For a three-phase system, we use three Multi RS Solar inverters. A BMS control system is used for the battery, which includes a DCL. However, this limit is either ignored or only applied per phase, causing the total sum of the currents to be exceeded. In a single-phase system, the DCL functions correctly.
Which is why the battery bank is supposed to be sized correctly.
In anycase you are where you are.
Some batteries will shut down a system when they are being used above DCL others do not. So it depends on the next question…
what batteries do you have
Are you running offgrid or on grid?
Some setting like keep batteries charged or near full will allow some assistance with solar to loads while protecting your battery bank.
In the future, the battery will be expanded, but even now I believe the Multi RS units should throttle back at the DCL.
At the moment, the battery does not shut down when the DCL is exceeded. It’s a 15 kWh DIY battery with a JK Inverter BMS. The DCL is set to 120 A, and Victron uses 95% of that, resulting in 114 A.
The system is grid-connected, with ESS set to 0 W.
Since I am connected to the grid, I believe the Multi RS should limit discharge to a total of 114 A. However, it currently appears that each Multi RS is allowed to discharge up to 114 A individually, without considering the combined current of the three units.
I haven’t verified this yet, but I’m about to set the DCL in the BMS to 20 A to see whether the Multi RS units will then limit to 60 A total.
Unfortunately, I haven’t yet had the time to reduce the DCL to an extremely low value in order to check whether the Multi RS Solar units throttle back to their individual DCL setpoints, rather than responding to the total consumption.
In the image, you can see that a total of 180 A is being drawn from the battery, even though the DCL is set to 114 A. The Multi RS units are set to external control, and the CVL is properly regulated for both the absorption and float voltages.
As I understand it, the DCL is intended to apply to the entire battery, not as a limit per individual Multi RS. It shouldn’t be an issue if one phase draws more current than the others, as long as the total remains within the DCL setpoint.
Other settings as well affect how much the battery is relied on.
For example the input current limit and how much you choose to cycle. If you have keep batteries charged and an input current of the load expected it may help. It means more grid reliance if there are larger load the solar does not cover but at thie point it is where you are untill thebank is expanded.
At the end of the day what loads the end users have on the system should be considerate of the battery bank choice. The Multi RS can’t be used like the MP2 where you can limit inverter power.
Thanks for your quick response!
When there is no grid available, I understand that the DCL may be exceeded.
In this situation, “Optimized without BatteryLife” is enabled up to 5%, with the ESS trying to keep grid usage as close to 0 W as possible.
In this case, the system can draw power from the grid during high loads, so exceeding the DCL isn’t strictly necessary.
Am I correct in understanding that the Multi RS does not take the DCL into account?
In a single-phase system, the DCL is enforced properly. That’s why I suspect that in a three-phase system, the DCL is applied per Multi RS, which could result in the DCL being exceeded three times in the worst-case scenario.
Most systems won’t which is why min DCL should meet the basic needs to the system.
The Multi RS can’t currently be set the same way the MP2 and ve bus systems can with regards to limiting inverter power which in turn limits DC draws.
I have a similar experience, although I have 3 x 16Kw batteries (Seplos 3.0) that can handle up to 200amps each (600amps total) while (3 x Multi RS is max 300Amps).
In my case the system offloads to grid at peak times at a rate of about 315amps, which is totally fine, except it produces more heat and thus noise. If I the system works with 225amps it makes a big diference in fan speed overal, so I would like to set my DCL to 225 as well.
When I started the build I was running on only 2 batteries (stil 400 amps is enough), but it happened once that the secondary battery shut itself off due to a lower charge as the main one. This resulted in the main battery drawing 300amps on it’s own for a while. The cirquit breaker somehow managed to stay live for a couple of minutes and then cut off. I had to replace the cirquit breaker cause it became unstable afterwards.
So in short, I see several reasons to suggest an option where we can tweak the DCL a bit with 3phase Multi’s too.
I thought setting Max discharge on each RS to 85A would resolve it for me, but it doesn’t seem to have any effect:
1 Like
I just adjusted the DCL to 14.2 amps (15 amps in the BMS). For the test, we had increased the consumption. The Multi RS then reduces its output power, causing the grid to supply more of the load. However, the DCL is now enforced per Multi RS instead of for the entire setup. In other words, the DCL is being exceeded three times in total.
As a result, the BMS disables discharging because the 15-amp limit is exceeded.
This doesn’t seem to be the intended behavior of the DCL, does it?
Thanks for your response, Martin!
I’m curious — if one battery in your setup gets blocked or shuts down, does the DCL automatically get reduced as well?
The battery settings are being exceeded because BMS control is being used — that’s why the 85-amp setting has no effect.
I have the same problem here: 3x15,5kWh batteries with JK bms connected through CAN to the Cerbo. The JK bms correctly reports CVL, DCL and CCL. It also reduces these values when one pack shuts down for whatever reason.
In my opinion a Multi RS system should follow these values as system-wide values and never exceed them except for short bursts maybe.
Hi Simon,
Thanks for your reply!
I completely agree — that’s exactly what DCL is meant to represent, as far as I know.
In the MP2 generation, it seems to work that way indeed.
Any news here?
I have a 3 phase Multi RS system with 2 battery modules and DVCC setup
105ah (16s) with 55a CCL and 55a DCL
315ah (16s) with a 150a CCL and 150a DCL
This results in 195A CCL and DCL
When discharged is blocking on the smaller battery the DCL is adjusted accordingly
I’m using JK-PB2A16S15P for both batteries. (150A BMS), they are connected with CAN.
I am running on-grid with DESS. So it actually not overdrawing because of load, most of it goes to grid.
At the time Max Discharge Current was disabled expecting DCL to take preference.
Hi Arjen,
Unfortunately, there’s no update yet. I’m also not quite sure where we should report this. I believe it can be considered a bug in the system. At least, I assume it works as described above on the MP2.
Hi @Boekel1 and others,
Unfortunately not to much I can say here on this topic right now, not because it is secret but just because I don’t have much first hand info.
Victron is aware that this handling of the DCL could be improved. There was a method explained to me that improves it developed by a Victron customer in Node Red, so there is a proof of concept.
I believe it involves another control loop development project that is being worked on internally, but then I am already at the end of my knowledge.
I would say Victron is aware, and progress is being made, but it is not the highest priority of those who are able to do it, and I have no timelines or expectations. If you wish to escalate it further because it is causing issues commercially, the path would be via your sales channel.
I’ll keep this thread in mind if I come across any further info.
Hi Guy,
Thanks so much for your response – good to hear that the Victron team is aware of it and working on improvements!
In the meantime, do you happen to have a link or any more information about the Node-RED solution that was developed by a customer?
I have a lead, and I will ask, but I cannot promise.
Hi! I have developed a solution in Node Red to respect the charging current set in dvcc even when DC injection is enabled. I have presented this solution in the following section:
As for the discharge current, I haven’t created a flow, but I think a loop can be made to automatically set the max invert power from the ess depending on the current parameter received on CAN from the BMS.
Thanks a lot for the link – I’ll definitely take a look!
I hadn’t thought of setting the maximum power output of the inverter myself. Probably an easier approach than what I was working on:
I was setting up a virtual BMS in Node-RED, where the DCL is adjusted dynamically.
The idea is to let the virtual DCL vary between a minimum of one-third and a maximum of 100% of the actual DCL.
As soon as the DCL is exceeded, the virtual DCL is set to the highest current (from the three Multi RS units) minus the excess.
If the DCL is undershot, the virtual DCL is increased by the available margin compared to the previous value.