are the metals which may be involved:

Stainless Steel -0.15 Volts. (#410) High resistance to Oxidation
Copper, Bronze -0.36 Volts. Oxidizes to brown, black, and then green
Iron / Steel -0.61 Volts. Low resistance, brown corrosion
Aluminum -0.79 Volts. Medium resistance, white corrosion
Galvanized Steel -0.98 Volts. Hot dipped or zinc-plated steel
Zinc -1.03 Volts. The least noble, most base metal


Stainless would resist oxidizing both itself and the steel frame (carbon steel == iron in this table), because it basically is already rusted - in a super thin layer on the surface, held really tight. But it is a terrible conductor for electricity, and should not be used.

Galvanized Steel (almost == zinc, it's coating) will rust, in comparison to the frame, and in comparison to the copper lug. It's a good conductor for electricity, and probably your best choice. With a Galvanized bolt, you should expect to replace it every 5-10 years (depending on humidity, and especially beach exposure to salt spray).

A plain steel bolt, if you could even find one, would have more electrical resistance than galvanized. You unfortunately have a very small area of contact between bolt threads and steel frame, that is where electrical resistance will be an issue.

Aluminum has better Galvanic potential, but you should probably use galvanized steel, because aluminum is weak. An aluminum bolt, after rusting against the steel frame, would have a very high probability of breaking off at the point when you remove the rusted one for replacement.
- - - - -
Daleyocum: I think that we did not have a ground loop in the original AS design. If we think of 3 different DC-powered components using the frame as a "secondary grounding bus, the only path which went to "battery -" went through the same panel bus (the "primary" grounding bus) on it's way to the panels.

I think that AS abandoned this scheme, and started to pull individual wires, because of galvanic corrosion - and also because vibration (from travel) can help cause those frame bolts to loosen up, if they have not already rusted really tight.

[rickst29]

Originally Posted by JonSchroeder

....Rickst29, your point about the fused Neg wires is a good one. I have no idea why those would be there. Clearly, it came from the factor like that, but it that doesn't mean its the right way to do it. I'm going to replace the current DC fuse panel with a modern one from BlueSea. I'll replace the replace the chassis ground wire with something heavier at that time. I'll also add a small non-lugged blade fuse block for the 7-pin connections to get rid of all the old style fuses.

Mini-ATC's fusses are easier to push and pull. And BattleBorn wouldn't like the idea of separate battery wires, separately fused: If you have two or even 3 identical batteries, they (and I) want them linked into a parallel string at the batteries - with only one big wire (with a big Breaker) going out to +12V distribution.

In the current scheme, any damage or lossening of wires on the 2 battery paths can cause one battery to be favored over the other, for both charging and discharging. With LiFePO4, even more than other battery types, they need to be kept in balance (for both charging and discharging). Getting rid of this panel is a good idea, go for it!

[JonSchroeder]

Thanks so much Rickst29 for reality checking all this for me.

Regarding the bolt, I used a zinc-plated Grade 8 Steel bolt (5/16"). I sanded the finish off a small circle around the tapped hole in the frame and included a locking washer between the 1st nut and the frame in the hopes of increasing the contact surface area. Because I expect the current flowing through this connection to be minimal at most times, I'm hopeful that this setup should last a while.

Regarding the 2 battery wires: That second battery connection (on both neg and positive side) is not being used. Even the previous lead acid setup had the two batteries paralleled before running to the panel. Seems odd that the panel would include that option, but who knows what AS was thinking back in the 80s.

Thanks again,
Jon

[rickst29]

[QUOTE=JonSchroeder;2276486
Regarding the bolt, I used a zinc-plated Grade 8 Steel bolt (5/16"). I sanded the finish off a small circle around the tapped hole in the frame and included a locking washer between the 1st nut and the frame in the hopes of increasing the contact surface area. Because I expect the current flowing through this connection to be minimal at most times, I'm hopeful that this setup should last a while.

Regarding the 2 battery wires: That second battery connection (on both neg and positive side) is not being used....[/QUOTE]
Grade-8 alloys ((type 8740 is most commonly used) tend to rust faster than Grade-5, but the zinc coating and dry environment should provide a good lifespan. I forgot to note that you used a 5/16 bolt, with relatively good surface area.


If it's not too late to re-do, I recommend that you add a galvanized flat washer underneath the lock washer. It would avoid the numerous, small rust-prone "gaps" - that might be present on the frame-to-washer surface. You could slightly over-size the washer, improving surface area. (More surface area, with tight contact, reduces rust and improves current flow.)


At the very end, after it has all been tightened up, you should apply a bit of "battery terminal" Lithium Grease (available at any car parts store) over the lower bolt and washers assembly, to slow down rust even more.

[JonSchroeder]

Hi everyone,

Just wanted to provide a quick update on my electrical project. While I'm sure there will be more upgrades in the future, I think I'm done with this first phase. I've now got everything installed and tested, and the system seems to be performing exactly as expected. The upgrade to LiFePO4 was costly, but man does that make a big difference in performance. If I'm ever low on solar, I can hook up the generator and have them topped off in no time. Also really happy that I made the decision to replace the original DC panel. The new blade-fuse design is so much cleaner, and I feel a lot more confident in all those DC connections now.

What to give a sincere thanks to everyone who commented, but a special thank you to rickst29, uncle_bob, and daleyocum. Your constructive feedback was enormously helpful, and gave me a lot more confidence in the decisions I was making.

Cheers Everyone!

Jon

[rickst29]

You're very welcome. Congrats on the completing the project!

[uncle_bob]

Hi

Glad it's working !!!

The biggest thing I've noticed going to the lithium's is that it has eliminated the batteries from the planning process. We started out with < 2 days worth of battery and it used to be a big deal.

Bob