4400w inverter / grid assisted solar charging

[ripple4]

Got the battery finished late in the day, with the BMS wiring done. Then, with that finally done, I got the EPEVER 100a solar MPPT/AC inverter power up. To cap it all off, I also plugged in the car and it was L2 charging! The harder the project, the sweeter the success. with the modded Nissan EVSE drawing 12.1amps@235v = 2850w the inverter was consuming 47.8v@63amps or about 3010w, 95% efficiency! Even 5:30pm et, it still had 2000w PV coming in, tapering off to 400w by 6:15pm.

The battery just was not going to fit in the aluminum box I had, so used an extra steelcase bookself and installed the battery in that. it just so happens to show off the battery while keeping things tidy.

https://ibb.co/R7NkNHP

the controls are installed on a wood panel mounted on an full motion TV mount. I opted for the ME-RTR instead of the ME-ARC control panel for some future plumbing if I ever install a second MS4448PAE inverter.

https://ibb.co/8gNBQL5

[ripple4]

update on the mppt/inverter voltage settings for the battery. i originally had end of discharge voltages that may have been too high, leaving unrealized capacity in the pack, so i took one module out and charge cycled it up to 3.5v and then down to 2.9 volts. i wanted to see where the "knees" of the discharge curve where for these particular cells.


https://ibb.co/KX5J10F

this is a 32 cell module so the advertised 6500mah cell capacity should have given a 208ah module capacity from 3.65v to 2.0v. but since i'm limiting the voltages to increase life, i knew i would not get all of that. and the seller might have overstated the capacity and/or sold me low quality seconds, who knows. I find that between 3.5v and 3.0v there is 155ah at what would be a 0.1C discharge rate. 155ah@48.3vdc is about 7.5kwh minus inverter and charger losses (3+12%) leaves about 6.4kwh which in the summer is about 27 miles range.

so my upper charge limit on the MPPT should be 52.5vdc and the inverter low voltage disconnect should be 45vdc.

[ripple4]

last night the pack was top balanced and today there was time for a DC->AC efficiency test. I see that the Nissan leaf is about 5.6% more efficient AC->DC converting with level 2 Vs level 1, which is assume is power related, not voltage related. in other words 240v6amp would have the same AC->DC efficiency as 120v12amp, is that right? But I'm finding with this grid-assisted setup, its more efficiency to invert at lower power, by about 4-5%, so looks like its a wash. its kind of cool that at L2-6amp and L2-11a the solar input =~= the battery output at 5pm in Ohio, but there is also the ability to quickly input with L2-16a, without a consequence. and I have grid-direct power too.

leaf AC->DC charging efficiency number:
https://ieeexplore.ieee.org/document/7046253

here is 240v 6A = 1400va


here is 240v 11A = 2650va
https://ibb.co/QNkwvRp


and finally 240v 16A = 3820va
https://ibb.co/7GMmHWV

[ripple4]

So I've shown this project has resulted in a working off-grid solar inverter system that can charge a 2012 leaf. However this idea was to be grid assisted, in other words use solar if it there, but grid or generator if its not. and if this happens automatically its the best of both worlds, selfish solar. Basically, the battery I made cannot fully charge the leaf energy wise. like I showed it has only about 7.5KWH in it, and the leaf uses ~10kwh/day from the utility on average year round, to drive 30 miles one way to work at 55mph. so it will exhaust the home-battery before the car-battery is full no matter what. but the inverter I have has settings to remedy this, and so here are the settings that make that work. the MS4448PAE and ME-RTR can be programed to connect the Edison or Generator power with the "VDC connect" settings, the trick is that the BMS will disconnect the battery if any cell gets below 2.1v, so balance is important (and i'm still understanding this in practice.) so I want to have the VDC connect voltage at 46v.

https://ibb.co/1MF5smm


Moving on from that to the system performance, I knew that these were used solar panels and for $0.30/watt I was expecting some significant power atrophy. They say silicon PV looses power at 1.5%/year. So I was pleasantly surprised to find the peak product yesterday was 4600W on panels rated new at 295w each/4720w array. Now, I’m not sure if that high number is a glitch of the measurement system or it really put that power out. Tellingly, the average production from 2-3pm was maybe 3900w.

https://ibb.co/82P9Vvy


Speaking of the measurement system the EPEVER tracer as shipped form amazon does connect to the PC with a USB wire and shows a snapshot of the current performance, with a trend line, but datalogging into excel requires a $20 box to buffer the data. Once I have that in, I can track performance with more than just pictures of the real-time chart.

The next steps for me will be continuing to understand the balance of the battery, to see if top or bottom balancing is better and if it can achieve balance by itself or if it needs to be managed externally. Also right now the only load I can connect is the Leaf, so I will run EMT and pull some wire over to my 10 circuit Xfer switch to start running household loads. This will allow me to run a window air conditioner and dehumidifier until I get the mini-split AC setup. Since this is a success and I have shown that, I don't see a need to continue documenting progress in this thread, except for issues that come up and questions. maybe I'll make a youtube video to reach more people.

https://ibb.co/Rh2HjV7

[ripple4]

I made some youtube videos of this project to share it more broadly.

https://www.youtube.com/channel/UC4vzgP ... 8KBOuMHF4w

on money side of things I've had the system running for about 40 days now and on my last electricity billing cycle my usage was about 277KWH less than it should have been, and i'm on track to create and use 500KWH this billing cycle. the real payback is the cash-purchased EV, the solar is really the cherry on top from a payback perspective. people i ask go back and forth if i should include the cost savings from the leaf in the solar project, some say its fair some say its two different things that need separate accounting. but i come down on the rubber meeting the road with monthly expenses, and i'm not buying a projected average $75 worth of power, nor buying $175 worth of fossil fuel, nor paying a $200/mo car note/payment, so in my case i say it all goes together.

on the functionality side, the magnum inverter is really pretty amazing. it starts and runs several hard loads, like my dehumidifier, while also powering 8-10 circuits in my house. to be open, the lights do dim a bit when a compressor comes on for a fraction of a second. i am looking at ways to smooth out the voltage sags. triplite has power conditioners that have an autotransformer that step up and down voltages, but i'm not sure how i would integrate that, since their product is a free standing consumer item. it may also be the case that voltage sags might be a price to pay for this type of install. in practice when charging the leaf i have been keeping it at 240v/6a because that allows all the other household circuits to run and still stay less than i'm comfortable with on the DC side draw, it will get up to 90 amps draw and the battery voltage sags a bit. it seems best to charge the leaf with my grid-powered handle when i need to quickly top up before going out again, and also once the battery goes down to 47VDC. if i let the lower KW usage home appliance run on battery i get 2KWH more out of it before it will transfer to grid. the data shows that now i transfer to battery from the grid at about 9am, run 8 circuits in the house including a 9KBTU window air conditioner from 11am-5pm, put about 6KWH in the leaf run home loads to about 4-5am.

my next steps are installing an efergy HUB energy monitor with a sub monitoring kit. i already have the EF2 kit and it reads total output and saves it to excell. it seems that, overall, if i multiply solar production by .875 it gets AC usage. that is like the average of the higher losses on the round trip battery in and out and the direct to load usage. also i bought a bunch of those A123 bus packs and am sorting though them to get my total battery capacity up 10-12KWH. the packs i have are only good for 12ah so i need 10 packs to get 5 more KWH. thanks for reading, Cheers.