LiFePO4 Battery Build

[dtfeld]

Nice David! I finally have all 12 of my cells now, and just last week purchased the solar controller, inverter/charger, shunt and Cerebro (Yep, all Victron so far). The only thing I'm deciding on now is the BMS, I haven't yet figured out which one I want to go with.

I also haven't decided which configuration I want to go with, I need to do that so I can start building out the box for them. It's likely going to be one of these two, but I haven't figured out the bus bar thing yet.

View attachment 121053

One issue I considered in deciding the config was weight. While true these cells are a lot lighter than LA, in reality a 4x280 or 4x304 cell config is about 50 lbs each. So in you config above, your battery is approaching 150 lbs. If its not strait into its final resting place, you need to figure out logistics. I opted for a dual 4x304Ah cell arrangement each with its own BMS. This allows for smaller batteries, and I can add additional batteries down the road as necessary.

If I were to choose an arrangement, I prefer the first option. If you purchased well made, closely balanced cells, your BMS should keep them well balanced ober the long term. Both offer pros and cons. Thats my "free advice". You know what that's worth.

 

[dtfeld]

Easiest way is turn those batteries upside down and start laying layers of fiberglass over them. But I’d probably build a wood mold around the batteries and fiberglass it instead. The wood can remain inside the fiberglass for structure and reduce the amount of glassing needed. Fiberglass is really easy to work with. It’s the gelcoat and the finishing that takes the most time. You won’t necessarily need that for this project though..Maybe even intergrate those tensioning rods into the box, and leave one end so that you can add tension to the batteries. Unless you can tension them with shims.

I prefer a plastic battery box with a seal, and the ability to add my own terminal posts. Somewhere, I found a company that make a spacer (about 1" tall) that goes between the box and the lid, allowing the mounting of the BMS directly in the lid. I cut down significantly on the internal wiring lengths. However, I have lost the link...

I have given up on the perfect tensioning system. I spent months reading and researching how much, how too, etc. In the end, I'm not letting perfect stand in the way of pretty good. I pre-tensioned with the fixture and taped tightly. Optimal? No but this is how the commercial batteries are supplied. I am accepting a slightly lower life cycle count for ease of assembly, and space savings.

With all that said, there batteries could easily outlive me. And if they don't? There will be a better, cheaper alternative when it comes time.

 

[jmauld]

I prefer a plastic battery box with a seal, and the ability to add my own terminal posts. Somewhere, I found a company that make a spacer (about 1" tall) that goes between the box and the lid, allowing the mounting of the BMS directly in the lid. I cut down significantly on the internal wiring lengths. However, I have lost the link...

All of that could be integrated into a custom box. I doubt you’re going to do this either way, but I would not use c.f. In this application. It does not function well as electrical insulation. Fiberglass does though.

 

[dtfeld]

All of that could be integrated into a custom box. I doubt you’re going to do this either way, but I would not use c.f. In this application. It does not function well as electrical insulation. Fiberglass does though.

The boxes I have a really pretty good for a marine/boat application if you were to stick with the Overkill Solar/RadioB 120A BMS. They are well suited to be taped to the end of the cells and wiring is easy. Would work really well if I had started a couple years ago with a 24V inverter.

However, with a 200A BMS, the configuration works better if you can get the BMS in the lid, but it takes a little more room, hence the spacers.

 

[dtfeld]

Found what I was looking for. If I were starting today, this would be on my list.

https://www.sunfunkits.com/product/20/sfk-battery-kit-v2-280-310-ah-12v-4s-lifepo4

 

[jmauld]

Found what I was looking for... alittle pricey, but all thing go together cleanly.

https://www.sunfunkits.com/product/20/sfk-battery-kit-v2-280-310-ah-12v-4s-lifepo4

Nice box. Wouldn’t you prefer to have your terminals enclosed as well? Also interesting that they mention your material choices as pla and abs. Are they 3d printing these boxes?

 

[dtfeld]

I think the spacer is 3D printed. The battery boxes are ABS, available from a number of sources. They are inexpensive except for shipping.

Im going to call/email see if they will sell a spacer and if it would fit.

 

[ttmott]

I think the spacer is 3D printed. The battery boxes are ABS, available from a number of sources. They are inexpensive except for shipping.

Im going to call/email see if they will sell a spacer and if it would fit.

Maybe make your boxes from Garolite sheet riveted together using aluminum angle. The top can be fastened with screws so it can be removed. Arrange the aluminum angle to serve as an external flange for sides, bottom and top; that way there are no protruding fasteners on the inside, just a clean box. A thin sheet of Garolite is very resistant to bending; super strong material and easy to work with. Garolite is also an electrical insulator and flame retardant.

I think 1/16" thick material would be just fine - https://www.mcmaster.com/plastics/multipurpose-flame-retardant-garolite-g-10-fr4-sheets-and-bars/

 

[ttmott]

Dave - on those BMS' - don't they have cooling fans? I wouldn't mount on the inside of the box as cooling may be an issue....

 

[dtfeld]

Dave - on those BMS' - don't they have cooling fans? I wouldn't mount on the inside of the box as cooling may be an issue....

No cooling fans. The heat sink is important only if you have balancing turned on as the "balancer" bleeds power from the high cells through a resistor creating heat. This function can be disabled, or only turned on/off depending on how you want to do it. Well matched and top balanced cells eliminate a lot of this. In my bench testing, these BMS don't even get warm to the touch.

Additionally, this BMS has several temperature probes with adjustable set points and this is really the biggest function of the BMS...if any charge/discharge or environmental temperature parameter, it disconnects the battery from the system.

About 99% of its function is a big safety switch.

 

[jmauld]

Any concern with taking it outside of its operating temperatures during a thermal event? I would make sure that you’re not going to limit it’s ability to function in the scenario where you need it the most.

 

[dtfeld]

Any concern with taking it outside of its operating temperatures during a thermal event? I would make sure that you’re not going to limit it’s ability to function in the scenario where you need it the most.

Not sure what your referring to WRT "taking it outside its operating temps during a thermal event"? This chemistry is pretty stable, so not much concern.

To prevent such things, the primary function of the BMS is to disconnect the battery from the system for a list of parameters, to keep it well with in all the current, voltages and temperature parameters specified by the cell Mfg. It is essentially a large safety switch.

What I'm building is essentially a modified version of what's commercially available, but designed with a high current carrying capability I need to operate with my inverter. The biggest difference, all components are sized for a 200A @12V current capability. Most of the commercially available batteries are 120A or occasionally 150A, primarily driven by the BMS's currently available.

If starting today, I'd tell you to put in a 24V Victron Multiplus and buy SOK marine grade 12V batteries --> https://www.us.sokbattery.com/product-page/marine-grade-12v-206ah-lifepo4-battery-sealed-plastic-box. 24V cuts the amps in half, and makes this a whole lot easier.

Unfortunately when I started with the inverter a lot of this stuff wasn't available or if available, was well beyond what I was willing to pay.

 

[jmauld]

Every time I see a semiconductor switch fail, it fails in one of two ways. Way too much current and it pops open. Too much heat or just a little too much current and it saturates closed, and is unable to open. I don't know what type of device that the BMS uses as a switching device, but I would plan on making sure they stay in their operating parameters. This is probably a fairly low risk anyway, because they should be designed with enough overhead to handle any overtemperature condition that they could protect against. If the battery ever reaches a certain temperature, it won't matter what the BMS does.

 

[GnrlPatton]

One issue I considered in deciding the config was weight. While true these cells are a lot lighter than LA, in reality a 4x280 or 4x304 cell config is about 50 lbs each. So in you config above, your battery is approaching 150 lbs. If its not strait into its final resting place, you need to figure out logistics. I opted for a dual 4x304Ah cell arrangement each with its own BMS. This allows for smaller batteries, and I can add additional batteries down the road as necessary.

If I were to choose an arrangement, I prefer the first option. If you purchased well made, closely balanced cells, your BMS should keep them well balanced ober the long term. Both offer pros and cons. Thats my "free advice". You know what that's worth.

Yeah, I am planning on building this out at home, testing it out and getting things configured, and then dismantling it to rebuild it on the boat. I've calculated the dimensions of my existing battery box that contains 4x 6v golf cart flooded batteries that are set up end to end. With the first option in the diagram, it will easily fit within the current footprint, even when I incorporate the various external items (BMS, switches, etc). The weight should also be less, but I'm less concerned about that (except for the moving them in part), as the battery box sits on the floor space between the 2 engines.

I am also leaning towards the first option, it seems easier to build that way, both with the straight bus bars and the compression options.

 

[dtfeld]

And the balancing current is limited to several mA, so it would take a month of Sundays for this thing to balance.

 

[jmauld]

DT, what's the purpose/benefit of the sealed box? Is it to protect the internals from the marine environment?

 

[dtfeld]

Every time I see a semiconductor switch fail, it fails in one of two ways. Way too much current and it pops open. Too much heat or just a little too much current and it saturates closed, and is unable to open. I don't know what type of device that the BMS uses as a switching device, but I would plan on making sure they stay in their operating parameters. This is probably a fairly low risk anyway, because they should be designed with enough overhead to handle any overtemperature condition that they could protect against. If the battery ever reaches a certain temperature, it won't matter what the BMS does.

Yes. BMS failure is a possibility. These are FET based. However, there are a lot of these devices out in the field, and I haven't heard of too many horror stories. I think the real risk is conducts/wiring devices from the batteries to the inverter/charger. They aint small. The inverter/charger also has its own safety cut offs.

The BMS really is a cell management device. Its looking at the individual cells and shutting off once the first cell gets to its preset limits. Prevent over discharge and over charge based on the weakest cell. Without this, the possibility of severely over charging one cell goes up, likely damaging the cell. In a really bad scenario, I guess the magic smoke might get let out. Same thing for temperature, too hot, or too cold and the prevents charge/discharge.

 

[dtfeld]

DT, what's the purpose/benefit of the sealed box? Is it to protect the internals from the marine environment?

Pretty much, but a couple of reasons.

1. Protect the internals from water/salt. Currently reside between the engines on the floor and can get washed down.
2. System designed to be expandable, so I'd like to just buy off the shelf.
3. Wanted to be able to place in existing marine battery hold downs.
4. individuals cells easier to manage due to lower weight.

There are a number of new boxes showing up, so I may have to try something a little taller (about 1" would do it), or machine a spacer. I'm liking this idea as it allows the BMS to be cradled and the wire runs short.

Or buck up (literally $$$) and buy the SFK box.

 

[dtfeld]

I contacted SFK about buying a spacer, but turns out the version of battery I bought wont fit as they are slightly taller than the previous generation of 280/304AH cells, and mine have welded in battery posts. They have a new kit (V3 as opposed to the V2) coming out that is intended to accommodate these taller batteries. I like what these guys are doing, and may revisit them in the future.

I've been looking at NOCO battery boxes for a while. A single 8D should allow both cell packs to be in one enclosure, and enclose all the wiring, BMS's shunts etc. Should also fit the space between the engines. The one box this doesn't check for me is keeping the battery in a "standard" size box. On the plus side, more room for for all the bits an pcs, and a little more airflow for the BMS. Additionally, I think I could add 1 more cell pack fairly easily.

Ordered one off Amazon. $99.

 

[dtfeld]

Got a response from SFK on their V3 including some photos. These look really nice especially if you want to stay in a standard battery footprint. I like this as I want to consolidate the inverter batteries over by the existing batteries. Downside is they are pretty spendy. Additionally coming out with a V4 that has a built in heater pad for those that live/cruise in colder climates.

However, for now, I’m going to try the NOCO 8D battery box and see what kind of performance and temperatures I get with the batteries down between the engines.