diy solar

diy solar

MPP LV-2424 uses only a fraction of available power to charge

jailcee

New Member
Joined
Feb 11, 2021
Messages
22
This past February I bought a LV-2424 I was recently able to install it at my off-grid cabin. It has been running for over a week now and I'm seeing an issue and would like to pick the brains of you folks if this is normal or potentially a faulty unit.

My system uses 4x 425 watt panels in a 2S2P configuration with a Voc 49.09 and Isc 10.78, Vmpp 41.39 and Impp 10.27 (qcells q.peak duo L-G8.3 425). This is a roughly 1.7kw array. It is connected to an 8s 280ah LiFePO4 battery with an Overkill Solar 24v BMS.

At full sun, my panel array should be generating around 20A at around 90V. The issue I'm seeing is the following:

At full sun the LV-2424 shows, in the PV1 voltage and charger power, around 88-90V input but only 200-290 watt output. I noticed that when the inverter is powered on and off or if the battery is disconnected and reconnected (through a breaker) the LV-2424 will then accept about 480 watts for a time, then switch back down to something in the 200-300 watt range. One time I also saw it ramp up and show up to 900 watt then I heard a switch or breaker go off in the unit and it completely cut off the watts to zero, then slowly let them come back up to the 200-300 range.

By looking at the BMS I see matching data in terms of charge energy, with the battery getting about 10 amps most of the time (at full sun), and up to 18 amps for a short while after a reset. The BMS can handle up to 100amps and the cells are balanced but low at 3.36 volts each so there is capacity in the battery to take charge. The LV-2424 is supposed to support up to 60amps of charge but I can't get close to that. It also seems like the LV-2424 is stuck in the "float" charge rather than the "bulk" charge, but even then is using a fraction of what the panels are generating.

Is this behavior from this unit typical or do I have a faulty unit?
 
Do you have any loads running? Are the batteries already fully charged?
What do see if you have 500W load running?
What are your settings?
 
The only load I've had on it is a small refrigerator which draw about 80w but I can certainly see what happens when I run a 1500w heat gun. What I did see about the refrigerator load is that when I remove it (by flipping a breaker in my AC distribution box) I did not see a difference in the power going to the battery. I'll try the heat gun this weekend.

Batteries are not fully charged with each cell currently at 3.36 volts (or 26.8v).

My relevant LV-2424 settings are:
11 - Maximum charging current: 80A
13 - Maximum utility charging current: 20A *(note that I am off-grid, no utility tie in)
14 - Battery Type: User Defined
17 - Bulk Charging Voltage: 28.4
18 - Float Charging Voltage: 27.2
19 - Low DC cut off battery voltage: 20.4 *(note this is the default for LV2424 and my BMS will kick in before then)
20 - Battery stop discharging voltage when grid is available: 23.0 *(note again I'm off-grid)
21 - Battery stop charging voltage when grid is available: 27
 
Update.

I've put on a bigger load with a heat gun and small appliances when the array is in full sun and I get around 1.5kw from the panels. So the panels are performing as expected, and the 2kw inverter can use that power.

The only question now is why more juice doesn't flow into my 24v 280ah battery. I'm guessing it's because at 27v it just can't accept much more current?
 
Using the xiaoxiand Android app to connect to the BMS I'm seeing 26.2v total, cells in the 3.23v-3.31v range. Telling me 9.15 AH (151w) remaining. :/
 
Update.

The only question now is why more juice doesn't flow into my 24v 280ah battery. I'm guessing it's because at 27v it just can't accept much more current?
I think thats your answer...your battery bank is charged, althought the extra power should be going to the inverter I would think if you have the inverter set up for solar priority (under the option for output source priority).
 
Using the xiaoxiand Android app to connect to the BMS I'm seeing 26.2v total, cells in the 3.23v-3.31v range. Telling me 9.15 AH (151w) remaining. :/
What voltage does the MPP show when you see 26.2V on the battery on the BMS readout?
 
Right now both read at 26.6v. It is 2:30pm on a cloudy day, BMS registering around 2A to 5A of charge current. PV is around 80v and 185w (I have the inverter on and a small refrigerator running).
 
Using the xiaoxiand Android app to connect to the BMS I'm seeing 26.2v total, cells in the 3.23v-3.31v range. Telling me 9.15 AH (151w) remaining. :/
This seems off to me. How's your BMS set up, and can you share some screenshots showing the SOC and voltages?

E: had something about absorb time, but doesn't seem like that model has that setting
 
Last edited:
I wish I could. I thought that perhaps my BMS setup wasn't right so I looked in the settings using the android app, from the overkill website. I saw that it was configured for a 100AH rather than 280AH battery so I changed that. I didn't see an improvement so I looked again and at some point the app crashed/disconnected while uploading settings.
The problem I have now is that although I can connect with the android app, I can't access parameter view or parameter settings. When I try to view parameters the app shows me a progress bar that fills extremely slowly and only ever gets to 49%. Likewise if I try to go into param settings the bar doesn't fill and sometimes just closes and goes back to the dashboard. I have since completely deleted and reinstalled the app on my phone and get the same result.
I seem to be effectively locked out of the BMS settings, even though I can connect to it through bluetooth still - it isn't a password issue.

I've sent overkill Solar an email, and posted to the Reddit sub about it as well but so far nothing.
 
Sounds like you have a BMS issue, that's for sure, however it shouldn't affect the charging and discharging as such.

I'd get a shunt battery meter to keep track of actual SOC.

What's the battery voltage in the morning? (before the solar starts trickling in)
 
Before the sun comes up discharge the battery some and see if the MPP will charge then, also a 100 amp BMS will not have a charge capacity of 100 amps it is more like half.
 
His system might be able to push a bit more than 50A of charge current on a good day anyway, so I doubt that'll be a problem...

He runs a fridge off his system, there should be enough discharge during the night that the MPPT should hit decent bulk charging pretty early in the day.
 
I've seen the battery reported as low as 25.1 v in the morning. Speaking with Steve at Overkill he suggests setting the charge limit to at least 28v. Because I've got my float charging set to 27.2v, and the charger never seems to get into bulk mode, it explains why I only see the battery get up to about 27.1v (even after two weeks of charging with no draw).

I'm not at the system now so I can't try anything until this weekend however I will set the float voltage to the same as my bulk voltage (28.4v) when I'm there next. At the very least it should allow me to get more power into the battery using the float amps.

I'm thinking more and more I might have an issue with my BMS that is limiting / cutting off bulk charging somehow. The fact that I currently can't get into the BMS settings with the Android app just adds to the issue. I could buy a used iOS device to use that app instead, or also try to connect with USB. To do the USB route I'll either need to order it from Overkill, or get something equivalent from Amazon. I'm in Canada so an Amazon order would be much faster. Has anyone got experience connecting to an Overkill Solar BMS using USB to TTL converter and does this unit (DSD TECH USB to TTL Serial Converter CP2102 with 4 PIN Dupont Cable Compatible with Windows 7,8,10,Linux,Mac OSX) seem like it would be compatible?
 
27.1V is still 3.38V per cell, and given enough time, you will reach nearly full SOC that way.
As far as I understand it, the MPP unit shouldn't switch to float until it hits the bulk target voltage for a certain period of time?

Set the MPP bulk charge voltage to 28V and see what happens.

Personally I'd get a shunt battery meter to keep track of battery SOC.
 
27.1V is still 3.38V per cell, and given enough time, you will reach nearly full SOC that way.
As far as I understand it, the MPP unit shouldn't switch to float until it hits the bulk target voltage for a certain period of time?

Set the MPP bulk charge voltage to 28V and see what happens.

Personally I'd get a shunt battery meter to keep track of battery SOC.
If the battery is not balanced the BMS will cut off the charge and the MPP will see a high voltage for a while then the MPP will switch to float is another scenario that can happen he also needs to check to see if his battery is balanced with a voltmeter.
 
@Wibla is there a shunt battery meter you recommend in particular? I know Will reco's the Victron Smart Shunt but I'm wondering if there are others (ideally on Amazon) that you think would work well for my setup?

@labeeman I had top balanced my cells initially but in action I've seen them be up to 0.07v out of balance (according to the BMS dashboard). I'll do a manual voltmeter check this weekend and report back.
 
@Wibla is there a shunt battery meter you recommend in particular? I know Will reco's the Victron Smart Shunt
Will recommends good stuff, I would prefer the Victron unit as well. It's not cheap, but you get what you pay for.

If the battery is not balanced the BMS will cut off the charge and the MPP will see a high voltage
I would think that the MPP would error out if it "lost" the battery during normal operation?
 
Has anyone got experience connecting to an Overkill Solar BMS using USB to TTL converter and does this unit (DSD TECH USB to TTL Serial Converter CP2102 with 4 PIN Dupont Cable Compatible with Windows 7,8,10,Linux,Mac OSX) seem like it would be compatible?
you can easily talk to the overkill BMS via serial...
I would recommend you avoid "clone" serial chips as they tend to drop the connection.
Clone chips, ftdi or even PL's, are just not reliable...
I have zero problems chatting with my overkill BMS using some custom python scripts built on top of a bms python program by eric poulsen.
FYI, I did have to make some changes to eric's code in that the BMS at times did not respond to commands unless I first sent out a block of null characters prior to any serial out cmd (self.s.write(serial.to_bytes(self.outzero))). Tried different serial devices, even the native port on a raspberry pi, port flushing, different threading...nothing worked until I thought of sending what amounts to a "buffer" flush inside the BMS by sending it a bunch of 0's and, well, poof...all fixed)
Eric provides source and the protocol definition so very handy..he does have some prebuilt programs for those that do not want to tweak things.
 
Back
Top