question

pf-1 avatar image
pf-1 asked

Combination of BMS control and internal charging logic

With a controlling BMS selected, the internal charging logic of the Multis, Quattros and MPPTs is overridden. Since most BMSs always charge up to the maximum set voltage and do not know about Absorption or Flaot, this is not an optimal solution.


On the other hand, a controlling BMS takes into account that there are dynamic values such as CCL and DCL, which are not used by the internal charging logic of the Multis, Quattros and MPPTs.


From my point of view, the optimum solution would therefore be a combination of internal charging logic in conjunction with the dynamic values CCL and DCL from the BMS.


What do you think? Is it worth asking victron about integrating such a solution?

cerbo gxbattery chargingBMSess discharging
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

5 Answers
bathnm avatar image
bathnm answered ·

@PF_1 Victron already have this. It is called DVCC. Have a read of the Venus manuals. For certain managed lithium batteries, this is turned on by default. The basic premise of the DVCC is to provide CVL, CCL and DCL to chargers and to distribute the charge current requested by batteries across all charging sources such that the charging sources do not exceed the current limit requested by the battery.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

pf-1 avatar image
pf-1 answered ·

@Bathnm as you can read here (chapter 11.1), the internal charging algorithm of the Multis, Quattros and MPPTs is disabled when a BMS is activated to control the charging logic and thus using CCL, DCL and CVL, and it is simply done what is told by the battery. In most cases this is the charging to the maximum set voltage.
Therefore, in this case, there is no Absortion time and no reduced voltage in Float mode, etc.

However, my approach is to retain the internal charging algorithm of the Multis, Quattros and MPPTs but to take into account the dynamic values from the batteries' BMS.

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

bathnm avatar image bathnm commented ·

@PF_1 If the battery is calling or a given charge voltage, what is the internal charge algorithm actually doing? The charger should supply that voltage for as long as the BMS requires it. As an example on the Lynx Smart BMS with a 24v Lithium Battery, it will generally call for 27v. Configurable within the Lynx BMS is how long to hold the battery at 28.4v (absorption) for balancing / SOC resetting, and how frequently to do a 28.4v charge cycle, for example every 30 days. Therefore the BMS will call for a 28.4v charge for a given time every 30 days or after a discharge below a given SOC threshold.

So with DVCC the BMS is going to always call the shots and the charger should follow. It does not matter what the charger wants to do, it should do what the BMS wants.

So again with this in mind, and most BMS will have the configurable parameters mentioned above, what would the internal algorithm actually do, the BMS is defining what to do.

Only one thing can call the shots. If the battery wants 27v, but the internal algorithm wants to do something different, what does the charger actually do?

If you want to reduce the charge current or charge voltage then these can be over-ridden in the DVCC configurations screen. What this would mean is for example if I configure 26.5v then that is what the chargers are asked to deliver, irrespective of what the battery wants. The reason being, only one thing can define what the charge goals are.

0 Likes 0 ·
Mike Dorsett avatar image
Mike Dorsett answered ·

Float is an archaic requirement related to lead acid batteries, that need a short period of overcharge to remove sulphation.

Lithium batteries don't need this, so there is likewise no requirement for a separation of Absorption voltage and float voltage. If you charger still requires the two voltages to be programmed then the float voltage only needs to be ~100mV below the Absorption voltage.

Even while under DVCC, the Maximum charger voltage is still that programmed into each charge controller's battery set-points. The charger will use the lower of the set point and the DVCC voltage.

For correct balancing, it is realy only the charge current that needs limiting, battery voltage set point should be marginally below the maximum to prevent any overshoot problems. The quantity of charge stored in the last 100mV or so of a lithium battery is minimal, so the does not led to any effective capacity loss, and helps reduce internal corrosion.

2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

bathnm avatar image bathnm commented ·

@Mike Dorsett If you are freeing to this statement in the manual, then it is actually referring to the Voltage and Current which an be set in the DVCC menu. If DVCC is used then the controllers just follow the voltage and current they are sent and do not use any of their own local configuration.


  1. If a CAN-bus BMS is connected and the BMS requests a maximum charge current that is different from the user-configurable setting, the lower of the two will be used.


screenshot-2024-01-07-at-090425.png

0 Likes 0 ·
screenshot-2024-01-07-at-090425.png (113.4 KiB)
pf-1 avatar image pf-1 commented ·

I disagree that float is an outdated requirement for lead-acid batteries.

Also LiFePo4 cells should not be kept at the high voltage needed to fully charge for a long time. Here too, it makes sense to reduce the charging voltage after a certain (absorption-) time to prevent premature ageing of the cells.

You can also find different absorption and float voltages in the VictronConnect app or in the VE.Configure-Software if you select LiFePo4 as the battery type.

0 Likes 0 ·
pf-1 avatar image
pf-1 answered ·

If your Lynx Smart BMS provides these options, everything is fine. Then you do not need any other option and your charger should indeed follow the specifications of the BMS.
But I have found that a lot of the BMS on the market are very limited (or "stupid") and usually only request the maximum set voltage and do not reduce the voltage to conserve the Battery when its full.
In this case, the internal charging algorithm of the Multi, Quattro or MPPT is superior because it provides this graduated charging process.
So far, you can only choose between the internal charging algorithm of the Multi, Quattro or MPPT and the control by the BMS. There is no setting in between (for the "stupid" BMS).

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

bathnm avatar image bathnm commented ·
@PF_1 Then disable DVCC as it does not do anything and use the internal charge mechanism. As I have tried to explain it is one or the other and there is no way to have both work. If the BMS is Stupid and asking for maximum set voltage and never reduces, but a internal algorithm has a reduced voltage then when does that charger follow the BMS and when does it follow it's own algorithm, Just switch of DVCC and use the internal algorithm. In such a case then make sure you have wired some form of ATC and ATD signal so that should things go wrong the BMS can stills signal a stop charge or stop discharge. You get what you pay for......
0 Likes 0 ·
pf-1 avatar image
pf-1 answered ·

@Bathnm Perhaps we have misunderstood each other. I'm not talking about the charging voltage, but about the charging and discharging current. (CCL and DCL, as described in my first post)
These values change if, for example, a battery (out of a stack of several batteries) blocks charging or discharging because a cell reports overvoltage or undervoltage. (NrOfModulesBlockingDischarge, NrOfModulesBlockingCharge, NrOfModulesOnline, NrOfModulesOffline)
These values should therefore also be taken into account when using the internal charging/discharging algorithm.
Currently, these values are largely fixed with the VictronConnect app or the VE.Configure software.

It would probably be a long discussion as to whether the charge control belongs in the charger, of which there is usually only one, or in the batteries, of which there are usually several per system. I'm happy to pay for what I need, but I don't like to pay more than once.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.