Then they’re stealing energy from the regeneration of my electric car. How dare they! I paid to get the car up that hill; I’m sure as hell going to extract every joule on the way down.
80 kilowatt-hours per hour. Snert.
There’s zero chance that a kilometer of the piezoelectric crap generates 80 kilowatts. A kilometer of reasonably busy road has perhaps 6 cars on it. A sedan at 60 mph uses perhaps 20 kW, but only around 5 kW of that is rolling resistance. So already we’re down to 30 kW, and that’s assuming perfect efficiency, which is nonsense.
Most of the losses are in the tire itself, not the deformation of the road. That alone cuts us down by a factor of 10, and then piezoelectric conversion inefficiency cuts us by at least a factor of 2, and more likely 5.
I’d be shocked if they could get more than 1 kW out of their 1 km stretch. And it would not greatly surprise me if their 80 kW-h figure is per year.
Really? Because this assessment report compares similar numbers from manufacturers’ and other tests:
Say, an output of 0.1 kW per km per vehicle per hour (conservative low end of the Innowattech figures), times 600 vehicles per hour, would produce 60 kW per hour per km of road surface. Which is not very different from the 80 kW per hour per km of road surface that you’re claiming is impossible.
[QUOTE=Dr. Strangelove]
A kilometer of reasonably busy road has perhaps 6 cars on it. A sedan at 60 mph uses perhaps 20 kW, but only around 5 kW of that is rolling resistance. So already we’re down to 30 kW, and that’s assuming perfect efficiency, which is nonsense.
Most of the losses are in the tire itself, not the deformation of the road. That alone cuts us down by a factor of 10, and then piezoelectric conversion inefficiency cuts us by at least a factor of 2, and more likely 5.
I’d be shocked if they could get more than 1 kW out of their 1 km stretch.
[/quote]
Well, that seems in line with the middle range 0.3-0.8 kW per km per vehicle per hour figures cited in the report.
[QUOTE=Dr. Strangelove]
And it would not greatly surprise me if their 80 kW-h figure is per year.
[/QUOTE]
How do you get from approximately 1 kW per km per vehicle per hour to only 80 kW-h in an entire year? Really, really low traffic volume?
No. A single vehicle does not spend an hour driving a kilometer. It only spends about 36 seconds, or 1/100 hour. So 0.1 kW/vehicle produces 0.001 kW-h/vehicle over the whole stretch. Multiply by 600 vehicles/hr and you get an average of 0.6 kW.
My efficiency numbers were already optimistic. The “Berkeley and Virginia Tech” numbers are probably accurate. The other numbers are simply lies (in particular the Genziko numbers). 13.6 megawatts for 6 cars? It is to laugh. It implies that each car is putting over 3000 horsepower into the road surface.
The other numbers are still implausible. There isn’t more than roughly 5 kW of energy per car even assuming perfect efficiency, but even the lowball Innowattech numbers claim 16.7 kW per car.
I think I see the problem: you’re assuming that this number of kW is per vehicle per hour. That is, you’re assuming that the base number is however many kW are produced by one vehicle in one hour of driving, so for a 1-km stretch the vehicle’s total output would be 1/100 of that base number.
But as you can see from the report I linked (Table 16), the base number is actually the total output for one km of road surface. Dividing that value by the authors’ standard traffic flow parameter of 600 vehicles per hour is what gives the kW per km per vehicle per hour figure. Data elsewhere in the report shows similar relations between car numbers and total output:
Those correspond respectively to values of about 0.33 and 0.81 kW per km per vehicle per hour, consistent with the tabulated values.
Of course, you may still personally believe that your estimate of 0.001 kW per km per vehicle per hour is more realistic than those values, and I’m not claiming to disprove that. I’m just pointing out that you can’t validly derive that number from the data I was quoting from Table 16.
[QUOTE=Dr. Strangelove]
The other numbers are simply lies (in particular the Genziko numbers). 13.6 megawatts for 6 cars? It is to laugh.
[/quote]
What do you mean, “6 cars”? Are you perhaps mixing up your own hypothetical situation of 6 cars simultaneously occupying 1 km of road, at a given moment, with the model in the report?
In other words, 51 MW per km assuming 2250 vehicles per hour corresponds to 13.6 MW per km assuming 600 vehicles per hour, which is mathematically consistent.
So that’s not 13.6 MW produced by 6 cars on 1 km of road surface: it’s 13.6 MW produced by 600 cars on 1 km of road surface, or about 22 kW per km per vehicle per hour.
[QUOTE=Dr. Strangelove]
It implies that each car is putting over 3000 horsepower into the road surface.
[/quote]
Nah.
[QUOTE=Dr. Strangelove]
The other numbers are still implausible. There isn’t more than roughly 5 kW of energy per car even assuming perfect efficiency, but even the lowball Innowattech numbers claim 16.7 kW per car.
[/QUOTE]
Certainly not in Table 16 they don’t. That lowball total output number is 100 kW per km, which divided by 600 vehicles per hour gives about 1.67 kW per km per vehicle per hour, not 16.7.
No–6 cars/km is the same as 600 cars/hr, using a nominal highway speed of 100 km/hr. One can quibble a little bit here but not even by a factor of 2, let alone the orders of magnitude needed to make the idea work.
I had guesstimated 6 cars before you linked to the report, but coincidentally they used the same effective value.
It may be mathematically consistent but it’s completely insane. For a continuous 13.6 MW, the cars currently occupying that section of road must be producing that much power. There are 6 cars on the road (i.e., 1 lane-kilometer) at any given time; hence, each one must be producing over 2 MW.
“kilowatts per kilometer per vehicle per hour” is the worst unit I’ve seen in a long time. Please, just convert to kW/vehicle to see what’s going on. You can do this by multiplying by an average speed of 100 km/hr. That is: x (kW/km)/(veh/hr) * 100 (km/hr) = 100x kW/veh.
So that almost reasonable-sounding “22 kW kilowatts per kilometer per vehicle per hour” is really 2200 kW/veh. Which is just what I said above, and an absurd amount of power.
I don’t know what kind of math you’re doing, but (100 kW/km) / (600 veh/hr) = 0.167 (kW/km)/(veh/hr). Multiply by average speed of 100 km/h and you get 16.7 kW/veh. Which, while not as insane as 2200 kW/veh, is still far over the possible value.
Well, it’s what’s used in the report. I agree it’s cumbersome, but I think it serves a purpose in avoiding mixups between the two roles of the kilometer (see below).
[QUOTE=Dr. Strangelove]
Please, just convert to kW/vehicle to see what’s going on. You can do this by multiplying by an average speed of 100 km/hr. That is: x (kW/km)/(veh/hr) * 100 (km/hr) = 100x kW/veh.
[/quote]
The thing is that the km in kW/km refers to the specific one-kilometer stretch of road surface that has the piezoelectric crystals in it. AFAICT, that unit actually represents the even more confusing “kWh/h per kilometer of road surface”: i.e., the total energy harvested in one hour from that one-km stretch of road with an average traffic flow of 600 vehicles per hour.
Then divide that number by 600 and you get the average output contributed by each of those 600 cars to the total harvested output from that 1-km stretch during that hour. That’s the number tabulated in Table 16 as kW per km per vehicle per hour.
Then when you multiply that number by 100 km/hr of vehicle speed, what you’re effectively doing is assuming that each vehicle is contributing that same average amount to each of one hundred 1-km stretches of piezoelectric-equipped road surface, rather than just one. Naturally, you end up with a number that’s 100 times bigger than the number tabulated in the report, and naturally, that strikes you as very unrealistic.
[QUOTE=Dr. Strangelove]
I don’t know what kind of math you’re doing, but (100 kW/km) / (600 veh/hr) = 0.167 (kW/km)/(veh/hr).
[/quote]
Whoops, yes it does! Sorry about the typo, I mistakenly typed 1.67 instead of 0.167, which is indeed the correct value in the table for the “lowball” Innowattech estimate of kW per km per vehicle per hour.
[QUOTE=Dr. Strangelove]
Multiply by average speed of 100 km/h and you get 16.7 kW/veh.
[/QUOTE]
Nope, now you’re unintentionally harvesting the average vehicle contribution from all those one hundred 1-km stretches of road again, instead of just the one assumed in the assessment report’s model. So of course the number you get is 100 times bigger than it should be.
Let’s look at the math from a different direction.
If a 1 km stretch of PE road receives 600 cars/hour and generates power at 13.6 MW, that means in 1 hour, 49 GJ are generated (13.6 MW * 1 h)
600 cars have passed by. So the road harvests 82 MJ for each vehicle that travels its 1 km length (49 GJ/600 vehicles).
But a gallon of gasoline is 120 MJ (LHV)! A 20 mpg car is going to burn <4 MJ of fuel to travel that kilometer.
(120 MJ/gal)(0.05 gal/mile)(0.62 mile/km) = 3.7 MJ/km
That fuel is already budgeted into the equation. You already had to go up the hill. Going down the other side doesn’t use much fuel, if any, but the cars have a lot of kinetic energy that can be harvested.
It’s designed to sow confusion among people that don’t trust basic unit analysis.
kW-h/h is not a unit of energy. It’s power, and independent of timescale.
Quite the opposite. By concentrating on the 600 number and ignoring the fact that at normal speeds, only 6 cars can be on that 1-km stretch, you’re coming up with a number that’s 100 times too low.
But it all works out if you just look at the units and correctly distinguish between rates and quantities.
Here’s another way of looking at it (though Ruken’s is good too). Do you agree that (kW/km)/(veh/h) rearranges to (kW-h)/(veh-km)? No outside information needed; it’s exactly the same unit, just expressed differently.
Would you agree that this represents the amount of energy (in units of kW-h) produced by a single vehicle-kilometer (that is, one car driving on one kilometer of piezo surface)?
1 kW-h is equal to 3.6 megajoules, and 22 kW-h is 79.2 MJ. But as Ruken said, a gallon of gas has 120 MJ. So somehow the road is extracting 2/3 of a gallon of energy from a car driving one kilometer.
The goal of this technology, if it even has a legitimate goal at all (a point I find doubtful) is to make good use of the energy which is currently wasted in deforming roads. One way to do this is to replace our current deformable roads with other deformable roads that can salvage a small fraction of that energy. Another way to do it would be to replace our current deformable roads with non-deformable roads that would prevent almost all of that energy from going into the road at all. The latter would be far more effective at the goal, and far cheaper.
If you really want to put electrical energy into the grid, then replace the roads with concrete, and then pass a law requiring drivers to plug their alternators into the grid for some amount of time at the end of each trip. With the right values for the requirements, you could generate more electrical energy this way, while costing the consumers less gasoline, and costing the transportation agency less money on infrastructure. This plan is better in every regard than the piezo-road plan. It is better even though it’s a terrible plan, which means that for the piezo-road plan to be even worse, the piezo-road plan must be absolutely abysmal.
How do you get from approximately 1 kW per km per vehicle per hour to only 80 kW-h in an entire year? Really, really low traffic volume?