RV battery/inverter troubleshooting after adding MPPTs

DIY
1

Thank you - they are now configured in that way.

I installed all 3 controllers, updated their firmware, joined them to the Bluetooth network from the shunt, and all 3 report that they are getting temperature and voltage from the shunt and that they are synchronized charging. In other words, their little network is all set up and works.

I then created a custom battery profile for my batteries and ensured all 3 were using it. In other words, the controllers should be all set for when I hook up the lines coming from the panels to the controllers. Right now, they are just powered off the batteries (a positive and negative going to each controller). No power lines from the panels.

Tomorrow, I’m planning on running the lines from the panels to controllers. So I thought I would drain the batteries a bit so I can watch them charge up with solar.

And that’s when the weird thing happened. I threw the breaker to the inverter, shutting off power to it, which makes it go inverter, with the idea to leave it like that overnight.

To my surprise, the inverter shut down shortly after. This is uncharacteristic. The batteries are full and should power the RV until well into tomorrow.

The only thing that’s changed is that I added the three solar controllers. I ran positive lines to each one and negative lines to the shunt side where all the loads are attached. That’s it. That’s the only change that was made.

Yet the inverter shut off when I killed the power feed to it.

What could cause this? Prior to this, any time you killed the power feed to the inverter it would run for hours depending on load. The only thing I did was add in the 3 solar controllers. Did I do something wrong?

2

You turned off the grid/ac input breaker?

What system is controlling your batteries?

I’m guessing perhaps your system doesnt like that you’ve disconnected your previously hardwired connections especially to your shunt.

3

Yes, the input breaker is the same as killing the grid.
I have a Multiplus 2x120 and Cerbo.

You must be right, if it’s nothing oddball then it must be the negative shunt connection. Perhaps it’s not right.

Edit: That’s what it was. Those 10 AWG Lugs I got are garbage. I need a busbar in a big way.

4

What kind of battery do you have?

5

I have (4) Trojan T105’s in series/parallel for 12V total. According to the manufacturer, 60A is the max charge for 4 of these wired this way.

6

If ist is true that they are flooded batteries, i hope that thea are installt outside of the RV. Or in a good vented compartment. If you can position the Panels in different directions you can one position to the noonday sun, one about 45° more to east and one about 45° more to west. So the peak current is lowered and the charging time is extended. Lead batteries need a long time for full charge.

7

They are flooded and they are in a vented compartment. What has become clear is that 24v and Lithium is the way to fly but this is in the future. For now, this is ideal and as far as I want to go.

I have two panels hooked up (routing the wires from the roof to the controllers turned out to be quite a bit of work) and I ran out of cable for the 3rd panel.

Early observation shows the incredible wisdom of your recommendation of running each panel off it’s own controller. In the morning, the rear panel puts out power, then, as the sun moves, it changes to the middle panel. So one can be putting out 12A and the other 1.8A - if they were in series this would have KILLED the array and it would have been worthless.

So, this was huge, and, even though it’s a lot of work to run 150’ of cable, it was essential to do this.

I have a bad connection on the M10 bolt going to the shunt, I did get the longer bolt, but there’s too much connected to it. In fact, the microwave will no longer run on the inverter because of a poor connection there. So even my preliminary solar numbers made be off because of the poor wiring at that point. It’s an obvious problem.

This brings me back to your suggestion about the busbar, it’s coming today, and I will install it and get the wiring perfect. At that point I will be able to accurately see the two panels in operation and use the inverter once more.

On Wednesday the extra solar wire is coming and I can then get the 3rd panel up and running.

I should have 4, or, preferably 5 panels as I have room for them. Experience (now) shows that I should have put one panel on the rear as it gets the most sun - and there’s no panel there! Because this is new to me the experience of actually seeing it working is essential and I’m learning as I go.

So I’m getting there, your advice and design suggestions have been absolutely essential, and I’m very pleased you took the time to help. It was very tough to run two sets of wires down from the two panels with the 3rd yet to come.

I had to pull the fridge, run them through the bathroom, and then pull the washer/dryer, run them behind there, and then pull the water heater. At that point they go down through the floor and into what’s now being called the Victron Compartment :slight_smile: They’re going to have to be covered with loom too.

So it was/is a lot of work to run those cables, but it was absolutely essential as I can already see that one panel can be high and one can be effectively off, and had they been done in series it would have made the array, and all that work, useless.

Pics to come once I get the busbar in and things cleaned up a bit.

8

Just a quick question to confirm, the negative wire from the battery side of the solar charger should go to the System Minus side of the shunt, correct? (That being the one where all the other grounds are).

I still have an issue. When I run the microwave on the inverter, it runs for a few seconds and then that entire circuit goes down for a few seconds and then resets.

It did NOT do this before adding solar. The only thing I can see it being is the only thing I touched, the System Minus side of the shunt. I took the cables off and hit them with a wire wheel on my drill - they are now shiny.

I installed a 300A bus bar for both negative and positive lines. So now the solar positives are going there as are the negatives and everything else that was formerly attached to posts on the batteries are now in their proper places. Everything works. Just not the Microwave on the inverter.

Microwave works fine on the inverter and even goes to assist. So it’s pulling 500 watts or so out of the batteries to assist when on the grid. This works fine.

An 800 watt load on the inverter works fine but a 1500 watt load on the inverter causes it to reset the circuit. It goes offline for about 5 seconds and then comes back.

It’s never done this before. This happened post-solar. So when it’s on the inverter (a MP 2x120) it’s getting it’s power from the batteries. This means the 2/0 positive and negative. They’re great! The inverter does cough up an error on the inverter (Low Battery Voltage - Inverter). Yet, the batteries show full. I’ve run these (recently) using the microwave for more than hour on inverter. It’s NOT the batteries.

Could this be a software configuration issue? I do have DVCC enabled with a max charge of 60A. I now have two of the solar controllers on VE.Direct cables and I dissolved the BlueTooth wireless network.

9

Edit: I found the problem. I had the low voltage cut out in the inverter set to 11.50v
This worked fine because…the batteries were fine.

Over the last two nights, I left the inverter on to drain the batteries so that I could have the joy of seeing the solar charge them up…they never got below 60% but I think what happened is that the batteries got ‘pushed over the edge’ and are now in need of replacement…

I lowered the low voltage cut out on the inverter to the default 9.30V and the microwave kept running and the battery voltage held at 11.45V

My guess is that the batteries are on the outs…after all this work…groan…

10

please make pictures of the wiring, especial of the batteries and shunt. And post the settings of the MPPs. Settings/Battery. There is a switch “expert mode”. switch it on, make screenhot, scoll down and make a screenshot.

do you have a battery hydrometer to test the specific gravity of the electrolyte? Did you install the Cerbo to see your installation in VRM?

11

The SOC of lead batteries can not be identified by only the Voltage. Without current the battery is half empty at 12V and full at 12,7V after resting time. At moderate current a full battery can be at 11V. At maximal Current a full battery can be at 10,5V. So i would set it to 11V.

If the batteries are not fully charged after discharge, they can build a hard sulfatic layer. This can be dissolved by charging a long time. Observe the voltage and current diagram in VRM/advanced. Here a typical diagram of a 24V/460Ah Gel battery installation:

image
image1106×661 45.2 KB

The current rised up to 100A at in a very sunny summerday and reached absorbtion voltage of 28V (in your installation the absorbtion voltage must be higher). Then the current fall down and at about 25A the MPP switched to float. Now the SOC was about 90% to 95%. But the battery is full at a current of C/200 (2.3A). If the charging is stopped every day in this moment, the battery looses capacity more and more. In the next diagram you can see that full charge is reached at 17:30, ten and a half hour after begin of charge by a very powerful Installation. (22 panels and 6 MPPTs). If there are bad solar conditiones, the load as to be lowered or switched off accurately timed for reaching full charge.

image
image1074×648 45 KB

The battery monitor synchrones too early so it leads to the “jump” at 95%. I set it so that the user of the installation will not be nervous.

12

Yes, I have a hydrometer :slight_smile: I will see if I can remember how to use it in short order :slight_smile:

I can also give you access to the VRM, I believe. Everything is set up there. The Cerbo is fully functional.

Just a quick recap, Multiplus 2x120 being fed by the grid, and now solar.

3 solar chargers, two of which are connected by Ve.Direct cables to the Cerbo. The Multi is connected to the Cerbo via RJ45.

The shunt provides STS, SVS, SCS to the bus (Temperature, Voltage, Current) through DVCC also on a VEDirect cable to the Cerbo.

The wiring is temporary, but accurate and tight, the rats nest will be cleaned up in due course!

2/0 Black from the multi to System Minus on the Shunt, the vehicle main ground also goes there, and the main ground from the negative (2 AWG) on the bus bar. Those connections are spotless and tight. 20mm M10 Bolt fully seated with washer and lock.

2/0 Red from the inverter goes directly to the battery bank.

The busbar positive starts with the 3 red power feeds from the solar controllers, along with various other things, and is connected with a 2 AWG red cable to the battery bank.

The busbar negative starts with the grounds from the solar controllers, along with various other things, and is connected with a 2 AWG black cable to the System Minus on the shunt.

(It’s basically same as it always was, just with everything cleaned up and connections moved from the batteries/shunt to the new busbar). In other words, logical and clean and there should be minimal restrictions (resistance).

I have never been able to equalize the batteries at the proper voltage (which is in the controllers). The Multi won’t do it, it tops out a 15.75 (?) and when I tried with the controllers it also didn’t do anything.

Hydrometer Readings (hard to be 100% accurate as tool is not very granular)

Front Right battery

1290
1300
1300

Rear Right

1280
1285
1285

Front Left battery

1290
1290
1290

Rear Left

1290
1285
1280

Broad Overview Showing Victron Compartment (left) and Battery Bay (right). Cerbo on left wall.

IMG_20260818_080552471_HDR
IMG_20260818_080552471_HDR1920×1446 1.11 MB

Battery Bay Wiring. First three on bus bars are solar controllers. Shunt on left wall.

IMG_20260818_080620016_HDR
IMG_20260818_080620016_HDR1920×1446 944 KB

Screenshot_20260818-091946.VictronConnect
Screenshot_20260818-091946.VictronConnect1080×2400 186 KB

Screenshot_20260818-091952.VictronConnect
Screenshot_20260818-091952.VictronConnect1080×2400 188 KB

Screenshot_20260818-091918.VictronConnect
Screenshot_20260818-091918.VictronConnect1080×2400 186 KB

Screenshot_20260818-091924.VictronConnect
Screenshot_20260818-091924.VictronConnect1080×2400 189 KB

Update: I finally got them to equalize using the solar controllers and manually doing starting it. It only lasts an hour, and it hasn’t yet hit the 16.20v (it’s about 15.7) but I think it’s going to work. You can hear the electrolyte bubbling…

There is a bit of a discrepancy between what the panels show for amps and the shunt. The panels show 7.5 amps each (x2 = 15A) and the shunt shows 12A. Don’t know if that matters or if it’s indicative of a loss somewhere..

13

I think it is the orange maked wire is connectet to plus of the left batteries. The current of this wire will not symmetic shared to the right and left batteries. It should be connected to one of the green marked points. Best direct to the right battery

image
image886×636 229 KB

If the electrolyte level ist correct, the spezific gravity should be better. For the next weeks the absorbtion duration should not be adaptive. The MPPs are looking for the resting voltage in the early mornig to calculate the absorbtio to only a half hour. That is not enough. Set it to “fixed” for 6 hours.

14

Good catch on the wiring! I changed the absorption settings as you suggested.
Also, on the wiring, I tried that but think I came up with a better solution - move the ground to the other battery and move the busbar feed to the other terminal.

It should be 6 of one and half-dozen of the other but it works better for terminal posts and routing. Along the way…I decided to recheck the battery terminals and found one loose!! (the one with the yellow circle). It had been loose for some time…

So we’re making progress it seems! I will re-test the specific gravity after a few equalization cycles and the electrolyte was NOT topped up before checking last time. I topped them off now. Many thanks for your continued help - it’s just great!

IMG_20260818_115122867_HDR
IMG_20260818_115122867_HDR1920×1446 962 KB

15

but now the wire between plus of the right batteries to the relais is at the wrong position. It delivers current from the starter battery.

This can be a reason for weak performance

16

That wire wasn’t changed - the boost solenoid should still work as it did. I can confirm the loose connection was problematic - and then some. I can’t for the life of me remember how that happened but it must have been some time ago.

I managed to finally get the 3rd panel wired up but I’m still missing my USB hub for the VE.Direct as it’s coming from Italy. So I can only ‘see’ the two panels working. I’m still learning (functionally) how it works.

I’m hoping to get new hydrometer readings once charging is stabilized. My thinking is that perhaps with that lose connection fixed and with the wiring now being diagonal, that perhaps the cells will balance?

17

But the current of this wire flows not diagonal. this is because you changed other wires. Before was only one wire (the orange marked) at the wrong connection. by changing minus, the other wire is now wrong.

I forgot to order the USB-C cable. Did you oder it?

18

Ahhh…I get it now. The solenoid you are looking at is there to join the coach batteries to the starting batteries - but only if the starting batteries are dead.

For example, if the coach batteries are too low to start the motorhome, you push a button on the dash, and then that solenoid activates and joins them together, while the button is pushed, so you can start the motorhome. It’s not normally used. I also have a separate maintainer/charger that uses the coach batteries to keep the starting batteries up. This existed before the Multiplus was installed so I just kept it.

I also forgot to order the USB cable but I can get one off Amazon very quickly. It’s just a USB A to USB C cable.

19

Just an update, my Quadport VE.Direct hub showed up today, and I quickly plugged the 3 solar controllers into it as well as the shunt and then plugged the USB C to USB A cable into the Cerbo.

I then dissolved the BlueTooth-based smart network. All is well ! And the hub puts on a nice light show too. I now have proper reporting on my Android display.

I’ve been working on cleaning up the wiring and am almost done. Tomorrow I’ll post up some pics - it’s a lot cleaner.

Now I just have to figure out how to get the system to prioritize solar above the grid.

20

Warning: in the image in #13, you can see several fire hazards! The upper green arrow points to the + busbar, where the battery feed is reduced in diameter without any protection whatsoever! The copper cable lugs will also soon develop resistance due to oxidation/corrosion.
There are far too many cable lugs connected to the shunt (we call this the “Tower of Babel”); one is sufficient for a feed to a ground busbar.
You can also see one of these cheap, rubbish circuit breakers at the top right, next to the round Guest main switch! Please replace it with something else, such as a Class T fuse!! Contact resistance in the switch will also lead to problems and increased resistance at the connections!

Get an electrician in!!