I’d like to check my understanding of grid feed in limits.
I currently have 32 solar panels driving two 4kw solar edge inverters and a grid limit of 9.3kw. The DNO may agree to a higher grid limit or they may not. I would like to add another 16 solar panels, a 5kw 48-volt quatro or multiplus II, a 10kwh battery and probably 8 mppt controllers (2 panels each). I would then want to set the grid limit to whatever the DNO will agree to.
The idea here is that as half the current panels face east and the other half face west, it is currently extremely rare to get anywhere near the 9.3 kw limit thus setting a limit less than the total power of the inverters (13kw) will have almost no effect on overall generation.
What I need to be sure of is that limits of say 9,10,11 or 12kw can be set, that they will take into account the output of the inverters they have no control of and that, when the batteries are full and there is excess generation, the mppt controllers will throttle back.
I think I would connect the existing inverters to the non protected output of the quatro to avoid any problems with disconnecting big loads. The only things connected to the protected output would be the house lights, heat pump and maybe some house sockets (light loads). Keeping going in the absence of mains is a secondary consideration.
It would be useful to know if there are people out there with a similar system.
A quick few observations. You will have to respect the factor 1:1 rule as explained on the Victron website. The amount of solar inverters on the ACOUT cannot exceed that of the inverter e.g. 4.4kW if you are looking at a 5 kVA MP II. With all the solar on the ACOUT, the total system power will be limited to the 4.4kW of the Quattro as the ESS will use frequency shifting to stop the solar inverters when they reach this limit.
If I understood it correctly, a Multiplus II has two AC out’s. One for non-critical loads which is just the mains in via a current meter and one for critical loads which, in the event of a mains failure, is isolated from the incoming mains and driven by the inverter. My intention is to put the existing inverters on the non-critical output so that they are directly driving the mains and, in the event of a mains failure, will shutdown… that’s what they do when there’s no mains. This means they will neither charge the battery nor drive the house in case of a mains failure. This is not ideal but does not impact the main objective of reducing CO2 emissions. I thought this arrangement avoided the problems of getting the existing inverters to shutdown when the batteries are full being referred to. The reason for connecting the existing inverters to the multiplus output rather than direct to the mains is to get the multiplus to register the exports the existing inverters are generating so that it would throttle itself back when the total exports are getting close to the DNO’s limit.
Factor 1:1 rule still not being followed!!!
Also why 8 mppt’s ? One mppt would suffice subject to all panels being faced same orientation & suitable sized, single mppt would throttle as directed by system in just the same manner as multiple mppt’s
It seems that one plan is to not connect anything at all to the AC2 output and only connect the critical loads (lights, heat-pump & a couple of sockets) to the AC1 output. The multiplus/quattro will then have to use a separate current transformer to measure the exports to the grid. Thus, starting from the grid, we have the 100a fuse, then the official input-export meter, then the current transformer, then, at this point, we connect to both the existing inverters and also the multiplus/quattro. This then gets back to my original question… I need to be sure that limits of say 9,10,11 or 12kw can be set for the total system export. I also need to be sure that we can prevent the batteries being used to feed the grid… so in effect a 0kw limit for export if the batteries are the source.
The alternative would be to go for a 10kw multiplus/quattro. The same questions remain of course plus the question as to whether the actual inverters I have got will play ball when the frequency is increased. There must also be the question of how the DNO will view a system which relies heavily on its configuration to stay within their limits.
The reason for 8 mppt controllers is that of the 16 panels, 2 will face south, 4 east and 10 west. This arrangement is dictated by the remaining space on the roof and requires at least three mppt controllers because of the very different light levels the east and west facing panels will get most of the day. One of these controllers must be for just two panels. The remaining panels could use fewer, bigger mppt controllers but this won’t save money and doesn’t have the advantage the fewer panels share one controller the nearer the optimum the result will be since the panels are never exactly matched and the light levels are never exactly the same… so a bit of muck on one panel reduces the output of one other panel rather than 9 other panels.
Is there a description somewhere of how mppt controllers talk to each other… dividing a varying max generation limit among several mppt controllers whose panels have different amount of light must be a bit tricky. It would be nice to be reassured that this is going to work.
2 x Mppt’s are controlled via MP2 GX in my case, JK BMS provides battery data readings to GX, from this data the mppt’s regulates the power going into the battery( subject to BMS & GX settings) and MPPT’s will ramp up if MP is using DC power to invert( once SOC drops ).
My MP2GX has 1no. VE direct( used to connect to 1no. Mppt) on the connection panel, it also has another unused VE direct internally ( used to connect to my 2nd Mppt).
You will have to investigate on connection of 8no( usb multi port splitter + 6no. Usb to VE direct cables maybe? This will add to costs as not cheap), on a side note the vrm dashboard will be very busy with displaying 8no!!.
Ps mppt’s as far as i know don’t talk to each other, they do what is required from available data from MP2GX / BMS to use available power as per settings.
Once set up you can see the charging Amp required drop on VRM - remote console - mppt - networked operation. This applies to all mppt’s connected, but no idea of how or why which individual mppt starts reducing first or last, it just works.
Once mppt’s are connected to GX they become slaves / do what they are told from GX device.
