Battery Drain

[Beowulf5g6]

Hey all!

I have two 100ah batteries, a renogy solar charge controller, and 720 watts of solar panels.

Everything was working great up until a week ago. Batteries would charge just fine. All of the sudden though, my batteries just drain. I went to work on the bus tonight and they were completely dead. The only variable that comes to mind was I added a renogy inverter. The inverter however, isn't even turned on at the moment though. So I'm not sure where to start troubleshooting. Any help would be much appreciated!

 

[nikitis]

Are these like car batteries?

If so these aren't great for longevity in solar applications, you need something like LiPo4's Lithium Ion type batteries. Same as in your cell phone but much larger. and can be drained down to zero.

Car batteries cannot be drained below 50-60% leaving only 40% usage. If you've been dipping below that you probably killed them.

 

[Jono14]

A clamp style ammeter can be your friend here

A very helpful tool in this situation is a clamp style ammeter which can cost in the $40-50 range.
This allows you to put the clamp around the battery cable and see the actual current flowing in and out of the battery at any moment.
In a healthy system, if you turn everything off, there should be zero current leaving the battery and the charge should last. But your system clearly has some sort of amp draw and the new inverter is the most likely culprit.

Inverters when on, even if nothing is being powered by them, draw an "idle" current which can drain a small battery system quickly. Renogy inverters note about 50 watts of idle current.

Quick math: 2 x 100 Ah 12V batteries hold about 2400 Watt hrs (Wh) of energy. Almost all of this is usable if they are LiFePO4 and about 50% is usable if lead-acid (as nikitis noted). The inverter uses 50 Wh every hour, or 1200 Wh in 24 hrs so will drain lead-acid batteries in 1 day and LiFePO4 batteries in 2 days.

So make sure your inverter is shut off when not using anything that needs AC and use an ammeter to confirm there are no other parasitic draws anywhere.

I'll also add, your battery bank is a bit undersized for your solar input. On a good summer day you might get 5 hours worth of solar (720 x 5 = 3,600 Wh) but a lead-acid bank that can only use/store 1,200 Wh will stop charging mid day - so a larger battery bank will store more of the energy your panels produce!

Cheers!

 

[Stinky-ClassB]

Batteries are a mystery. I have 2 x 100A lithiums. Looks to me like one has gone bad. For no reason. Well at least they are now half the price I paid for them.

 

[nikitis]

Batteries are a mystery. I have 2 x 100A lithiums. Looks to me like one has gone bad. For no reason. Well at least they are now half the price I paid for them.

Some lithium batteries do not come with BMS's and a cell could go bad in them and make the whole battery go bad. a BMS is supposed to monitor each cell and if one is lagging, it will hold charging on the good cells to force the bad cell to take a charge, and if it's too bad for too long should route just the good cells and blocking the bad cell with some diminished capacity.

Sounds like your Lithium ones didn't have a BMS, a cell went bad, and now it acts like the whole battery is bad likely.

Not all Lithium Ion batteries have BMS's. They cost more for the BMS but it should handle better and last longer as a result of it handling the cells better over the course of it's life.

 

[Stinky-ClassB]

No, my batteries do have BMS. I've been in touch with the mfg. I'm half way through a 5 year warranty. Call me cynical but I don't expect much from that. For the time being I'm doing some testing. Separated the 2 and put each under an equal draw. They charged equally and so far they are drawing down equally. I expect one to fail around 12V.

 

[Stinky-ClassB]

Update on my LFP batteries. I was right in my analysis. Both batteries charged up fine and drew down to 12V in sync. (I separated them at the positive terminals and connected a headlight to each) Under 12V one started to lose voltage faster. Eventually the headlight connected to it blinked out at around 10.5V. The good battery continued on at 11.8V So it's an internal problem and the bad battery was killing the good one in that the total voltage of the 2 was only as good as the bad one. I'm holing my breath to see if they make good on the warranty.

 

[Albatross]

I would get two--it happens more often with LiFePO4 batteries than with PbSO4 batteries, but it is still common for a bad battery of a pair to mess up the chemistry of the other one in a pair. It's kind of how like people with a bad leg or knee with in very short order wind up with two bad knees/legs.

 

[Stinky-ClassB]

Ah Jeez, I broke my hip 2 years ago. You figure I should just trade them both in for a wheel chair?

 

[Beowulf5g6]

Thanks everyone for the feedback so far, sorry for the late response I've been travelling out of the country. Just got back last week and here's what I've done:
Replaced the solar charge controller.
Changed the solar panels from series to parallel.

None of that has worked. I currently have a battery charger hooked up to the batteries to see if that does anything but my battery meter isn't showing any change:

To clarify, the batteries are LiTime 12V 100Ah LiFePO4, less than a year old.

A very helpful tool in this situation is a clamp style ammeter which can cost in the $40-50 range.
This allows you to put the clamp around the battery cable and see the actual current flowing in and out of the battery at any moment.
In a healthy system, if you turn everything off, there should be zero current leaving the battery and the charge should last. But your system clearly has some sort of amp draw and the new inverter is the most likely culprit.

Inverters when on, even if nothing is being powered by them, draw an "idle" current which can drain a small battery system quickly. Renogy inverters note about 50 watts of idle current.

Quick math: 2 x 100 Ah 12V batteries hold about 2400 Watt hrs (Wh) of energy. Almost all of this is usable if they are LiFePO4 and about 50% is usable if lead-acid (as nikitis noted). The inverter uses 50 Wh every hour, or 1200 Wh in 24 hrs so will drain lead-acid batteries in 1 day and LiFePO4 batteries in 2 days.

So make sure your inverter is shut off when not using anything that needs AC and use an ammeter to confirm there are no other parasitic draws anywhere.

I'll also add, your battery bank is a bit undersized for your solar input. On a good summer day you might get 5 hours worth of solar (720 x 5 = 3,600 Wh) but a lead-acid bank that can only use/store 1,200 Wh will stop charging mid day - so a larger battery bank will store more of the energy your panels produce!

Cheers!

Is this the meter you're talking about? https://a.co/d/8XIM40A

And here is a picture of my setup so far:

If anything looks off please let me know!

 

[Jono14]

Looks like a nice system and I expect it will work well.
A few thoughts:

Yes, that meter you linked to should work. The beauty of a multimeter is you can dig into the different parts of the system. Set it to DC volts and then you can check the input and output voltages for the SCC to be sure your panels are wired correctly, then you can test down at the bus bars to be sure your switches are on and finally at the batteries. The clamp can be put around any single wire in the circuit to measure current flow at that point which can tell you a lot.

The main issue for me from the photos is that your negative battery charger clamp (I assume those are on the clamps on the battery posts) should be on the other side of the shunt. I expect that your batteries are getting charged from the battery charger but your shunt and meter are not recording the flow because they are getting bypassed. The 11.9V shown on the meter suggests each of the 4 cells per battery is just under 3 volts which is holding some level of charge (maybe in the 5-10% range so they may already be holding 10-20 Ah).
You can just take the cover off the negative bus bar and put the negative clamp on that. I expect your meter will start counting the charging flow at that point.

Note that your meter will be off from the actual battery charge. Generally these meters allow you to reset the charge percentage to 0% or 100% when you know you are there, some do it automatically when they reach a threshold, so after fully charging you can get the meter aligned to reality (200Ah = full).

This does not explain why the solar charge controller is not charging however because that is hooked up correctly and should be recorded by the shunt. The simplest check is making sure both switches and both breakers in the path of the solar current are on. If SCC charging is still not happening, getting your new meter out and testing each location in the chain would be my next step.

Looks like you are almost there!
Jon

 

[Beowulf5g6]

Looks like a nice system and I expect it will work well.
A few thoughts:

Yes, that meter you linked to should work. The beauty of a multimeter is you can dig into the different parts of the system. Set it to DC volts and then you can check the input and output voltages for the SCC to be sure your panels are wired correctly, then you can test down at the bus bars to be sure your switches are on and finally at the batteries. The clamp can be put around any single wire in the circuit to measure current flow at that point which can tell you a lot.

The main issue for me from the photos is that your negative battery charger clamp (I assume those are on the clamps on the battery posts) should be on the other side of the shunt. I expect that your batteries are getting charged from the battery charger but your shunt and meter are not recording the flow because they are getting bypassed. The 11.9V shown on the meter suggests each of the 4 cells per battery is just under 3 volts which is holding some level of charge (maybe in the 5-10% range so they may already be holding 10-20 Ah).
You can just take the cover off the negative bus bar and put the negative clamp on that. I expect your meter will start counting the charging flow at that point.

Note that your meter will be off from the actual battery charge. Generally these meters allow you to reset the charge percentage to 0% or 100% when you know you are there, some do it automatically when they reach a threshold, so after fully charging you can get the meter aligned to reality (200Ah = full).

This does not explain why the solar charge controller is not charging however because that is hooked up correctly and should be recorded by the shunt. The simplest check is making sure both switches and both breakers in the path of the solar current are on. If SCC charging is still not happening, getting your new meter out and testing each location in the chain would be my next step.

Looks like you are almost there!
Jon

Thanks Jon! I moved the negative clamp to the bus bar and I see some movement now.

New multimeter should be here today so I can start investigating further!

 

[Beowulf5g6]

I removed the battery charger and turned the solar back on. Here is what it's doing:

 

[Jono14]

So it looks like the solar is charging a little. 14.6V for the panels might be appropriate if the panels are shaded or sun is at an extreme angle but it seems low if they are in good sun.
Do you know the Vmp and Imp rating of the panels (voltage and current at max power)? If you are getting full sun on all panels there could be a wiring problem.
With your new clamp ammeter, you can see what the current is coming from each panel, and you might disconnect the panels plugs and test Voc (open current voltage) for each panel to be sure they are good.
Also, you can confirm the connections are good through the chain by showing that voltage is pretty much the same on both sides of each switch and breaker (always possible to get a bad one that is introducing voltage drop through resistance.)

 

[Beowulf5g6]

The imp is 8.61A and the vmp is 20.9V. It is cloudy today but tomorrow should be a sunny (and hot) day here in Vegas. Clampmeter still hasn’t arrived.

Would you recommend I run the panels in series or parallel?

 

[Jono14]

So there are 3 ways to arrange your 4 panels. From your posts I am figuring they are 180W each and you have 4 of them.
1) You can run them all in series (4S)
2) You can run them all in parallel (4P)
3) You can make 2 series strings and then parallel the 2 strings (2P2S)
When you parallel the strings, the voltage stays the same and the amps increase, when you put panels in series, the voltage multiplies and the amps stay the same.

So if all panels are operating at max power (never actually quite happens):
Option 1 (4S) will be running at about 84V and 8.61amps
Option 2 (4P) will be 20.9V and ~34 amps
Option 3 (2P2S) will be ~42CV and ~17A

Your charge controller (Renogy MPPT 60A) can handle 140V input by my web search so any of the 3 options are possible.

The main trade off is that parallel is better with shading (if 1 panel gets shaded, the other 3 can still operate fully) but uses higher amps meaning larger voltage loss.
For example, at 34 amps, if you are running 10 AGW wire over a 40 foot round trip, you would have voltage drop of about 6% and loose about 45W to heating the wires. But if you use 6 AWG wire, the loss is about 2.6% and only 20W lost to heat.
For option 1, with just 8.6A, the 10AWG voltage drop is just 1.6% and loss of 3W to heat - 6AWG would lose <1% and almost nothing to heat.

So if all the panels will be in great full sun all the time (no shading issues) then I'd rec 4S for the low energy loss.

But if shading is more of an issue, like the panels are on a roof with other equipment that will shade different panels at different times, then 4P is probably better if you can use thicker wire (the Renogy can handle up to 4AWG). Also, if your runs are <40 feet round trip the issue is relatively smaller.

One thing I skipped is that you really should use Voc (open circuit voltage) when calculating max volts and factor in the panels getting cold (not a typical Vegas issue, but maybe driving north?) because that boosts the panel voltage. I doubt the temp correct Voc is anywhere close to 35V - but that would be reaching the hazard limit of your SCC (140V) if planning a 4S configuration.

 

[Beowulf5g6]

Ok so I tried using the new clamp meter but just getting 0 readings with everything. Not sure what I'm doing wrong. Here is the setting I'm using:

I also switched all the panels back to series. The charge controller says they are producing 4 watts. I then tested each panel individually and it came up as 0 watts. So I'm starting to think again that the panels are all bad.

 

[Jono14]

A few ideas:
- Panels are probably not bad as that is rare.
- In the picture, the meter is set to DC volts which is fine but that only works with the wire leads you plug into the bottom of the meter. (The clamp part only works for current in Amps)
- Plug the leads in, take off the little plastic cover at the bottom of the Rover MPPT so you can access the 4 screws connecting the wires and use the leads to measure the voltage across the left 2 screws (this is the solar panel input voltage) - in sunlight it should be in the 16-22V range for a single panel or up to 80V if you have all 4 in series - (that is getting to be a more serious voltage that can hurt so be careful.)
- If you are getting zero, leave the black negative lead on the 2nd screw from the left and move the positive red lead to the input nut on the upper breaker. If you see voltage there, the switch or breaker is not working or not set correctly.
- If you still get zero volts coming in from the panels, go out to the individual panels when in sun and try disconnecting them and test the 2 leads of 1 panel with the 2 leads of the volt meter. Only if that shows zero would I question the panels.
- Assuming you get a proper voltage on a single panel, you can also connect the 2 leads of the panel to itself (briefly) creating a short circuit and switch the meter to the 2nd setting "40A" and confirm it shows "DC" on the screen (toggle with the "FUNC" button if not) then put the clamp around the wire to see a current. In bright sun, you should get something approximating the Isc rating, maybe in the 9A range, but I'd be happy with anything over 4-5A.
- Once you can show a proper voltage for your panels making it to the MPPT input, then you can look at the output (right 2 screws) voltage and current from the charge controller and you should be able to explore why that might not be working. Make sure the batteries are not fully charged so that the MPPT is not just appropriately shutting down charging.
- There may be some settings of the MPPT that are not set correctly yet which could prevent charging as well (battery type and voltage thresholds etc.)

 

[Beowulf5g6]

A few ideas:
- Panels are probably not bad as that is rare.
- In the picture, the meter is set to DC volts which is fine but that only works with the wire leads you plug into the bottom of the meter. (The clamp part only works for current in Amps)
- Plug the leads in, take off the little plastic cover at the bottom of the Rover MPPT so you can access the 4 screws connecting the wires and use the leads to measure the voltage across the left 2 screws (this is the solar panel input voltage) - in sunlight it should be in the 16-22V range for a single panel or up to 80V if you have all 4 in series - (that is getting to be a more serious voltage that can hurt so be careful.)
- If you are getting zero, leave the black negative lead on the 2nd screw from the left and move the positive red lead to the input nut on the upper breaker. If you see voltage there, the switch or breaker is not working or not set correctly.
- If you still get zero volts coming in from the panels, go out to the individual panels when in sun and try disconnecting them and test the 2 leads of 1 panel with the 2 leads of the volt meter. Only if that shows zero would I question the panels.
- Assuming you get a proper voltage on a single panel, you can also connect the 2 leads of the panel to itself (briefly) creating a short circuit and switch the meter to the 2nd setting "40A" and confirm it shows "DC" on the screen (toggle with the "FUNC" button if not) then put the clamp around the wire to see a current. In bright sun, you should get something approximating the Isc rating, maybe in the 9A range, but I'd be happy with anything over 4-5A.
- Once you can show a proper voltage for your panels making it to the MPPT input, then you can look at the output (right 2 screws) voltage and current from the charge controller and you should be able to explore why that might not be working. Make sure the batteries are not fully charged so that the MPPT is not just appropriately shutting down charging.
- There may be some settings of the MPPT that are not set correctly yet which could prevent charging as well (battery type and voltage thresholds etc.)

Okay so I measured the voltage from the rover as you suggested, and it was 79 volts. I then tested everything down to the batteries and it was all 14.3 volts.

The only setting I changed in the MPPT was the battery type and capacity I think. Everything else I left as the default.

 

[Jono14]

Cool - that's all encouraging!
Assuming you are still seeing minimal charging from the panels, it seems your batteries may be full. If they are truly at 14.3 then they are functionally at 100% and your Renogy MPPT is appropriately holding off sending them any further current.

Alternatively, if your battery charger is still connected and running, it may be holding them at 14.3 and they are not quite full. Your battery monitor should be able to give you a hint as to how full they are (but remember it missed a bit of the charging when the battery charger was initially on the battery directly.) So if it is saying 80% and the voltage is 14.3, might be time to manually reset the monitor to 100%.
Either way, when trying to sort out the MPPT function, disconnect the plug-in charger so they are not competing while you are doing diagnostics.

If you think the 100% theory makes sense, you can test the MPPT by turning on the inverter and run something high powered for a while to run the battery down a bit (the monitor can track the progress) and then give the MPPT another chance when the batteries have some space.

It actually might all be working!
Jon