New garage time!

[fatjay]

Took the drone up to get a picture of the array as it stands. Once I figure out how to go around the chimney, I'll be putting another 21 panels into the array for 3 sets of 15 to 3 inverters.

 

[fatjay]

To get the most out of the lifepo4 cells, they need to be clamped at ~700-1600lb of force. At the same time, they need to be allowed up to 3% expansion. The cells are 570m end to end, so 3% is 17mm or 1.7cm. That's not a small amount.

I ordered some aluminum compression plates. I'll drill some holes and run all-thread between the two ends with a bolt on one end. On the other end, I'm going to get a fender washer and bolt, and put a valve spring under the fender washer. A valve spring is 250lb to compress and has a range of travel of about an inch.

So that's my plan for the moment. When I built this I wasn't taking into account the plates, or springs. So it's going to be a bit tight.

Fingers crossed.

 

[ning]

Given that the batteries may have varying temperatures, do they themselves expand at all?
Also temperature cycles may allow for fastener movement, though that may be mitigated by the all-thread also expanding and contracting.
Still, I'd consider using locktite, or some sort of lock nut (nylock or double-nut).

 

[Kyle_in_Tex]

You might research belleville washers to get your force with less space.
McMaster-Carr

You need about up to 3/4" of movement. With these, you can put some on each end to save space maybe?
you stack them in opposing pairs to get your force/distance needed like this,
()()()()

 

[fatjay]

Given that the batteries may have varying temperatures, do they themselves expand at all?
Also temperature cycles may allow for fastener movement, though that may be mitigated by the all-thread also expanding and contracting.
Still, I'd consider using locktite, or some sort of lock nut (nylock or double-nut).

The batteries don't really change temperature with use, only expand/contract as an effect of the gasses forming and getting re-absorbed.

You might research belleville washers to get your force with less space.
McMaster-Carr

You need about up to 3/4" of movement. With these, you can put some on each end to save space maybe?
you stack them in opposing pairs to get your force/distance needed like this,
()()()()

I can't see if it says where the specific tension is provided. And even once flattened, it still needs to provide a degree of expansion. How much force does it take the washer to be flattened?

 

[fatjay]

Well, its a start. At 18 inchpounds the springs are exhibiting 175lb of force. 6 springs will be 700lb. At full compression it's 250lb of force, or 1500lb total.

I just needed to get this far so I could mark my outer holes.

The springs are from the valve train of a 1997 corvette.

 

[Kyle_in_Tex]

The batteries don't really change temperature with use, only expand/contract as an effect of the gasses forming and getting re-absorbed.

I can't see if it says where the specific tension is provided. And even once flattened, it still needs to provide a degree of expansion. How much force does it take the washer to be flattened?

A stack of these is what is used in a cnc mill's spindle to provide strength to the pull collet which grips the tool's pull stud.

It will take some engineering to get the right number, diameter, thickness of each washer. They are spring steel and will work for a long long time.

Belleville Spring Washer Equation and Calculator

A Belleville spring or washer is formed into a conical or cone shaped geometry. The slight or aggressive conical shape gives the washer a spring characteristic and action. Belleville washers are typically used as springs where the spring action is used to apply a pre-load or flexible quality to...

www.engineersedge.com

 

[airbiscuit]

This is a big project. How much of all this did you know in advance, and how much are you learning as you go along?

 

[fatjay]

This is a big project. How much of all this did you know in advance, and how much are you learning as you go along?

I'm just learning as I go and hoping it doesn't all fall down

 

[fatjay]

Now with the springs the mounts are to wide. To turn the orientation, 17" deep to 27" deep, I needed more brackets, which means more floor. So I pumped that out tonight.

 

[murphy1244]

What are the springs supposed to do?

 

[fatjay]

What are the springs supposed to do?

Apply compression, while allowing room for expansion.

As the cells charge, gasses form inside, causing bloating. This causes the cells to expand in the middle. This is bad. With applying 700-1600lb of clamping force, the gasses form, but are uniform throughout the cell, not just in the middle.

Attaching a rod with bolts on both ends and wrenched down to 10ft/lb provides 1600lb clamping force. But when the cells expand, up to 3%, that clamping force increases with no where to go and can damage the cells.

Having springs allows me to clamp down and create pressure to have uniform cell expansion, while allowing for the expansion itself.

 

[fatjay]

New posts up, orientation switched. Problem solved.

Got a template made and banged out 2 plates. Ran the all-thread through, covered in 3/8" tigon tubing over it, mounted springs, and it's ready for cells.

 

[airbiscuit]

I wonder how many DIY solar installations are done incorrectly?

 

[fatjay]

I wonder how many DIY solar installations are done incorrectly?

This is what I know to date. This information may not be 100% accurate. I may learn new things in the future that contradict this information. But here goes.

Probably a good amount. But there's varying degree's of wrong. Most of which result in a decreased battery life. Maybe 5000 cycles instead of 8000 cycles? So 13 years instead of 21 years. And what is the calendar age for them? 15 years? 25 years? Does it even matter?

I've spent $4500 on batteries, which is considerably less than most people, totaling 60kwh. However, I want them to last as long as possible

Commercial installers/builders use banding. Which is probably ok for the most part, but people have found that since the swelling occurs in the middle, it creates stress on the outer sides. Under the blue plastic coating is an aluminum shell, which is linked to the + terminal. If that blue plastic coating gets damaged, you now have + connected to anything that comes in contact with it.

The other risk of swelling is that top link bus bar. It's pretty rigid, and clamped down pretty well. If the cells expand and are permitted to misform more than just a bit, that's a lot of stress on that top post, which is basically spot welded on. I've heard a lot of reports with these snapping off.

So the method of my compression is to attempt to minimize risk of fire/damage, while maximizing cell life. However, calendar aging there is really nothing I can do about that.

Here are a couple other solutions that people have come up with. Not saying they're all right, they were assembled in the past 3ish years. When lifepo4 started becoming commercially viable. And people have learned a lot in that time.

 

[fatjay]

Here's a video demonstrating the swelling of uncompressed cells.

 

[fatjay]

Got the first pack loaded and torqued to spec. I'm out of eyelets to create the top balance leads, so I'll head to the store tomorrow.

I also picked up 4 of the latest and greatest JBD 100a BMS. I particularly like these because they have ring termianls on B- and C-, as opposed to soldered leads.

Added bonus is RS485, UART, and CANBUS communications. Fingers crossed that one of them will be compatible with with the growatt inverters.

 

[airbiscuit]

As critical as it is, you'd think the battery manufacturer would sell a tensioning kit.

 

[fatjay]

As critical as it is, you'd think the battery manufacturer would sell a tensioning kit.

If the cells go **** up early, they sell more cells. And this stuff is strait from china. There's no american company supporting it.

There are a couple battery boxes out by 3rd party, but they're just hitting the market from china now, but they're questionable quality, and the restraints are sheet metal, just a bit flimsy.

Swapped out the springs for more rigid springs with higher travel rates.

 

[fatjay]

Step by step, top balance wires, made one by one.