Battery Dis/charging start power address 33148, 33149

[mmierke]

Hello, I accidentally changed the values for those addresses and tried to identify the standard values.

1. It would be nice if you could provide some default values for those registers and maybe some explanation about their purpose
2. It would be nice if you could add that kind of info either to the yaml and/or the dashboard

[danielclau]

You can just diff the file in the github repo to find the original value.

[mmierke]

I was looking into the yaml file, but was not able to find a default value for those attributes.

[Gnarfoz]

These parameters are supposed to represent the threshold at which the EMS starts charging or discharging the battery.
As such, IMHO, just set it to a value you find acceptable. Most likely that's 0 for both, meaning it will use the battery as soon as there's a need. That's why you bought a battery in the first place, right? ;-).
If I recall correctly, both had an initial value of 10 W or 20 W on my SH10RT-v112, but I'm not sure why that would make sense. To allow some leeway so that the EMS doesn't act quicker than the MPPTs can? Maybe.

However, I'm not sure I fully understand the inner workings.
I have ~15 kWp PV generation installed, with a SH10RT inverter, which has an AC limit of 10 kW. Normally, this means I can only use 10 kW of those potential 15 kWp.
However, on sunny days, while at this limit, it is possible to charge the battery, because that is DC and as such does not fall under the AC limit. Meaning, it's possible to use up to 15 kW (10 kW AC + 5 kW DC).
Now, it would seem obvious that to easily achieve this, one would set the Battery charging start power to for example 9900 W so it only starts charging when that threshold is crossed. Also, set the Battery max charge power to something low enough (1 kW perhaps) so that the battery doesn't take away too much of the power that is currently being used to max out the AC capacity. Adjusting this dynamically is what other discussions / issues / forum posts refer to as "dynamic battery load control".
But what is that threshold based on? PV generation? Before or after inverter losses? Surplus/export power?

If I set this to 9000 W, while generating 10 kW and exporting 9.4 kW of those 10, it does start charging the battery, but at a measly ~15 W.
It seems that the further I lower the threshold value (8000, 7000, 6000), the higher the charging power climbs, but it's still a low number, 100~150 W. See images below.
If I lower it to 5000 W, suddenly the charging power jumps to 1 kW (the Battery max charge power mentioned above).
If I then slowly increase it back to ~6100 W, the charging power again goes down drastically, to around ~300 W.
I'm not sure what to make of this behavior. 🤷‍♂️

[mmierke]

Maybe you should give it a try. Maybe I'm wrong as well, but that is what I'm observed so far.

[mmierke]

In case the additional power is below the trash hold, the converter is using the grid instead of the battery.,

[Gnarfoz]

Indeed, I normally wouldn't need to adjust this all the time. I'm currently trying to figure out if it's a good method to achieve the desired "charge the battery at noon, where the AC limit normally causes some wasted power" thing described above.
I looked a bit more closely this morning and found the following: up to 5900 W, the value works as expected. Charging (blue line) starts once (PV power - load power; cyan line) exceeds this value (orange line). Starting from 6000 W, however, it stops working as expected. (PV - load) crosses 6000 W with nothing happening, until I reduce the threshold to 5900 W (around 10:36 and 10:50), which of course immediately causes the charging to start, since (PV - load) is already over that threshold. Raising it back to 6000 W stops the charging. Only much later, at 11:35, where the cyan line is around ~9800 W, charging begins.

The cyan line is actually (MPPT1 voltage * MPPT1 current + MPPT2 voltage * MPPT2 current - Load power), the regular "Total DC power" lags too far behind and includes internal losses of the inverter. This one pretty closely lines up with when charging starts (below 6000 W).

Either this is a bug, or maybe the value is not actually set in Watts, or... something.
5900 W is relatively close to the specified maximum charge / discharge power (5760 W) of my Sungrow SBR096 battery, but I'm not sure why that would come into play here.

[mmierke]

I'm not sure if this kind of micromanagement is a good idea. The outcome in case of a rainy afternoon is not nice. Lucky me, I only have 10kW installed and a 10kWh battery. With your approach, you can save battery capacity x 8ct (In case you are not on the spot market). In the worst case, you have to by energy from the grid, meaning battery capacity x 29ct in my case.

[Gnarfoz]

The potential financial gain is indeed negligible (0,071€ / kWh), but it's just nice to know you've found a way to deal with the mid-day clipping.
It's more a question of "can it be done" not of "should it be done". ;-)
I want to learn as much as I can about what this expensive device I bought can do. This experiment seems well-suited for that.
Things that seem like they're supposed to be doing one thing but turn out to do another are particularly interesting.

Of course, there are other ways to achieve this, but if I can use built-in functions, then why not? :-)