Trying to fix a low voltage cutoff issue with my RV based inverter

[skftw]

I have an RV which recently had the original propane fridge fail. I replaced it with a "GoWise Power" 1000W pure sine wave inverter and a residential type compressor fridge. I've been having problems with the inverter cutting off when first starting the fridge. Right now the RV is running dual 6V 210Ah GC2 size batteries which are only about 18 months old. There's 300W of solar on the roof and a Victron MPPT charge controller. According to the Victron app, I'm getting a full charge each day.

Measuring the fridge with a kill-a-watt, it draws about 46W when running. Startup is hard to catch, but it seems to be around the 300-450W range for less than 2 seconds. The refresh rate of the kill-a-watt makes it hard to see the exact value, but it's very brief.

After a night with the furnace going, I've occasionally had issues where the inverter hits the low voltage cutoff when the fridge tries to state in the morning. The batteries still have plenty of energy to crank the big generator over, but that early morning voltage droop is irritating the inverter for the 2 seconds the fridge is drawing more power. I have the inverter mounted as close to the batteries as physically possible.

I had a few thoughts on what to do to remedy this:

1. See if there's a way to disable/modify that low voltage cutoff on the inverter. I've seen a thread here to do that on a different model. Given that the batteries are still able to crank the generator over quickly, I don't think I'm lacking in actual capacity; it's just the sag of the lead acid batteries under any sort of load gets them very near that 10V cutoff.

2. Look into some sort of DC-DC boost converter to give the inverter a more stable supply. Finding one that's 80-100A capable seems to be a challenge.

3. Upgrade to lithium batteries. The increased Ah rating is good, but what I'm really after is that flatter voltage curve. The Victron MPPT controller can handle them and I'd pair it with a good BMS. I dont think I'd need much else to complete the upgrade.

Any other suggestions for this setup? Also, is it recommended to leave the batteries and solar connected when the RV is stored, allowing the Victron controller to do its thing? Or is it better to disconnect everything when the RV is not in use to keep the cycle count low?

 

[Rocketman]

One other thing I though of; are the wires from the battery to the inverter undersized in any way? Probably not... I think you pretty well have the answer - now which path...

on converting to lithium- you need to look into the converter/charger Some of those don’t have good settings for lithium.

One other possible method to keep the fridge running. Install a Victron Smart Battery protect (bp65) on the furnace power - so the furnace cannot run down the batteries too much at night. You might have to play with the settings to find that sweet spot where the furnace will run - but not too much. Also, I would wire it with a bypass so if it was too cold - you could keep the furnace on. Ok maybe a larger lithium battery is sounding better and better.

As far as winter storage - I have done both. I like how easy the solar panel makes it to leave the power on. Maybe someone else has a better opinion.

 

[skftw]

The wires I would say are borderline if the inverter is actually seeing full rated load (1000W/85A plus whatever efficiency losses). They are 4AWG and only about 4ft from the battery terminals. Since the peak wattage of the fridge is less than 500W I don't think that the wires would be choking the inverter, but it's definitely possible.

I just checked and the Victron Smart Solar 100/30 I have does have a preset for LiFePO4, and also has adjustable voltages for each charging phase. I think that will be ok, especially paired with a good BMS.

The battery protection on the furnace circuit is an interesting idea. That certainly could be the quickest fix...Personally I prefer leaving the furnace off or set very low and just bundling up at night, but my guests aren't always as understanding. I'll do some more research on that option. It does sound like tearing the bandaid off and just building a more expensive lithium pack would be the best solution for the long term.

Appreciate the help!

 

[Rocketman]

For the lithium battery - yes the Victron solar charge controller will be just fine for lithium.

What I was trying to say is the converter could be an issue - so when you plug your trailer into 120v - the converter changes that 120v ac to 12v dc. Many of the cheaper ones will overcharge lithium batteries (because they never go into a float mode).

 

[skftw]

Oh, thats a good point. With the age of the trailer I can just about guarantee it's not lithium compatible. When I run the generator I usually see it running around 13.5-13.6V. I don't think it ever goes high enough to damage the lithium cells, but it certainly isnt going to charge them very well.

Would a good BMS protect the cells from a charger that attempts to overcharge them?

 

[Rocketman]

As long as the lithium’s are not full- that voltage will be fine. It will just very slowly charge them. However, if you camp at a place that has electric hookups and connect to the power- then my understanding is the batteries can get overcharged - even at the 13.6 volts. (The problem with this overcharge voltage is it will not trip the bms). The batteries will just keep taking the charge until the cells bloat. (I could be wrong- so you may want to do more checking on this issue).

Good Luck

 

[boondox]

One thing you might try is to log the watt/hours out of the battery and see if that tracks with the charge state you think you have. Meaning, make sure you really do have sufficient capacity left to maintain voltage when the fridge starts.

Once you are in the fault condition, before you charge the batteries, test the actual voltages at the battery and at the inverter when the fridge attempts to start up. This will tell you if the wiring is suspect. Another thing you might try is sub in another inverter and see if the problem persists.

I would want to track down the source of the problem before buying new batteries. Otherwise the problem may well continue.

 

[Ampster]

I usually see it running around 13.5-13.6V. I don't think it ever goes high enough to damage the lithium cells, but it certainly isnt going to charge them very well.

I only charge my 48 volt pack to 54.4 volts which is the same 3.4 volts per cell that you would get at 13.6 volts. That is about 85 to 90 % capacity for my cells.

 

[skftw]

One thing you might try is to log the watt/hours out of the battery and see if that tracks with the charge state you think you have. Meaning, make sure you really do have sufficient capacity left to maintain voltage when the fridge starts.

That's an excellent idea. I unfortunately don't have a current shunt to accurately monitor it, but perhaps I should buy one before doing the full upgrade. I think there's capacity left over. My reasoning is as follows:

- These are 210ah batteries with approximately half that (105ah) usable without fatiguing the batteries.
- The solar controller is reaching its absorption and float stages midday, so it doesnt seem to be struggling to keep up.
- Assuming I have the full 105Ah available, the fridge's 45W (or ~4A) load should be able to run for about 26 hours on its own (if it ran without cycling).
- The furnace is on a 15A fuse, but I haven't measured its actual current. Assuming its the full 15A, I'm dealing with a combined 20A load.

That drops my usable capacity to only 5 hours, however the furnace and fridge don't run anywhere near full time. I would estimate their duty cycle to be far less than 50%, which gives me at least 10 hours. The capacity is definitely questionable if they're both running a lot but so far that hasn't appeared to have been the case.

 

[Ampster]

Peukert had a lot to say about voltage sag, especially at lower SOC.

 

[boondox]

While your calculations and reasoning of battery condition seem sound what you need are measured facts. I would say that 50% daily discharge is going to result in shorter battery life. I've been off the grid on golf cart batteries for a long time and that is my experience. I have my bank sized so that with my average nightly use I am at 75% SOC in the am and I got 10 years out of my last bank. In any case, finding out how much power you are using nightly is a good idea both for this issue and for the future.

I just noticed that you are running golf cart batteries, which means flooded, right? If so, get a hydrometer like an E-Z-Red. That will tell you for sure what your SOC is. I also have an old school load tester which is very handy to have, maybe look around for one. If they are not flooded then measuring power in and out combined with the tests I described above should help you find the issue.

 

[skftw]

Peukert had a lot to say about voltage sag, especially at lower SOC.

All the more reason to move to lithium.

I would say that 50% daily discharge is going to result in shorter battery life. I've been off the grid on golf cart batteries for a long time and that is my experience. I have my bank sized so that with my average nightly use I am at 75% SOC in the am and I got 10 years out of my last bank. In any case, finding out how much power you are using nightly is a good idea both for this issue and for the future.

Even more evidence that I should probably upsize to a good lithium pack. I dont think I have the room to carry four of the golf cart batteries to get my capacity where I'd want it to be. I'm not full timing in the RV so it's at least not seeing that load every day, but I do try to use it 2-3 days per month.

I just noticed that you are running golf cart batteries, which means flooded, right?

Yup, flooded batteries with removable caps. I did check them this morning to ensure they weren't low on electrolyte, but I haven't checked them with a hydrometer. I'll pick one up and add that to the toolbox.

 

[Hedges]

Insufficient battery cables was mentioned.
Poor connections? Plug in a heavy load like space heater. Check for voltage drop across each connection. After a while, check for hot spots.

Maybe add a big capacitor across battery.

Use a timer to turn off fridge/freezer at night, only run when sun is up and has been charging.

A 50W compressor ought to have 250W surge, shouldn't be any problem for a 1000W inverter.
It would seem to be voltage drop.
Check also connection of AC, try fridge wired directly to inverter. if AC voltage sags, motor starts slower.

 

[boondox]

Make sure to get a temperature compensated hydrometer. Easier to use.

I'd by no means try to talk you out of a LFP battery. But I'd want to milk what I could out of your batteries first, they are pretty new. A few excursions to 50% SOC a month shouldn't trouble the batteries too much. I'd also want to find the root cause of the issue you are having. Otherwise you might find that the trouble continues even with new batteries.

 

[Ampster]

All the more reason to move to lithium.

Yes. Comparing kWhr discharge capacity of Lithium to Pb does show that other advantage as well. Since Lithium has very little Peukerts, they actually have more than twice the useable capacity. Comparing Amphours discharge, masks that.

 

[skftw]

Maybe add a big capacitor across battery.

I was thinking this could be a possibility. Is that something that others have done to combat an issue like this?

I'll run some of the other tests mentioned to see if I can isolate it to a poor connection. My initial configuration for this had the inverter near the fuse panel inside, but over the RV's long runs of 6 gauge wire the voltage drop was extreme, even with fully charged batteries. Moving the inverter up front next to the batteries fixed that. I'm not sure I can get them any closer but will check to make sure no connections have vibrated loose.

 

[Rocketman]

I just had a thought... on your Victron solar charge controller 100/30 app - in the history tab it shows the lowest voltage the charge controller saw on the batteries (it only keeps the history for 30 days). What is that saying the low voltage was on mornings there was a problem vs mornings there was no problem.

That I think will help with the problem.

 

[skftw]

Throughout the 4 day trip I'm seeing lows of 11.44, 11.75, 11.19, and 10.97. For the day preceeding the 10.97 reading I don't think I got a full charge. That one day the Victron was in Bulk mode the entire time. All other days had some Absorption and Float stages.

 

[Hedges]

Throughout the 4 day trip I'm seeing lows of 11.44, 11.75, 11.19, and 10.97. For the day preceeding the 10.97 reading I don't think I got a full charge. That one day the Victron was in Bulk mode the entire time. All other days had some Absorption and Float stages.

Sounds like "Cutoff" could be working just fine, and "Low Voltage" is the issue.
Those aren't voltages you want to see except during heavy loads.

 

[skftw]

Yeah. Even if they only occur for a couple seconds, its probably too close. I ordered some LiFePO4s today. Will have to get these through a few more trips in the meantime, but I dont think we'll be needing the furnace much more in the meantime.