I just want to add my 2 cents. I'm a liveaboard doing a world circumnavigation on a 55ft monohull with my kids. I built a 24V 840Ah EVE cells battery. My advise is to keep it as simple as possible as the battery and power system is actually a thing you want to build and forget.
I choose the Batrium BMS (cost about 1000USD) and all other equipment is Victron. The boat has an 11kW generator, and I've a Quattro 120A charger / inverter and a Skylla i 100A charger. So I charge at around 200A. A Cerbo GX with display is used as central hub. The alternator is controlled by a Wakespeed WS500.
Some advice from living 8 months on board:
- If possible use CANBus to communicate. The Batrium BMS is fully compatible and shows all information on the GX display. The system uses DVCC so the BMS is in charge of communicating to the chargers and sets their voltage /amps. This is more advanced than you would think. The Batrium BMS can balance each cell with 2A, so when balancing is needed it lowers the amps of the charger to a much lower value to be sure it can be burnt off.
- Check how your BMS triggers balancing. There are multiple strategies (above a certain cell voltage, auto levelling during charging based on voltage difference, ...). If you balance incorrectly or with to much current, you'll be inefficiënt and burn off a lot of heat. Chargers have no clue on balancing, so forget "bulk, float, storage, ..." modes. The BMS should be in charge and the whole system has to be aligned properly.
- Not all equipment can talk through canBUS (eg Victron Skylla-i). There you'll need a hardware shutoff wire. This trip should be at a lower level than the big critical shutoff. If not, you're alternator will spike the voltage for a split second and it will destroy your equipment.
- Charge / discharge relays won't work with Multiplus / Quattro chargers/inverters as they both charge and discharge
- Hardware issues:
- Most BMS measure voltage on each cell. These measurements have inaccuracies and can drift over time.
- Does the BMS allow for voltage recalibration or voltage offset? (Batrium does)
- What will you do if a cell runs away, take spare cells with you on the boat as you cannot fly them in. I made this mistake and can only get spare cells through shipment on a ship. These cells are not allowed on planes (even cargo).
- Same for BMS hardware. I have 20% spare "blockmons" with me. These are the distributed cell monitors that report to the central Baterium Watchmon unit.
- You need to protect the system on multiple levels:
- Cell level: voltage (will you measure each cell, or will there be cells paralleled)
- Shunt level: voltage and amps (you need to protect the max amp draw to protect your contactors, eg max 500A)
- Heat: you need to monitor balancing heat, the BMS will need to be configured to allow for external fans to be activated
- ...
- The Batrium has several layers of protection than can all be configured. In my setup:
- @ 3.5V balancing starts, output amps of charger is reduced to allow for dissapating heat. If you're not using canbus, the charger will keep pumping full amps into the battery.
- @ 3.6V charging is stopped, this is communicated through CANBUS and through a relay to shutoff the "dump" equipment
- @ 3.65V critical stop is triggered, and this shuts the big 500A continuous relay at the bank itself
Jan,
Thanks for your real life feedback. Also 100% lifeaboard on a 10years World circumnavigation.
Batrium is a great BMS but also very expensive.
A lot of good and valuable points here.
Some comments:
Multiplus/Quattro: you can switch on/off charger with remote control AUX1 and inverter with AUX2, instead a massive relay/SSR you just use a 5Euro small optocopler to cut in LCV or HCV event.
Spare cells: well a 2 sided sword…you need to keep them at 20-40SOC and they won‘t fit with the rest of the pack as this degrades while spare is lying around unused. My solution is stay 12V (high current loads are within cable length of 60cm from the bank) as 4 cells in series drift apart much less then 8, if you then also parallel cells you multiply chances to drift. also if 1 or 3 cell dies I only loose capacity of 4 cells. 2nd part of my solution is oversize the bank and use your spare eg enough is 560AH for me but I added 4 more to 3p4S 816AH (Lishen grade A 272AH, has 1C rate compared to EVE with 0.5C), so I have 4 spares and if one or till 4 cells run off I just reconfigure the bank. Also lowers the c-rate per cell for discharge. From experience for a 24V system I would not parallel cells and get 8 Winston cells in the capacity needed.
Hardware:
Cell voltage inaccuracies and drifting: even the cheapest BMS measure the cell voltages correctly, that’s kindergarten even for a cheap Daly. If you have inaccuracies then something is definitely wrong at the heart of your system and it’s Bad practice to correct that with the BMS…you need to fix it. Contact or oxidation/corrosion issue
- bad crimp of the sensing wire lugs is issue no1. These are often tiny wires, make sure you have the right tinned ! lugs for that cable size. After crimping you can solder only!!! the side facing inwards to the lug with a tiny! bit of tin to the lug to make sure it has proper contact. The other crimped end need to stay flexibel, no tin here at all!
- 2nd bad contact of the lug with the busbar, only use tinned lugs and tinned or silver plated Cooper busbars…the supplied ones are crap.
- 3rd issue these cells have Alu terminal that immediately oxidize at air and this oxide on Alu is very hard and bad conductive. Same applies to the cheap Alu busbars(See 2)…you need to sand with 2000grid wet or black scotchbride and then clean with 99%, after this put No-OX-AL or similar to insure good contact while air is sealed off to prevent oxcidation.
- 4th issue: Cooper busbar,sensing lug, serrated flange nut…no washers. Nut tightened to manufacturer of cells spec…EVE i think 8nm, check please. Secure nut with a drop of any nail polish available (from wife, girlfriend or get cheapest at supermarket).
- 5th issue: if sensing wire is not supplied, use tinned stranded marine grade wire and all wires from the same coil and exactly the same length.
If drifting cell measurements over time you have a contact issue, mostly oxidation or corrosion over time, again fix and NOT correct it with the BMS.
1) your bank is in salty marine environment, so your battery box has to be air tight
2) use no-OX-AL on all Alu surfaces in contact
3) bad contact see 4) above, only busbar, lug and serrated flange nut and re tighten the nut to highest allowed spec and secure with nail polish. no loctide here!!!, the nail polish stays flexible and fixes the nut but if you apply slight force it breaks much easier then loctide
4) seal from air/humidity busbar, stud, lug, I use liquide rubber product called plastidip that can be easily pulled off if you need to
BMS parameters: for liveaboard circumnavigation from a safety standpoint I find your settings too aggressive, you don‘t need the last 5% capacity and use the upper leg of the charge curve. Important they are measured correctly at the cells with correctly well connected sensing wires.
Stop charging at 3,5V or better 3,45V, leave yourself more room for errors plus LifePo4 likes to stay between 30 and 70SOC and not full…
3,65V is 500A relay disconnect all charge sources in HVC (that would be correct)? or disconnect the whole bank= incorrect as you need loads to drive bank asap below 3,65V and therefor shouldn’t shut all off.
Suggest a 3rd 3,75V as Desaster disconnect all. My Electrodacus BMS has separate stop charging, HVC and a fault condition settings, batrium should have them too.
Just my 2 cents…
Happy sailing
Christian