1
\$\begingroup\$

I recently purchased a new RV with a 12V system that has the following components:

  1. A single lead acid battery
  2. The usual 12V lights, pumps and other loads
  3. A solar charge controller
  4. A 120VAC to 12VDC charger (500W) for when the RV is plugged in

I'm adding a 24V LiFePO4 (8S 280Ah with a BMS) battery bank and later a 2kW 24V inverter and I'd like to be able to charge it from the 12V system, but also have it provide power for the 12V loads.

The reason to go with a 24V battery bank configuration is twofold. First, it's to minimize the amount of current going from the battery to the inverter. A 2kW inverter would require 166A which is just a silly amount to me. 80A is more doable, it will keep the conductor size down and have less losses at the terminals. The second is that it's my understanding that it's better for the cells to wire them in series as they will share the load equally and it should keep them balanced better.

Obviously I'll need some DC-DC converters. I've searched for "bidirectional dc-dc" and it appears to exist as a concept, but there are almost no ready-to-buy products available. Ideally I'd like to have a converter that will charge the 24V bank when the 12V side is at 13.2V and above as that means that either the shore charger is plugged in or the solar is producing more power than what's being consumed, and, vice versa, when the voltage drops below 13.0V, it should pull power from the bank to power the loads and float the lead-acid battery.

Could this be accomplished with two DC-DC converters wired in parallel? It seems like there's the danger of a feedback loop. The 12V-24V converter should also only turn on at a preset supply voltage (13.2V in my example) and I'm yet to find a device like that. Anything else I should look out for?

Another option is to separate the loads and the chargers on the 12V side so that loads are only connected to the step-down converter and the chargers to the step-up converter, but it would require running quite a lot more wires due to the location of all of those things.

\$\endgroup\$
9
  • 1
    \$\begingroup\$ It really sounds as though you are seeking a commercial product recommendation to solve this. But you are also willing to consider combining several different commercial products if that gets you there. What I don't see much of here is a design question, though I think answers could certainly refocus on that rather than commercial recommendations. You could also consider a way to stack two 12 V systems to get 24 V, while unstacking them for charging purposes. But I can't even tell if that's something you'd consider from what you write. \$\endgroup\$
    – jonk
    Commented Nov 25, 2022 at 22:07
  • \$\begingroup\$ I guess what I'm really asking is whether I'm missing something obvious. It seems like a common problem to have, but the lack of commercial products is usually a sign that some of my assumptions are wrong. If I understand your suggestion correctly and you mean wiring the lithium bank to be switchable between 8S and 4S2P, it seems like it would be a manual process and not allow me to use the inverter while charging from solar. Not ideal, but getting a 24V solar charge controller would make it a workable solution. \$\endgroup\$
    – Gunchars
    Commented Nov 25, 2022 at 22:27
  • \$\begingroup\$ 12 V has been around since I was born. And there are four times as many people on this planet since then. So a long time and then some. 24 V is a little more recent and while remodeling a 12 V design to work as a fully functional dual DC system may be a thing of sorts, it may not be sufficiently so in order to attract large investments. \$\endgroup\$
    – jonk
    Commented Nov 25, 2022 at 22:33
  • \$\begingroup\$ Reading between your lines, I think you are saying you want a 24 V battery system that can be charged from a 12 V charger while also using it to supply 12 V loads, as well. Yet it is 24 V. Details matter and I don't think you provide enough, yet. But perhaps it's just my own failures in reading well. I'm not sure if you plan to add 24 V loads, or not. Probably, or else why are you bothering at all? But what do I know? Why are you switching to 24 V? I don't know. Is it copper wiring costs? Anyway, I think you could write a lot more about the larger goals and it would help some. \$\endgroup\$
    – jonk
    Commented Nov 25, 2022 at 22:37
  • \$\begingroup\$ I very briefly mentioned it's to run a 24v inverter, but I just edited the question to expound on that aspect. Anyway, your mention of "a dual DC system" lead me to some promising results as it looks like semi-trucks now commonly come in such a configuration. Sometimes it's just a matter of having the correct language. Thanks! \$\endgroup\$
    – Gunchars
    Commented Nov 25, 2022 at 23:07

2 Answers 2

0
\$\begingroup\$

it's better for the cells to wire them in series as they will share the load equally and it should keep them balanced better.

They don't. They do get unbalanced. That is why you MUST have a battery management system, not only to protect the cells, but also to balance the series string.

Could this be accomplished with two DC-DC converters wired in parallel?

No. I'd be like a dog chasing its own tail.

separate the loads and the chargers on the 12V side so that loads are only connected to the step-down converter and the chargers to the step-up converter

That's the way.

Have three buses:

  • 12 V bus connected to the car alternator, non-critical loads
  • DC-DC charger from 12 V to 24 V (charger = current limited)
  • 24 V battery with a BMS
  • 24 V bus connected to the inverter
  • DC-DC step-down converter from 24 V to 12 V
  • 12 V critical loads
\$\endgroup\$
2
  • \$\begingroup\$ I know that even series cells will go out of balance which is why I did get a BMS with a passive balancer, but are you sure that series vs parallel makes no difference in this regard? One poorly crimped terminal in the parallel configuration can create a voltage drop of several hundred mV affecting the current draw from the strings while in the series configuration it's not a good thing either, but at least the current stays the same. \$\endgroup\$
    – Gunchars
    Commented Nov 25, 2022 at 23:52
  • \$\begingroup\$ In parallel it will actually balance itself a bit because cells that are are discharged less will have a higher voltage and so will be discharged quicker. In serial the imbalance is stable and will get worse. Bad connectors have nothing to do with that. \$\endgroup\$
    – Boldar
    Commented Nov 26, 2022 at 1:05
0
\$\begingroup\$

It is possible to use a buck converter in both directions. If the duty cycle is 50% and the batteries have their nominal voltage, there is no energy transfer, just component losses.

Small variations in the duty cycle create an effective current upstream or downstream. Such a system needs a closed loop current control regulation to avoid coil saturation.

The simulator allows playing around with different duty cycles using P1.

This is not a real world schematic, it just shows a concept. Three such buck converters driven with 120° phase shift can deliver 500W without ugly high currents per MOSFET.

There are topologies with better efficiency, but to make them bidirectional is high art. LT has some controllers to build 2 quadrant current sources like this, often used to drive TEC elements.

Transient response is not important in this application, so a simple microcontroller can create the needed wave forms as well.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.