[subutai]

I'm planning on building a solar system that outputs ~2100 W peak to charge my car completely off grid. I drive a stock Tesla Model 3 LR and I don't plan on modifying the car, so I will likely use the default Universal Mobile Connector to charge off AC (EU schuko connector, 230V).

I'm a total beginner to DIY but I've been reading a bit, and as far as I understand, the solar would have to charge an intermediary battery first, and then that battery can output power to charge the car. As far as I can tell, the Tesla charger is limited to 5 A minimum at 230 V, so the battery and inverter would need to be able to carry at least 1150 W sustained.

The solar also outputs 2100 W peak, so I assume the intermediary battery would need to be able to handle that kind of input.

I initially looked at getting something off the shelf like a Jackery 1500 battery, but that is limited to 600W input from solar (I'm assuming the size of the battery is what limits the input?)

My questions boil down to;

• How large would the intermediary battery need to be?
• What voltage should a system like this run on?
• Do I need any other parts other than solar controller, battery, inverter?
• Is there any off the shelf system that could handle this, either partially or completely (and that is not outrageously overpowered or expensive!)

I would really appreciate it if someone helps me walk through the math as well so I can rerun the calculations with different battery and solar setups

 

[floydr]

• How large would the intermediary battery need to be?
• What voltage should a system like this run on?
• Do I need any other parts other than solar controller, battery, inverter?
• Is there any off the shelf system that could handle this, either partially or completely (and that is not outrageously overpowered or expensive!)

10-20 kWh capacity minimum (ideal would be at least 2 times thesize of the tesla battery), Also depends on where you are located ,how much driving per day you do . number of hours of sunshine (usable sunshine) per day you get. The 2100w solar system is too small to keep up with the charging. 2100 x 5hours max production is just 10 kW a 21000w system is 105 kW
How often you want to charge the car ? The assumption is that you would charge during the day for a solar charge system to work. But if you up the size of the battery bank to 160 kWh storage capacity and only use 20 kW per day that gives you 7-8 charges you could use the car during the day charge it at night.
51.2-51.8v 16s LiFePO⁴ 14s NMC, any of the 3.6-3.8v nominal lithium cells. these voltages work with most 48v lead acid based inverters
There are inverters that take higher voltages.
A bms is required for all lithium based ESS
Tesla powerwall, LG RESU, There are many most have a high price. DIY for large systems one of the best resources around for DIY powerwalls is Second Life Storage & Solar
later floyd

 

[brian_]

The 2100w solar system is too small to keep up with the charging. 2100 x 5hours max production is just 10 kW a 21000w system is 105 kW

That's really confusing when you don't use the right units of measure for power and energy. This was presumably supposed to be:

The 2100 W solar system is too small to keep up with the charging. 2100 W x 5 hours per day max production is just 10 kWh per day, a 21000 W system is 105 kWh per day​

How often you want to charge the car ? The assumption is that you would charge during the day for a solar charge system to work. But if you up the size of the battery bank to 160 kWh storage capacity and only use 20 kW per day that gives you 7-8 charges you could use the car during the day charge it at night.

Good point, but again one of the units was mangled:

But if you up the size of the battery bank to 160 kWh storage capacity and only use 20 kWh per day that gives you 7-8 charges you could use the car during the day charge it at night.​

 

[brian_]

If the car is home every day, and if the car's battery capacity is high enough that you don't need to worry about charging it every day, then the intermediate battery at worst only need to store one day of solar generation, to be transferred to the car after the end of the day. As floydr noted, that might be 10 kWh. If you want to use the solar system to power anything else off-grid, I would think that you would want about that much energy storage anyway.

 

[reiderM]

Yep, with only 2100w of solar, this is a waste of time and money. It'll take well over a week for the panels to output enough energy to fully charge your tesla. You'll probably want a system that is at least 8000w.

 

[cricketo]

What is ~2100 W peak ? Is that your total wattage of the PV array, your calculated estimate based on the geographic multiplier ?

 

[floydr]

Good point, but again one of the units was mangled:

Funny thing is just before I posted this I changed the the kWh to kw. a duh moment
Since it looks like the Op is in Sweden 5 hours per day might be stretching the hours per day of solar production in winter, early spring, late fall.That is why I suggested a 160 kWh battery bank and 20kW solar array.
Later floyd

 

[cricketo]

Funny thing is just before I posted this I changed the the kWh to kw. a duh moment
Since it looks like the Op is in Sweden 5 hours per day might be stretching the hours per day of solar production in winter, early spring, late fall.That is why I suggested a 160 kWh battery bank and 20kW solar array.
Later floyd

It is Sweden, but Sweden is not Lichtenstein either - it stretches over 1000 miles South to North, making the geographic adjustment a significant part of the answer.

 

[OR-Carl]

So as mentioned, this is going to be a slow way to charge a car, but I am going to assume that you know this, and your intention is to use the big battery pack of the car to basically do opportunity charging, and not try and rely on a 2kw array to completely fill your battery every day.

If you were connected to a grid, I would agree that this was going to be more hassle than it was worth, but if you want to charge off-grid, then the situation is a bit different, as there are not really a lot of other options.

The part I am not sure on is if you can tell the EVSE or the onboard charger to only deliver/accept a certain amount of current. With my Jerry-rigged truck, I can just tell it to charge at whatever level I want it to, and then set that at just below what my panels can produce. If that is possible with your setup, then you should be able to just size the inverter to be able to deliver your peak power, and put in battery with enough capacity to run the charger for like maybe an hour? Just enough to give you a bit of buffer for clouds or whatnot. It will likely mean you have to babysit it while it is charging, but it should work just fine.

If you cannot throttle down the amount of current, then you would likely need to increase the size of the array and inverter to handle whatever the charger is rated for. I dont really know anything about real electric car chargers, basically you need to recreate whatever sort of circuit it needs, have enough panels to run that load, and enough batteries to pick up the slack for a little while. Also, as mentioned, you will need plenty of time to babysit it. But, you get the satisfaction of knowing you can charge up your car in the event of a zombie apocalypse. The self-driving feature will be nice when you want to hang out of the window and fire your machine gun with both hands.

 

[brian_]

The part I am not sure on is if you can tell the EVSE or the onboard charger to only deliver/accept a certain amount of current.

An AC EVSE is just a smart on/off switch - it can't control current except to shut it off entirely, and to tell the vehicle how much is available; the onboard charger controls the charging rate. The good part of this is that the onboard charger should stay within what the EVSE says is available, within the limits of the J1772 protocol presumably used.

 

[brian_]

Yep, with only 2100w of solar, this is a waste of time and money. It'll take well over a week for the panels to output enough energy to fully charge your tesla. You'll probably want a system that is at least 8000w.

If the solar system can deliver - for instance - about 10 kWh per day, that's enough to drive the car at least 50 kilometres per day in reasonable conditions. If that meets the owner's needs (whether it is 50 km each day, or a 200 km trip every four days, or whatever), then the system will work.

Seasonal variations will be the big problem, with winter both greatly reducing energy production and substantially increasing energy demand (for the same distance driven).

 

[cricketo]

I brought it up in another similar thread, but folks can also take a look at this :

SUNNY BOY-US

The Sunny Boy-US features Secure Power Supply, which now delivers up to 2,000 W of opportunity power when the grid goes down and the sun is shining. Installation has never been easier thanks to SMA’s Installation Assistant, direct access via smartphone, and integrated DC disconnect. Independent input channels, each with OptiTrac™ Global Peak MPP tracking, mean hundreds of stringing configurations while solving the challenges of complex roofs and shading. Through innovative improvements, the Sunny Boy US helps you decrease costs throughout all stages of the project cycle.

So basically it may be possible to build a charging setup that will operate opportunistically without a battery altogether. It may not work as well, but it will be a lot easier / cheaper to build.

 

[brian_]

I have no idea where subutai is in Sweden, but insolation in most of southern Sweden is over 1000 kWh per year, per kW of array capacity. So a 2100 W array should yield over 2100 kWh per year, or 5.7 kWh per day (the 10 kW estimate would not be valid as a year-round average). As noted above, with the entire country at least 55° latitude, the seasonal variation is huge.

This is a trend (should display for two complete years of 2020 and 2021, but tweak the sliders if it doesn't) for a commercial solar farm in southern Alberta, a bit further south than the southern tip of Sweden: Brooks Solar
The large-scale pattern of output shows the changing length of day; the day-to-day variations within that are due to weather. If you zoom right into to a good day you can see that output goes right up to the rated 15 MW capacity, but of course only at mid-day in clear weather. The annual output of this facility has been about 21 GWh per year, or about the 1400 kWh per kW of array capacity expected in its location, but a mid-summer month produces several times as much energy as a mid-winter month.

 

[cricketo]

https://iea-pvps.org/wp-content/uploads/2021/10/National-Survey-Report-of-PV-Power-Applications-in-Sweden-2020.pdf

 

[subutai]

Thank you for all the brilliant replies. I know the system is quite small, but as @brian_ has pointed out, with 5 solar hours I estimate to be able to drive at least 50 km per day in the summer charging only off solar. I have already considered the size of the solar system and the implications for how many kilometers that will get me, but I appreciate everyone chiming in to confirm my calculations.

For some context, and yes I know I am crazy, but I am planning to make this system entirely portable. The plan is to drive south one summer and challenge myself to see how far south I can get using only solar that I bring with me. I know it's ridiculous, but people have done crazier things . And also, solar radiation average per year only counts if the system is stationary, right? The main purpose of my system is to be used during the summer, when even Sweden gets a lot of solar radiation.

This is also why I considered the Jackery 1500 because it's reasonably packaged and portable. A powerwall or such a massive 8 kWh+ system is not an option (and way out of my budget). And as some have pointed out, the solar will only output ~ 8 kWh in a day at most. So really what I'm getting at is I'm trying to estimate the lower limits of a battery system. If the battery fills up that's fine, I will take that opportunity to charge the car. I know it's very unpractical, but that's fundamentally what makes this a challenge!

@cricketo I'm not entirely familiar with the language, but as far as I understand, peak is what the solar system puts out at noon. It's probably redundant language. I plan on building a system with 12x175W panels if that's more clear!

@OR-Carl The Tesla on-board charger is able to limit how much current it draws from the grid. I have no idea how it works, but the range is between 4 A and 16 A on the European grid. That's why I estimate the min inverter output to be around ~1200W.

 

[john61ct]

You already have a big battery in the Tesla, how many kWh?

That 50km,you think uses how many kWh?

the Jackery 1500 is how many Wh?

price per kWh?

8kWh is not a big battery for this use case?

Where are you carrying the 12 panels?

Are you planning to deploy them by the side of the road in a spot with no trees?

 

[subutai]

You already have a big battery in the Tesla, how many kWh?

That 50km,you think uses how many kWh?

the Jackery 1500 is how many Wh?

price per kWh?

8kWh is not a big battery for this use case?

Where are you carrying the 12 panels?

Are you planning to deploy them by the side of the road in a spot with no trees?

The Model 3 LR has about 68 kWh of usable battery. If I avoid driving at highway speeds, I think I can get consumption down to 140 Wh/km, so 50 km would require ~7 kWh (+charging losses) or about 10% of the usable battery. In theory I could park the car in an open space and set up the solar and let it charge for a week no problem. Then I could drive up to 350 km in one go.

Jackery 1500 - it's in the name. It's a 1500 Wh battery. Price per kWh? Not sure what this question means so please clarify! In terms of 8 kWh storage I think it's overkill, since the battery would just be a buffer between intermittent/unpredictable solar and the car's battery. I doubt the battery would ever get close to full if I do this challenge. Capacity is not the limiting factor for a portable system, charging rate is.

How to carry the panels is a solvable problem. I have plenty of ideas but nothing set in stone. I'd preferably park somewhere with no shade for charging yes Thanks for all your questions!

 

[floydr]

I like a challenge ( why else would I be building an Ev) good luck on the challenge. When I first responded I assumed that this was a stationary set up my bad. Since it is a portable system build a system like @cricketo suggests could work. Solar panels to an off grid solar pure sine wave inverter 230v eliminate the battery. https://www.amazon.com/s?k=300+watt+solar+panel+flexible a couple of the 1200w kits , an offgrid/hybrid 2.4-3 kW pure sine wave 230v solar inverter to AC EVSE. if necessary put a small 2 kw LiFePO battery in the system. The flexible solar panels 300 watt are 1.2m x .5m x 2.5mm + junction box 2kg . There should be room in a model 3 for 8 300w flexible solar panels.
This is for one summer? Not a multi year summer project. From what I have read the flexible panels have problems lasting for very long. Good luck
Later floyd

 

[OR-Carl]

Interesting! I see it is more of a "The Martian" situation than "Mad Max", but I think the basic idea seems valid. I suggest you do some experiements to see how low you can actually get the consumption to go by driving slower so you have an idea of what to expect. There is probably a sweet spot between reducing drag and not having a bunch of angry cars behind you as you putter along. At least most of Sweden is flat, so you wont have to worry too much about hills. You dont happen to live in Dalarna, do you?

Flexible solar panels would save a lot of space and weight, and they do wear out after a couple of seasons. I have some that I think are maybe 4 years old, and they dropped to about 50% output. That Jackery power unit seems like it would likely work, but you could also put something like it together yourself and have more flexibility to do repairs and or upgrades.

Your basic components should include a battery, an inverter, a charge controller and some sort of meter. You would be wise to think about how you are going to rack your solar panels too. I am currently running my shop with a solar panel propped up on buckets, but it does not give optimal results, to say the least. A 2000W array will be fairly large. To get the most sun, tracking the sun would help, which will be easier if you are not having to fiddle with a dozen panels leaning against rocks and things 😀 . Also, MC4 connectors are sort of a hassle to undo, and it might be worth thinking about swapping them for some other suitably sized quick connector.

 

[cricketo]

12x175W

I suspect that will not provide 2100W peak at noon. That is just the nameplate capacity, but that's where the the geographic multiplier comes in, which basically takes the latitude into account. For example for Oregon, which is significantly further South, 0.8 is used. So if I wanted 2100W of peak output I'd actually need a 2500W worth of panels. Though to be honest I don't know if they take the angle into account - typically stationary systems are installed at 30 degrees, which is sort of an average angle for year round operation. My optimal summer angle would be about 20 degrees I believe.

Either way, I do have some experience with the mobile solar systems too, and I would say even 2kW will be quite a pain to position at an optimal angle and without shading