Any plans for Dynamic current based off load

1

Like the title says, are there any plans on implementing dynamic current based off real time loads? My system has dynamic loads mostly idle around 10A and I can’t set a final charge current any lower than about 15A because the system would be discharging. We need function that reads the current usage and dynamically adjust current/voltage to compensate.

I brought this topic up earlier this year and never really got a good resolution, is this something planned down the line or should I move on from Batrium? I really love everything else about the system, the hardware, the design, the ecosystem but really need some kind of dynamic current control based off load. I know other systems can do this why not Batrium?

I have played around with dynamic voltage but it’s not the feature I need.

2

Hi t37,

As I mentioned in my last reply, the charging response in most inverters is slow enough that we can’t really control current dynamically.

Generally, inverters can be configured in a few different modes depending on your strategy. Victron is no exception here, ESS settings can have quite an effect on how the system behaves.

Could you share more about how DVT fell short for you? Can you email us your serial number so we can take a look at your config and check nothing is amiss?

Keep in mind that if we ask for 10A, the loads are 10A, and solar is capable of generating, say, 30A, Don’t you think it is the inverter’s job to do the maths and modulate solar respectively? It’s the only one that knows all 3 values.

3

James, thank you for restoring my question and responding.

As I mentioned in my last reply, the charging response in most inverters is slow enough that we can’t really control current dynamically.

There are other systems and software that can though, is this a hardware limitation? *

GitHub - Louisvdw/dbus-serialbattery: Battery Monitor driver for serial battery in VenusOS GX systems this is software that can do just that, it installs on the Victron system and listens to values published on the d-bus and will write new values accordingly. I have first had experience with this and if an active system with a dynamic load has the load change then it will alter the CCV of the Victron DVCC to match. For example, if I tell the software that at 99% SOC I want the charging current to be 1A and my load is 10A then CCV will be set to 11A if that load suddenly changes to 45A then the value will be changed to 46A so the overall affect is that the charging rate is 1A. Is it more of a philosophy approach that you’re taking that precludes you from using this method? Victron has made their protocol open to manufactures for a while and there is no reason with the proper serial connection you couldn’t do the same thing. Even the cheapest JBD/JKBMS mossfet BMSs from China are starting to come with this ability…

Generally, inverters can be configured in a few different modes depending on your strategy. Victron is no exception here, ESS settings can have quite an effect on how the system behaves.

I’m not using ESS and I don’t have a gird connected system, this doesn’t apply to me.

Could you share more about how DVT fell short for you? Can you email us your serial number so we can take a look at your config and check nothing is amiss?

The new DVT system does work but it’s not what I need. It still won’t allow me to charge the last 1-5% at a low current value.

Keep in mind that if we ask for 10A, the loads are 10A, and solar is capable of generating, say, 30A, Don’t you think it is the inverter’s job to do the maths and modulate solar respectively? It’s the only one that knows all 3 values.

Honestly, no. I feel that way of thinking is outdated. Especially when dealing with Victron… when you are using their system and have DVCC enabled the VenusOS device becomes the “brains” and it is aware of all the needed values and allows the BMS to alter them. If you’re goal is have a Victron integrated device then you should adopt the philosophy of Victron and fully embrace their methods, which would allow the BMS to set charging values based on real time data… it’s how their very expensive BMS works… I mean what’s the point in having a “smart” BMS if it doesn’t do smart things?

I really do want the Batrium system to work, I believe your hardware design is superior to all others but the software design, implantation, and approach is frustrating as an end user. I don’t mean to be negative or anything and I do apologize if my frustration is coming out but I feel it’s important for anyone doing research to know the limitations of the product. When I first did my research and I read about the Victron integration I wrongfully assumed it was actually integrated and not just a follower device that listens for critical messages to respond to. If anything comes out of this conversation perhaps updating the wiki on your site to the limitations that the Victron integration has.

Thanks again for your response.

4

HI t37,

Apologies, I’ve only just had the time to re-read the last thread
https://community.batrium.com/t/victron-cerbogx-multiplus-ii-3k-actual-values-ns/

I had forgotten you had fluctuating DC loads. This post in the Victron system mentions a specific setting that seems to do what you want:

5

James, I believe that any battery monitor in the system can provide this function.

This line

The power shown in the DC system box is the sum of power reported by all battery monitors configured as such.

This just indicates that the DVCC system can use multiple smart shunts to know all DC loads. The shuntMON is the battery meter in my system, so the DVCC system already knows what it needs to know from the shuntMON. Without looking at your code (because it’s not open source) is it possible that the CCL value being sent by the watchMON is stepping on/over riding this function? It seems that the DVCC system should do this on its own without BMS control but with BMS control enabled it’s possible that it defers to the BMS for this function. So using the “remote” feature in your design the current value sent is what the GX device is choosing to use, since it defers to the BMS.

I do have an extra Victron smart shunt laying around and I can test this but I’m guessing that it won’t make a difference because DVCC is already aware of all DC loads (because I have the shuntMON installed as a master shunt in the entire system).

6

Hi t37,

I would like to ask, what is exactly the problem in your system, that makes controlling the charge voltage not sufficient for correctly charging your batteries? Normally this is totally sufficient in all systems, I know of. I do understand, that you have dynamic DC loads. I do have very dynamic loads, too, and it is always sufficient that Batrium reports a charge voltage limit to the Victron system and Victron does the rest. Why is this not sufficient in your case?

Kind regards,
Kleini

7

How are you overcoming your DC loads being greater than charge current? Are you using a Victron system? Care to share screen shots of all your settings Batrium and Victron?

8

James,

I just did about an hour’s worth of testing with a spare smart shunt and I was not able to get the GX system to do current compensation. Everything I tried had no effect and it appeared to always follow the CCL set by the BMS. I believe that when a BMS is integrated into the system the value it applies to the d-bus is the value followed by the DVCC subsystem. Thus it is up to the BMS to calculate a dynamic CCL based on current load.

The link you sent while promising I believe applies to systems without a BMS but having multiple chargers that carshare current/voltage through DVCC.

9

IMG_2829
IMG_28291920×1440 274 KB

I am missing out or just got lucky. I have a trio of victron’s communicating through the Cerbos hub to the batrium core-K9 system measuring 160 cells. the load is a trio of Schneider Cognex XWPro+ inverters running my home. The comms are set to victron gx mode and batrium setting in the cerbo. I have no issue with current on almost charged batteries unless the load exceeds the solar output. If the amps are more positive than 2 amps on the DC shunt, I slightly charge and equalize. If the current goes above preset the solar chargers dial back slowly. My charge max cell charge is 3.6 volts and goes to limited mode at 3.45 volts and 10 amps. Dividing this current across 160 batteries allows the resistive load to make a good difference with minimal overheat and if you had 100 amp swing for. a second it is only 1.6 amps at the cell that is rated for 160 amps. Until I hit a low voltage of 3.40 and the full charge kicks back in. I don’t push my batteries at all and they will die from old age. Heavy load swings only kick in full charge quicker and will also allow it to go into off mode and drop back into limited and repeat at a slower frequency as the PID’s take over. Am I lucky or missing the point of the discussion.