Deciding to DIY my solar battery was one of my worst mistakes ever

[kolek]

This is an anti-DIY message specific to DIY batteries, so I apologize, but my message for the masses is, unless you know what you're getting into, consider buying a pre-assembled battery.

Deciding to DIY my solar battery was one of my worst mistakes ever.
I'm still not done yet, and so far the project has wasted about 12 million hours of my time.

Originally I was going to buy fully wired batteries in a box from Amy, but was told the boxes were too big and couldn't be shipped to my country. So stupidly, I figured I could do all the boxes and wiring myself.

Even for pro-level handymen, to do this right, with all the research required, the fixtures and compression and flexible busbars and sizing and building the box and top balancing and the BMS and the fuses and wiring... etc. etc. this is a really big project. For the record, I happen to be a pretty crappy handyman. So maybe for some of you this is a walk-in-the-park.

But my message for the masses is, unless you really really know what you're getting into, buy a pre-assembled battery. For the vast majority of people who are working for a decent wage, when you figure in the cost of your time, a DIY battery doesn't save you anything. I guess if building one of these is your idea of a good time, then go for it. But in my case, I'd rather spend the time doing other things.

/rant

Edited to fix typos

 

[Magrathea]

I think it's more of a hobby for people rather than a means of saving money, I've never built a battery but used to build computers, I wasn't really doing it because I wanted to save the 10% or so, I did it because I enjoyed it. I'm building my first power system right now, and while I know it's not the same thing as building a battery, I'd probably break even after figuring cost of tools and cabling compared to something like an ecoflow system. But that's just boring consumerism, this is fun for me, and I suspect many people here.

If it's not your thing, well hey! You just learned something about yourself, that's neat.

 

[kolek]

I've never built a battery but used to build computers

I've built lots of computers and servers, they take me about 1 hour from start to finish. Building a solar battery is not in the same universe.

>If it's not your thing, well hey! You just learned something about yourself, that's neat.
Well, I kind of knew it was not going to be my thing. I should have figured out a way to ship the finished batteries.

Thanks for your comment!

 

[Magrathea]

No I get that, building a computer is easy, I didn't mean to compare the complexities, but the broad strokes of it all. But with the amount of time you've spent, I bet you've learned some stuff along the way, so maybe not a complete waste of time

 

[kolek]

No I get that, building a computer is easy, I didn't mean to compare the complexities, but the broad strokes of it all. But with the amount of time you've spent, I bet you've learned some stuff along the way, so maybe not a complete waste of time

Yes, that's true, I have learned a lot. And if I ever have to fix it, I'll be in a much better position than if I had bought finished batteries. Still not worth it though IMHO.

 

[kolek]

I should mention I've also had a massive amount of assistance from Todd @ncsolarelectric and many of the other geniuses on this forum, otherwise I quite literally would not have been able to even attempt the project. So a huge "thank you" to this forum for making the project at least somewhat plausible.

 

[cdsolar]

I get it. For some it is easy, for others, not. Kits are an option as well. That way, you have all you need.

Here is one site:

Sun Fun Kits DIY LifePo4 Lithium Battery Kits & Ready Made AGM / Lead Acid Replacement Batteries for RV, Golf Carts, Solar Storage, Back Up Power, Hurricane Outage . Made in Baton Rouge, LA, USA Ready to Ship out no overseas delay! Sunfunkits.com

Sun Fun Kits DIY LifePo4 Lithium Battery Kits & Ready Made AGM / Lead Acid Replacement Batteries for RV, Golf Carts, Solar Storage, Back Up Power, Hurricane Outage . Made in Baton Rouge, LA, USA Ready to Ship out no overseas delay! Sunfunkits.com

www.sunfunkits.com

 

[Brucey]

I've built lots of computers and servers, they take me about 1 hour from start to finish. Building a solar battery is not in the same universe.

>If it's not your thing, well hey! You just learned something about yourself, that's neat.
Well, I kind of knew it was not going to be my thing. I should have figured out a way to ship the finished batteries.

Thanks for your comment!

Building pcs has become easier over time. When I first started assembling pcs back in the 486/Pentium days there was always the dark arts of autoexec tweaking and drivers to deal with. Nowadays things are much simpler, pop the parts together install os, drivers updated online and off you go.

I think with batteries that is also happening. I haven't built any yet (hoarding two jk BMS), but seems like with the China enclosures and BMS sense wire circuit boards available things are heading in that direction, I e. You no longer need to crimp 18 sense leads and risk getting one out of order and smoking BMS etc, just connect BMS harness into BMS.

 

[42OhmsPA]

This is an anti-DIY message specific to DIY batteries, so I apologize, but my message for the masses is, unless you know what you're getting into, consider buying a pre-assembled battery.

Deciding to DIY my solar battery was one of my worst mistakes ever.
I'm still not done yet, and so far the project has wasted about 12 million hours of my time.

Originally I was going to buy fully wired batteries in a box from Amy, but was told the boxes were too big and couldn't be shipped to my country. So stupidly, I figured I could do all the boxes and wiring myself.

Even for pro-level handymen, to do this right, with all the research required, the fixtures and compression and flexible busbars and sizing and building the box and top balancing and the BMS and the fuses and wiring... etc. etc. this is a really big project. For the record, I happen to be a pretty crappy handyman. So maybe for some of you this is a walk-in-the-park.

But my message for the masses is, unless you really really know what you're getting into, buy a pre-assembled battery. For the vast majority of people who are working for a decent wage, when you figure in the cost of your time, a DIY battery doesn't save you anything. I guess if building one of these is your idea of a good time, then go for it. But in my case, I'd rather spend the time doing other things.

/rant

Edited to fix typos

What specifically did you find so difficult and time consuming?
What were your biggest DIY projects before trying to build a battery?

 

[timselectric]

It doesn't have to be that complicated.
But you definitely need to be comfortable with diy projects and electricity.

 

[SeaGal]

Building pcs has become easier over time. When I first started assembling pcs back in the 486/Pentium days there was always the dark arts of autoexec tweaking and drivers to deal with. Nowadays things are much simpler, pop the parts together install os, drivers updated online and off you go.

LOL... lucky you... I had to use a soldering iron to build my first computer. Not just for connecting bits together, but soldering all the components to the PCB... Yes, I am that old

 

[ChrisG]

I’ve built batteries for both my primary residence and cabin. Enjoyed it and liked the digging into how all this works together. For long term I think buying a battery would be easier, specifically if something happened to me. Family understands a good amount and I have documented everything but they don’t have the passion I do for this topic.

 

[1201]

ive done a lot of diy. and building a diy battery may be the easiest thing ive ever done.

the 16 cells already came with busbars. I had to crimp on ends to the jk bms leads. I encased it in some plywood and it was done.

literally one of the easiest projects ever.

 

[Brucey]

LOL... lucky you... I had to use a soldering iron to build my first computer. Not just for connecting bits together, but soldering all the components to the PCB... Yes, I am that old

Those skills no doubt gave you an advantage when building your batteries.

 

[Mattb4]

Like a lot of projects it depends on what you are happy with when you get done. For one thing the arguments around needing compression as a big deal or not convinced me to skip that when building my own. Yes this possibly could effect longevity but it is something I simply accept. I found in my own 24vDC battery build one of the more difficult aspects was procuring all the needed components to achieve the build. Considering I am a cheap bastard I did not wish to pay for high priced boxes so had to come up with a solution from common components readily available (used a plastic tool box).

Doing all the assembly is pretty much recipe following and being meticulous about how things go together. Fiddly detail stuff.

However I would agree that simply buying an all ready made battery would make more sense for most folks.

 

[kolek]

What specifically did you find so difficult and time consuming?
What were your biggest DIY projects before trying to build a battery?

Mainly research is what's time consuming. Currently I'm trying to figure out the correct torque figure for the battery fixture I created. One person on this forum says:

"300 kgf looks to me to be 660 lbs of clamp load. 25 in/lbs of torque on 5/16-18 threads"

another guy says:

"The spec from EVE was 300 KG force which rounds off to 660lbs. Battery face is approx 6.85"x 7.874" = 53.94 sq inches
660lbs/53.94sqin=12.23 lbs per sq inch
Divide 660 by 4 bolts that's 165 lbs Axial (clamping) force per bolt.
Using 4 course 1/4 in threaded rods that should equate to roughly 8 INCH pounds torque per bolt. Realistically, that's a snug twist of the wrist on a regular nut driver for the average build mechanic."

So 1 person says 25 in/lb, the other says 8 in/lb, I don't know which is right. If anyone in this thread happens to have authoritative information, that would be great. My fixture is using 6 threaded rods.

 

[shvm]

One person on this forum says:

"300 kgf looks to me to be 660 lbs of clamp load. 25 in/lbs of torque on 5/16-18 threads"

another guy says:

I will be the person who says you do not need compression for LFP, For DIY Solar applications.
Using a good enough insulator (thin 1-2 mm EPE Foam/acrylic/fibreglass sheet) should be enough.
I would instead focus my efforts on ensuring that your terminals and busbars are clean, if not lightly sand and clean them properly.

 

[kolek]

Building pcs has become easier over time. When I first started assembling pcs back in the 486/Pentium days there was always the dark arts of autoexec tweaking and drivers to deal with. Nowadays things are much simpler, pop the parts together install os, drivers updated online and off you go.

I think with batteries that is also happening. I haven't built any yet (hoarding two jk BMS), but seems like with the China enclosures and BMS sense wire circuit boards available things are heading in that direction, I e. You no longer need to crimp 18 sense leads and risk getting one out of order and smoking BMS etc, just connect BMS harness into BMS.

@Brucey Interesting info... can you point me in the direction of a "BMS sense wire circuit board" for the JK-BMS?

 

[hwy17]

My Orion build left me feeling a bit like @kolek in the middle of the process. It dragged on for months.

In the end I am satisfied with it and feel like I have something more than I could have got out of a prebuilt. But I am not finding the energy to want to do it again right away.

 

[offgrid-curious]

Mainly research is what's time consuming. Currently I'm trying to figure out the correct torque figure for the battery fixture I created.

I can agree with this. I went down the rabbit hole of compression research before deciding like Mattb4 that I'm just not going to bother with it for my DIY battery project. For my use case I don't think it is worth the trouble. I'm not charging and discharging my cells at high C rates. Maybe my cells will only last 70% of the cycles they're rated for .. that's fine with me. By that time I'm sure we'll have cells that are better and cheaper than what we're using now anyway.

 

[kolek]

A DIY battery is a lot more doable if you don't stress over the details. Throw on the regular busbars that come with the battery. Grab any BMS. Don't bother to top balance. Don't use a torque wrench. Don't bother with compression. Don't build a case. Don't worry about giant sparks when you connect the battery to the inverter. Don't bother calculating the wire thickness. Maybe almost electrocute yourself a couple times. Even then the project would still suck. It would just suck a lot less.

Unfortunately, I tend to be way too much of a perfectionist, to the point of just being stupid. The net result is I get practically paralyzed trying to figure out simple things like, "what is the exact best material I should use between the cells" and I spend 3 hours researching it. When you go at it from that point of view, it's like a neverending project.

Edit: Note the above is not "advice." It's just what some people do who don't stress the details. Personally, I can't operate like this.

 

[Tulex]

I think if you don't mind spending the money, and you can get what you want in a premade battery, and you aren't doing it for the fun, then certainly buy.

But I'm not made out of money, I didn't have room for standard batteries, and I do like projects. On top of that, I know the quality of the build.

Beyond that, much of the discussion about what has to go into a battery is overdone, at least in my opinion. You don't really need compression, you don't need flexible busbars, they don't even need to be in a container. There are so many examples of what people have done on here that it's just not that hard. Follow their lead, use some common sense, it just works.

All that being said, I know a guy who installed a new door knob assembly. When I stopped in, he asked me why the door wouldn't close. He had installed the latch assembly backward, yet he couldn't see that. So, DIY isn't for everyone.

 

[1201]

A DIY battery is a lot more doable if you don't stress over the details.

Throw on the regular busbars that come with the battery.

. They are solid copper

Grab any BMS.

Does that make it easier to assemble?

Don't bother to top balance.

if you diy and use your savings on more battery, at a low c rate, they will top balance themselves.

Don't use a torque wrench.

Are there people out there who are really using a torque wrench to build their batteries?

Don't bother with compression.

some prebuilt batteries are not compressed

Don't build a case.

. Ive never been able to figure out the sense In buying a $500 case for a $2000 battery. That $500 can buy a lot more battery

Don't worry about giant sparks when you connect the battery to the inverter

Terrible advice

. Don't bother calculating the wire thickness.

More bad advice

Maybe almost electrocute yourself a couple times. Even then the project would still suck. It would just suck a lot less.

 

[SilverbackMP]

I designed and built all of my boxes…bit time consuming but fairly easy with a track saw.

My electrician and I wire the first one together, and I was able to use that as a model to wire the rest. Without him helping, I might have been fked.

On compression, I say it need some, but nothing crazy and I damned sure am not going down the rabbit hole that I’ve seen others go down. Mine consists of three row of strapping tape on a group of eight cells and 1/4 rubber floor material lining a solid box with flexible cutting board material used as shims to get a tight fit.

 

[42OhmsPA]

Mainly research is what's time consuming. Currently I'm trying to figure out the correct torque figure for the battery fixture I created. One person on this forum says:

"300 kgf looks to me to be 660 lbs of clamp load. 25 in/lbs of torque on 5/16-18 threads"

another guy says:

"The spec from EVE was 300 KG force which rounds off to 660lbs. Battery face is approx 6.85"x 7.874" = 53.94 sq inches
660lbs/53.94sqin=12.23 lbs per sq inch
Divide 660 by 4 bolts that's 165 lbs Axial (clamping) force per bolt.
Using 4 course 1/4 in threaded rods that should equate to roughly 8 INCH pounds torque per bolt. Realistically, that's a snug twist of the wrist on a regular nut driver for the average build mechanic."

So 1 person says 25 in/lb, the other says 8 in/lb, I don't know which is right. If anyone in this thread happens to have authoritative information, that would be great. My fixture is using 6 threaded rods.

I can understand and relate to where you're coming from with some of your points. I definitely spent the most time in the research and planning stages for my first build and over-analyzed many things thanks to some excellent discussion on this forum.
Spent hours building my own bus bars that would handle a 2C discharge, back and forth on compression, terminal torque etc.
The second expansion was so much easier using common sense along with what I learned the first time.
I'm currently working out a deal for 48 cells for the third expansion.
I've saved thousands of dollars building my own, yes my time is valuable; however the knowledge gained is invaluable.

My takeaways after putting a few MWh through my packs.
-Compression isn't necessary if cells are used in low C rate applications. Lashing straps to hold them work great for me.

-Torque the cell terminals by hand and use common sense (only do this if you've been wrenching for years and have a feel for it).
I suppose using a torque wrench on the first build gave me a better feel for it as well.
-Insulators between cells are absolutely needed, I like flexible cutting boards.
-JBD BMS's are great and would be awesome with active balancing.
-Sparks should be non existent unless your welding, cutting, grinding... Use a resistor to precharge.
-At least double whatever capacity you think you need.