Charging one battery from another using a Solar charge controller?

[Karottemkopf]

I had an idea of how to charge one battery from another with higher voltage until their voltage is equal.

Connect the one with lower voltage at the batter connection of a Solar charge controller like normal.
Connect the higher voltage battery to the solar PV input.

As I understand it, Solar charge controllers are just DC/DC converter, so it should not matter if the Input voltage comes from a Battery or PV panel.

The only possible issue I see is that the Battery could supply at lot off current, but they have an overcurrent protection that is trigger when the MOSFETs inside overheat.
Maybe to be safe, adding a break or fuse is a good idea in case the overcurrent protection fails.

As anyone ever tested this?

 

[sunshine_eggo]

Using a MPPT as a battery charger?

can I use a 75/15 as a 24v to 12v battery charger? both batteries are Lifepo4.

diysolarforum.com

In almost all cases, MPPT input voltage has to be notably higher than charge voltage.

 

[Karottemkopf]

Using a MPPT as a battery charger?

can I use a 75/15 as a 24v to 12v battery charger? both batteries are Lifepo4.

diysolarforum.com

In almost all cases, MPPT input voltage has to be notably higher than charge voltage.

I would be using a PWM SCC.

 

[sunshine_eggo]

Nope.

 

[Vigo]

I just replied to your post in this other thread with the answer to that exact question: https://diysolarforum.com/threads/using-a-mppt-as-a-battery-charger.41303/page-5

post #82 i said:

The problem is the controller doesn't know anything until AFTER the first 'pulse', and if there is very much voltage difference between input and output, and the input is a battery capable of supplying hundreds, perhaps into the 1000a range, that first pulse = you let the smoke out. I've verified this personally..lol

If you think about a 12v typical car battery, hooked up to something with 12 less volts.. almost a dead short.. lots of them will touch 1000a in that scenario. Some bigger ones will do 2000, some bigger AGM claim 3000+a. How many microseconds of that does it take to flash boil the guts in the transistor trying to flow that current? One. 1 microsecond. 36v into 24v, same deal, doesn't matter. You've got a source with huge ampacity hooked to something with much lower volts and very low resistance in the circuit = instant excitement.

It can be made to work with much finicky bs. It would be easier to skip the bs and buy an mppt controller with current limit settings, as they 'ramp up' versus simply shorting the input to the output.

 

[sunshine_eggo]

Pretty sure no PWM will adjust duty cycle to regulate current. It's my understanding they only regulate voltage via duty cycle. They must be rated for whatever input current is attached.

 

[Vigo]

I recently bought 2 of these (a 20a and then an 80a) which do have a current limit setting:
PWM scc

However, i still fried the 80a one due to the issue i explained. Yeah, that's recent learning i'm recounting there!

I don't know functionally how it is regulating current. I do know based on my clamp meter that they are both somewhat inaccurate on the amp reading and the 80a is a lot more inaccurate than the 20a model.

I've been using the 20a model with an adjustable 0-24v meanwell knockoff power supply as input, set to 16a for a week or two, then upped it to 18a and neither device has failed yet. The SCC gets pretty hot, the power supply does not.

I took apart the 80a model and it really is built 3-4X bigger than the various cheap 'small form factor' ~$20 pwm controllers i've opened up. It is very plausibly not a POS. However, i haven't tested it anywhere close to 80a yet. A sufficiently large charge source that isn't a battery should be arriving tonight, so more to follow.

But i CAN CONFIRM that it will not sufficiently regulate current on the FIRST pulse to keep from frying if you hook up a stupidly overpowered input to it. Like my 48v house pack.

 

[MattiFin]

I would be using a PWM SCC.

Extremely bad idea. PWM SCC works properly only with current sources like solar cell. Battery will provide way too much current without limit.

 

[Karottemkopf]

I recently bought 2 of these (a 20a and then an 80a) which do have a current limit setting:
PWM scc

However, i still fried the 80a one due to the issue i explained. Yeah, that's recent learning i'm recounting there!

I don't know functionally how it is regulating current. I do know based on my clamp meter that they are both somewhat inaccurate on the amp reading and the 80a is a lot more inaccurate than the 20a model.

I've been using the 20a model with an adjustable 0-24v meanwell knockoff power supply as input, set to 16a for a week or two, then upped it to 18a and neither device has failed yet. The SCC gets pretty hot, the power supply does not.

I took apart the 80a model and it really is built 3-4X bigger than the various cheap 'small form factor' ~$20 pwm controllers i've opened up. It is very plausibly not a POS. However, i haven't tested it anywhere close to 80a yet. A sufficiently large charge source that isn't a battery should be arriving tonight, so more to follow.

But i CAN CONFIRM that it will not sufficiently regulate current on the FIRST pulse to keep from frying if you hook up a stupidly overpowered input to it. Like my 48v house pack.

I think i will put a power resistor ion the middle and in my case there should never be much power flowing as both are 24V batteries.

 

[Will Prowse]

I used to do silly stuff like this and it didn't work. Due to mppt algorithm, and no current regulation on battery, it will push to absolute limit and typically shut down. Last time I tried this was battery to battery charging a Tesla 24V pack. Caused an over current safety trip.

I do not recommend trying this ever. Very dangerous. The PV input is designed for use with panels, not a battery.

 

[Vigo]

I think i will put a power resistor ion the middle and in my case there should never be much power flowing as both are 24V batteries.

As long as the state of charge isn't massively different between the two!

I do not recommend trying this ever. Very dangerous. The PV input is designed for use with panels, not a battery.

So far i have only experimented with PWM controllers on 12v systems. I agree that it could be pretty dangerous if you did something that caused a overheat or short leading to fire, especially IF it didn't happen the first instant you hooked it up..

But given how absurdly cheap some of these PWM controllers are, they beg to be messed with for 'off label uses'. For example, you can buy a complete solar controller for only a few bucks more than you can buy a single adjustable voltage switch module which is just a populated board with no case, no screen, can't be adjusted without holding an actual tool, and it's just ONE setpoint switching between two things. A solar controller on the other hand has multiple adjustable setpoints for both high and low voltages, between three things (pv/batt/load), along with the case, the screen (sometimes backlit), and 'no-tool' adjustability through the UI. As far as i know it's by far the cheapest available form of voltage 'window switch' with the level of functionality it has and when hooked up to control external devices like relays etc can be used to control some pretty tricky stuff. It doesn't really have to be hooked up to panels or batteries since it can't tell one from the other anyway. It could be looked at as just a fancy adjustable switching device you can get for $12 with one day shipping.

But of course if you use it for unintended purposes you have to take ownership of the risks, and i think someone in your position has to be especially sensitive to people taking everything you do as a recommendation or 'implicit advice' regardless of it never being expressed that way. So i understand your stance on this.

 

[Risky Rob]

I was wondering if this was possible. Glad I didn't try it.

 

[Karottemkopf]

I used to do silly stuff like this and it didn't work. Due to mppt algorithm, and no current regulation on battery, it will push to absolute limit and typically shut down. Last time I tried this was battery to battery charging a Tesla 24V pack. Caused an over current safety trip.

I do not recommend trying this ever. Very dangerous. The PV input is designed for use with panels, not a battery.

It seems that I'm not the first person to to have the idea that a SCC might be a cheap DC/DC battery charger.
But with allthe feedback I've gotten I won't try it.

In my case both are 24V battery but the might be at different SOCs. Parrallel connecting over a power resistor to let them equaliz semes like a safer option.
It is less efficient but probably still more efficient then going through the inverter and then a battery charger in my case where there is a very large difference in capacity between the two batteries.

 

[OzSolar]

Many years ago, with Ouback's blessing I've used an Outback MX60 charge controller to charge 12v batteries from 24v batteries for some testing that we needed to do in the shop. It was rather clumsy and there are other ways to do it but it's all we had at the time.

 

[guerby]

I used to do silly stuff like this and it didn't work. Due to mppt algorithm, and no current regulation on battery, it will push to absolute limit and typically shut down. Last time I tried this was battery to battery charging a Tesla 24V pack. Caused an over current safety trip.

I do not recommend trying this ever. Very dangerous. The PV input is designed for use with panels, not a battery.

What would you recommand to transfer energy safely (and not too $$$) from one battery to another?

 

[Vigo]

What types of batteries are they? Depends on how far apart their voltages are, what their capacities and chemistries are, whether you can just parallel them to each other. A lot of same-nominal-voltage, similar capacity batteries can be paralleled to each other even with very different SOCs because the 'charge rate' would still not be dangerous or abusive. But there are certainly lots of scenarios that fall outside that window, and any time one battery is 10V+ higher than the other (24 to 12, for example) you REALLY should not ever directly parallel them.

 

[Karottemkopf]

What types of batteries are they? Depends on how far apart their voltages are, what their capacities and chemistries are, whether you can just parallel them to each other. A lot of same-nominal-voltage, similar capacity batteries can be paralleled to each other even with very different SOCs because the 'charge rate' would still not be dangerous or abusive. But there are certainly lots of scenarios that fall outside that window, and any time one battery is 10V+ higher than the other (24 to 12, for example) you REALLY should not ever directly parallel them.

The stationary battery is a 8s LFP 20-28.8V 20Ah.
The battery that will be connected and disconnected, frequently a 12s 21-28.8V 280Ah lead acid battery in a electric forklift.

 

[stienman]

If you study what an MPPT charge controller does, you'll find that it won't work well - or possibly at all - with a battery as an input.

However, using old school trickery, you might be able to make it work with significant losses.

Put a filament lightbulb in series with the input. This will consume more power the more current goes through it, and by the brightness you'll be able to tell if your losses are high. The MPPT algorithm will find a decent happy point, and will charge the output cells accordingly. Set the charger's current limit according to the output cell's ability to accept a charge. The lightbulb will act as a fuse and burn out if it's passing "too much" current - and figuring out whether you want a 60w, 100w, 250w, or perhaps a 12v vehicle headlamp will be an exercise left to the reader, but you can always experiment, and as long as you have proper fusing on the input and output (separate from the bulb itself!) then you aren't likely to damage anything in the process.

But your efficiency isn't going to be much better than a DC-->AC inverter with an AC-->DC charger following to do the conversion, both of which you probably already have. Getting a purpose built DC-->DC charger doesn't seem cost effective unless you're going to be doing this often (such as charging a vehicle battery from an ESS battery, skipping the whole DC-->AC-->DC conversion), but having a wide range input DC battery charger on hand may prove useful at other times - such as the iCharger X6 or X8.

 

[MattiFin]

What types of batteries are they? Depends on how far apart their voltages are, what their capacities and chemistries are, whether you can just parallel them to each other. A lot of same-nominal-voltage, similar capacity batteries can be paralleled to each other even with very different SOCs because the 'charge rate' would still not be dangerous or abusive. But there are certainly lots of scenarios that fall outside that window, and any time one battery is 10V+ higher than the other (24 to 12, for example) you REALLY should not ever directly parallel them.

And quite often same-nominal voltage batteries with similar chemistry don't charge at all if connected parallei. Or charging is extremely slow.
Required charge voltage above 10% charge level is higher than voltage of the full battery and charging won't happen or is extremely slow

https://www.custommarineproducts.com/uploads/1/4/3/7/1437708/published/lifepo4-discharge-curves_2.png

Lead-acid is even worse, empty battery jumps above 13 volts almost straight away and full battery under load is only 12.8v

 

[Karottemkopf]

If you study what an MPPT charge controller does, you'll find that it won't work well - or possibly at all - with a battery as an input.

However, using old school trickery, you might be able to make it work with significant losses.

Put a filament lightbulb in series with the input. This will consume more power the more current goes through it, and by the brightness you'll be able to tell if your losses are high. The MPPT algorithm will find a decent happy point, and will charge the output cells accordingly. Set the charger's current limit according to the output cell's ability to accept a charge. The lightbulb will act as a fuse and burn out if it's passing "too much" current - and figuring out whether you want a 60w, 100w, 250w, or perhaps a 12v vehicle headlamp will be an exercise left to the reader, but you can always experiment, and as long as you have proper fusing on the input and output (separate from the bulb itself!) then you aren't likely to damage anything in the process.

But your efficiency isn't going to be much better than a DC-->AC inverter with an AC-->DC charger following to do the conversion, both of which you probably already have. Getting a purpose built DC-->DC charger doesn't seem cost effective unless you're going to be doing this often (such as charging a vehicle battery from an ESS battery, skipping the whole DC-->AC-->DC conversion), but having a wide range input DC battery charger on hand may prove useful at other times - such as the iCharger X6 or X8.

There are other reasons why i don't want to use the Battery charge from AC.
- The as the second battery that is connected and disconnected frequently has much lager capacity, it would always discharge the first completed and then there is no more energy for anything else
- I want the option to power the inverter from the second larger battery when it's connected

Efficiency is difficult to compare because the two conversion have a constant power loss.
Compared to connecting them in parallel over a small resistance only waste power during voltage equalization, and then I can connect them directly and no more power will be lost.