diy solar

diy solar

DIY BMS design and reflection

I added the DPB specs:

And the general description:
 
Is it possible to include the option of providing overvoltage and undervoltage disconnect signals in place of the Disconnect & Precharge Board?
 
Yes, there will be no hardware modification needed but it will require custom software though ;)
 
  • Like
Reactions: Cal
A very interesting option would be to enable low voltage disconnect through the BMS but provide a signal for high voltage disconnect.

This scenario wouldn't require the back-to-back fets within the bms. The design is a lot simpler. Since your disconnect is at the low side of the battery, the shunt needs to be located at the high side of battey.
 
A very interesting option would be to enable low voltage disconnect through the BMS but provide a signal for high voltage disconnect.

That's possible with the current design: disable the disconnection for the over voltage and use the aux output for your over voltage disconnect signal.


This scenario wouldn't require the back-to-back fets within the bms. The design is a lot simpler.

You can do a custom DPB with only half the mosfets but I don't think it's worth it.


Since your disconnect is at the low side of the battery, the shunt needs to be located at the high side of battey.

Why?


NB: you can't use the redundant hardware protections with all of this custom stuff.
 
You can do a custom DPB with only half the mosfets but I don't think it's worth it.

Why?

NB: you can't use the redundant hardware protections with all of this custom stuff.
Why don't you think it's worth eliminating back-to-back fets? A single fet (mechanized as dual port) can now do the job four fets do in a back-to-back scenario. That's a major design improvement.

My error in regards to needing a high-side shunt. The shunt can remain where it is.
 
Because the proper way would involve redesigning the majority of the board. You can always do it the easy way by not populating half the MOSFETs and replacing them by bridges, but that's not really ideal.

Why one instead of 4? it would be one instead of 2.
 
Why one instead of 4? it would be one instead of 2.
Say design criteria is a unit1 for fet resistance through the switfch. Each fet has a resistance of unit 1. For a common port design you would require 4 fets, as 2 back-to-back fets have a resistance of 2. The design requires two sets of back-to-back fets to achieve unit 1 resistance. The dual port design requires just one fet to disable the load. The smart way to disconnect charging sources is at their inputs Therefore no high current is involved when disabling the charging input. There may be exceptions, but I believe that's the general case for motor homes.
 
I see, but here the main constraint is a thermal one, so you could only remove half the MOSFETs, not 3/4 of them ;)
 
Last edited:
Why don't you think it's worth eliminating back-to-back fets? A single fet (mechanized as dual port) can now do the job four fets do in a back-to-back scenario. That's a major design improvement.

My system has a Victron inverter/charger - making a 2 port BMS useless.

There is absolutely, positively nothing wrong with back-to-back FETs. That is even more true if you absolutely need a single port BMS.
 
My system has a Victron inverter/charger - making a 2 port BMS useless.

There is absolutely, positively nothing wrong with back-to-back FETs. That is even more true if you absolutely need a single port BMS.
If you have the 3000W Multiplus, it has two inputs that can be used for different purposes. One for charge enable, one for load enable.

The 2000W Multiplus only has a single input. I am using mine for charge control so the battery monitors temperature sensor can control charging.

I only going down to 24V, so I am comfortable using the inverter's low voltage cutoff function control that. My BMS will still disconnect on low cell voltage.
 
If you have the 3000W Multiplus, it has two inputs that can be used for different purposes. One for charge enable, one for load enable.

The 2000W Multiplus only has a single input. I am using mine for charge control so the battery monitors temperature sensor can control charging.

I only going down to 24V, so I am comfortable using the inverter's low voltage cutoff function control that. My BMS will still disconnect on low cell voltage.
the 3000VA unit still only has one high-current connection that is both charge/discharge, right?

Are you talking about the remote control inputs?
 
I see, but here the main constraint is a thermal one, so you could only remove half the MOSFETs, not 3/4 of them ;)
Exactly, main constraint is thermal, which is limited by on resistance. Four bidirectional fets have same resistance as one single direction fet. Ratio is 4 : 1.
 
Exactly, main constraint is thermal, which is limited by on resistance. Four bidirectional fets have same resistance as one single direction fet. Ratio is 4 : 1.
How are you proposing to accomplish a 1 port BMS switch without back-to-back FETs?
 
Exactly, main constraint is thermal, which is limited by on resistance. Four bidirectional fets have same resistance as one single direction fet. Ratio is 4 : 1.

For the total thermal power, yes, but the main limit is the individual thermal power and it'll be too high if you put only 1/4 of the MOSFETs because then you'll have double the current per MOSFETs so 4x the dissipated power per MOSFET.


How are you proposing to accomplish a 1 port BMS switch without back-to-back FETs?

He wants to do a separate port BMS with the discharge side MOSFETs switched and the charge side low current signal controlled.
 
He wants to do a separate port BMS with the discharge side MOSFETs switched and the charge side low current signal controlled.

Ok - so that means the charge cutoff relies on a signal to the charger to stop charging?
 
Ok - so that means the charge cutoff relies on a signal to the charger to stop charging?

Yep. I don't like that solution for my system (as previously discussed way too many times on other threads) but if someone wants to use it on their system then it's fine of course. The customization possibilities are one of the big advantages of having a modular system which is open source ;)
 
Yep. I don't like that solution for my system (as previously discussed way too many times on other threads) but if someone wants to use it on their system then it's fine of course. The customization possibilities are one of the big advantages of having a modular system which is open source ;)

I am not sure I trust my inverter/charger to stop just because I said so.
The biggest concern is that it requires a positive signal to stop (as far as I know). If the wiring came loose - the system would never stop charging no matter what the BMS said about it.

I would rather have an active signal required to allow charging so that if anything was disconnected or damaged in some way - the immediate result is that it stops charging. Not sure if that is an option on Victron units since I have not yet used the remote enable/disable features.
 
Yep, same reason why e-stops are pretty much always a NC loop ;)

Do inverter/charger usually work this way though?
....currently looking at my Victron Multiplus 3k's to see how they work.

If I am going to rely solely on an external device to behave a certain way - I want to make sure that device has rightfully earned my trust.
 
Back
Top