MPPT power drops when battery voltage drops due to high load

Question

question

mboye asked Sep 6, '21

MPPT power drops when battery voltage drops due to high load

My cabin is equipped with 2x 100Ah LiFePo4 batteries, a SmartSolar 100/50 MPPT, and a Phoenix Inverter 12/1200. The system performs very well, but recently I noticed that the MPPT appears to get confused during high loads where the battery voltage sags a bit.

The below screenshot shows the effects of a induction cooktop cycling on and off in order to keep a certain temperature while boiling potatoes. The MPPT output is steady at maximum output until the cooktop turns on, at which point the output collapsed to 0 watts. At the same time the battery voltage drops due to the 1000w load of the cooktop. The MPPT output does not recover until the cooktop turn off again.

Once the potatoes were cooked and the cooktop turned off, the MPPT output returned to its maximum output level and the output remained stable.

Is this expected behavior? It would be ideal if the MPPT actually delivered maximum power when it is needed the most i.e. during high load.

MPPT Controllers Phoenix Inverter

1630957023786.png (304.1 KiB)

1630957359757.png (296.9 KiB)

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mboye answered · Feb 04, 2023 at 06:14 PM

I eventually solved my problem and though I would share my solution.

In my original system I was using a bus bar from Blue Sea Systems with only 2 studs and this forced me to connect both the inverter and solar charge controller onto the same stud. When running a large 230V load, the solar charge controller experienced such a significant voltage drop that it likely caused the MPPT algorithm to restart, thus not providing any solar power when the large load was on.

I rewired my system using a Victron Lynx Distributor, and the inverter and SCC no longer share the same connection point on the bus bars. The screenshot below shows how the output of the SCC barely drops when the large AC load hits the inverter. The battery voltage also drops much less, so overall I am very happy with my improvement and the Lynx bus bar.

1675534262962.png (431.1 KiB)

2

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Alexandra ♦ commented · Feb 04, 2023 at 06:26 PM

@mboye

Thank you for coming back and sharing the solution and cause of the issue.

0 ·

klim8skeptic ♦ commented · Feb 04, 2023 at 06:29 PM

Thanks for the feedback.

0 ·

Modifications Space

Node Red Space

Answer 1 · 2021-09-09T06:47:15Z

Thiemo van Engelen (Victron Energy staff) answered · Sep 09, 2021 at 06:47 AM

Hi

When looking at the time scale of the graphs, it seems that the high power burst are rather short and the control algorithm in the MPPT is not built to respond that fast.

On other thing that I notice is that the voltage drops quite a lot. I would not expect the battery voltage on the MPPT terminals to drop that much. How is everything wired? Is the MPPT directly connect to the battery or is it connected to the inverter battery terminals? Can you perhaps make a picture of the wiring?

This also related to your problem, because when the voltage drop on the MPPT terminals would be limited, the produced power would also be more constant.

Kind regards,

Thiemo van Engelen

1

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mboye commented · Sep 14, 2021 at 06:33 PM

Hi,

The high load generated by the induction cooker came in burst approximately every 10 seconds as far as I recall.

This is drawing of my system:

Last weekend I checked the voltage in a couple of points while brewing coffee to investigate the voltage drop a bit further. I was unfortunately not able to observe the SCC behavior because it was not a sunny day.

Measurement location	Load	Voltage
Battery terminals	0 A	13.3 V (SoC was 60%)
Battery terminals	98 A	12.85 V
Battery switch	98 A	12.4 V
Inverter terminals	98 A	12.23 V
SCC terminals	98 A	12.27 V

The voltage drop from the battery to the inverter under load is 0.62V. Is this considered an acceptable loss?

I realize that the 1.5m of cable from the battery switch to the battery is not ideal, and I am planning on shortening it at some point. However, according the the Phoenix 12/1200 manual 35 mm2 should be more than good enough for the distance between interverter and battery.

The busbars are rated for 150A continuous current so they should not be heating up much.

The inverter and SCC share the same studs on the busbars. Could this cause problems under load?

0 ·

cabin-solar-system.png (45.4 KiB)

Answer 2 · 2021-09-09T11:14:15Z

raymiller answered · Sep 09, 2021 at 11:14 AM

Over the last two days my Smart MPPT 100/20 has been behaving in a similar fashion. I'm not sure what the load was doing but the solar was going up and down due to full sun and then isolated clouds. The output dropped to 2W and stayed there, I managed to reset it by briefly opening the PV breaker and on two occasions, the MPPT output power returned to normal operation.

It was as though the MPPT just locked up and needed a reset. I've not noticed this behaviour before. I'll try to go over my VRM data to see what is going on.

2

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

raymiller commented · Sep 09, 2021 at 11:33 AM

VRM data recorded interval 10 minutes note the two peaks in PV voltage as the MPPT goes into a low power state for unknown reasons, just after the second peak is where I briefly opened the PV breaker which restored the output current and lowered the PV voltage. It does seem the Smart MPPT 100/20 controller is shutting down for some reason?

0 ·

1631186718295.png (71.5 KiB)

JohnC ♦ raymiller commented · Sep 09, 2021 at 01:56 PM

Hi @RayMiller

In your case it could be just the mppt tracking, which it does every 10 minutes, and takes a few seconds to complete. You may have been catching the edge of it when it happens to line up with your 10 minute sampling.

This is mine today, clear sky, constant load (per-minute sample)..

It's normal, and can be seen quite clearly in VictronConnect, with it's per-second sampling.

And clouds can change where it tracks too. For me that pv V can range from 60 to 115V.

0 ·

1631195011333.png (15.7 KiB)

question