@dfaber in the old forum, I asked that the grid setpoint parameterized in the ESS also be taken into account for active DESS. In the spring of this year it was still ignored, and you replied as follows:
In this case, as the ESS grid setpoint is known and we could use it, I did create an issue for it in our system to investigate the best way to use it. Note that this is no guarantee that it will get implemented.
But as far as I could observe recently, is it already being considered? I have not tested this further, but I have parameterized grid-setpoint -10W in the ESS and according to my observations, the average grid consumption during battery use is now actually at a slight feed-in with an average of approx. -10W. So does this actually seem to have been implemented in the meantime?
I’m using Grid Setpoint (-65W) in order to compensate self consumption of serially connected Voltronic King inverter between Victron grid meter and MultiRS Dual Tracker.
Also system is on Dynamic ESS.
On graphs I see grid setpoint does work.
Let me know what actual concern, I’d try to find it in my setup to help you
Thanks, all good here. I just wanted it confirmed that grid-setpoint actually works when DESS is active.
Originally grid-setpoint was ignored when DESS was active, it was always assumed as 0W. Now it is obviously taken into account. That’s a good thing, and I’m happy that Victron has now implemented it!
…did you test it?
This is definitely not working for me.
For example setting it to -1000W does not have any effect here.
Also, I find DESS much slower to react as with local ESS and my EM540 GridMeter…Performance is more like with an EM24.
That said, I am unintentionally buying 2kWh more per day compared to local ESS and Grid-Setpoint set to 0W.
These days, when PV is still more than enough available, my normal setpoint is -50W which saves another 0.9kWh per day with local ESS (normally I “buy” 2kWh per month, not per day)…no effect on DESS. @dfaber can you confirm or comment?
It seems that DESS does take the grid setpoint into account within certain limits! Here last night with grid setpoint -20W. On average, there was actually always a slight feed-in:
I’ve been looking at the code. ESS setpoint does not seem to be used.
In some scenarios "0" is hardcoded and in other scenarios it doesn’t use a setpoint directly like ESS does, but instead uses a very large setpoint along with a MaxDischargePower override which is calculated from pv output and consumption. This calculation is done every 4 seconds, so it’s not as reactive as using a setpoint either.
Conclusion: small negative grid setpoints such as -10 or -20W work in DESS, high negative values such as several minus hundreds or -1000W are obviously limited to significantly lower negative values. In my opinion, this makes sense when using DESS!
So for me this is a positive conclusion, thank you Victron!
I see that smaller values work, but I don’t see a sense in “crippling” this featzre.
I have 2 heatpumps with a separate tariff which are connected in a cascading fashion between grid and house (this tariff is significantly lower than the dynamic tarif, except for very few occasions in winter).
Hence, when battery SoC is too low to serve both, heatpumps plus house, I need to excempt the power drawn by the heatpumps from the coverage of the battery.
The only way to achieve that is to set the grid setpoint to a positive value (the power draw of the respective heatpump, which is value between 900w and 2.9kW in my case).
Edit: OK, I just tested a larger POSITIVE value (like +2500W) and this still works!!
However I still do not see the value in the way it is limited on the negative side.
DESS attempts to calculate the best possible course of the target SOC based on the solar and consumption forecast. The original algorithm naturally assumed a setpoint of 0, which is correct if the control behavior is ideal.
Setpoint <> 0 naturally causes interventions in this control algorithm, which was not originally taken into account. In this respect, I am very satisfied with small deviations from 0W.
