EEVblog #632 - Solar Roadways Are BULLSHIT!

[EEVblog]

Dave debunks the Solar Roadways project that generated over $1.2M on Indiegogo, as well as almost another million dollars in US government funding.
In this video Dave does the solar generation calculations to see if this system is viable from an energy generation viewpoint, and also how much power the LED's might typically consume.

Datasheets & Links:
http://www.solarroadways.com
Sunpower E18 Solar Panels: http://hvce.com/admin/content/uploads/Sunpower230.pdf
Dave's solar system data:
http://pvoutput.org/intraday.jsp?id=22501&sid=20402&dt=20131104

Thunderf00t's videos:

[zapta]

Bullshit? With the right marketing that can get them tons of tax payers' money.

[GeoffS]

Was it just me or was there a hiccup at about 23:50?

[Legit-Design]

Spotted many hiccups at 23:00 and watching after that. Also the "sample text" ? at the end. BUT I DON'T MIND. AWESOME VID 

Please, please more. Gotta milk the cow while the milk is hot?

[EEVblog]

Bullshit? With the right marketing that can get them tons of tax payers' money.

No doubt!
Already nabbed them $850K of taxpayer $$$$$$

[mux]

Although I'm with you on the general gist that this doesn't really seem like the best way to go about engineering solar panels or a road, the calculations you do are pretty handwavy to say the least, using fairly bad metrics to try and calculate things.

For instance, you're using some ballpark LED spacing and individual LED power for your LED calculation. That just seems unnecessary. It's much easier to work from visibility principles. Let me try:
In clear daylight, current road markers (white reflective paint) get the required minimum of 10:1 contrast ratio by dispersing light at nearly 100% efficiency. That means the light coming to your eyes from these lights is in the order of 10k lux surface brightness. 10klux is 10klumen per square meter. First google hit (http://uttipec.nic.in/writereaddata/mainlinkFile/File208.pdf) tells me that road lane markings are supposed to be 3m lines with 4.5m gaps, 150mm wide. That's equivalent to a 0.06m wide continuous line, or 60 square meters per kilometer of road. The side markers are going to be 150mm contiguous lines, so for both markers on a 2-lane road that's 300 m2 per km, making a total of 360 m2 of marker surface per km. Times 10klumen is 3.6Mlm of total light output.

Assuming perfect conversion efficiency and assuming the use of the best commercial LEDs available at the moment for general lighting (Cree XM-L, 151lm/W raw), that's 23.8kW of power in full daylight.

Now, the same calculation as for your insolation applies for the drive strength of those LEDs. You're absolutely not going to run those at full strength all day, that is just an unreasonable assumption to make. You can simply use the same calculations you used for solar insolation to get the effective run time. Running them all day at variable strength is, like insolation, going to be equal to running them at full strength for 7 hours. And over the total year, you can use the same x0.6 multiplier to account for seasonal changes in daylight brightness. This means the total AVERAGE power required is going to be 23.8kW x (7/24) x 0.6 = 4.17kW. This boils down to 100kWh/day, or about 3% of the daily production. Less than cable losses.

If you're going to do calculations to try and prove your point, do them well or don't do them at all. Again, I agree with the general premise but I think your math is severely lacking.

[EEVblog]

Was it just me or was there a hiccup at about 23:50?

Not for me?
Athough Youtube did through up an error.
Might still be processing or something.

[Zbig]

I also noticed what seems like some editing glitches starting around ~22 minutes mark.

[KD0CAC John]

It's another attempt to create negative propaganda about solar - not Dave's video , but the guy that came up with this .

[GeoffS]

Was it just me or was there a hiccup at about 23:50?

Not for me?
Athough Youtube did through up an error.
Might still be processing or something.

Perhaps I just shouldn't watch videos after midnight 

[EEVblog]

Assuming perfect conversion efficiency and assuming the use of the best commercial LEDs available at the moment for general lighting (Cree XM-L, 151lm/W raw), that's 23.8kW of power in full daylight.

So pretty close to my estimated 30kW.
These numbers have to be hand-wavey, because real data doesn't exist for the very small view angles and thick glass we are talking about here.
It needs serious empirical measurement.

Now, the same calculation as for your insolation applies for the drive strength of those LEDs. You're absolutely not going to run those at full strength all day, that is just an unreasonable assumption to make.

Yes, I know. But I figure it's going to come out in the wash. Because during the day you are going need almost ludicrous amounts of light. To the point where it may simply not be possible to view that at all (as I noted in the video)
You are seriously forgetting the issue of very shallow angles here.
I get what you are saying, but I'm not convinced your numbers are any better, or that you are comparing apples with apples here.

You can simply use the same calculations you used for solar insolation to get the effective run time. Running them all day at variable strength is, like insolation, going to be equal to running them at full strength for 7 hours. And over the total year, you can use the same x0.6 multiplier to account for seasonal changes in daylight brightness.

Yes. My 0.5W figure was designed to take into account the variation, and would be an average over a day.
I'm not buying the season change in brightness. I don't think it will impact much.

[/quote]
 This means the total AVERAGE power required is going to be 23.8kW x (7/24) x 0.6 = 4.17kW. This boils down to 100kWh/day, or about 3% of the daily production. Less than cable losses.
[/quote]

Where did your nighttime consumption go?

And once again, it's all pointless discussion if you can't see these things in the daytime anyway, which I suspect will be the case.
If you want to prove me wrong here, do some real testing with LED's show us. I think you'll be in for a bit of shock.
I recon you'll still need a hefty amount of power at night time as well. I stick with my figure of 0.5W average.
Not to mention visibility at night time with the car headlights on.

In any case, even without any LED consumption at all, you will find the ROI of these panels is absurdly low.

[EEVblog]

I also noticed what seems like some editing glitches starting around ~22 minutes mark.

Ok, I'll check this again in the morning.

[mariush]

Was it just me or was there a hiccup at about 23:50?

Not for me?
Athough Youtube did through up an error.
Might still be processing or something.

Nope, at around 23:00 you started a sentence and probably didn't like how it came out and decided to start again and you forgot to cut out that part when editing. Then you cut again at around 23:50.
And there's the "Sample text" black screen instead of the usual "footer" video with the eevblog text and lightning sparks and noise.

Video works, just looks rushed to upload.

[EEVblog]

Nope, at around 23:00 you started a sentence and probably didn't like how it came out and decided to start again and you forgot to cut out that part when editing. Then you cut again at around 23:50.
And there's the "Sample text" black screen instead of the usual "footer" video with the eevblog text and lightning sparks and noise.
Video works, just looks rushed to upload.

Damn, ok. Too late to fix now.
I might take offline and fix tomorrow morning.

[Legit-Design]

Maybe it's a great way to keep idiots laughing at editing mistakes when they don't have to even try and say the number or whole project is going to work? I really hope there are those other videos coming.

[firewalker]

Maybe they ?nvented a more efficient way to convert solar energy to electric power! 

Alexander.

[max_torque]

They also market these "led panels" as being "safer" than conventional road markings?  er, how?  In the UK with have had passive "catseyes" for over 50 years.  They are simple reflectors, with a built in cleaning mechanism, and they work perfectly and are cheap.  Being able to "actively" illuminate a road is no better than a simple painted white line, and the ability to "reconfigure road markings"? again er, why?  Motorists have enough trouble with fixed lane markings, being able to modify them isn't going to help matters.

In the UK, a kWh of electricity currently costs around 15 pence (£0.15), so each square meter of roadway, if we take Daves figure of around 350wh/day is worth around just 5 pence per day (not including any power consumption by LED/road heating or other such pointless addenda).

And the killer point.  We are NOT short of space to mount conventional solar panels.  IE lack of space is NOT the factor that is preventing a more widespread uptake of solar power generation.  Therefore, "solving" an non problem is not going to actually having any effect on the situation.

For example, in 2014 it is completely technically possible to design a special system of parachutes and catch netting to allow people to get downstairs by just jumping out of an upstairs window.  However, as most people just use the stairs to do this, that "technically" feasible solution is just as pointless as "solar fracking roadways"!......  ;-)

[Hole]

They simply maximize problems.

Assume solar cells placed on the roof are the optimum. They add substantial surface stress, distribution loss, reduced efficiency, "magic grid" management loss and so on.

Heating them to melt the snow? Good idea, but where comes the energy from when there is snow on the panels?

Looking at the current state of our streets (at least here in Germany) building and maintenance costs enough money. Now implement high tech and pay that... And nothing against street workers  construction worker but I assume it will be a hard job to trains Joe-concrete-truck-driver to wire up solar street panels.

Why not start with adding solar to all roofs. Produce where it is needed. After that, we can conquer the streets.

[EEVblog]

Video will be back up after I fix the edit and re-upload tomorrow morning, maybe by 10am.
Yes, I really rushed this one. A rare one where SWMBO let me stay back to finish and upload it.
FAIL.

[firewalker]

How much the glass only would cost? Glass isn't cheap. At least the type needed for the job.

Alexander.

[MrAureliusR]

Video will be back up after I fix the edit and re-upload tomorrow morning, maybe by 10am.
Yes, I really rushed this one. A rare one where SWMBO let me stay back to finish and upload it.
FAIL.

Yeah, when I tried to reply to a comment the video disappeared!

BTW, their website has a ridiculous 'Clearing the Freakin' Air' page where they try to "bust" some myths surrounding their product -- without including a SINGLE piece of data ANYWHERE on the page.
Nice try guys, but no engineer would fall for that.

[Rentium]

God damit i was just about to watch it   

Guess i'll have to wait 8:30 hours untill 10 A.M. 

[Bored@Work]

Nice try guys, but no engineer would fall for that.

And the guy is an engineer. Either the worst engineer ever having managed to get an EE degree (unaccredited prestigious university anyone?), or a scammer.

[AlfBaz]

Let 'em go Dave!

It'll be good for our economy as I envisage another mining boom to supply them the raw materials to make all that shit

[firewalker]

Any rough numbers for the price of the glass (per square meter) they would need?

Alexander.