Kazetsukai's Build

[BeNimble]

A significant investment for off-grid. Look forward to seeing how they do.
I would add that the straps are straight solid, usually best to have some flex in them.

[kazetsukai]

Originally Posted by BeNimble

A significant investment for off-grid. Look forward to seeing how they do.

Its hard to do something like a capacity test without the whole bank hooked up- any synthetic load will take forever to drain, and charging them from empty would take forever as well.

Hopefully I can provide some meaningful benchmarks when the inverter/solar are hooked up.

Originally Posted by BeNimble

I would add that the straps are straight solid, usually best to have some flex in them.

I was thinking of threaded rod / strut, to hold them down, with sheet metal on the ends and rubber mat directly around the cells to provide some level of flexibility. Maybe even just ratchet strap them down that way... definitely taking suggestions here!

[kidharris]

Originally Posted by kazetsukai

11 boxes arrived an hour ago...
Attachment 56682Attachment 56688

So now begins a saga on this LiFePo4 upgrade. I have lots of work ahead of me here:

• Measure all of the cells out-of-box and write their voltages down.
• Top balance all of the cells, which will take who knows how long
• Re-measure all the cells and write their top balanced voltages down
• Remove any cells that significiantly deviate from the rest for return/replacement
• Will have to skip the capacity test.
• Have to prepare the battery box for the new cells by insulating for temperature and dampening for shock, secure mount/etc.
• Much of the right hand side of the electrical cabinet will need to be redone.

Unboxing, box came with a little vanilla envelope containing hardware:
Attachment 56681Attachment 56680


Taking the top foam layer out, you are greeted by four cells.
Attachment 56679


They really spent some money on shipping materials. Nice thick foam all around, no chances taken.
Attachment 56678Attachment 56677


Each battery has a sticker with specifications and what seem like measurements at the time of shipping. It seems like there is a decent QC process going on at the very least:
Attachment 56684


Measuring the cell myself, I get 3.31V which is near the optimal storage voltage.
Attachment 56683


So now I'm taking measurements of the box to see what orientations I can fit them in...
Attachment 56685Attachment 56686Attachment 56687


Not much else today, these are heavy, impressive looking and I see no signs of wear or damage.


Look forward to top balancing and limited capacity testing!

Looking forward to a few more details and cost of the batteries in a future post. Thanks. I'm getting a lot of my solar education from you.

[kidharris]

Originally Posted by kazetsukai

Its hard to do something like a capacity test without the whole bank hooked up- any synthetic load will take forever to drain, and charging them from empty would take forever as well.

Hopefully I can provide some meaningful benchmarks when the inverter/solar are hooked up.



I was thinking of threaded rod / strut, to hold them down, with sheet metal on the ends and rubber mat directly around the cells to provide some level of flexibility. Maybe even just ratchet strap them down that way... definitely taking suggestions here!

Balloons/air bags/inner tubes for adjustable pressure or odd shaped clamps/hold downs/spacers? Too flexible or too crazy? inner tubes come in a lot of sizes and shouldn't be too hard to modify.


For instance a bicycle inner tube could be used like a bungee cord and then pressurized to tighten or provide a space filling property or a air/water seal. I can see how you would have really think about whatever hold down system to channel airflow to prevent/provide heating/cooling. Bonus could be light weight and not electrically or thermally conducting


Inner tubes can also provide too much pressure if you don't understand how that pressure is additive according to contact area. For instance 10psi applied over an area of 100 square inches is a 1000 pounds of pressure. Getting too eager could crush your batteries or wreck the structure holding them. I've seen crusher and jack designs that used inner tubes. Just a warning. Don't underestimate them.

[kazetsukai]

Originally Posted by kidharris

Looking forward to a few more details and cost of the batteries in a future post. Thanks. I'm getting a lot of my solar education from you.

I did a cost breakdown here for someone here:
https://www.skoolie.net/forums/f49/b...tml#post435229

They're 3.2V, 280Ah cells, roughly $80-$90 per cell pre-shipping. Post shipping very competitive with AGM, you can frequently beat AGM in capacity/buck.

Thanks for the inflatables idea, I think I'll go ratchet straps + rubber mat for now.

[kazetsukai]

So... the supplier ships one nickel plated copper bus bar along with two set screws and two nuts for every cell. For 4 cells, this means 4 bus bars, 8 set screws/nuts.


This is great for wiring series, you even have one extra bus bar (you only need three to wire a 12V 4S pack):


However for parallel you need more bars, one per terminal per cell. In my desired 3P configuration this means I have 4 bars, but need 6 just to do the parallel run- then ideally another 3 to make the series connection to the next parallel group.

The other problem is that in 3P the bus bars provided are not ideal. I'd have to overlap them creating a potentially weak electrical connection:


So yesterday I got the wild idea I could manufacture some of these things myself, using commonly available 6063-T5 aluminum alloy strips. This may end up being a fruitless endeavor electrically, I don't know. But so far its looking pretty good.

I took two bus bars, overlapped them on a short piece of strip and marked the holes to get an appropriate bar length for three cells:


This will sit flat on the cells unlike the two single cell bars:


And so I scaled this up- making all the cuts. I had enough metal for 28, I need 32 of these:


Post cut, angle grinder with a pad to take the edges off, still a few burs though. Need something finer, but for now this will do the job.


Post drilling/full prototype:


One hole is larger- I switched down in bit size. I want the holes larger than the set screws for some play but the first hole was a bit too big for my liking. Its my hope these will work well enough for the time being... if not I can ship one of the prototypes to a local CNC shop and they can make me nickel-plated copper equivalents.

Any thoughts (aside from "aluminum doesn't conduct as well")? These are 3/4" wide, 1/8" thickness, so thicker than the originals by 1.5x-2x. Maximum theoretical load would be 125A (6000W inverter maxing out, typical loads will be well under 2000W or 41A). Being 48V I think my use case is quite tame compared to some 12V applications.

[cdrobbins77]

Looks like you'll be at the gathering this weekend in indiana? Just wanted to say I'm looking forward to seeing your build in person. I'm very happy with the way my build came out but I love following along with yours.

[kazetsukai]

Originally Posted by cdrobbins77

Looks like you'll be at the gathering this weekend in indiana?

Yeah, with any luck . Busy day tomorrow for us, doing an oil change at first light then off to Indiana. I imagine we'd roll in sometime between 2PM and 4PM.

Originally Posted by cdrobbins77

Just wanted to say I'm looking forward to seeing your build in person. I'm very happy with the way my build came out but I love following along with yours.

Bus 54 was it? I think I saw you thread a long while back, very clean interior build-out. You and I went for some very different aesthetics! Likewise, I looking forward to seeing your build as well.

[cdrobbins77]

Originally Posted by kazetsukai

Yeah, with any luck . Busy day tomorrow for us, doing an oil change at first light then off to Indiana. I imagine we'd roll in sometime between 2PM and 4PM.



Bus 54 was it? I think I saw you thread a long while back, very clean interior build-out. You and I went for some very different aesthetics! Likewise, I looking forward to seeing your build as well.

Yes, that's the one. @senorbussenstien on Instagram.
We'll probably be arriving 7pm or a little after

[kazetsukai]

The underbay storage box I installed early in the build has served a variety of purposes up to this point. First for some lead acid batteries + charge controller, then random RV stuff, then the inverter. Each phase came out of necessity and now is no different, but I think its new job might end up being the final one: battery box for the new bank.


With the remaining 12 cells on their way in a week or so, I finally got some motivation to pick this project back up. I did some final measurements and concluded the interior cabinet used to house the Tesla packs just couldn't fit a 16S3P bank, no matter how "clever" I got with the arrangement. They had to live downstairs. Even if they could live in the current battery compartment, this arrangement would be safer/more manageable.

Design concerns with the modifications:

• Ensuring space for a good layout
• Securing the packs down
• Mitigating excessive vibration
• Climate control for heat
• Climate control for cold

Being very hot out here in KY, we put up a tarp canopy to shield us from the menacing sun. Created a nice cool corridor between the bus and the adjacent workshop, this work would have sucked without it.



Leaving the last one for later in the year, I looked at the current layout of the box. The forward section stored random RV supplies and the rear section housed the inverter. The batteries would be installed towards the front, secured by wood.


This leaves room for the blower fan in the rear.


The box was secured by four threaded rods that went through the interior of the box, from the floor joists of the cabin up top down to superstrut beneath. The two rods in the back of the box were in the way, so I had to relocate them to outside behind the box. I did this by cutting the existing rod out then running a small length of strut from the front strut support to just behind the box, supporting it fully as before using a new rod. Not sure why I didn't do this to begin with. This allows me to run two strings of 3P8S rows parallel to one another and join them easily at the front.


Then the inverter had to come out. It was screwed onto some wood supports bolted through the floor of the box, those had to go as well.


I likely wont be reinstalling it, but rather its replacement where the Tesla packs currently live.

*salute*


Can't wait for the new one to be installed- proper thermal management, smart functionalities, automatic transfer switch, shore charging!


Next up was covering up that large aluminum surface area to avoid potential mishaps or problems... I chose rubber mat for the task:


Facing the pattern down, this gives me a nice smooth surface area to work with that will do well with vibration, insulates the cells from he aluminum behind it and all of the holes/screws that threaten to scratch/puncture stuff.


Next I needed an anchor point to strap the cells down, small strut at each end of the rubber mat did the trick:



Above was all yesterday. Today we redid the bus bars to be 6-position instead of 3, to make both the series and parallel connections in one bar.



Each of the 16 3-cell packs will be taped together using heat vinyl press tape, moved into position and then connected using the bus bars. Hope to get all 14 cells I have on-hand in next weekend.

[kazetsukai]

Yesterday, FINALLY, the rest of my batteries arrived. 2.5 months, not too happy about that. First shipment took less than a month.

Well, we got to assembling that right away:


This is done by taping three packs together for each cell in the 16-cell series. I'm using both vinyl press tape and electrical splice tape- the latter being rubber, I think should help cushion the cells together. You can see my masking tape-covered torque wrench in that shot- taking every precaution assembling these- and avoid careless shorts at all costs.

Up close you can see my custom, 6-position aluminum bus bars, which make both the series and parallel connections. Going to torque these down to 60 inch-lbs on every stud.



For freezing protection, I'll be installing these tank heaters around and inbetween cells. They have a built-in thermostat, but I'll add my own so that I can put a little warning light in the cabin somewhere to tell me the batteries are being heated.


The 250A Smart DALY BMS is pretty beefy!


For the balance leads I am building some cover supports:


These have channels cut into them to organize the wiring:


This will keep the wiring neat and clean, and will allow me to cover the top of the batteries entirely. Once covered, I will ratchet the whole bank down. Day one of assembly ended like this:


Cannot wait to finish this and be on the new inverter + bank. Once assembled I will take some time to top-balance these by charging the pack to a modest voltage (54V?) cranking it .1V at a time until all of the cells reach somewhere around 3.55V. I'll introduce the bench power supply as needed on a per cell basis, this bank is too big to top balance with every cell in parallel.

[Bert06840]

Oooh wow, that looks awesome!

Did you have to pay any import duty?

[kazetsukai]

Originally Posted by Bert06840

Oooh wow, that looks awesome!

Did you have to pay any import duty?

It was probably baked into the cost.



These things have gone up since I bought mine- mine were $90-ish each, now they're around $120 each.

[kazetsukai]

Taking a short break, got the lower half of the banks' balance leads installed. The BMS leads are way too short, so I'm aggregating them on these rails:



The leads come out at planned places along the rail:



All of the cells can be accessed at the end where the BMS will be:



Installed:



Was debating installing fuses, doesn't seem to be industry practice and I wouldn't want the additional resistance to throw off the BMS' readings.

[kidharris]

how much will the battery weigh? Is it all gonna be on one side?

[Bert06840]

This is music to my proverbial ears. I ended up also not fusing my balance leads, because the batteries will instantly vaporize any wire thinner than a thumb. So you end up with, shall we call it… a distributed fuse?

Will you stay away from non-freezing temperatures?

[TJones]

That is a big beautiful battery!

Ted

[kazetsukai]

Originally Posted by kidharris

how much will the battery weigh? Is it all gonna be on one side?

I read the spec sheet for this, at 11.5lb per cell, that's 552lb for all cells. Lithium IRON Phosphate alright... Yeah, one side since its the underbelly storage.

Originally Posted by Bert06840

This is music to my proverbial ears. I ended up also not fusing my balance leads, because the batteries will instantly vaporize any wire thinner than a thumb. So you end up with, shall we call it… a distributed fuse?

Will you stay away from non-freezing temperatures?

I installed RV heat pads, and the box does have a blower for the hotter weather. I'm just going to have to see how much maintenance the box is heat wise...

Originally Posted by TJones

That is a big beautiful battery!

Well here's where we finished off yesterday:


Balance rails in, every nut torqued down to spec -they say under 9Nm, I was putting them down around 5Nm. Still, using a torque wrench I stripped I think two screws. What the frick...

Today, aside from installing the BMS/etc I have to rip out the right hand side (supply side, not the automation side) of my DC in the electrical cabinet. Its a mess that needs cleanup...


The positive leads you see floating midair are from the box/not hooked up to anything. You can see the electrodacus BMSes on the very right hand side... White cable in the middle is ethernet from outside.... I'm on the fence about keeping that capacitor... the black box is filled with ethernet.

Just needs tidying up in there. I also want to install a vent to put the hot air in there outside....

About to say goodbye to the Tesla packs, too- new inverter and half of the mess above is moving here. Maybe I'll build an E-lawnmower or E-UTV with them, idk.

[kazetsukai]

So yeah, yesterday we went ahead and swapped over our battery bank from the Teslas to the new EVE LiFePO4 bank. It went mostly without a hitch... until it came to getting the BMS situated. More on that later.

What went well
As a part of this I took the time to redo some of the electrical cabinet. Didn't go crazy this time, but I wanted to clean up. Before/after:


Overall pleased with how that turned out... there's more work to do but I didn't do it yesterday for some of the reasons I lay out later.



What didn't go so well
I was going to install my new inverter, but it was damaged in shipping:


Given the choice to have them send me a new terminal output block, or ordering a new one on Amazon, I chose the latter: Wednesday. I slid the inverter into position without hooking anything up, along with my 50A inline surge protector.


I also ordered a pair of RV roof vents, one to go into the electrical cabinet. Before the inverter living there it was getting awfully hot in the cabinet, and that heat was leaking into the rest of the cabin. I'm going to make it so air can circulate from outside, to in the cabinet, to back outside directly. The other vent will likely go in the kitchen.


What sucked
After getting the BMS (Daly Smart 250A) into place I found it was completely unresponsive. I had heard rumors about a charge current needed to connect, so I set the MPPT to 56V and eventually bluetooth became available.

Looking over the settings, it was all pretty standard issue. However I'd say that the cell voltage protection was a little on the extreme side- LVD was 2.2V per cell, OVD was something like 3.65V per cell. Balance started at 3.2V. Overall I think that's fine, the BMS shutdown is a last resort error condition, not normal operation. Normal operation, I want the MPPT to decide charge shutoff, for the Victron BatteryProtect to decide DC side shutoff, and for the inverter to decide its shutoff voltages.

Still, the BMS was not allowing charge or discharge... after getting really frustrated I realized there was a fault code for low temperature- reading -40C. Obviously to me this means the temp sensor wasn't hooked up, so I went looking for it:


Nope, nothing. No temp sensor. I reached out to DALY on Amazon, but I was SOL... so I decided I'd hook the battery up without a BMS and be very careful/conservative until I could find a replacement. And so the day ended...


The next day
Well they got back to me and asked me for info/screenshots/etc, so this morning I gave them to them. But I also got the idea of trying to hook up one of my Arduino thermistors to the NTC port:


Voila! It works!


But I can't talk to the BMS via bluetooth anymore.... I'll need to setup RS-485 to talk to this from now on using Daly's protocol document. At least I have some basic protection. The cells all read 3.29V at arrival, and they were paired up with two others, so I'm pretty sure a rapid drift scenario is off the table.


From here
We'll see how deep this seemingly bottomless pit of energy goes over the next week or so. Once the inverter is fixed (its just a bar that needs replacement up front) I'll be pulling the plug to see how long it takes before I need to plug back in. I'm starting a new job soon, and will likely be installing a second split in the upcoming month or two.

[kazetsukai]

By golly I have 11.5kWh into the bank and the voltage hasn't budged. LiFePO4's charge curve is so flat compared to LiCoO2...