Solar chargers & BMS cutoffs

[CaptainRivet]

regarding load/Charge concept you can then go 1 level deeper:
I then splited the load bus into must-have/emergency=what are essential in an emergency and rest bus. The emergency load bus is galvanically isolated with a 30A victron 12V to 12V Victron phönix DC2DC converter (protecting and stabilizing voltage to 13,8V) and will get soon a small LTO buffer battery. Like this its independent from rest of system and running on the LTO independently when all the rest is down.
Additionally that victron DC2DC is the last thats gets disconnected at 2,5V by BMS. Everthing else gets disconnected at 3.0V as i also use the bank as starter for engine.

[sailorboy1]

Originally Posted by krid2000

Why are we not concerned about BMS cutoffs and then solar chargers causing similar voltage spikes? Does it not happen?

Thanks,
Dirk

It doesn't happen/cause a problem.

But the real question is why would even have your solar programmed to be able to trip your BMS? The same question applies to your alternator and charger. Tripping a BMS should happen as much as blowing your main battery fuse. It should never happen.

[CaptainRivet]

Originally Posted by sailorboy1

It doesn't happen/cause a problem.

But the real question is why would even have your solar programmed to be able to trip your BMS? The same question applies to your alternator and charger. Tripping a BMS should happen as much as blowing your main battery fuse. It should never happen.

Because BMS are battery management systems nowadays means a) managing the battery bank in daily operation and b) being prtoection device in case daily operation parameters failed. Its the only device in your whole system that knows the cell voltages and state of charge of each pack so its the perfect device to manage the whole system. In daily operation. If then operation parameter like end of charge are passed then its protection parameters kick in. Depending on system design eg your MPPTs parameter are 1st or 2nd line of defense and BMS 3rd.
If you have a bad cell and 200mv devation even at 13,8V that an MPPT sees one cell can easily be above 3,7V...only the BMS knows.
The pure BMS protection systems you can only find in drop ins.

[CaptainRivet]

Originally Posted by sailorboy1

It doesn't happen/cause a problem.

But the real question is why would even have your solar programmed to be able to trip your BMS? The same question applies to your alternator and charger. Tripping a BMS should happen as much as blowing your main battery fuse. It should never happen.

Wrong, you can kill several MPPT if you disconnect its output under load and that can create spikes.
Victron is protected and even optimized to be cut off like that as that protects the diodes of solar panel compated to disconnect under load on the input side.
Thats the reason you simply isolate galvanically your expensive electronics with a isolated DC2DC converter which stabilzes also the output voltage to a set threshold if you use a high quality one like victron phönix. Like this nothing can happen. This also valid for large agm or gel banks...

[sailorboy1]

Never mind

By all means do whatever you want and worry about it

If your bms shuts down your batteries you probably earned whatever happens

[tanglewood]

I think there is a lot of blending of charge/load management with a BMS protective disconnect.


Charge/load management can be done many different ways, and if done correctly will keep a battery operating well within normal operating ranges, and will prevent a BMS protective disconnect from ever happening.


The BMS protective disconnect is the last line of defense in case charge/load management fails, or if there is some other fault. ABYC and ISO require this.


If I am following CaptainRivet's "combo bus", I don't think it meets the requirements of a BMS protective disconnect. In fact, it seems to have been created specifically to avoid having a disconnect device which is indeed expensive at a higher current ratings, and instead only controls the charger/load. That's a perfectly good way to do charge/load management, but I don't think it's a suitable replacement for a BMS protective disconnect.

[sailingharry]

Originally Posted by CaptainRivet

What cells are you using? 100A on a 800AH bank should not cause deviations if good quality cells, if they are not so well match B what most really get you will need an active one. i source mine via an NPO that gets all 2 weeks 2 container cells for EV and power walls.

Cells and a source are still very much in flux. But since they tend to be a long lead item, I should probably make a decision sooner rather than later!
Currently I am finding myself looking at EVE 280 cells, probably in a 3p4s configuration.
With regard to the active balancer, I am likely to follow the route it sounds like you did. Build and install using passive balancing, and if a problem develops it is inexpensive and easy to remedy.

[CaptainRivet]

Originally Posted by tanglewood

I think there is a lot of blending of charge/load management with a BMS protective disconnect.


Charge/load management can be done many different ways, and if done correctly will keep a battery operating well within normal operating ranges, and will prevent a BMS protective disconnect from ever happening.


The BMS protective disconnect is the last line of defense in case charge/load management fails, or if there is some other fault. ABYC and ISO require this.


If I am following CaptainRivet's "combo bus", I don't think it meets the requirements of a BMS protective disconnect. In fact, it seems to have been created specifically to avoid having a disconnect device which is indeed expensive at a higher current ratings, and instead only controls the charger/load. That's a perfectly good way to do charge/load management, but I don't think it's a suitable replacement for a BMS protective disconnect.

You understand me wrong with the combo bus. The protective disconnect is switching the multiplus off via its remote through the assistent via BMS This is tested to work min 10000times by victron R&D. On The combo bus only both or 3 multis are connected but because i stage them via the generator assistent instead paralleling them switching one off, switches them all off. And that 2 or victrons fail at the same time is very unrealistic, a mechanic one fails 50 times before.

As combo bus disconnect device must work both ways charge and inverter you cannot use a SSD and mechanical 400A relays are very unreliable plus need a lot current, using latching relay brings additional complexity. blue sea high quality gives you 300 cycles...had bad experience with the mechanical quality ones and avoiding them wherever i can, 1 that speced 400A melted close with 280A load and another where contacts burned over time which created high resistance which got extremly hot...if i wouldn't realized manually the high voltage drop and disconnected it it right away that could have caused a fire. Prefer the tested Victron assistent...

[CaptainRivet]

Originally Posted by sailingharry

Cells and a source are still very much in flux. But since they tend to be a long lead item, I should probably make a decision sooner rather than later!
Currently I am finding myself looking at EVE 280 cells, probably in a 3p4s configuration.
With regard to the active balancer, I am likely to follow the route it sounds like you did. Build and install using passive balancing, and if a problem develops it is inexpensive and easy to remedy.

Would take the EVE 304AH cells, the terminal connection are much better then the 280AH one and they can also handle more C-rates. Price is maybe 20Euro higher per cell,totally worth it.

[CaptainRivet]

Originally Posted by tanglewood

I think there is a lot of blending of charge/load management with a BMS protective disconnect.


Charge/load management can be done many different ways, and if done correctly will keep a battery operating well within normal operating ranges, and will prevent a BMS protective disconnect from ever happening.


The BMS protective disconnect is the last line of defense in case charge/load management fails, or if there is some other fault. ABYC and ISO require this.


If I am following CaptainRivet's "combo bus", I don't think it meets the requirements of a BMS protective disconnect. In fact, it seems to have been created specifically to avoid having a disconnect device which is indeed expensive at a higher current ratings, and instead only controls the charger/load. That's a perfectly good way to do charge/load management, but I don't think it's a suitable replacement for a BMS protective disconnect.

Combo bus additional will meet ISO and ABYC. It just splits in a 3rd bus reducing current and therefor risk.
Like this load and charge bus have lower current and can be switched of via a normal SSR, which you can't if you connect a combo device...

[tanglewood]

Originally Posted by CaptainRivet

You understand me wrong with the combo bus. The protective disconnect is switching the multiplus off via its remote through the assistent via BMS This is tested to work min 10000times by victron R&D. On The combo bus only both or 3 multis are connected but because i stage them via the generator assistent instead paralleling them switching one off, switches them all off. And that 2 or victrons fail at the same time is very unrealistic, a mechanic one fails 50 times before.

As combo bus disconnect device must work both ways charge and inverter you cannot use a SSD and mechanical 400A relays are very unreliable plus need a lot current, using latching relay brings additional complexity. blue sea high quality gives you 300 cycles...had bad experience with the mechanical quality ones and avoiding them wherever i can, 1 that speced 400A melted close with 280A load and another where contacts burned over time which created high resistance which got extremly hot...if i wouldn't realized manually the high voltage drop and disconnected it it right away that could have caused a fire. Prefer the tested Victron assistent...

Thanks for confirming. I did understand you correctly. My point is that in the eyes of ISO and ABYC, turning off the multi is not the same as a disconnect device, and is not a substitute. The disconnect device needs to disconnect EVERYTHING from the battery. All loads, all charge sources, everything. For a high power system, it does indeed take a large contactor or lots of parallel FETs, and I agree that the contactors are big and expensive. But it's required if you want to comply with either standard.

[CaptainRivet]

Originally Posted by tanglewood

Thanks for confirming. I did understand you correctly. My point is that in the eyes of ISO and ABYC, turning off the multi is not the same as a disconnect device, and is not a substitute. The disconnect device needs to disconnect EVERYTHING from the battery. All loads, all charge sources, everything. For a high power system, it does indeed take a large contactor or lots of parallel FETs, and I agree that the contactors are big and expensive. But it's required if you want to comply with either standard.

Switching off the victron by its low power side and assistant is in line with ISO definitly, my system does that and was checked by surveyor to comply.
You must be able to disconnect everything, that can be done on the low power side or with high current contactor interrupting the high side or a combination of both. It stands nowhere it has to be a seperate device that does that.
Also if i can choose between a disconnect process that professionally tested to work >10000 times and a seperate where only the device itself is rated for 300times i know what i choose, doesn't matter what any standard states or i if a standard requests specifically the 300times device i ignore this standard as obviously not sufficient.
A majority of commercial BMS does that via CAN.

So if you have a charge bus which has 3 MPPT and a shore charger all victron connected you can either switch all devices off with their remote (as you have to pull them high to be on if power disconnects or remote cable disconnects the device is switched off) at once or use a SSD to disconnect the charge bus BUT that means the devices are still on, just high power side is disconnected. With victron via remote that disconnects the high side and then switches the whole device off.

Same on multi: if assistent cuts off charging that means inside the victron the 230V pass through relay disconnects the charger and the whole high power side physically disconnected. For inverter assistent switching it off is the Victrons 12V relay that disconnects the whole inverter side physically off. Better then they are on and an external cuts off the high power side and victron multi still run.

[CarlF]

I agree with Sailorboy. I have never heard of a conservatively programmed Victron solar controller causing a BMS disconnect. Just set the absorption voltage to 14.2v or lower.

Automatic disconnect of alternators and other charge sources is usually not possible if you have drop-ins - which I would guess is over 90% of installs in the last several years. ABYC E-13 came out too late to slow the market switch to drop-ins.

The cost-effective solution to “disconnect” an alternator is to simply program absorption voltage to 14.0v. That way it is not charging as the battery approaches 100% so it does not cause or can be damaged by a BMS disconnect (although most modern alternators have avalanche diodes which survive a surges just fine).

[tanglewood]

Originally Posted by CaptainRivet

Switching off the victron by its low power side and assistant is in line with ISO definitly, my system does that and was checked by surveyor to comply.
You must be able to disconnect everything, that can be done on the low power side or with high current contactor interrupting the high side or a combination of both. It stands nowhere it has to be a seperate device that does that.
Also if i can choose between a disconnect process that professionally tested to work >10000 times and a seperate where only the device itself is rated for 300times i know what i choose, doesn't matter what any standard states or i if a standard requests specifically the 300times device i ignore this standard as obviously not sufficient.
A majority of commercial BMS does that via CAN.

So if you have a charge bus which has 3 MPPT and a shore charger all victron connected you can either switch all devices off with their remote (as you have to pull them high to be on if power disconnects or remote cable disconnects the device is switched off) at once or use a SSD to disconnect the charge bus BUT that means the devices are still on, just high power side is disconnected. With victron via remote that disconnects the high side and then switches the whole device off.

Same on multi: if assistent cuts off charging that means inside the victron the 230V pass through relay disconnects the charger and the whole high power side physically disconnected. For inverter assistent switching it off is the Victrons 12V relay that disconnects the whole inverter side physically off. Better then they are on and an external cuts off the high power side and victron multi still run.

Well, I would disagree with your and your surveyor's interpretation of the ISO requirements. ISO TS 23625-2021 says "disconnects" need to be done with a contactor, defined as a "relay or switch". It does specifically allow multiple contactors, with individual control, but they need to be contactors. I don't think turning off an inverter is the same as disconnecting its DC power connection. The Victron inverters do not contain a DC contactor. Turning something off is not the same as disconnecting it. Turning something off is sufficient for load/charge control, but it is not sufficient for voltage or temperature triggered cutout.



I also think your understanding of what the inverter does is incorrect. It shuts off the charging function, but I don't think it opens the 230V relay. Doing so would also shut off pass-thru. There isn't a separate relay for the "charger" because there isn't a separate "charger". The same electronics provide charging and inverting.

[tanglewood]

Originally Posted by CaptainRivet

Combo bus additional will meet ISO and ABYC. It just splits in a 3rd bus reducing current and therefor risk.
Like this load and charge bus have lower current and can be switched of via a normal SSR, which you can't if you connect a combo device...

To be clear, I don't have any objection to your multiple buss approach, though not how I would or have gone about building systems. The issue is the difference between turning something off vs disconnecting it. They are not the same, and I believe both standards require a disconnect on certain events. ISO specifically allows multiple contactors for different busses, and although ABYC doesn't call it out, I don't see why it wouldn't be acceptable as long as it behaves as specified. The key is that each needs to be a contactor, defined by ISO as a switch or relay.


BTW, I'd be interested in seeing the test report on the Victron testing of the on/off function. Would you have a link to that?