Cinergi's 28 kWh / 4 kW Solar / 10 kW inverter RV build

[cinergi]

Got the other rubber feet ... This looks good. I'll need to test that this still allows for 4mm of movement (the feet will be bolted down).
Despite the two square tubes supporting the cells, it still bows a little bit. I'll cut a piece of plywood and drop that in there and see what happens....

 

[cinergi]

Oh, I also ordered a SSR and the Gigavac MX14 contactor which will reduce my relay consumption from ~5Ah/day to 1.3Ah/day. I'll post more about that if I like it and it works out. The only downside of the SSR is that the minimum enable voltage is 48.

I also have the Riden 6018 power supply coming next week.

Still waiting for my 4 Lishen 272 cells.

 

[fafrd]

You've got it!

This particular spring can travel 0.5 inches. The rate is constant.

I have repeatedly measured a maximum of 2mm of movement in a 4s configuration. The first run of 100% to 0% SoC will result in larger movement but after that, I've not seen more than 2mm.

Note that they charge shipping if you only order $16 worth.

After spending more time than I want to admit reviewing springs on Lee Springs catalog, I think I’ve convinced myself to select a slightly different spring than the one you chose.

LHL 625C 08 is 20% more expensive than LHL 625D 03 ($5.08 vs $4.03) but 8 will cost $40.64 all-in versus $58.74 for the ‘cheaper’ spring because of the shipping and $20 handling fee on orders below $40...

It’s also quite a bit longer at 2.24” solid versus the 1.02” of your spring, so my choice won’t be attractive for anyone with limited length constraints.

But for those with the advantage of 625C 08 is that it has a rate of 122lbs/inch, less than 1/3rd the rate of 625D 03 at 380lbs/inch. And this means it will only give up ~1.4psi over the full 4mm working range.

Fully-compressed, it is at 12.2psi so if no headroom is needed, it can go from 12.2psi at 100% SOC to 10.8psi at 0% SOC.

Or if you want to leave some margin to prevent the possibility of any over expansion damaging a cell, you can back off 2mm from solid at 100% SOC which will result in 100% SOC being compressed at 11.5psi and 0% SOC still being compressed at 10.1psi.

You are seeing only ~0.5mm/cell of expansion from 0% SOC to 100% SOC on your fixture, but I’ve measured closer to 0.75mm/cell with no fixture and others have reported expansion of 1.0mm so I feel better about having a spring solution that can adjust to my particular pack once I start compressing it.

Because I have the extra 1-1/4” of space for that longer spring, I’ll have a total usable range of over 17mm before I hit the recommended ‘minimum’ of 6psi...

 

[cinergi]

I was also entertaining the longer springs but you're right, I chose this one due to my space constraints. Nice work!

 

[Hedges]

I was also entertaining the longer springs but you're right, I chose this one due to my space constraints. Nice work!

You've got box tubes at both ends.
By recessing springs inside the box and putting springs at both ends, you would double spring length and shorten the assembly.

 

[fafrd]

I was also entertaining the longer springs but you're right, I chose this one due to my space constraints. Nice work!

When I ran into that longer spring option, I figured as much.

For roughly the same cost, just seems like those longer springs provide a larger safety margin at both ends. Starting off, I can tighten the springs to ~9psi at ~50% SOC and I can see how much compression I get at 100% SOC, tighten so that the remaining margin is reduced by half (or tighten to solid and then back off 1-2mm for a safety margin), and then see how much the spring expands at 0% SOC.

Anywhere above 9psi or even 8psi @ 0% SOC, I think I’m good to go (and will just leave the remainder as a margin for unexpected expansion at 100% SOC).

These longer 625C 08 springs have a total working range (free length - solid length) of 1.36” or 35.54mm, compared to the 625D 03 spring you chose with a working range of 0.48” or 12.2mm.

 

[cinergi]

You've got box tubes at both ends.
By recessing springs inside the box and putting springs at both ends, you would double spring length and shorten the assembly.

My original plans didn't include the box tubes or springs.. I've already added a lot of space. I'm still OK. And I don't have the equipment (or, let's be honest, the willpower!) needed to recess those springs. Going from 8 packs of 4 to 4 packs of 8 saves a lot of room, too.

All that said, I'm quite happy with the spring choice. I stay between 8.5 and 12.5 PSI -- well within the "good" range of 6-17. Even with all the room in the world, I see no need for the longer ones (especially now that I've personally observed how much movement there actually is).

 

[fafrd]

My original plans didn't include the box tubes or springs.. I've already added a lot of space. I'm still OK. And I don't have the equipment (or, let's be honest, the willpower!) needed to recess those springs. Going from 8 packs of 4 to 4 packs of 8 saves a lot of room, too.

All that said, I'm quite happy with the spring choice. I stay between 8.5 and 12.5 PSI -- well within the "good" range of 6-17. Even with all the room in the world, I see no need for the longer ones (especially now that I've personally observed how much movement there actually is).

How convinced are you that the box tube was necessary?

I’m thinking about going with 1” HDPE plastic for my ends and not sure whether I’ll need to bother with reinforcement of using plastic that thick...

 

[cinergi]

How convinced are you that the box tube was necessary?

I’m thinking about going with 1” HDPE plastic for my ends and not sure whether I’ll need to bother with reinforcement of using plastic that thick...

All I know is that the plywood was bending, so it definitely wasn't applying force where I needed it most (in the center). Turns out that the box tubes make great handles, too

 

[cinergi]

OK, I've built the first of 4 packs of 8 cells with the final (I hope!) hardware configuration:





I was able to pick this up and maneuver it more easily than I thought. It's heavy for sure, but I will be able to carry it to my truck and put it in the truck bed and then move it from the truck to the RV. It's not flexing under the weight and is quite secure. I'm pretty happy with this setup. I think it will have no problems flexing 4mm at the feet (I still need to test that). Next I'll be getting some anderson connectors and finding out the best way to connect these batteries in series while still maintaining fully independent 8-cell packs which can be swapped out as a whole unit without disconnecting anything on the cell terminals.

I'm playing with wiring loom for the BMS leads (the loom there is way too big - just some stuff I had lying around for testing the idea). I'll probably notch the plywood for running the BMS leads (which will have a 6-pin molex connector on them - in the pink bag on the table) and perhaps the 2 sets of 1-gauge wires which make up the positive and negative connection points. The molex connectors have 6 positions so I'll connect every group of 8's negative to position 1 even though I'll only use it on cell group 1 and 3; it'll allow me to use any of the 4 packs in any position. 5 of the 6 pins will be used. Off to buy some Anderson connectors ...

These cells haven't been under compression for a while so they're relatively fat. The extra space under the bottom of the left cell will shrink once I cycle these to 0 and back to 100. I'll try to measure before and after because the initial compression is going to be a lot more than 4mm.

All the other parts (plywood, threaded rod, CALB bus bars, neoprene, angle iron, box tube, etc) are now in the garage ready for me to built all 4 packs.

RV is due March 30th ... it's getting closer! I can't wait!

Oh, BTW, Victron Connect now supports configuring and updating multiple units ... Happy Dance! I've had to use the old software over the CAN Bus to make changes or update firmware etc. I need to get this system back up and running so I can try that!

I also got a Touch Screen and I'll see how long of an HDMI cable I can connect it with. I hope it can be long so I can place it in the RV interior "command center" area.

I also have a Riden 6018 on the way ...

And I have 2 200 watt 24v panels on the way both for testing while I'm here and for ground deployment in case I need it in the wild.

And I got a torque wrench that should work for the cell terminals.

I'll also want to pick up the Champion dual fuel inverter generator and see if I can run the system off of that (in power assist mode).

All the while, I still need to get a gooseneck ball, iPhone mount for the truck, and who knows what else. Busy! Which is great, especially these days. I'm quite thankful that I have this project, that I can iterate on its design, and share it here.

 

[fafrd]

OK, I've built the first of 4 packs of 8 cells with the final (I hope!) hardware configuration:

View attachment 36474



I was able to pick this up and maneuver it more easily than I thought. It's heavy for sure, but I will be able to carry it to my truck and put it in the truck bed and then move it from the truck to the RV. It's not flexing under the weight and is quite secure. I'm pretty happy with this setup. I think it will have no problems flexing 4mm at the feet (I still need to test that). Next I'll be getting some anderson connectors and finding out the best way to connect these batteries in series while still maintaining fully independent 8-cell packs which can be swapped out as a whole unit without disconnecting anything on the cell terminals.

I'm playing with wiring loom for the BMS leads (the loom there is way too big - just some stuff I had lying around for testing the idea). I'll probably notch the plywood for running the BMS leads (which will have a 6-pin molex connector on them - in the pink bag on the table) and perhaps the 2 sets of 1-gauge wires which make up the positive and negative connection points. The molex connectors have 6 positions so I'll connect every group of 8's negative to position 1 even though I'll only use it on cell group 1 and 3; it'll allow me to use any of the 4 packs in any position. 5 of the 6 pins will be used. Off to buy some Anderson connectors ...

These cells haven't been under compression for a while so they're relatively fat. The extra space under the bottom of the left cell will shrink once I cycle these to 0 and back to 100. I'll try to measure before and after because the initial compression is going to be a lot more than 4mm.

All the other parts (plywood, threaded rod, CALB bus bars, neoprene, angle iron, box tube, etc) are now in the garage ready for me to built all 4 packs.

RV is due March 30th ... it's getting closer! I can't wait!

Oh, BTW, Victron Connect now supports configuring and updating multiple units ... Happy Dance! I've had to use the old software over the CAN Bus to make changes or update firmware etc. I need to get this system back up and running so I can try that!

I also got a Touch Screen and I'll see how long of an HDMI cable I can connect it with. I hope it can be long so I can place it in the RV interior "command center" area.

I also have a Riden 6018 on the way ...

And I have 2 200 watt 24v panels on the way both for testing while I'm here and for ground deployment in case I need it in the wild.

And I got a torque wrench that should work for the cell terminals.

I'll also want to pick up the Champion dual fuel inverter generator and see if I can run the system off of that (in power assist mode).

All the while, I still need to get a gooseneck ball, iPhone mount for the truck, and who knows what else. Busy! Which is great, especially these days. I'm quite thankful that I have this project, that I can iterate on its design, and share it here.

Looks great, but the only thing I’m not understanding is your reference to it flexing 4mm ‘at the feet’ - do you mean the base of the farthest cells to the left sliding 4mm on the wooden base? It looks to me like the rubber feet and the base will remain fixed in place and it is just the cells that will be moving (more to the left, less to the right). Is there something I am missing?

 

[cinergi]

Looks great, but the only thing I’m not understanding is your reference to it flexing 4mm ‘at the feet’ - do you mean the base of the farthest cells to the left sliding 4mm on the wooden base? It looks to me like the rubber feet and the base will remain fixed in place and it is just the cells that will be moving (more to the left, less to the right). Is there something I am missing?

The plywood sides, angle bracket, and feet are all part of the spring compression and will thus move. The wooden platform on which the cells sit are floating on top of those angle brackets.

 

[fafrd]

The plywood sides, angle bracket, and feet are all part of the spring compression and will thus move. The wooden platform on which the cells sit are floating on top of those angle brackets.

Oh, now I see. Did you consider mounting those feet to the bottom of the lower piece of plywood?

Also, have you ever looked into the maximum load those 5/16” threaded rods can take? I’m having a dickens of a time finding any specifications on the Home Depot 5/16” plated threaded rod, but I’ve seen a few references to 65,000 psi for that low quality of steel and the central cross-section seems to be ~0.12225”, so that should mean over 3000lbs per rod (so no problem for even a single rod to handle 1000lbs), but I’m having difficulty confirming...)

 

[cinergi]

Oh, now I see. Did you consider mounting those feet to the bottom of the lower piece of plywood?

Also, have you ever looked into the maximum load those 5/16” threaded rods can take? I’m having a dickens of a time finding any specifications on the Home Depot 5/16” plated threaded rod, but I’ve seen a few references to 65,000 psi for that low quality of steel and the central cross-section seems to be ~0.12225”, so that should mean over 3000lbs per rod (so no problem for even a single rod to handle 1000lbs), but I’m having difficulty confirming...)

Hrm, I don't think I saw that option as I iterated into this design. I'd have to think about whether the plywood would be capable of supporting the 100 pounds across ~2 feet without flexing. If I didn't have height restrictions, the 1" box tube would work as an I-beam of sorts for that. Definitely doable. Damn it, now you have me thinking about another design revision LOL. I do like this idea though ...

I haven't even tried to look for what 5/16" can handle. I can tell you that I had the prototype constructed with 1/4" and that was OK. Threaded rod is used a lot in unistrut/superstrut construction ... so I found this just now: https://www.homedepot.com/p/Superst...eaded-Electrical-Support-Rod-ZR1028/100133690

900 pounds for 1/4" ...

 

[fafrd]

Hrm, I don't think I saw that option as I iterated into this design. I'd have to think about whether the plywood would be capable of supporting the 100 pounds across ~2 feet without flexing. If I didn't have height restrictions, the 1" box tube would work as an I-beam of sorts for that. Definitely doable. Damn it, now you have me thinking about another design revision LOL. I do like this idea though ...

I haven't even tried to look for what 5/16" can handle. I can tell you that I had the prototype constructed with 1/4" and that was OK. Threaded rod is used a lot in unistrut/superstrut construction ... so I found this just now: https://www.homedepot.com/p/Superst...eaded-Electrical-Support-Rod-ZR1028/100133690

900 pounds for 1/4" ...

Yeah, that one is made by Superstrut, while the 5/8” zinc-plated rods I have are made by Everbilt. Probably similar/OK but it’s a PITA that Home Depot doesn’t have full specs published for all of their threaded rod...

 

[Hedges]

Yeah, that one is made by Superstrut, while the 5/8” zinc-plated rods I have are made by Everbilt. Probably similar/OK but it’s a PITA that Home Depot doesn’t have full specs published for all of their threaded rod...

McMaster-Carr and Graingers have a lot of technical detail in their catalogs.

This doesn't give recommended max load (as Hope Depot does) but it does give psi strength.

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

Home Depot probably doesn't know the specs of whatever lowest bidder of the day is delivering.
I bought 3/4" galvanized pipe. Cut it and tried to thread, my Rigid die wouldn't start. Bought different brand at Orchard and had no trouble.
For years, they only sold channel nut with a spring attached to conveniently hold it in place in the channel. Crimp to hold spring deformed threads, so bolt couldn't be tightened without galling and seizing. I had to re-tap every one before use.

Their customers want the cheapest of everything, and they deliver.

 

[fafrd]

McMaster-Carr and Graingers have a lot of technical detail in their catalogs.

This doesn't give recommended max load (as Hope Depot does) but it does give psi strength.

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

Home Depot probably doesn't know the specs of whatever lowest bidder of the day is delivering.
I bought 3/4" galvanized pipe. Cut it and tried to thread, my Rigid die wouldn't start. Bought different brand at Orchard and had no trouble.
For years, they only sold channel nut with a spring attached to conveniently hold it in place in the channel. Crimp to hold spring deformed threads, so bolt couldn't be tightened without galling and seizing. I had to re-tap every one before use.

Their customers want the cheapest of everything, and they deliver.

Yeah, with McMaster, you know what you’re getting. I’d seen that 50,000 psi spec and I believe that translates to about 2000lbs load on a single 5/16” rod, so I think 500lbs should no problem for Home Depot’s Everbilt rod. I guess I’ll just stress mine to that level and hope for the best...

 

[cinergi]

In a 4 rod configuration, each rod is only seeing max 200 pounds (more like 150).

 

[fafrd]

In a 4 rod configuration, each rod is only seeing max 200 pounds (more like 150).

Yeah, I’m considering going with a 2-rod configuration (so 330lbs @ 12.1psi) and leaving margin for 150% @ 100% SOC means each of 2 rods being able to handle ~500lbs...

 

[eranrund]

Very nice project!
There's one thing I am failing to understand - if the feet are attached to the side pieces, and the side pieces are expected to move, and the feet are anchored to the floor of the RV, how is the movement supposed to happen? What do you expect will flex?
Also, do you think the four feet are sufficient to anchor this much weight? I'm about to embark on a similar project (16 cell pack inside a van) and am debating how to use a spring side mechanism like you did but also really anchor the pack to the floor/wall so that it can't go flying in case of a bad accident.