4400w inverter / grid assisted solar charging

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
i used thick foil tape with acrylic adhesive to seal the gap i left between all the panels. i left about a 3/8" gap panel-panel, i only needed 1/32" of an inch to cover me from a CTE point of view. That said the glass does not seem to be sealed very well to the aluminum frame, and some drips come through. to stay grounded keep in mind The walls are 8' of screen so a robust wind will blow rain into the area a good ways, water always wins.

 
Here is a idea how the PV combiner box finished up. I have two, two-pole DC breakers that switch PV+ and PV- to completely isolate PV from the circuit. then i have two parallel MOV packages for each PV+. On the panel layout, I ended up going with a 8-series arrangement instead of 6-series like i had planned before, which put the OCV at 362VDC. which is really close to my 240vAC MOV clamping voltage of 390VDC, so i stacked two MOVs in series to get a clamp voltage of 780vDC, which is fine for shunting lightning surges to ground and well below the 2000v rating of the PV wire i used. i'll have another ~580vDC MOV at the inverter inputs too. I ended up useing a MOV intended for a 120vAC and 240vAC in series to get 585vdc clamping, measured with a fluke 1000v insulation tester. the ac and dc rating of MOVs are different because the AC rating accounts for peak to peak voltage plus about 14%, where DC the RMS and peak are the same. typing that all out might make it seem like I may be a little overboard on surge protection, and that might be true.

https://ibb.co/dtpS8FH



https://ibb.co/zXX929R




there is also a labeled PV DC disconnect at arm level and outside facing for first responders to use, I have the breakers in the roof-mounter combiner box acting more like a switch for service in that box. in the event of a really good surge the MOVs will explode and need to be replaced.

https://ibb.co/Pcg8QBk

 
The AC and non-battery DC wiring is completed, all 10awg stranded thhn building wire, no lamp cord. one idea that worked well was how i did the cable tray on the bottom. they don't sell real cable trays at my local hardware store so i made one out of 6x 3/4" knock out handy-boxes, this is needed to keep the fan inlets free for airflow and the little conduit pieces are so close to the inverter and Xfer switch its little-finger safe.

The battery is still a work in progress, 286x of the the LiFePo4 battery cells came in, still waiting on a few more, the 120a BMS, and cell level fuses too. i was thinking that i would encase the battery in 4x large ammo cans, that way 3 of the ammo cans are a useful 12v 4s size and the 4th one which is 3s can hold the BMS. the plan is 32p 15s overall, 208AH capacity and 120amp max discharge.
 
Well, for those waiting for their 'I thought it was too good to be true' moment this is it. the 3-in-1 solar converter did not work out. I connected the PV, and the box made some unsavory noises, and then turned into a short circuit. no idea what happened, other than error code 9 on the LCD, but the array had 311vdc and when I close the breaker it goes to 2.5VDC, just like a short. its not to ground, PV+ and PV- are shorted.

So I found a different way to do it, significantly more expensive though. instead of a 3-in-1 unit I'll try to get higher feature-content individual parts. most importantly, I have a epever 200 VDC, 100a MPPT 48v charger coming, which has a generator start and inverter load function. Since the solar input voltage is much less I'll need to break the solar array into 4S-4P panel arrangement instead of 8s2p, I'll just need to add a few more feet of PV wire and terminate some extra MC4s, no big deal.

the battery is taking shape, my 32700 cell endcaps are on thingaverse. https://www.thingiverse.com/thing:3691963
 
if i connected the 3-in-1 box wrong, then i still don't know what that wrong thing would have been. some MPPT controllers have specific order for connecting the battery first and then solar etc, and some are positive common. but as far as i can tell i did it right. once it opened it up i found the BUS+ and BUS- connectors are shorted out, so the FETs of the inverter this are probably bad. the seller claims they will send a new board, i'm not holding my breath.
 
Update on the battery for the solar Leaf charging project. I could not find ammo-cans that fit the bill to hold the cells. some where close, like the so-called FAT 50, but with 2 cubic feet of battery, it was a problem. so I ended up buying an aluminum trailer storage box from amazon, it was too close a fit and needed a 2" extension because the handles stick in not out, but its moving along at least. I can 3D print 2-3 end caps per day.

outdated info removed for clarity. see page 5 for finished battery case.

this is much larger project than I had first anticipated, if not for intestinal and financial fortitude I would have long since quit this idea. Afterall, it just does not cost that much to charge with grid power! these are the pros, they make power for a living! The whole idea of the cheap 3-in-1 solar gizmo got me started, but with the crap quality, I found out how much it really costs to do this, the payback is not going to be as few years as I first thought, but I'll still make back 2x my money in the life of the system.
 
There's also 40mm grenade cans. I use those for welding rods because they are a little deeper and wider than a 50 cal ammo can but I can fit 18 inch long welding rods in them.
Then there are 20mm ammo cans. They are much bigger.
My next "12v" LiFePO4 battery I build will probably go in a 40mm grenade can.

It pays to buy the good stuff.
The only alternative energy stuff you can kind of go cheap on is stuff that's not electric and doesn't have moving parts. Which narrows it way down to panel racking. But if you live some where with sea spray or is always wet, you probably have to buy the high end stuff.
 
not exactly the same thing, but along those lines check out that Jehugarcia built in his portable power pack. He has also built some home made power walls.

https://www.youtube.com/watch?v=bn0KtdrpoEM
 
I like that guys' videos, i had not seen that one yet, good stuff.

Doing what he did for a 48v 200ah lifepo4 battery would be very cost prohibitive, a much easier, faster option would have been to buy 8x 100ah 12 LIFePo4 batteries and gang them up, they already have a BMS and are rated for higher peak current draws. for example, the so called mighty max on amazon has 12v 100ah batteries, 8 of which would recreate what i have here with much less effort, at $30/ah/48v. the other option would have been the 30x of the 3.2v 100ah brick cells at $23/ah/48v. this setup i'm working on should be $9/ah/48v, but its so much more work. lead acid is cheaper up front, i learned, but has 1/3 the life, so it ends up costing 2x as much if the time period is long enough.

avoiding imported parts is a nice goal, my leaf was made in asia, so i'm not off to a good start, right. In general its very hard to find the exact part domestically made at all, let alone for a competitive price. often I give the imports a shot, and if they don't work go premium on that item. there are many layers of protection to get refunds on imported parts that fail, esspecally from good market places like amazon, and as a last resort, credit card charge backs.
 
He gets those 18650 for real cheap somehow, I haven't watched all his videos, but I know he gets a few dud and has to check them all and find the good ones.

I know when my 400a @ 48v bank of L16's goes I am switching to Li and be done with lead acid :)
 
New inverter and new MPPT installed, also there are only 12 of the 30 battery modules to go, the end is in sight! the epever MPPT is the 200v 100a version with auto-generator start. This USA-made inverter and highly-rated MPPT update cost 4x what the original 3-in-1 box cost, but based on the weight, its 3x times heavier too (22lb vs 73lb) in that dollar per pound perspective, its close I guess!

 
this is what i came up with for the battery wiring. i used automotive 10awg wire becuase the insulation is soft and easy to work with. this connects 8 cells together, and then i have 4 of these wires parallel. each 10awg conductor can conduct 30amps, so that's 120amps continuous current flow. which i figure is less than the 110amps DC 48v the 4400w inverter will draw at full power.

i put the little glass fuses on each cell so that if i shorted it out, it would not cause a fire. the cells were very hard to spot weld to i'm not really sure what black magic goes into getting strong spot welds on these because i have a good spot welder and even with 2,000amps the first weld was the only strong weld, after the first weld the second and third welds were weak and i can pull the nickel strap off with my bare hands.

the soldering is not easy either. the metal of the battery sucks the heat out of the iron, and the smooth finish of the cell metal prevents solder (88pb10sn2ag) from sticking, so hitting it will a stone or sanding drum got the solder to stick, but it took two irons at full power to flow the tinned solder around the fuse leads. but once its on, i cannot pull it off. on the negative side i used a small U shaped bare tinned copper 16awg wire to connect the cell to the 10awg wire.

then once i have these all done, i'll use 30amp rated terminal strips to connect each of the 4 + leads to the next modules 4 - leads. that way if i ever want to i can easily rewire it to 12v or 24v. also the BMS wiring will be really simple since all connections are pulled out on terminal strips. it would have been alot easier to buy 4x $900-1500 12v 200ah lifepo4 batteries and just used 3 jumpers to hook it up, this way i'm showing is not the easy way to do it.

 
Next time I mess with LiFePO4 cells I'm going with 50 or 100ah cells, assuming that the pack is going to be that big.
1st and 2nd battery was 18650 cells, my 3rd battery amp20 prizmatics.
I liked working with the big cells a lot more.
Not messing around with a bunch of little batteries any more.
 
It's cheap and doesn't contain the slave mined environmental disaster known as cobalt.
 
I agree that the ever popular 18650 is a non-starter. after testing 5-6 different kinds (no *fire ones either, branded ones) none of them could provide more than 1.5ah with the limited voltage ranged needed for long life. the really large format cells can be a value, I always am on the lookout for 30ah+ deals, but if you make a chart with cost on the y axis and (real) amps hours on the X axis, the minimum point was these 32700 cells. if someone can link to a 48v/200ah/2000+cycle battery for even close to the $1800 this method cost, that's good info to get in this thread, please share.
 
Back
Top