[pchew]

I have a 100w solar panel on top of my car...to tend a 12v battery. It's got a Dewalt battery charger, mikrotik ltap, and raspberry pi hooked up to it. Little hotspot with multiple sims and resource server(mainly just for fun). Anyone that can do basic math should immediately realize there's just not enough area to make an appreciable difference in regards to mileage.

[barnas2]

The Prius Prime solar panel roof I think can net 3-6 miles a day under ideal conditions (which we're probably close to here in Arizona). I think that's a little more than people would expect, but still only applicable in niche conditions (tiny daily commute, or a longer non-daily commute). I think the math works out to ~4-6 years to break even for the cost of adding the solar roof assuming $0.15 per kwh, which isn't terrible.

If solar tech gets more efficient or cheaper, I think it starts becoming a much more attractive option in some areas. If you get into the 10+ miles per day range, that would cover a lot of peoples commutes in certain areas.

[thegrim33]

13.6 kWh battery. 39mile EPA range. Equals 2.87 miles of range per kWh. Leaving it out for 8 hours straight, on a sunny day, in LA, netted 915 Wh. Or, 2.86 miles. [0] Not 3-6, 2.86.

2.86 miles of charge, but only if left outside, uncovered, in full sun, on a fully sunny day, for a full 8 hours, in a place that gets effectively the maximum amount of solar radiation per day out of anywhere in the entire country.

Now, do the same experiment anywhere else in the country, that doesn't get max solar radiation, or that can't get full sunlight for full 8 hours, or where it's cloudy at all, or rainy at all.

2.86 miles per day is the practical MAXIMUM, given perfect conditions. For the average scenario it'd be some fraction of that.

The 6 miles figure is what they said you'd get if, in addition to perfect conditions, "if the sun shifted its orbit" (?) and gave perfect sunlight for 12 hours straight. Which is a number which should obviously not be thrown around as if it's obtainable.

The fact that they're quoting numbers about what range you'd get if the solar system was constructed differently also makes me doubt the impartiality of their experiment and the numbers they provided.

[0] https://www.motortrend.com/features/the-2023-toyota-prius-pr...

[cornholio]

> 2.86 miles per day is the practical MAXIMUM, given perfect conditions

In your particular setup.

A typical car can expose about 3 square meters of lateral area for those same 8 hours, and receive 3 kW of irradiance. multijunction cells can exceed 50% efficiency, so we're talking about a theoretical upper limit of 12 kWh electric per day.

That would require a vehicle totally covered in cells, including the windows, so not very practical, but adding up to 30 miles/50 km per day is nothing to sneeze at.

We could also imagine all sorts of solar receivers that engage during parking and inflate the apparent surface within the limits available, track the sun etc. to maximize energy.

[labcomputer]

This is all wrong.

> multijunction cells can exceed 50% efficiency

The maximum demonstrated efficiency of a multijunction cell, in a lab, WITH CONCENTRATION is less than 50%. Commercially available cells are lower.

Concentration is an important caveat for two reasons:

First, it implies that you are collecting light from a larger area than the PV panel itself. Second, efficiency grows with increased irradiance (so efficiency will be lower without concentration).

> 3 square meters of lateral area

Lateral area is meaningless. It’s all about area perpendicular to the solar axis. Unless you are driving a box van or a big pickup truck, there is zero probability that you can put 3 kW of irradiance on your panels. Neither of those vehicles will achieve kWh/mile numbers anywhere close to a Prius.

In practice, you need to halve the efficiency and more than halve the collection area you quoted. You also need to account for conversion losses.

[cornholio]

This is a theoretical exploration of the hard limits, not an engineering design.

The multijunction theoretical efficiency limit is 87% with infinite junctions, and over 50% with a practical number of junctions. There's nothing stopping you from creating a miniatural concentrating solar device that focuses the light from a 10 cm^2 area onto a 0.5cm^2 cell, we haven't seen such devices because the cost and extra mass exceed what you get from the efficiency gains when you can simply increase area; a very area constrained application with high power requirements might change that.

> It’s all about area perpendicular to the solar axis. Unless you are driving a box van or a big pickup truck

Again, what stops the top hood and engine cover of a Prius from raising at an angle and tracking the sun, perhaps even unfurl additional area? what about the area of the doors and windows?

Current solar cars can drive 1000 km per day with an average speed approaching 100km/h. It doesn't seem completely out of the realm of the possible to achieve 50km in an hour for a passenger car that can expose similar area while parked.

[labcomputer]

> This is a theoretical exploration of the hard limits, not an engineering design

You replied to (and even quoted) a comment explaining the practical limits, then doubled-down and said multi junction cells can exceed a number which has never been experimentally demonstrated.

And here you are again saying that multifunction cells can achieve 87% efficiency.

An ICE engine can achieve 100% thermal efficiency with the right Th and Tc. That has about as much relevance to the discussion at hand as 87% efficient solar panels.

Let me be clear: the most efficient cells that have ever been experimentally demonstrated under any conditions are less efficient than the number you originally stated. In real world conditions the number is half of what you originally.

> Again, what stops the top hood and engine cover of a Prius from raising at an angle and tracking the

I guess I’m just feeding the trolls here.

[conjecTech]

EPA range tends to be pessimistic for EVs as it assumes you are always traveling at highway speeds. Even small reductions in speeds can make EVs much more efficient since drag is quadratic. A quick google search shows Prius prime owners reporting 4-5.5 miles/kwh, so the 3-6 mile range is entirely plausible.

[bangaladore]

> EPA range tends to be pessimistic for EVs as it assumes you are always traveling at highway speeds.

EV EPA range historically has been overstated. However, the water is muddied because the EPA doesn't really force the manufacturers to give an accurate number. A manufacturer can choose a highway only test, but then also arbitrarily decide to derate the value (EPA example is 70%). A manufacturer can choose to include city driving in the rating and weigh it accordingly and also derate the value (if they want).

Tesla traditionally (still the vast majority of new and used EV market share) has been the only manufacturer that uses the highway + city driving tests. People then get surprised when the car cannot do the full range at 85 MPH.

All in all, this is the EPAs fault. For EVs they really need two numbers, city driving range and highway driving range. EVs are so much more efficient than ICE that speed makes a huge difference given there substantially smaller energy density.

https://www.epa.gov/greenvehicles/fuel-economy-and-ev-range-...

[slaw]

EPA's city cycle speed is an average 19.6 mph, highway speed is an average of 48.3 mph. City range weighted at 55% and the highway range at 45%.

[bangaladore]

I don't think it's that simple. IIRC you don't have to do the city testing.

https://www.epa.gov/greenvehicles/fuel-economy-and-ev-range-...

[labcomputer]

Regardless, the EPA test cycles have unrealistically low speed profiles.

[pchew]

Everyone is also glossing over the distinction that regardless of the actual amount, it's not at an actual voltage that can charge the battery to add mileage. You can hypothetically say that because it's offsetting the power usage from the AC that it could theoretically be saving that battery usage...but there's so many gross assumptions being made that it's a pointless statement to make, and it's all out the window the second the car starts the ICE side of the hybrid drive system for even an instant.

[Dylan16807]

> Everyone is also glossing over the distinction that regardless of the actual amount, it's not at an actual voltage that can charge the battery to add mileage.

Neither is the voltage when you plug it in at home. The car has a unit specifically to convert the voltage.

If you're saying they didn't connect the right wires for that, that sucks but is easily fixed.

> it's all out the window the second the car starts the ICE side of the hybrid drive system for even an instant

Nah, doing a drive where it's 99% solar power and 1% "burned an ounce of gas to maintain the engine for the month" is fine.

[Dylan16807]

That article sure has 49 pictures, none of which show more than the very edge of the solar panel.

But looking at some proper pictures, it covers most of the midroof with 7x8 tiles of solar. But you could fit a good percentage, even sticking with a design where a huge amount of the roof into the trunk is all glass. And there are no panels on the hood. So that's an easy doubling right there, more with an average car roof shape.

[schiffern]

  >13.6 kWh battery. 39mile EPA range... 915 Wh... 2.86 miles, not 3-6

If we're picking nits, the usable capacity is only 10.5 kWh (11.5 kWh with AC-to-DC losses), so it should be 3.40 miles. Not 2.86, or even 2.62 (= 39/13.6 * 0.915).

[bgnn]

I wonder how much extra range you would get if one leaves the car in the shade so that it doesn't get super hot and there is no need to turn on the AC hard? I bet it's more than 2.86 miles.

I believe having a carport and house roof covered with solar panels + (PH)EV is the best option.

[beAbU]

Does the extra 3-6 miles factor in the need to now run the AC much more aggressively because the car will be hot from sitting in the sun all day?

If this quoted number comes from the manufacturer itself, then I think the answer is "no".

[thinkalone]

The initial use of solar on the Prius was to power a ventilation fan while the car was parked, and the current version seems to specifically be designed to provide power to the air conditioner while driving. But, I also can't imagine the difference between cooling down the cabin is much different from parking in the sun or in the shade - you'd be running it continually to achieve "room temperature" during the entire drive either way.

[saltcured]

You can't imagine that air conditioning power draw varies with the heat load that it is working against? As a heat-pump, it takes more energy to move more energy.

In the old days, they used duty cycle to adapt to the changing load. Modern ones do things like varying compressor displacement or compressor speed to adapt to the load. Variable frequency inverters are used to efficiently drive electric compressors.

The variable displacement trick is used in ones mechanically linked to internal combustion engines. It can vary the compression stroke to account for different load as well as different engine speed.

[geerlingguy]

Watching power draw on my Leaf with LeafSpy, the AC seems to use between 500-1000W (maybe more sometimes, but that's just off the top of my head from a few times running it while driving).

At the low end maybe achievable with a full rooftop covered in solar panels, but probably not adequate at 1kW+.

[danans]

> Does the extra 3-6 miles factor in the need to now run the AC much more aggressively because the car will be hot from sitting in the sun all day?

Most cars are already sitting in the sun all day.

[aziaziazi]

Not here in European cities where they’re either within a multi story park or in the side of a half day shaded tiny street.

[sdenton4]

Yah, it's a great point that the whole scheme is predicated on very questionable land use policy...

[parsimo2010]

You are not getting the 3-6 mile per day boost if your car is parked in the shade.

[tstenner]

That's his point. You won't get any reasonable charge because you (mostly avoid parking your car in direct sunlight

[gambiting]

What kind of European cities are you talking about lol, no offence but I hate this generalisation of "European" anything as if Southern Spain has the same culture and architecture as Poland or Lithuania.

[gunalx]

It is a somewhat fair generalisation of urban europe to have more innside parking or tall ish buildings that give shade.

[lupusreal]

To my thinking, the best use of a solar panel on a car is running a low power AC unit all the time whenever the car is in the sun. Parking in the shade often isn't possible.

[kgermino]

I don’t think you’d have to run the AC any more aggressively with the solar panels than with a traditional steel roof?

If you’re suggesting it wouldn’t work in a garage, that’s obviously true (and another factor in whether car solar makes sense) but many (most?) people park their cars outside during the day anyway. I for one can’t remember the last time I parked under cover

[singpolyma3]

That's still 3-6 fewer miles worth of charging to do from more expensive sources. Even if it can't come close to covering your full use it's still covering something

[IAmBroom]

Not if it won't recover it's own costs.

[cma]

It may still lose on this, but you would also want to include the externality costs that the consumer doesn't themselves bear for whether it is worth it overall.

[dzhiurgis]

Cost is convenience.

Manufacturing cost is like $40 cells, $20 electronics and $200 in glass fusion, mounting, etc.

My car eats 200W just being online so it would be useless to charge from solar.

[roamerz]

There is going to be a parasitic drag loss to figure into it as well. I think the only way to accurately calculate that would be in a wind tunnel or maybe an amp meter with a before and after installation under identical conditions.

[jeffbee]

The Prius Prime solar roof is a $610 option available only on the top XSE trim level, so a hypothetical buyer is paying ~$7500 to access this effectively negligible amount of energy.

ETA: and the fact that this option is tied to the significantly less efficient 19" wheel package, instead of the standard 17" wheels, means that this will never, ever be a net benefit.

[mattmaroon]

Not if they were getting that trim level anyway.

[AdamJacobMuller]

I just started doing this with my car, mostly to add a camera/temp monitoring for when I leave my dog in the kennel in the car (she's well watched over, please don't fret over it).

I'm hooking it up via starlink specifically so it works in remote areas with no cell coverage too.

Monitoring and proxying everything via an RPI as well. Victron DC-DC inverter to keep the bluetti battery pack charged with bluetooth relay boards so we can turn loads (camera/starlink/others) on/off programmatically (it only turns the starlink on when there's no good/known wifi for example).

Fun project, combines software dev (which I'm fairly good at) with hardware work (which I'm less) and my dogs (which I'm a big fan of).

[theoreticalmal]

It’s interesting to see how many people in this thread have the same idea. I’m doing something similar and learning that hardware is just plain hard

[ben_w]

The maths says that the *mean* number of miles driven by a vehicle is surprisingly low, and that tiling the surface of a car can get to about 80% of that *mean* in places where the car is just left out on the street and not shaded parking.

But!

That's a practical consideration at the level of "should a government require EV makers to design the roof, bonnet, doors etc. to be tiled in PV in order to reduce, but not eliminate, the induced extra demand on the grid" and definitely not "should I personally bolt a small, fixed, PV panel and inverter into my EV as an aftermarket DIY job?"

The former gets wind-tunnel tests for efficiency, QA, designed around all the other safety concerns cars have e.g. crash safety.

The latter, doesn't.

[jollyllama]

Very nice. How long does that tend to stay alive for? And what kind of cold weather conditions do you have to contend with?

[andoando]

If only we didnt start off with having 3000+ lbs of metal to move a 100-200lb person as a design limitation