The competing brand – and I think we all know which one that is – couples the PV on the HVDC side (not on the LV side, low voltage). So it is a lot closer to having the PV AC-coupled, and also a lot closer to the topology used by the new Multi RS inverter. So in some ways, it is not so much a feature, it is an outcome of the design decision they made. They essentially get it for free almost
Further on this, it seems to me the issue is large loads, and not necessarily input vs output (although that is an easy way to draw the line).
To illustrate, here is a scenario. Imagine there are a number of small loads that are not essential (and thus on the input side), but I nevertheless want to power them from the battery over night (and this I am not making up, this was the case in my previous house), but at the same time I do not want a large load (geyser) to deplete my battery, yet I still want to use PV to heat that geyser during the day…
The target here shouldn’t merely be to do what the other brand does. It should be to do better…
What gets my goat even more, is that this isn’t even an issue in places where you are allowed to feed into the grid. You simply feed in all your excess, and if the geyser is on at the time, it gets first pick of that excess. End of story. Or you fit one of those PV diverter devices (very popular in the UK) if you want to get even more fancy. So again, do you see how this is kinda sorta a very local type of problem?
Firstly, I hope I am understanding your issue correctly.
So I’ll restate what I think you want.:
You want your charged batteries not to be depleted by heavy evening usage and would rather use the grid to power these items. Which in turn allows you to face the rest of the night with a relatively high state of charge.
I think if you used “scheduled charging” to an X percentage that enveloped your heavy item evening usage times. I think by doing this the batteries would not be allowed to deplete during this evening peak period.
This would essentially force the Grid to do the heavy lifting.
Edit: I used this technique, to cope when my batteries were kaput and I knew the electric assist of the solar thermal geyser was timed to come on before dawn. The batteries reported a high SOC but within 5 mins of heavy load the inverter would switch off. ( The batteries weren’t holding a charge). By scheduling a charge to coincide with the geyser timer, I was able to avoid this PITA.
Cap the max drain on the batts at night, by setting the max inverter watts in a Cronjob.
Push up the SOC throughout the day, for evening use i.e. spread the usage in other words.
Bad weather days, focus on charging the batts, not powering loads, or the batts will stay in a low SOC the whole day and night, week.
All the above are based on battery usage, controlling the inverter watts base on that. KIS.
@plonkster, this is a stupid … problem to have. BUT, it is a problem, nonetheless, it should be to do better …
Take one idea at a time, refine it to perfection, non-developer terms, then add the next idea, rinse repeat. Once the logic is clear, the screen design is done, then get into the real technical side on HOW to program it. Maybe it is as simple as “battery usage and adjusting inverter watts” to accomodate.
Did not say that at all. Simply used existing screens to propose a UX that would give the user the control for those two scenarios, added under a new menu option under ESS, say called “Scheduled Charge/Discharge”.
Or even better, not under ESS. A new main menu item, linked to say Cronjobs that are not lost when upgrading the firmware.
The Quattro Schedule charging menu looks a bit different. Does not allow me to set Inverter Watts as you have it above. Also does not have and end time, rather a duration. This is how it looks
The 3rd scenario that affects most on bad weather days, that NodeRED takes care of for me, limiting Inverter power based on battery draw, to match the max inverter to max the the panels can produce.
I.e. if 100w, or more, is drawn from the batteries for more than 10min, drop the max inverter watts by the draw from the batteries. This brings the inverter max watts closer to the panels, for bad weather days.
Reverse: If the batteries are being charged again, after 10min, increase the inverter watts with what goes into the batteries accordingly.
Yes, DC Watts and AC watts are not the same, but that is finer details to figure out IF this idea resonates for most.
Did use Scheduled charging before to make some of this work, in the end, I went Cronjobs and NodeRED, as it is a dynamically fluid situation depending on the scenario.
It almost sounds like you want to move the zero-point based on time of day
Normal day: zero-point should be at the main breaker.
Night: zero-point should be at the inverter input. Or a second meter on the grid input, after the main breaker + meter, but with a chance to branch off before the inverter input.
Does that make sense?
Main => Meter 1 => Heavy non-essentials
breaker => Meter 2 ===========> Light non-essentials
===========> Inverter ==========> Essentials
During the day meter 1 should be at zero, at night you want meter 2 at zero.
Now Venus supports multiple ET meters, so you need to switch which one to use.
But… Meter 2 is already included / built-in to the Multi as long as you use AC-Out2 for your light non-essentials.
The trick is just switching the meter based on time, but surely that can be done with a cronjob? Then you don’t need another active control loop, you just move the zero-point.
Edit: simpler picture:
Main => Meter => Heavy non-essentials
breaker => Inverter (AC1) => Essentials
(AC2) => Light non-essentials
At night you switch zero point from grid-meter to inverter itself.
That’s about 90% of what it is. Except, I can imagine that on a very cloudy day, the same would be expected. In other words, it is not so much time based, but “how much PV”-based. During such a cloudy day we then still expect any excess PV to go to the hard-hitters.
So it is more like having two points of control. The battery is zeroed to the one, and the overall (PV and PV-inverters included) is zeroed to the other one.
When TTT first pitched the idea to me the way I though of doing it was to have a daytime schedule and during that time you slowly increase the min SOC of the ESS by 5% while the SOC is < 95% for the PV to be used to charge the batteries, and then after the schedule you drop the min SOC to the normal value.
