LiFePO4 reference diagram, Alternator version

[krid2000]

Originally Posted by s/v Jedi

So how are you going to control it? Can you put the Contura switch that comes with the RBS in your panel? It is a really nice switch…

For now I'll probably just use a spare on/off switch on my panel, much more convenient than the mechanical switch I currently use on the 12V side at the verrrrrry forward end of the v-berth. Two flies, one stone.

My boat has its house battery bank under the v-berth by design (built 1985, when I was still hacking my ZX Spectrum). In addition to the above change, my plan for the upcoming season is to test moving to an electric galley setup, using just my Duxtop induction hub and an electric kettle. The installed CNG system works great, but refills outside the US are nigh impossible.

If that test turns out successful, I'll do a larger electric refit switching to an all electric galley, LiFePo4 etc. Right now I'm trying to find space for the inverter close to the batteries to make that test setup work. Looking at the space constraints under the v-berth, it just dawned on me today that I might then rather move the new energy system aft, closer to the bulkhead just aft of the companionway. Keeping a small crank AGM battery up front for the windlass, but having much more space to do a proper installation.

Jedi, thanks for all the inspiration I've drawn from your messages, electrical plans and build-photos . I suspect I'm not the only one! More questions to come for sure.

Thanks!
Dirk

[stanhoney]

Here are some measurements of the Bluesea 7713 switch. Some of these characteristics are not perfectly clear from the instructions:

The operating current of the switch is drawn from either stud terminal A or B, not from the red Control input.

The continuous operating current of the switch is specified as <13mA. I measured 4mA.
The changing state current is specified as <7amps. I measured a peak of 2.5A.
Again, this current comes into the switch from either of the A or B stud terminals, and then goes out via the black ground wire.

The red Control input draws 4.3mA. There is no increase or transient in this current when the relay is changing state. This is simply a “logic” input to the 7713 switch. There is no need to protect the circuit that drives this (red Control) input with a “freewheeling”, “snubber”, or “flyback” diode, as is customary for circuits that control a solenoid.

If the yellow toggle is turned to the right the switch is locked off forever, no matter what the red control input is. The yellow toggle must be manually turned to the left to resume automatic operation.

If the yellow toggle is turned to the left, AND the yellow button is pushed in, the switch is locked on forever, no matter what the red control input is, and through power cycles on the A or B stud terminals. To resume automatic operation, the yellow toggle must be manually turned to the right (turning off the switch) and then to the left again.

Note that this is different from the older 7712 switch. In that version, if the yellow toggle was turned to the left and the relay state (i.e. the pushbutton) was in a different state from the red Control input, the relay state would change state to match the red Control input after 10 minutes. AFAIK, the 7712 switch is no longer available.

Total current usage of the 7713 switch:

When on, 4mA from either the A or B stud terminal, and 4mA from the red Control input, for a total of 8mA.
When off, 4mA from either the A or B stud terminal.
Of course these measurements are from just one sample.

[s/v Jedi]

Originally Posted by stanhoney

Here are some measurements of the Bluesea 7713 switch. Some of these characteristics are not perfectly clear from the instructions:

The operating current of the switch is drawn from either stud terminal A or B, not from the red Control input.

The continuous operating current of the switch is specified as <13mA. I measured 4mA.
The changing state current is specified as <7amps. I measured a peak of 2.5A.
Again, this current comes into the switch from either of the A or B stud terminals, and then goes out via the black ground wire.

The red Control input draws 4.3mA. There is no increase or transient in this current when the relay is changing state. This is simply a “logic” input to the 7713 switch. There is no need to protect the circuit that drives this (red Control) input with a “freewheeling”, “snubber”, or “flyback” diode, as is customary for circuits that control a solenoid.

If the yellow toggle is turned to the right the switch is locked off forever, no matter what the red control input is. The yellow toggle must be manually turned to the left to resume automatic operation.

If the yellow toggle is turned to the left, AND the yellow button is pushed in, the switch is locked on forever, no matter what the red control input is, and through power cycles on the A or B stud terminals. To resume automatic operation, the yellow toggle must be manually turned to the right (turning off the switch) and then to the left again.

Note that this is different from the older 7712 switch. In that version, if the yellow toggle was turned to the left and the relay state (i.e. the pushbutton) was in a different state from the red Control input, the relay state would change state to match the red Control input after 10 minutes. AFAIK, the 7712 switch is no longer available.

Total current usage of the 7713 switch:

When on, 4mA from either the A or B stud terminal, and 4mA from the red Control input, for a total of 8mA.
When off, 4mA from either the A or B stud terminal.
Of course these measurements are from just one sample.

That’s great info, thank you! From a technical standpoint the 7700 is much better but when the BMS expects an auto-releasing solenoid then this is the best option. That said, someone posted schematics to convert the signal so a 7700 could still be used on a standard REC BMS without their own converter.

[CaptainRivet]

@Jedi:
How do you deal with the 2 different batteries Winston and Li-Time in your setup?
The Li-Time really need their 3,65V so the balance properly while the winston are fine with 3,45-3,55V.

The Li-time are controlled and switched off by their internal BMS but the Winston are managed by the REC so its eg easy to cut them off earlier.

But that your install diagramm doesn't allow that because eg You mention in post #1 that cut off relay are in series for alternator.

And why does the first battery eg your winston get a HVC if its shut down earlier by BMS (because eg end of charge is lower or you got a runner) when the 2nd LI-Time gets just charged further because it needs its higher balance voltage? The Winston are already disconnected and HVC should be equal for all batteries/banks in the installation.

[s/v Jedi]

Originally Posted by CaptainRivet

@Jedi:
How do you deal with the 2 different batteries Winston and Li-Time in your setup?
The Li-Time really need their 3,65V so the balance properly while the winston are fine with 3,45-3,55V.

The Li-time are controlled and switched off by their internal BMS but the Winston are managed by the REC so its eg easy to cut them off earlier.

But that your install diagramm doesn't allow that because eg You mention in post #1 that cut off relay are in series for alternator.

And why does the first battery eg your winston get a HVC if its shut down earlier by BMS (because eg end of charge is lower or you got a runner) when the 2nd LI-Time gets just charged further because it needs its higher balance voltage? The Winston are already disconnected and HVC should be equal for all batteries/banks in the installation.

My diagrams are not my installation… I actually create diagrams to help others and call these my reference diagrams. I don’t charge LFP with the alternator nor do I have a REC BMS

[CaptainRivet]

Originally Posted by s/v Jedi

My diagrams are not my installation… I actually create diagrams to help others and call these my reference diagrams. I don’t charge LFP with the alternator nor do I have a REC BMS

how do you have them then connected and do you run them on seperate end of charge voltage, the Li-Time need 3,65V to balance properly due to manufacurer recommendations. Assume you don't run the winston that high..

[s/v Jedi]

Originally Posted by CaptainRivet

how do you have them then connected and do you run them on seperate end of charge voltage, the Li-Time need 3,65V to balance properly due to manufacurer recommendations. Assume you don't run the winston that high..

My LiTime batteries only have cells in series so every cell is guaranteed to get the same charge and discharge so they don’t require constant balancing.

If I want to balance them I can simply run them up to a higher voltage while taking the Winston offline, or I just keep it online as it allows up to 4.0V per cell.

[svloki]

Great work on the reference diagram, very helpful.


Can you please tell me what the devices are between the Victron DC-DC chargers and the distribution panel? I'm guessing contactors. A spec would be great.

[s/v Jedi]

Originally Posted by svloki

Great work on the reference diagram, very helpful.

Can you please tell me what the devices are between the Victron DC-DC chargers and the distribution panel? I'm guessing contactors. A spec would be great.

Thank you. They are fuse boxes. I like the BlueSea SafetyHub: https://www.bluesea.com/products/774...150_Fuse_Block

[svloki]

Got it thanks.


Can you explain the need for the alternator disconnects in your diagram? I would think the Victron DC-DC chargers provide a path for the alternator output to the start bank in the case of a lithium bank disconnect.

[newhaul]

Originally Posted by svloki

Great work on the reference diagram, very helpful.


Can you please tell me what the devices are between the Victron DC-DC chargers and the distribution panel? I'm guessing contactors. A spec would be great.

Fuse blocks / distribution panels .

[newhaul]

Originally Posted by svloki

Got it thanks.


Can you explain the need for the alternator disconnects in your diagram? I would think the Victron DC-DC chargers provide a path for the alternator output to the start bank in the case of a lithium bank disconnect.

Don't know about Jedi but on my boat the alternator cut out switch in the exciter wire is to allow maximum engine hp to the shaft when needed.

It takes approx 1hp to run a 30 amp alternator at maximum output

[svloki]

I think Jedi has the BMS triggering alternator field disconnects. Good design for sure, save the bridge if BMS disconnects the lithium bank.



I'm just thinking the alternators may be able to stay online if enough output goes through the DC-DC charger to save them in a shutdown event.


Has anyone tested this scenario?

[wholybee]

Originally Posted by svloki

I think Jedi has the BMS triggering alternator field disconnects. Good design for sure, save the bridge if BMS disconnects the lithium bank.



I'm just thinking the alternators may be able to stay online if enough output goes through the DC-DC charger to save them in a shutdown event.


Has anyone tested this scenario?

Do you mean the alternator connected direct to the LFP, and then a DCDC to charge a Lead Acid? I am not aware of anyone testing this, but I expect with a significant load dump the DCDC would be smoked. Probably why it has not been tested, the expected result is loss of expensive equipment.

Consider, first if the Lead Acid is charged and the DCDC in float, then almost no current would flow to the Lead Acid, and a high voltage would be applied to the DCDC input, smoking the DCDC.

Even if the Lead Acid were dead and accepting 30A of charge, it will never exceed 30A. So if the alternator was outputting 120A, the BMS disconnects and the alternator now is only outputting 30A, that 90A drop is going to raise the voltage on the input of the DCDC.

The whole key to protecting from a load dump is that there is somewhere for that current to go. It might only be a few milliseconds, but you need to allow the current to rise.

[svloki]

Originally Posted by wholybee

Do you mean the alternator connected direct to the LFP, and then a DCDC to charge a Lead Acid?

Yes the reference diagram shows that connection scenario.

I take your point, very short duration pulse but still too much for the charger to handle. Also seems no way the fuses between the charger and the lead acid batteries would be fast enough to protect from such a short pulse. Good system.

I saw a schematic with BMS-controlled contactors to disconnect solar panels from the Victron MPPT controllers as well. Same logic here on load dump?