question

info7060 avatar image
info7060 asked

Does MPPT know it current maximal output without load?

Hi,

I'm still waiting for my solar system to be installed but already writing home automation system scripts, so the answer to this will help me a lot.

Let's say that load is 300W only. But there is enough sunshine that MPPT can draw 1000W from the panels. Does MPPT know it, or is it possible to know only after load increase to more than 1000W?

In other words, do I need to increase the load until I know I draw more than sun can provide or is there a better way?

I guess MPPT has no way of knowing it but I want to be sure.

Thanks a lot!

MPPT Controllers
2 |3000

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

4 Answers
joepah avatar image
joepah answered ·

MPPT controllers always optimizing input voltage and current to maximize output.

It senses battery voltage, and will either bulk charge, boost or float.

' Voltage sensing
An optional Smart Battery Sense or battery monitor measures the battery terminal voltage and sends this via Bluetooth using the VE.Smart [28] network to the solar charger. If the battery voltage is less than the solar charge voltage, the solar charger will increase its charge voltage to compensate for voltage losses '

15 comments
2 |3000

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

info7060 avatar image info7060 commented ·

Thanks for the answer. But if the battery is full for example and there is no other load, MPPT wont output all it can get from the solar panels because there is no place to send it, right? So it does not know what is the potential. Or am I missing something?

0 Likes 0 ·
nickdb avatar image nickdb ♦♦ info7060 commented ·

It won't know what the potential is until it tries to scan under load.

You could get yourself a light sensor which will tell you what is available and use that as the reference.

1 Like 1 ·
info7060 avatar image info7060 nickdb ♦♦ commented ·
That is what I thought I would need to do, or just try start something and check if battery or grid is used. The idea is to consume as much of the production as possible without using grid and without excessively using battery. I have a set of electric heaters that I can programmatically control. So I can change the load pretty neatly, but to do that I would need to know what is possible to draw.
0 Likes 0 ·
nickdb avatar image nickdb ♦♦ info7060 commented ·
I track a rolling average of PV generation via node red, it's not perfect but seems to work as long as there are regular loads. If everything goes idle it starts to tail off. It's a work in progress.
0 Likes 0 ·
info7060 avatar image info7060 nickdb ♦♦ commented ·
Thanks for the idea.
0 Likes 0 ·
usernamepasswordbs avatar image usernamepasswordbs info7060 commented ·

If the MPPT won't know whether more production is possible without "scanning under load" (not sure what that means) is there instead a way to track when the MPPT is reducing output due to there being not enough load on it? I think it would most likely "know" when it is doing this because it is an active thing that it does.

I want to switch a load only when the MPPT is doing this output reduction but cannot find an integrated method. I have read of people using light sensors which is ingenious but that does mean going outside of the system.

0 Likes 0 ·
secessus avatar image secessus usernamepasswordbs commented ·

is there instead a way to track when the MPPT is reducing output due to there being not enough load on it?

Normally when the controller is reducing output you will see panel voltages > rated Vmp.


0 Likes 0 ·
usernamepasswordbs avatar image usernamepasswordbs secessus commented ·
Thanks for that input. I have been watching the panel voltage a bit just because there is a relay setting for panel voltage. I don't suppose you have a detailed way to make use of it for avoiding lost production?
0 Likes 0 ·
kevgermany avatar image kevgermany ♦♦ usernamepasswordbs commented ·
The multi is more interested in output voltage than current. So every scan cycle it adjusts to maximise current at the output voltage. As panel output drops rapidly if it's not running at optimum voltage, the tracker compensates by altering it's draw from the panels to maintain an optimal panel voltage.
0 Likes 0 ·
usernamepasswordbs avatar image usernamepasswordbs kevgermany ♦♦ commented ·
I have noticed that the voltage of my 6-panel string for example goes to about 220v when the MPPT is "throttling". Normally it varies between 190v and 205v throughout the day.
0 Likes 0 ·
kevgermany avatar image kevgermany ♦♦ usernamepasswordbs commented ·
Yes, quite normal. But it's the battery that's throttling, causing the PV voltage to rise. All the MPPT is doing is controlling output voltage.
0 Likes 0 ·
usernamepasswordbs avatar image usernamepasswordbs kevgermany ♦♦ commented ·

But the battery is really only a load. It can draw as much current as the MPPT will give up to the moment when bulk charging is finished, then it suddenly becomes a much smaller load.

Battery or no battery the MPPT would still "throttle" if load is less than production at any given moment. If another load could be switched on automatically at this point....

0 Likes 0 ·
mark-a avatar image mark-a usernamepasswordbs commented ·

You can detect if your panels can deliver more power by watching the battery voltage. When it is at float voltage level they are more or less fully charged and the charge current drops steadily. This can be seen in the right half of the screenshot attached. At this point it's likely your panels can deliver more power.

Now when you add a load and the voltage stays high you are drawing additional power from your panels and that's what was happening at the right end of the curves in the screenshot. But if the battery voltage drops your load draws more power than your panels can deliver and you are reducing the power which goes into your batteries.

So you can do something similar the MPPT does to optimize panel output: successively add loads until the battery voltage drops and if so remove them until it's at float voltage again.

screenshot-detectthrottledpv-20230322t123750.png


0 Likes 0 ·
usernamepasswordbs avatar image usernamepasswordbs mark-a commented ·
Thanks for that. My batteries are 48v lithium with a built-in BMS, will the same method work ?
0 Likes 0 ·
mark-a avatar image mark-a usernamepasswordbs commented ·

CLARIFICATION: While the method is independent of the battery type (esp. as it does not draw any conclusions about the state of charge from the battery voltage) it works only if the power from the panels goes into the batteries first. This is the typical scenario for off-grid applications like on a boat or RV but not necessarily when you are (permanently) connected to the grid.


The numbers will be different but not the principles. I took more screenshots today when I manually implemented the method adding loads when the batteries where at float voltage for some time and removing them after clouds came and the voltage dropped.

UPDATE: WRT the BMS. Increasing the battery voltage above limits will damage it. No BMS will do that. So if you are at a high SoC and still charging your battery voltage will be at float level no matter of the technology inside. Once you remove the charging current even without a load attached the battery voltage will slowly drop to its nominal value. The only point is you need to have access to the true battery voltage.

screenshot-loadaddedinfloatstatewithmoresolarpower.pngThe batteries are at float voltage at 10 °C. The spikes on the right are from a load added. The voltage stays high afterwards i.e. the panels were able to deliver more power.

screenshot-cloudsandloadswitchedoff-20230325t12491.pngLater the panels are shaded by clouds and the battery voltage drops. After disconnecting the loads the voltage rises again. The battery current rises until the float voltage is reached again.

0 Likes 0 ·
bigbadbob76 avatar image
bigbadbob76 answered ·

I have looked into this as well, my problem was when using the PV voltage to measure available power I switched in additional loads which then dropped the voltage.

Better way was to have a separate small PV panel driving into a fixed load resistor and a current sensor. this gave me an estimate of solar irradiation at any given moment.

I have yet to use that information to switch in additional loads, maybe this summer.

Alternatively, if you have the cash, Victron sell an irradiation meter that can be interfaced to the GX device and report on VRM.

2 |3000

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

usernamepasswordbs avatar image
usernamepasswordbs answered ·

So I suppose the load would keep dropping out once the voltage reduced. I never thought that far ahead! The panel voltage seems useful just because the MPPT has the relay ready and waiting and it has a minimum "on" time.


2 |3000

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

usernamepasswordbs avatar image
usernamepasswordbs answered ·

I am thinking that if I change settings to allow grid infeed (currently switched off) then the MPPT will never need to throttle back so the infeed current could be used to switch on loads.

Edit: I mean that the control system would be constantly measuring in-feed to grid and constantly switching loads in or out to keep in-feed at -50w just like the Multiplus does now.

2 |3000

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