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

[LeighC]

Thanks for all your help, you make a lot of very good points, a lot of things to are too smart for their own good today.
I appreciate your good knowledge & it will help me to make a very good decision, I think I will stick with what I have got & buy a larger latching relay & one with manual override.
Cheers - Leigh

[pretzel]

Originally Posted by CatNewBee

Quattro is set to 14.4V Absorption, shortest time, 13.5 FloatAttachment 223255Attachment 223257Attachment 223258

Hello CNB,
hope you guys are fine and healthy. Saw your battery thread in the german RV forum. I wonder whether you have any plans to upload your python scripts for extra info in venus OS to Github one day? Would be really neat to start from there . There was one other guy in the RV forum asking for this.

Originally Posted by CatNewBee

Yes, you can integrate the REC via CAN Bus and stop charge by cell voltage setting, you can set up a distributed network of REC - CAN - Cerbo - Serial Bus - Quattro, Cerbo - Serial P2P to solar, P2P to BMV etc. and hope for the best. It is easy. But you still have to configure each unit to work properly independently if the data connection fails.

I actually rely on the data connection to work. My Victron MPPT does not have an remote on-off input. Sometimes I use my battery main switch to cut off the power of my RV. The Victron MPPT is then disconnected from the battery but still gets power from the solar panels during the day. However there is no raspberry pi/venus os communicating with the MPPT anymore. From the app I am able to see the Victron MPPT then goes into an error mode saying BMS connection failed and that it stopped charging. So there is a safety measure in place.



What seems to be funny: Even with the error in place there is still enough juice to power my compost toilet fan even with no battery connected to it. Quite handy actually Never tested the max possible current however ...

[CatNewBee]

Hi, well, I don't intent to publish it on GITHUB, I did it quick and dirty while learning about the Venus and reverse engineering some of the scripts. It was never intended to be published, also the changes I did were severe, not only for the REC, but also for the whole OS based on try and error. It runs fine, but it's not updateable.

It included the 1wire libraries, I2C libraries for temperature, moisture and pressure sensors, a newer python version side by side, newer serial drivers, video and touch screen patches, all extracted from the Raspbean and patched into the Venus OS for my other hardware without a proper recompile with dependencies. The scripts are developed and debugged in that environment and have dependencies to this libraries and to each other by the underlying extended data model.

One can port them back to the older python version used in the original Venus and leave off all rPi specific extensions, but its a PITA.

Also the changes are done on multiple levels, new services to communicate with the hardware and feed the dbus, changes/extensions to the venus dbus routines to aggregate the data and changes in the qml-UI to read and show the sensor data up.

It is not too much, and I can deal with it / live with - because I know what i have done, but it's nothing for the wild outside without that background.

It is just not a clean implementation, but rather an experiment that workes well here, but may not work with a newer Venus version - and I am not developing it further at the moment, so cannot support or help others get it running on their hardware.

So GIT is probably not the right place to publish this stuff yet.

But I can encourage you to take a look at the Venus and make your own changes according to your skill level.

Btw, it is not healthy / supported to run the MPPT without a battery connected, you may get hardware issues. Better disconnect the panel side by a relay and keep the controller connected to the bus.

[CatNewBee]

A screenshot when running the generator during the day

Input

5000W from Generator
1360W from Solar

AC Loads:

4500W charge to the battery
- 1743W AC loads
- 1000W watermaker
- 750W Cristek's chargers

DC loads:
300W fridges, freezer, house loads

The Cristek's contribute net charge current of 36A, while the rest is powering the house loads.

Total charge current to the LFP:
331A
- 195A from Quattro
- 100A from Solar
- 36A from Cristek chargers

Just to show how a complex system works.

Also see the pack voltage of 13.6V at SOC 60% and 500Ah to be charged. If you limit charging to that voltage, you would miss a lot of charge current and likely will not get your battery full.

[jmh2002]

Cool

[CatNewBee]

Cooking on battery. 2 induction plates in full power use (2x level 9, no boost function on), a pan to fry meat and a veggie pan.

[jmh2002]

Originally Posted by CatNewBee

Cooking on battery. 2 induction plates in full power use (2x level 9, no boost function on), a pan to fry meat and a veggie pan.

Nice info, thanks

[aybabtme]

Blue Saga looks awesome, my dream built out!

[CatNewBee]

April in Saint Martin, Caribbean just for reference on energy harvesting on 1650Wp.

[T1 Terry]

Originally Posted by CatNewBee

Also see the pack voltage of 13.6V at SOC 60% and 500Ah to be charged. If you limit charging to that voltage, you would miss a lot of charge current and likely will not get your battery full. Attachment 235710

The slightly elevated voltage is due to the high charge rate. This is why it is still advisable to use the 3 levels of charging Boost/bulk set at 14v for when the high current is coming in, 13.9v absorption limited to 15 mins and adjust from there depending on the current flow when the system drops to float voltage at 13.8v for solar and 13.7v or even 13.6v for mains/generator and possibly DC to DC, really depends how much you run the motor as to whether you use 13.8v float or long motor runs when the system would be in float mode, 13.6v.

T1 Terry

[aybabtme]

Doesn't the REC BMS communicate in real time it's allowable charge current on CANBUS?

[newhaul]

Originally Posted by T1 Terry

The slightly elevated voltage is due to the high charge rate. This is why it is still advisable to use the 3 levels of charging Boost/bulk set at 14v for when the high current is coming in, 13.9v absorption limited to 15 mins and adjust from there depending on the current flow when the system drops to float voltage at 13.8v for solar and 13.7v or even 13.6v for mains/generator and possibly DC to DC, really depends how much you run the motor as to whether you use 13.8v float or long motor runs when the system would be in float mode, 13.6v.

T1 Terry

Except lfp /lfpy dont like absorption nor do they need it . I prefer to not kill my bank

[T1 Terry]

Originally Posted by newhaul

Except lfp /lfpy dont like absorption nor do they need it . I prefer to not kill my bank

Charging at 14v in bulk/boost will result in the system dropping back to absorption mode early, that is why a short absorption cycle is needed. If you have a proper BMS the charging will be cut if a cell goes over 3.6v no matter what charging regime is being used, so there is zero chance of "killing" your lithium bank .... of course if you don't have a proper BMS that monitors cell voltages ..... good luck, nothing will stop your lithium battery bank from being killed. A low end of charge voltage choice just means the cells won't fully charge, the lowest cell with just keep getting lower and it will be the first to hit zero voltage while the other cells still hold above 3v. The next draw on the battery pulls current backwards through the cell that was completely drained and the cell is as good as dead, they can't be recovered from a reverse current flow. Less damage occurs in the upper voltage range than the low to zero voltage range, capacity is lost and cell resistance increases with a cell over voltage event, total cell failure occurs with a reverse current flow event.
If you aren't monitoring the cell voltages you really don't have a clue how the over all health of the battery is going

T1 Terry

[more]

Originally Posted by CatNewBee

we had to run the generator frequently to recharge the batteries, make water and hot water, cook and bake etc. It was often either overcast or the sails shaded the solar panels, so we cannot rely on solar, also the crew was really not in water saving mood, we had our daily hot showers and produced a mess in the galley to fight boredom, created fancy meals, baked cakes and cookies, variations of bread and on and on, run the washing machine and whatever... We run the watermaker in total for 63.5h, producing 6350l of fresh water for the crossing or 265l per day. Other boats make the whole passage with 300l of fresh water...




- 7h on the engines (14l Diesel) for 2800nm,
- 47.6h in total on the Onan generator (94l Diesel) - so a total of
- 108l diesel for the whole passage or 4.5l/day.
- genny used about 2h/day to charge 500Ah, make water, hot water

Having a partly bored crew, the passage turned in a cooking and baking tournament and having all electric galley with all the toys to play, that meant a huge total consumption of 327kWh on the passage,

- 91kWh comming from solar
- Solar yelded in average 3.8kWh/day, not too bad.
- 236kWh from the generator, that produced in average over 5kW - quite efficiently.

- 63.5h on watermaker in total
- produced and used 6350l of fresh water for showers, dishes, laundry, toilets.
- average water consumption 265l/day,
- total energy for water making about 63kWh.

- Freezer and 2 fridges consumed (18..23A) about 6.5kWh during the passage
- instruments, autopilot, vhf, radar..., maybe another 6kWh.
- Hot Water (1.2kW element, 1h/day) accounted for about 30kWh,
- 220kWh were used on laundry, induction cooking, baking every day fresh bread, baking christmas cookies, cakes, entertainment...
Attachment 231761

Good,installation. but if I be on your place I be use every excellent your BenQ 96-cell panels Uoc 60V with 1 MPPT 30A

[newhaul]

Originally Posted by T1 Terry

Charging at 14v in bulk/boost will result in the system dropping back to absorption mode early, that is why a short absorption cycle is needed. If you have a proper BMS the charging will be cut if a cell goes over 3.6v no matter what charging regime is being used, so there is zero chance of "killing" your lithium bank .... of course if you don't have a proper BMS that monitors cell voltages ..... good luck, nothing will stop your lithium battery bank from being killed. A low end of charge voltage choice just means the cells won't fully charge, the lowest cell with just keep getting lower and it will be the first to hit zero voltage while the other cells still hold above 3v. The next draw on the battery pulls current backwards through the cell that was completely drained and the cell is as good as dead, they can't be recovered from a reverse current flow. Less damage occurs in the upper voltage range than the low to zero voltage range, capacity is lost and cell resistance increases with a cell over voltage event, total cell failure occurs with a reverse current flow event.
If you aren't monitoring the cell voltages you really don't have a clue how the over all health of the battery is going

T1 Terry

Ok let's start with this you are recommending charging to 3.5vpc well thats just missing about 20% of capacity.
Next a bms with a 3.6vpc HVC ? A proper bms with balancing will not even go into balance mode until 3.65vpc .
As to out of balance cells . Well mine have been top balanced 4 years ago and with my charging regime they are still all within .002vpc of each other. I still have yet to rebalance.
Once a year I charge to 3.6vpc for the annual load capacity test. I still get 110% of rated capacity.
Honestly I only check each cells voltage during the annual capacity monitoring.

Question just to be sure we are all talking apples and oranges correctly what is the makeup of your house bank system?