I'm converting my 1979 Cape Dory 30 sailboat from a 14 hp diesel to a 10Kw 48V electric motor (the motor is ordered) ... the article that was attached to one of the threads say that going with cells over 200Ah for a marine application is not a good idea.
Welcome to the build your own battery club! I know just where you're coming from - I'm looking at my glorious old YSE12 and thinking that sooner rather than later I need to either rebuild or repower. Now I'm charting a course for repower and have likewise concluded that 48v is the only way to go. But I will approach this incrementally by first adding electric auxiliary propulsion, a pair of these:
Caroute S400-48V 180LB Electric Boat Trolling Motor with speed controller 180 Pound Large Thrust Saltwater For DIY
Only when I have solved all the other issues, including battery, solar array and controllers, will I feel brave enough to hoist the old diesel out and bolt on a 10kw motor. That's just me, I applaud you for jumping right into the thick of it. Many of the issues we have to worry about are exactly the same. So...
Solar array. I currently do just fine with a 200 watt array of two compact panels in parallel, with a cheap PWM controller. But this will not do for 48 volts, I will need to go to at least four panels in series. Reason: the solar controllers we use are buck converters, they can only reduce voltage, not boost it. So you need just shy of 60 volts to reach 100% SOC, 3.65v. (16 x 3.65 = 58.4) Four panels can just barely do that, especially in the morning or evening or on overcast days. The other day I saw 16 volts, 1.5 amps in the morning. If I only had three panels I would just be throwing away that precious 24 watts because it would be ten volts shy of the battery voltage.
At high noon you will do better, probably 20 volts, and three panels will be enough to feed your batteries. But for a good chunk of the day it won't be enough. This one is clear: four panels or more to feed a 48 volt system, otherwise there will be large chunks of the day when you get no charging at all.
Battery. I have my own reasons for going with smaller cells, but I would really like to know the specific reasons in the article you mentioned, can you find it again? For me, it is mainly about manageability. Sixteen 280ah cells weigh around 200 pounds, and that is more than I want to wrestle with as a single unit. Also awkward to fit into the odd shaped spaces I have available. My first prototype pack will be 80-100 pounds, roughly 5 to 7 kwh. That will be enough for sea trials of my electric auxiliaries, and also limit my risk during my learning period.
Battery management system. I have pretty much settled on the Overkill 16s:
16s BMS 100a LifePo4 Battery Management System for 48v DIY Batteries
Reasons: the documentation is worlds better than the Daly. In particular, the way this controller handles high voltage cutoff is spelled out clearly: don't disconnect the load, just disable charging. Thank you, that's exactly what I want. The possibility that my boat could shut down entirely just because some cell got fully charged makes me shudder. I can't say for sure that this is what the Daly does, but there's no documentation so how would I know? I just can't live with that.
Another huge reason for going with Overkill: the open source aspect. This is big with me. I want to really know what is going on inside and I want to add computer intelligence as time goes by, because that's what I do. You can sort of do that with good published specs such as Victron's, but maybe not exactly the way I want. And I certainly can't do it with Daly's undocumented black box.
For your project, the Overkill's 100 amps is not enough. But why not go with smaller cells and build multiple packs, each 100 amps? Two packs with 120ah cells should be fine. You can always add more later.
A major benefit of multiple smaller packs is redundancy. Sooner or later a cell is going to fail. With one huge battery pack, you aren't actually dead in the water because it's a sailboat, but close to it. Going to need a tow at least into the marina.
Downside of smaller cells: they cost a few cents per amp/hour more. They require more components - each needs its own controller, fuse and switch. More wires, more assembly. But on the other hand, that's the fun part, isn't that why we're in here? And oh, less than half the price of store-bought. And you can craft the design to fit your boat elegantly.
Battery enclosure. So this is where I get to the fitting elegantly part and it's what I've been obsessing a lot about lately. I've been hoovering up papers and forum threads, trying to answer questions like: is compression really a thing? What about heat dissipation? So I will share some thoughts/opinions/tentative working designs now.
Compression: much internet ink has been spilled about it but little engineering insight is provided. I learned that LFP cells do swell, certainly when stressed and to a lesser but measurable extent in normal operation. For example, your 10kw motor with two 5kwh battery packs will discharge at 1C, full power. Well below the 2-3C rate where swelling seems to be reported in the literature. There is talk of dendrite formation and delamination in cases of extreme swelling. Cells forced apart can stress and damage the terminals.
OK, in all of these cases I don't see how compression actually helps. I do see that fixing cells in position is important, both to protect the terminals and to generally reduce wear. So on to my design.
My battery is going into a dry part of the boat, so a waterproof box will not do much for me other than to take up space, add weight and make the battery harder to inspect. I am uninterested in compression, so I will do away with the tie rods and hefty end plates. Instead, I will fabricate an aluminum box to go on each end of my 2x8 cell pack and build a frame with four pieces of angle stock, bolted to the end boxes. The BMS goes in one of the (open on the outside) end boxes. With this approach I don't put pressure on the cells, but I do fix them in position. That should take care of the terminal damage issue and also eliminate wear due to chafing.
I am still working on details. Thermal dissipation concerns me, however as I have found little to no data to shed light on the question, I think I am just going to build a battery pack and find out. I might put rubber washers between the cells as spacers to let air circulate between the cells. Thermal dissipation is not the only reason for this - it might improve toleration of vibration and it could prevent shorting between aluminum battery cans should holes develop in the delicate looking plastic cover.
Well, I really need to post a drawing. Maybe here, or in show and tell? I do want to get some feedback on my design decisions, and I hope there's something useful here for you. Executive summary: I am building a frame, not a box, a press, or a medaeval torture device.
Legacy 12 volt subsystem. Nearly all of the existing 12 volt subsystem is going to stay: starter motor, cabin lights, electronics, etc. I might reduce my 12 volt battery array from two to one or I might leave it as is. The main change is the solar charging system, it becomes 48 volts. So I need to charge the 12 volt system from the 48 volt system. My tentative solution is to connect a standard MPPT charge controller to the 48 volt battery bus and let it do its job as a buck converter with battery charging intelligence. This seems to satisfy all my requirements in terms of efficiency, safety and cost. If anybody out there has a better idea I would love to hear about it.
inverter. This is where we get serious payback for all the investment in 48 volt systems. The inverter gets a whole lot more elegant. Suddenly, 2000 watts of continuous power is not involving short, scary, thumb-sized cables and massive fuses and the device shrinks to a reasonable size. Yay.