3

Not wishing ever to charge a (prospective) house-battery from the grid, naively, I think of a DC-coupled battery as a sort of capacitance between existing solar panels and their inverter.

Again, naively, I imagine a DC-side battery should be able to smooth the supply to varying load beyond the existing inverter. The inverter shouldn't be able to tell there's a battery upstream.. AC load downstream of the inverter (including export to grid) shouldn't care where the inverter got its DC from.

But domestic installers' information sites always insist that it's prohibitively difficult to retrofit a DC-coupled battery to existing panels and their inverter, and that an AC-coupled battery, with its own inverter, is much to be preferred.

Can anyone explain why?

1 Answer 1

2
+50

DC battery as a capacitance is not a good idea.

Firstly, a DC battery cannot be simply connected to the output of the solar panel like you would connect a capacitor. The DC battery would be destroyed by lack of charge control algorithm. To actually not destroy it, you need a PWM or a MPPT charge controller. PWM is notoriously inefficient, wasting a lot of energy, so by adding the battery plus a PWM charge controller behind the inverter you lose so much energy you'd be better off without a battery.

So, this limits us to MPPT.

MPPT charge controller extracts the maximum amount of energy from a solar panel to charge a battery the fastest you can charge it. If you have DC loads, or AC loads run from an inverter, this is what you want.

However, with a grid-tie inverter the grid is stiff. You can output practically any amount of energy to the grid. The grid-tie inverter extracts as much energy from the solar panels (and your MPPT-charged battery it knows nothing about) as it can. So it prefers to keep the battery empty.

So with a grid-tie inverter this is what will happen:

  1. MPPT charge controller charges your battery as fast as it can from solar power
  2. Grid-tie inverter extracts energy out of the PV+battery system as fast as you can, and since it has a battery, it can extract energy very fast. So grid-tie inverter is working at it maximum power all the time. This prefers to keep the battery empty. The only location where the battery is not empty is that if sun is shining directly on the panels and if you have more panel capacity than inverter capacity. But, it can only happen during cloud-free middays.
  3. There is a blackout.
  4. Your battery is empty during the blackout since the grid-tie inverter is producing power as quickly as it can all the time, keeping the battery empty. Your battery helps nothing during the blackout.
  5. Even if your battery wasn't empty, the grid-tie inverter would stop its production when it sees no grid connection. With a grid-tie inverter, you never get any power during blackouts, even if you have a full battery, since it won't allow feeding power to the grid, there being no grid.

How a DC battery could theoretically work is that if you feed all power through a charger and an inverter like in an online UPS. But this is very inefficient, expect about 70% round trip efficiency. So you don't definitely want this, since your electricity bill will be increased by 1/0.7 = 1.43x. Also the charger and inverter need to be sized according to your fuse size which makes them very expensive.

AC coupling like what is done in for example Victron MultiPlus is far better.

The AC coupled battery has its own inverter. It also measures all currents going through it, measures whether grid is online and can connect/disconnect your isolated electricity island to/from the grid.

So an AC coupled battery can vary its charging and discharging. It can for example run optimized self-consumption algorithms which ensure that electricity produced during daytime is used during nighttime for 0% grid electricity usage, completely eliminating your power bill (apart from the fixed fee). It can also run algorithms to ensure that at least 50% battery state of charge remains for blackouts.

Also, the whenever a blackout happens, the AC coupled battery can notice this and use a relay to disconnect your electricity island from the grid. Then the AC coupled battery maintains the frequency on your isolated electricity island, meaning the grid-tie solar inverters still continue to produce electricity. When the grid becomes available, the AC coupled battery slowly starts to drift the phase and voltage of your isolated electricity island to match that of the grid, after which it connects the relay to return back the grid electricity.

If the solar inverters ever generate too much energy, the AC coupled battery uses its main primary inverter to slowly ramp up the frequency in the isolated grid, to signal the solar grid-tie inverters that it's overloading, at which point the solar grid-tie inverters slowly start to decrease their production, seeing the signal.

Besides, you don't always need an AC coupled battery. You can buy an inverter-charger that can be connected via DC to as large battery bank as you like. For example Victron MultiPlus is just an inverter-charger, and you can connect a lead-acid AGM/gel or a LiFePO4 battery system to it via DC.

1
  • Thanks so much for this.. I've had to look up a couple of abbreviations, fine. It strikes me this could be the answer to quite a range of questions, so full is the explanation. Commented Dec 3, 2023 at 16:46

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.