Merged LiFeYPO4 1000Ah Winston prismatic cells and all electric galley...

[missourisailor]

CatNewBee,

I have seen your drawing. And by the way, nice job with them.
And hence the reason I asked the question.

I too am using the ML-RBS's.

Because they operate with a pulse current, I talked to Blue Seas about what harm or effect would happen if a solid input signal was applied instead of a pulse.
They said a few hours or maybe a day, most likely would have no effect on them other than getting rather warm. However after I mentioned that they could be energized for weeks if not months at a time, they would in no way stand behind this kind of use.

Hence I had to build a "One shot" circuit to control them.

From your drawing/s It looks like you are supplying a constant input (voltage/current) to your ML-RBS's.

I was just wondering how your ML-RBS's are holding up?

[CatNewBee]

No, look at the drawing, it is a short pulse of 50ms created by the capacitor in line with the input. When powered on, the capacitor pulls up the input and than no current is flowing when it is charged, so the input is then low again.

When the signal goes down, the capacitor discharges quickly over the resistor and the LED to accept the next signal if necessary.

It is a 100 microfarrad capacitor, discharging takes about 500ms. Without the Resistor and the LED it would take a minute or longer until the capacitor looses enough charge and can pull up the input on a subsequent high.

You can test it easily with a 12v power supply, connect red and ground and attach a capacitor one side to plus and the other to the R (brown) respectively S (orange) contact, the relay will flip once, but as long as the capacitor is charged, you cannot repeat the switching. Short the capacitor and try again, it will switch again.

So the whole logic is a capacitor and a resistor for discharge, the LED is a bonus to have a free feed back on the current input state.

You loose 10mA per resistor and glowing LED when running, so my interface draws 50mA continous, not a problem with solar and 1000Ah.

If you connect a static signal, the ML-RBS draws around 10 times more energy - approximately 100...120mA.

[missourisailor]

Originally Posted by CatNewBee

No, look at the drawing, it is a short pulse of 50ms created by the capacitor in line with the input. When powered on, the capacitor pulls up the input and than no current is flowing when it is charged, so the input is then low again.

When the signal goes down, the capacitor discharges quickly over the resistor and the LED to accept the next signal if necessary.

It is a 100 microfarrad capacitor, discharging takes about 500ms. Without the Resistor and the LED it would take a minute or longer until the capacitor looses enough charge and can pull up the input on a subsequent high.

You can test it easily with a 12v power supply, connect red and ground and attach a capacitor one side to plus and the other to the R (brown) respectively S (orange) contact, the relay will flip once, but as long as the capacitor is charged, you cannot repeat the switching. Short the capacitor and try again, it will switch again.

So the whole logic is a capacitor and a resistor for discharge, the LED is a bonus to have a free feed back on the current input state.

You loose 10mA per resistor and glowing LED when running, so my interface draws 50mA continous, not a problem with solar and 1000Ah.

If you connect a static signal, the ML-RBS draws around 10 times more energy - approximately 100...120mA.

Ahhh,
I see the cap/s. now.
Us Yanks, draw electrolytic's different. Hence why they did not jump out at me.

Very cleaver way to do it.
I chose the much more complex way, using logic circuits and timers.

I like you way much better. In fact, I'm going to do some testing and change my triggers to caps.

Good job!!!

[CatNewBee]

Thanx, my first idea was also to use monoflops, but than I recall my old days projects where I used to use capacitors and resistors for power-on reset signals on CPU circuits and CMOS / TTL logic to have a defined boot impulse. It is a pretty similar principle.

I gave it a shot and found out, the ML-RBS have internaly digital inputs, so I tried various capacitors, 47microfarrad was the smallest that worked reliably in 9V power supply, then I doubled to 100microfarrad for safety. You can use larger capacitors, but it would add unnecessary stress on the inputs when charging and discharging, because the currents and the signal length is then higher.

I wanted to have safe switching even when the batteries are ways too low.

[CatNewBee]

Winston Cells testing by the Chinese Army...

https://www.ev-power.eu/blog/Tests-a...ety-tests.html



The photo shows the punch-through simulating a shooting from 45 mm calibre anti-tank gun.

Fully charged cells are punched through simulating the shoot-through from a military weapon. The purpose of the test is to see the safety and stability of the lithium cells.

As visible from the photo, the cells remained complete, did not burn, did not set on fire and did not explode. This kind of extreme abuse proved that superiority of the Winston Battery lithium cells. These cells are suitable for all kinds of applications including the heavy duty and military grade installations.

Hope this helps, when trying to scare people away from LFP usage

[hzcruiser]

Originally Posted by CatNewBee

Winston Cells testing by the Chinese Army...

https://www.ev-power.eu/blog/Tests-a...ety-tests.html

Hope this helps, when trying to scare people away from LFP usage

Thanks for the link, CNB! It's a new technology and people are understandably sceptical. One important factor to their confusion is not knowing there aree different "flavours" of Lithium batteries with different safety profiles.

So far, the press is quite often oblivious to this, too, which doesn't help either.


Having said that, I'm certainly not pushing anyone towards LFP. If they're happy with lead and enjoy the seemingly lower cost they can stick with it as long as they don't ask me to help them moving those weights around their boat.

For me, every LA battery I'm taking out has been replaced by LFP, one by one. That way I can get the max life time from the LAs while not having to shell out a big sum up front to replace them all at once.

[hellosailor]

If those Winston cells won't stop a 45mm. antitank round, what use are they? Useless.

[AndyEss]

Originally Posted by hzcruiser

Having said that, I'm certainly not pushing anyone towards LFP. If they're happy with lead and enjoy the seemingly lower cost they can stick with it as long as they don't ask me to help them moving those weights around their boat.

For me, every LA battery I'm taking out has been replaced by LFP, one by one. That way I can get the max life time from the LAs while not having to shell out a big sum up front to replace them all at once.

Ditto for me. I still have 2 x 6v FLA batteries in my house bank - and my starting battery which is FLA.

The starting battery will probably be the last to go.

One of the benefits of LFP weight and size, is that small, low AH batteries can be placed near loads, far from the house bank. I replaced my hand pumped toilet with an electric, macerator toilet. The wiring to the head (at the bow of the boat) was inadequate, and I was preparing to rewire. Running new wiring in a wood rich Hans interior is not easy at all. Instead I put in a small 20AH battery near the toilet, wired like a capacitor in parallel with the inadequate run. The macerator pump runs fine now because voltage drop is minimal.

My back heartily thanks the existence of ever more affordable LFP batteries.

[SFH]

CatNewBee,

I am using your design as inspiration for my own project. I believe I have come to grip of most of it but I have a few questions that I hope that you can clarify

1. In one of the circuit drawings you have shared the Charge Relay OVP ML-RBS connects directly to the charge bus (Solar, 2xAlternators and Christec)

2. On the other circuit drawing that you have shared, the charge control circuit showing the connection between the BMV712 and the CHG ML-RBS also connects directly to the charge bus but here the drawing shows that the Charge Relay OVP ML-RBS connects in series as you state in the desciption "The legacy charger solenoid connects in series to the OVP (over voltage protection) charge solenoid, that shuts off the solar controller.

3. Which one is correct? My guess would be 2. but then I cant see why you need the OVP circuit since the BMV712 will disconnect from charging at the preset charge point - long before you would reach a high voltage situation

4. I read it that you have kept your existing battery isolator but in the drawings you have shared you route all charge sources - legacy and solar - directly to the charge busbar. Have I misunderstood that? How do you charge your start battery and bow truster bank?

My own logic would say that I route all my charge sources to the input side of the Battery Isolator and connect the "house bank" output from the Battery Isolator to the BMV712 controlled ML-RBS and then keep the other outputs from the Battery Isolator to the start and truster banks as-is. somewhat like what you write below. Am i wrong?

Many thanks in advance

//Steen

Originally Posted by CatNewBee

Attachment 170649


Here the interfaces between the REC ABMS, the Victron BMV712 and the bi-stable 500A solenoids ML-RBS.

The circuitry and programming do the following:

the 2 relay outputs of the ABMS are used as smart fuses, they cut charge and discharge circuits off if the cell voltage goes out of an acceptable range.

the Quattro is considered a load and the solenoid is connected to the UVP (under voltage protection)

The legacy charger solenoid connects in series to the OVP (over voltage protection) charge solenoid, that shuts off the solar controller.

Legacy sources are the factory Cristec charger and the outputs of the charge distribution / battery separation diodes of the alternators.

This sources are turned on and off by the BMV712 relay based on a programmable SOC (switch on when SOC is below 50% and off when SOC reaches 80%). They are only used if the solar is unable to handle the SOC alone.

In regular operations (SOC between 51 and 100%) they only charge the spiral cells AGM start batteries.

[CatNewBee]

Originally Posted by SFH

CatNewBee,

I am using your design as inspiration for my own project. I believe I have come to grip of most of it but I have a few questions that I hope that you can clarify

1. In one of the circuit drawings you have shared the Charge Relay OVP ML-RBS connects directly to the charge bus (Solar, 2xAlternators and Christec)

2. On the other circuit drawing that you have shared, the charge control circuit showing the connection between the BMV712 and the CHG ML-RBS also connects directly to the charge bus but here the drawing shows that the Charge Relay OVP ML-RBS connects in series as you state in the desciption "The legacy charger solenoid connects in series to the OVP (over voltage protection) charge solenoid, that shuts off the solar controller.

3. Which one is correct? My guess would be 2. but then I cant see why you need the OVP circuit since the BMV712 will disconnect from charging at the preset charge point - long before you would reach a high voltage situation

4. I read it that you have kept your existing battery isolator but in the drawings you have shared you route all charge sources - legacy and solar - directly to the charge busbar. Have I misunderstood that? How do you charge your start battery and bow truster bank?

My own logic would say that I route all my charge sources to the input side of the Battery Isolator and connect the "house bank" output from the Battery Isolator to the BMV712 controlled ML-RBS and then keep the other outputs from the Battery Isolator to the start and truster banks as-is. somewhat like what you write below. Am i wrong?

Many thanks in advance

//Steen

that was the original design.

there are 2 charge relays in series. the first one next to the plus bar disconnects everything charging as last line of defense, now programmed to 3.75V cut off cell voltage. it will never cut in normal operation.

The second relay is operated by the BMV and can be configured to your liking based on calculated SOC.


The new design, not fully tested and explained yet in the thread, has a signal wire from the opto coupler output relay to both, the Victron Quattro and the Victron SmartSolar controller to stop charging gracefully if one cell exceedes 3.65V considered 100% SOC. This can happen on high current charging when a cell gets out of balance because of inner resistance variations. The BMS reports differences of up to 0.035 mOhm between the cells. But this could also be a result of conductivity variations on the poles.

I have observed this on my battery with balanced cells (difference 3...4mV) on a small current, but out of balance on high charge currents of 250A and more of up to 30-40mV between the highest and lowest cell. Usually I stop charging at 14.2V, or 3.55V. The opto-coupler signal would shut the chargers without disconnecting the bus. Some solar controller do not like it to be disconnected by brute force from the battery when under full load. You can also use the opto coupler signal to disconnect the solar panels from the controller instead disconnecting the battery if your controller does not have a remote control input.

Regarding the isolators, they are in place as original, only the house bank wire was cut off by the solenoid controlled by the BMV. The starter battery wiring remained untouched.

[BrissyBrew]

Originally Posted by CatNewBee

The water heater is really a power hog.

It takes 110A and needs 1h 20min to heat 40l of water, that translates to almost 150Ah for hot water (it is a 1200W / 220V unit). Will see how to handle this on the way. Probably run it only when the solar goes to absorption or maybe earlier, when SOC goes over 95% at the end of charging...

I am putting electric hot water in with a 400AH @24V pack running via victron multiplus inverter 230v, any ideas of how I might program it so I can dump excess solar power to hot water always, disconnecting solar seems a waste if I could be heating the water up.

[CatNewBee]

Originally Posted by BrissyBrew

I am putting electric hot water in with a 400AH @24V pack running via victron multiplus inverter 230v, any ideas of how I might program it so I can dump excess solar power to hot water always, disconnecting solar seems a waste if I could be heating the water up.

If you have a Victron BMV, you can use the buit in programmable relay contacts based on SOC to turn on the water heater by an 230V relay suitable for the current expected. It depends on your set up.
Alternatively with a Victron MPPT controller you can use the programmable relay contacts to turn on the water heater when the controller goes to float.

[john61ct]

Be aware, load contacts are pretty limited amps

[senormechanico]

The MPPT load contacts can drive a big relay coil which switches the tank element.

It's not rocket science.

[BrissyBrew]

Originally Posted by senormechanico

The MPPT load contacts can drive a big relay coil which switches the tank element.

It's not rocket science.

Interesting, so I would want to heat hot water when SOC is higher on the battery bank, say 90% SOC then start heating hot water via inverter. Might have a look at the multiplus see if there is a way to program the inverter (could do so on battery voltage SOC maybe).