5V voltage difference for MPPT needed for startup or normal operation?

Q&A and troubleshooting
, ,
1

My system runs with a 2s2p PV configuration (1620 Wp) and a 16s LiFePo battery using an MPPT SmartSolar 150/45. Last summer, this setup worked without noticeable issues. This summer I noticed that the power output stays well below expectations (at about 50%-60%). Initially I suspected an MPPT hardware issue so I replaced the MPPT with a brand-new device. This did not change the situation.

I discovered the following today:

I noticed that the MPPT keeps the string voltage always approx. 5V over the battery voltage. The Vmp of the cells is 31.2V or 62.4V - so this is close to the battery voltage (~54V) but well above what Victron requires for startup.

After noticing this I confirmed by re-wireing my setup to a 3s configuration. This puts the Vmp well above the battery voltage. I found that the output power reached an expected level.

According to Victron a 5V Voltage difference is needed for MPPT startup but I did not expect that the MPPT maintains this constraint after starting up. Instead Victron outlined here around the 6:00 minute mark that 1V is sufficient https://www.youtube.com/watch?v=uTMVmW7GTus

I remember that I updated the firmware in spring so I suspect that firmware changes are the reason why my setup worked last summer but acts up this year. (*)

All tests were done on a perfect day in the middle of a heatwave close to the summer solstice around mid-day.

Is the 5V minimum voltage difference between PV voltage and battery voltage is mandatory for proper MPPT operation on the current firmware? And if so, would this explain the reduced performance I observed with the lower-voltage PV configuration?

(*) I tried to downgrade to an older Firmware version but the MPPT does not properly connect anymore to the VictronConnect app. It is stuck in “Fetching Data”. MPPT reboots did not change this. The used app version is v6.34 on an Android Pixel8a. (I crosschecked on a different Android device as well)

This is the situation running 2s2p. You can see the 5V voltage difference which limits the max power in this case:

image
image1009×1286 72.8 KB

This is the situation running in 3s.

image
image1170×1342 88.1 KB

2

You are aware that panel voltage drops as the panel temperature rises? Likewise as panel temps drop below 25c (77f) panel voltage rises.

Where I live panel temps get up to 80c (176f), resulting in panel voltages dropping by `8% or so (depending in panel temp coefficient specs).

Better do your PV array calculations again. Or use Victrons mppt calculator.

3

The MPPT needs Vbat +5V on the PV side in order to start tracking the MPP. After that, the PV voltage needs to stay above Vbat +1V to continue running the MPP algorithm.

While i would also expect the algorithm to properly track at a very low PV-to-Vbat delta, its very well possible that it still struggles. I mean, it did continue to work and charge, but just not very well.

Also, yes use the MPPT calculator to calculate your system, you may be able to run all four panels in series.

4

I think this is not true. It may have been true in the past or it may be a misinterpretation.

The manual states “1b) The PV voltage must exceed Vbat + 5V for the controller to start. Thereafter the minimum PV voltage is Vbat + 1V.” This makes statements about controller start and stop points. It does not state that the controller tracks the MPPT point in this range

Additionally it may have been true in the past. This behavior could have changed with a recent FW version.

From today: My string voltage was about 10V higher then the battery voltage today in the morning (Cold Cells, Empty Battery). This means the MPPT should have started to track the MPP. Over the course of the day when the cells heated up and the battery voltage increased the delta of 5V started to be the limiting factor again.

5

Ok, i can understand the argument here. That might be true. Just because the MPPT has started charging does not mean that it can actually track an MPP at that low voltage level. MPP algorithm surely is something that could change with the firmware.