Float charging batteries: what really happens?

[yahastu]

A typical off grid solar system looks like this:

PV Array -> MPPT controller -> Battery bank -> 240v AC inverter -> loads

Suppose that the system is currently float charging:
"Float charging, also called trickle charging, consists of charging the battery at the same rate that it is discharging. This just keeps the batteries fully charged."

I am curious what is going on inside the batteries when they are float charging like this. Does the current basically just skip over the battery terminals from the MPPT controller directly to the 240 VAC inverter in this case, rather than being converted into chemical energy inbetween?

If so, does this mean that the battery size/chemistry in no way impedes the flow of current - meaning that while float charging, even a really slow charging battery would not present a bottleneck, because the current isn't really going through the battery anyway?

EDIT: The reason I ask is because I want to add a non-critical high amp loads panel that is controlled by a current relay to the PV generation. This way, I can plug non-essential high amperage loads (eg, EV charger) into a special breaker box which is only activated with power when the PV generation capacity supports it directly, without putting strain on the battery system.

I know that I could do this using an AC Coupled system, but I prefer to avoid AC coupled systems if possible because most of them are designed for grid based systems and there can be compatibility issues. So, I'm wondering if this would just work in a regular off grid design.

 

[anorlunda]

Does the current basically just skip over the battery terminals from the MPPT controller directly to the 240 VAC inverter in this case, rather than being converted into chemical energy inbetween?

Yes, that's basically it. Only a small current, typically 1 A, flows into the battery countering self discharge. The rest of the current bypasses the battery.

https://batteryuniversity.com/learn/article/all_about_chargers

Some charge controllers include two sets of outputs, one to the batteries, and a second output for diversion loads. Charging the batteries takes priority, but any power not needed for that is diverted for other purposes. That kind of control is continuous, not binary as in relay open or closed. For example, in phase 2 of charging, the power to the battery is ramped down. Therefore diverted power should ramp up. Also, as generated power varies randomly, such as when a cloud passes over a solar PV, the battery's needs are held constant and the diversion load is varied. It would not be easy for you to duplicate the smarts of those controllers.

Resistance hot water heating is an ideal type of diversion load.

 

[yahastu]

Some charge controllers include two sets of outputs, one to the batteries, and a second output for diversion loads. Charging the batteries takes priority, but any power not needed for that is diverted for other purposes. That kind of control is continuous, not binary as in relay open or closed. For example, in phase 2 of charging, the power to the battery is ramped down. Therefore diverted power should ramp up. Also, as generated power varies randomly, such as when a cloud passes over a solar PV, the battery's needs are held constant and the diversion load is varied. It would not be easy for you to duplicate the smarts of those controllers.

Resistance hot water heating is an ideal type of diversion load.

So would you just need to connect the diversion load terminals directly to a second 240 vac inverter then?

 

[anorlunda]

So would you just need to connect the diversion load terminals directly to a second 240 vac inverter then?

Yes, you could do that. But "second inverter"?

In the simplest case, 100% of the DC power goes to the charge controller. A fraction of that goes to charge the batteries and the rest goes to the diversion port to which you could connect to the one and only inverter.

But you may not want to size the charge controller to handle 100% of the power generated. In that case, yes you could use a less simple rig with two inverters. But be careful of paralleling them on the AC side. See
Control of parallel inverters in distributed AC power systems with consideration of line impedance effect

 

[yahastu]

Yes, you could do that. But "second inverter"?

In the simplest case, 100% of the DC power goes to the charge controller. A fraction of that goes to charge the batteries and the rest goes to the diversion port to which you could connect to the one and only inverter.

But you may not want to size the charge controller to handle 100% of the power generated. In that case, yes you could use a less simple rig with two inverters. But be careful of paralleling them on the AC side. See
Control of parallel inverters in distributed AC power systems with consideration of line impedance effect

If I were to connect the diversion leads to the primary inverter then I would have to also connect my diversion loads to the primary inverter, and if the diverted current was less than the load, it would draw from the battery..which is exactly the situation that I'm trying to avoid.

The battery is there to ensure that essential loads can be maintained in the steady state. If I am fast charging an electric vehicle I don't want to be sucking my battery backup dry to do so. Thats why I was thinking of adding a current relay to the MPPT output, and I would essential only have the EV breaker enabled if I know that my generation can keep up. Your idea of connecting it to the diversion loads is intriguing but I don't see how it would work without running it to a separate isolated inverter.

 

[anorlunda]

it would draw from the battery..which is exactly the situation that I'm trying to avoid.

No. It would draw from the charge controller, not the battery. Remember that the diverted load port on the charger bypasses the battery.

Edit: I should have mentioned. This whole topic is about uninterruptible power supply (UPS). You are making a UPS as a DIY project. You should look up UPS design for more on-target advice.

 

[yahastu]

No. It would draw from the charge controller, not the battery. Remember that the diverted load port on the charger bypasses the battery.

I don't understand what you mean. A single inverter has only a single 240v ac outlet which powers the circuit breaker. How are you proposing to distinguish between loads that are allowed to draw from the battery backup vs loads that are only allowed to draw from the diversion?

 

[Tom.G]

No. It would draw from the charge controller, not the battery.

This discussion sounds like the "charge controller" and the "MPPT" are assumed to be the same thing. In any case, same or different devices, the "diversion load" would connect to one of them; that is, before the battery.