Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Plan: Off grid solar with a Model S battery pack at the heart

This site may earn commission on affiliate links.
For intermittent use you can get ~20kWh of lead acid batteries for <$2k at Costco. It wouldn't be worth it if you're going to cycle these daily but they're cost effective if only used for power outages.

Keep in mind the accelerated discharge rate of lead acid at higher power draws due to high internal resistance. The rating on lead acid is usually something like 1/20C or some ridiculously low power draw to get that listed capacity.

The Tesla pack cells will discharge at 1C+ before internal resistance is a large enough factor to derate capacity. In my setup I'll be discharging at maximum rates of around 1/4 to 1/3C... an average of less 1/40C... meaning I will always have nearly 100% rated capacity available for use (minus small safety and degradation-prevention offsets on each end of the SoC scale, probably 4-5% on each end so maybe 8-10% capacity kept locked away for safe keeping)
 
Keep in mind the accelerated discharge rate of lead acid at higher power draws due to high internal resistance. The rating on lead acid is usually something like 1/20C or some ridiculously low power draw to get that listed capacity.

The Tesla pack cells will discharge at 1C+ before internal resistance is a large enough factor to derate capacity. In my setup I'll be discharging at maximum rates of around 1/4 to 1/3C... an average of less 1/40C... meaning I will always have nearly 100% rated capacity available for use (minus small safety and degradation-prevention offsets on each end of the SoC scale, probably 4-5% on each end so maybe 8-10% capacity kept locked away for safe keeping)

Agreed... Lithium Ion cells are FAR superior to lead acid... I actually expect lithium ion to completely displace lead acid in the next 10-15 years... BUT for at least the next 3-5 years for emergency use (<10 cycles/yr) lead acid is the more economic choice. A 10kWh bank is enough to get my house through the night until the sun is up again in the rare event of a grid failure.

Spending ~$1k for 10kWh of batteries you only use a couple times a year seems like a lot of money... until it's 110F outside and your power is out... and you have solar panels you can't use...
 
For intermittent use you can get ~20kWh of lead acid batteries for <$2k at Costco. It wouldn't be worth it if you're going to cycle these daily but they're cost effective if only used for power outages.

That may be but he also has the nice breaker boxes, neat cabling, solar panels, inverters, etc. To have all that in my house wouldn't be cheap. I don't even have a room to dedicate for that without building an addition. I guess if the equipment could be put in a 8x8x2 space I could give up a wall in the garage and move some storage items to another room or to the shed but I still expect my substandard wiring would make it a mess (house was built in 1972, didn't get a breaker box until 1996 and they used the cheapest cruddy brand of breakers). It's not utter nightmare wiring but there cabling has no slack, isn't oversized (might even be undersized), and so on.

I'm sure as I convert my ICE vehicles to EVs I'll end up spending money on electrical work but there is no way my existing infrastructure in this house will ever look as nice as his without gutting the wiring and starting over.
 
I guess if the equipment could be put in a 8x8x2 space I could give up a wall in the garage and move some storage items to another room or to the shed but I still expect my substandard wiring would make it a mess

... how about 4' x 18" x 3'?

IMG_0681.JPG


I've still got a little straightening up to do but this steel rack can easily accommodate ~20kWh of batteries. The whole system cost <$4k (Batteries, wiring, AC-coupled inverter.... and steel rack). By far the largest cost saver is not using charge controllers... my grid-tie AC inverter couples to the off-grid inverter/charger. It's not intended for daily use like the system wk057 has built and when lithium is cheaper I'll be the first in line. This is a back-up system for use during a grid failure.
 
... how about 4' x 18" x 3'?

View attachment 72223

I've still got a little straightening up to do but this steel rack can easily accommodate ~20kWh of batteries. The whole system cost <$4k (Batteries, wiring, AC-coupled inverter.... and steel rack). By far the largest cost saver is not using charge controllers... my grid-tie AC inverter couples to the off-grid inverter/charger. It's not intended for daily use like the system wk057 has built and when lithium is cheaper I'll be the first in line. This is a back-up system for use during a grid failure.

Using the inverter(s) as charger(s) in a UPS style setup is good for that type of setup, for sure.

Hmm... mismatched batteries... no self discharge issues? Way back when my father and I built a lead acid based mini off grid system (~1kW?) we used a hodgepodge of batteries (including car batteries, golf cart batteries, boat batteries, you name it) and ended up having self discharge issues as they would try to self balance. (I know this now, didn't understand then when I was ~10, lol) Eventually we replaced them all with brand new 6V deep cycle types. My father kept this system up and running using it for some lighting and some ceiling fans and small loads like a small TV. The batteries lasted about 12 years before they had degraded to be nearly useless (~5 minutes of power... lol). The replacement cost was pretty pointless since the original system never even broke even on material costs after 12 years, so dumping more into it would be a losing battle. He still has the wiring and all of his extra "Solar" 120V outlets we put in around the house that run back to his 1.5kW 24V inverter. If I end up with a spare Tesla pack module or two I may revive his sytem. His lead acid bank was ~12 batteries for about 13kWh of storage. Two Tesla pack modules are almost that (~11kWh) in ~1/8th the form factor.

For example, just four recycled Tesla pack modules could replace your 20kWh lead acid bank with a ~21kWh li-ion bank.

As for space... yeah, this project is utilizing basically an entire room. It's about 15'x10' with about 8'9" from the concrete floor to the floor joists above. Everything installed with NEC working clearances in mind. Honestly... could probably use a bit more space to make things easier, but, it's coming along. 170kWh of lead acid batteries would definitely NOT fit in this room with all of the other equipment, where a rack of Tesla modules will fit nicely in an enclosure in the corner when I'm done.

Anyway, as has been requested, here are some messy phase 2 work-in-progress pics:

Inverters and charge controllers all mounted. Yet unconnected wire pulls hanging everywhere...
2015-02-14 15.39.39-1920.jpg



Some junction boxes, one will be where the connections from the actual PV panels comes into the house eventually. Notice all of the labels. The one on the wall to the right will be where I tie in the AC input wiring from the grid panels to the inverters' AC input. Going to use 6/3 NM for the inverter AC input and output to their respective junction boxes. (33A max continuous output, so 6/3 NM is more than enough).
The batteries will be where the orange ladder is eventually. They're last, though. I need to move that gray PVC box on the wall there (some wiring that ties the old lawn sprinkler system to the newer one... likely can cram it into a small box on the block wall later).
2015-02-14 15.39.22-1920.jpg


Epic EMT bending work (before mounting the box above). :cool:
2015-02-14 00.01.02-conduitcrop.jpg


More progress today I hope. Hoping to get the rest of the non-battery DC wiring pulled at least, and LV comm wiring, then work on the AC ties to the inverters. Then finally get to the battery racking/enclosure.
 
Last edited:
Great project! Did you weigh one of the 16 modules (5.3kWh)? I searched this and your other threads but could not find it. Just wanted to compute kWh/lb for the Model S without the Al armor/bms/wiring etc. Thanks.
 
Great project! Did you weigh one of the 16 modules (5.3kWh)? I searched this and your other threads but could not find it. Just wanted to compute kWh/lb for the Model S without the Al armor/bms/wiring etc. Thanks.

I actually did finally get around to weighing a module when working on the design for my storage rack.

Weighing myself, then weighing myself holding a module yielded a net weight of 56 lbs (25.4 kg) for one module (drained of coolant, but that should be negligible), lower than I had originally estimated (~75 lbs). That comes out to about 10.55 lbs per kWh... which is awesome. That will give me a total of 1,792 lbs (nearly one ton) of modules for my 170 kWh set.

So, the modules themselves in an 85 kWh Model S are about 900 lbs. The other ~300 lbs is everything else for the pack.

Edit: For comparison, a common 1.35 kWh lead acid battery use in solar setups (never mind that this capacity is only available at ~1/20C...) weighs more at 67 lbs, or just under 50 lbs per kWh. So 170 kWh of lead acid would be about 4.25 tons (about 8500 lbs). Yikes. This lead acid battery is about 10.375" x 7.125" x 10.875", so about 804 cubic inches. The Tesla modules measure at about 26"x11"x3.5" for about 1001 cubic inches. So that's 188.4 in^2 per kWh for Tesla and 595.6 in^2 per kWh for lead acid... 3.16x the volume per kWh.

Summary:

Tesla (LiCoO2):
56 lbs - 5.31kWh - 1001 in^2 - 10.55 lbs per kWh - 188.4 in^2 per kWh

Trojan T105-RE (Lead Acid):
67 lbs - 1.35 kWh - 804 in^2 - 49.63 lbs per kWh - 595.6 in^2 per kWh


Tesla is ~69% less volume per kWh and ~79% less weight per kWh.
Or put differently, Tesla is 4.7x more kWh per lb and 3.16x more kWh per in^2. :)
 
Last edited:
Using the inverter(s) as charger(s) in a UPS style setup is good for that type of setup, for sure.

Hmm... mismatched batteries... no self discharge issues?

Two separate banks... the Trojans are from my previous back-up system and are ~4 years old... I found the Golf Cart Batteries at Costco for $84/ea and thought I'd give them a try... I'll probably keep them separated when not in use to avoid the self-discharge problems... The older Trojan bank is running ~1v lower than the other bank open circuit...

Summary:

Tesla (LiCoO2):
56 lbs - 5.31kWh - 1001 in^2 - 10.55 lbs per kWh - 188.4 in^2 per kWh

Trojan T105-RE (Lead Acid):
67 lbs - 1.35 kWh - 804 in^2 - 49.63 lbs per kWh - 595.6 in^2 per kWh

Agreed... wish I had ~10kWh of LiCoO2 instead of ~10kWh of PbSO4... but the 10kWh of PbSO4 cost <$800 including core fees... not sure where to even start looking for 10kWh of affordable LiCoO2 :crying:

Sounds like I might be able to get some from Tesla in a few months from Musks comments on the Q4 conference call :biggrin:
 
LiFePO4 for about $2800 for 10 kWh EVTV Motor Verks Store: CALB SE100AHA 100Ah LiFePo4 Cell, Lithium Battery Cells, SE100AHA Much more durable than LiCoO2, safer too, and easy to assemble. Used Volt and LEAF packs might be another option for cheaper lithium variants.

This is the second time in this thread I believe that someone has referenced this site's LiFePo4 cells.

"LIMITED QUANTITY LIQUIDATION SALE"

I wouldn't really consider that a benchmark for the price of a storage pack based on LiFePo4 cells, personally.
 
One could use the same argument for the price of a salvaged Tesla pack based system as well. If used salvaged packs are an option why not "new" leftover cells? Get deals where you can. Sometimes used packs of LiFePO4 turn up for sale when people upgrade their packs.
 
That may be but he also has the nice breaker boxes, neat cabling, solar panels, inverters, etc. To have all that in my house wouldn't be cheap. I don't even have a room to dedicate for that without building an addition. I guess if the equipment could be put in a 8x8x2 space I could give up a wall in the garage and move some storage items to another room or to the shed but I still expect my substandard wiring would make it a mess (house was built in 1972, didn't get a breaker box until 1996 and they used the cheapest cruddy brand of breakers). It's not utter nightmare wiring but there cabling has no slack, isn't oversized (might even be undersized), and so on.

I'm sure as I convert my ICE vehicles to EVs I'll end up spending money on electrical work but there is no way my existing infrastructure in this house will ever look as nice as his without gutting the wiring and starting over.

Your mention of space reminds me of JB Straubel's comment at the storage symposium that footprint matters. I think people often think that density doesn't matter much for static storage, but it does, for practical reasons and cost reasons.
 
This is the second time in this thread I believe that someone has referenced this site's LiFePo4 cells.

"LIMITED QUANTITY LIQUIDATION SALE"

I wouldn't really consider that a benchmark for the price of a storage pack based on LiFePo4 cells, personally.

It isn't that far off. 3 years ago I bought (8) 100 ahr Winston cells at retail for $880. I have cycled them over 500 times and they still give 7% more capacity than their new rating. I expect 3000 cycles before dropping to 80% of rated capacity.

DSC02937.jpg

DSC02972_zps579f42ff.jpg
 
So that's 188.4 in^2 per kWh for Tesla and 595.6 in^2 per kWh for lead acid... 3.16x the volume per kWh.

Summary:

Tesla (LiCoO2):
56 lbs - 5.31kWh - 1001 in^2 - 10.55 lbs per kWh - 188.4 in^2 per kWh

Trojan T105-RE (Lead Acid):
67 lbs - 1.35 kWh - 804 in^2 - 49.63 lbs per kWh - 595.6 in^2 per kWh


Tesla is ~69% less volume per kWh and ~79% less weight per kWh.
Or put differently, Tesla is 4.7x more kWh per lb and 3.16x more kWh per in^2. :)

You are talking volume, so the units should be in^3, no?

Excellent progress wk057... Epic EMT bends indeed. ;)
 
....we need an off-grid inverter that can handle ~400vdc...

Yep. This was the reason I ended up opting to break the pack down in the first place. The only commercially available inverters I could find that would accept 400+VDC were grid tied inverters made for solar input, which would not work out very well. In my testing, even after defeating the anti-islanding provisions and getting a mini-grid started, among other things, the grid tie inverters have nothing that will regulate their DC usage. They try to pull as much power as is available from the DC source and output it on to the existing AC on the other side. So, in my testing, if I had loads >= the output of the grid-tie inverter on my mini-grid, then it was happy. As soon as the load dipped below the combined output of the small off-grid inverter (generating the mini-grid) and the grid tie inverter output, the grid tie inverter would push it's AC output voltage above spec and cut off entirely. Granted, this was a small scale proof-of-concept, using mostly second hand components, but it proved to me that grid tie inverters were not the answer.