the use case
- 600w of panel
- a 40A controller
- has recently swapped out a dying lead-chemistry bank for LFP
Their main concern is whether or not the solar can charge the lithium sufficiently. It’s not the topic of this backchannel post, but IMO if the loads remain the same and the solar was charging the lead fully then it it will be more-than-sufficient to charge the LFP. Lithium’s charging efficiency is practically 100% while healthy flooded lead-acid is ~85% and AGM/gel ~90%. Toward the end of their life the respective efficiencies might be as low as 70% / 80%.
You’d want a 60A controller
First off, I have a saying about people who tell you what you want.
Secondly, 40A is appropriate for 600w of panel in most use cases. 60A is never appropriate. Let’s look at the math.
The worst case scenario is lowest normal battery voltage combined with optimal solar harvest.1
The most conservative approach2 formula is RATED SOLAR WATTAGE / NOMINAL BANK VOLTAGE = SOLAR CHARGE CONTROLLER AMPS
- lead bank: 600w / 12.0v = 50A
- LFP bank: 600w / 12.8v = 46.9A
Already we can see that 50A is the largest we would need under any forseeable circumstances. 60A is a waste of money, space, and weight.
In the real world panels don’t make rated power under normal circumstances.
After bucking losses, wiring losses, cell temp derating, etc, MPPT-powered systems typically make ~85% of rated at local solar noon and under good conditions. So let’s work the formula again with this in mind:
- LFP bank: 600w x 0.85 / 12.8v = 39.8A
40A is looking pretty good about now.
As discussed before, local solar noon and low bank voltage rarely intersect.3
Absent huge loads, when LFP banks have active charging they are typically ≥13.4v. Let’s do the formula again with this in mind:
- LFP bank: 600w x 0.85 / ≥13.4v = ≤38.1A
but cloud effect!
Yes, cloud-edge effect is real. It is also highly transient. We do have to size PWM controllers for such events, but MPPT handles them nicely.
twitter comment thread for this post