Redoing the highway math
I haven’t had the time or the patience to repeat the roll-down test again. However, I found an old picture file I took out of the Web a few years ago that has all the measures of my car, and, most importantly, it has a perfect frontal area projection! 
With a little help from “The Gimp” and some basic maths I was then able to get a very precise frontal area for my car: 2,30 m2.
Then I used the spreadsheet again and this time the values appear a little bit better:
Cd = 0,364
Crr = 0,013
I wouldn’t put my hand in fire for the trueness of these values, but they seem very likely.
So, taking Cd = 0,364 and Crr = 0,013…
The force to maintain the car at 120 km/h on a horizontal road would be:
- Velocity = 120 km/h = 33,3 m/s
- F(air) = – Cd * Frontal_Area * 0,5 * Air_Density * V^2
- = -0,364 * 2,41 * 0,5 * 1,22 * 1108,89 = -593,38 N
- F(road) = – Crr * Mass * Gravity
- = -0,013 * (1080+120+30) * 9,81 = -156,86 N
- F(air+road)@120 = - 750,24 N
So the necessary power would be:
- P(120km/h) = F * V
- = 750,24 * 33,3 = 25 kW (~34 HP)
And the equivalent necessary torque:
- T(120km/h) = F * Wheel_Radius
- = 750,24 * 0,273 = 204,82 Nm
So each motor would have to continuously handle 12,5 kW and 102,41 Nm. Pretty easy for today’s technology.
Now let’s think about this for a minute.
The “GM Volt“, it is said, will have:
- a 16 kWh battery pack (next-generation Lithium-Ion), with a useful charge window of 50% (8 kWh);
- a 120 kW motor (3-phase induction);
- a 53 kW generator.
This means a fully charged “Volt” battery (8 kWh) would propel my car at 120 km/h for 19,2 minutes (60*8/25) or 38,4 km, which is not that bad. And the on-board generator would charge the battery in just 9 minutes (60*8/53), so I would enjoy almost 10 minutes of silence followed by 9 minutes of engine roar, and so on. Hmm… it does have the potential to become irritating, if the engine cabinet is not sound-proof enough.
Of course, if I also had the Volt’s electric motor and permanently slammed on the gas (”redlining” the system) the full 120 kW would totally discharge the battery in just 4 minutes (at an unknown speed) and the generator would not be capable of recharging the battery on time, effectively stopping the car!! (That won’t happen in the “Volt”, I hear it has a “bypass” mode, so the generator can feed the motor directly if the battery is kaput. But it will still limit the car to 53 kW of power, if that happens.)
I really like these numbers. There is a nice symmetry to them: the generator is rated at half the power of the motor. I wonder what kind of recipe GM followed to get at this result… city driving cycle statistics? Must be. It also makes my life simpler in building my hybrid: I can build 3 equal motors, and use one as the generator (coupled to the combustion engine). 
July 28, 2008 at 3:46 pm
Hi Vasco,
Great site! I’m still slogging through it all. I too, am building my own car. However, I’m going for a pure EV, and plan on building a “bolt on” system for those long trips. So I guess I’m going down a similar road. You building a self-contained hybrid and me building a two piece hybrid. I like my approach because most travelling is done over short distances and the added weight of a hybrid drains the system faster.
Your number here looks pretty right. Everything I’ve researched so far puts 25kW as the power needed to drive a modern car. A 120kW motor is way overkill. To paraphrase my arch nemesis: no one will ever need more than 64kW of power for a car.
However, my own in flux design is shooting for 25kW (for a motorcycle) to 65 kW for a sedan.
The 120kW is probably not a bad number, as it’s highly improbable anyone could overload this engine and kill it. That may be why they chose 120, and I’m betting the controller won’t let the engine ever get close to 120kW. Making the generator much smaller is also smart. I would almost be inclined to go with 25kW as a minimum for the generator. That way if something goes wrong with the battery system, you can drive the motor(s) from the generator directly with good response – as opoosed to powering a car in “crippled” mode.
Your site has given me some great ideas! I hadn’t considered in wheel motors much before, but it makes great sense for a motorcycle! My trike design iswas a direct drive e system using an axle.
Lastly, I think you abandoned axial motors too soon. I’ve been looking at how some home-grown windmills are using these as generators. They are using disk brake cylinders which are two plates with internal support. They do however have to use special tools to (dis-)assemble the motor parts. Due of course to the great magnetic force.
July 30, 2008 at 11:21 pm
Hi, Celtic! Welcome.
Yeah, pure EV is looking more and more the way to go… I’m thinking about it too. I may give up on the ICE later on…
As to the “bolt-on” generator, have you considered a “trailer” setup? I mean, put the ICE/generator set in a small trailer that is simple to attach to / release from the car.
Or you could use a pushing trailer (google for “EV pusher” and you’ll see what I mean).
The numbers here are for “steady state” usage, not for maximum spikes; I expect that the car may need 2 or 3 times that occasionally, to push uphill or just start from a red light with a heavy load. However, those spikes are temporary and so I don’t think they should be a heating problem. Or course, electronic thermal control is a must, just in case.
Making the generator power equal to the motor(s) is an option. I’ll see if it makes sense when I’m designing the battery pack.
Hub motors are pretty much the standard for production bikes. It makes little sense to carry all that transmission nonsense in such a small vehicle, when you have so little volume to put the batteries in. In a car, the advantages that I see are also the extra free space and weight capacity.
Axial motors are so advanced I still don’t “get” them well enough. Its easy to build something that works, but I’d like to make them over 90% efficient, and that means really understanding all the tricks about halbach magnet layouts, twisted windings, and so forth, and I still haven’t gotten into it as deeply as I should. I especially doubt how I could get enough flux and therefore torque from ironless core motors, which is the top approach I’ve seen around…
Radial motors are way simpler to understand because they are more “traditional”.
Keep us updated on your project, will ya?
Cheers!
August 14, 2008 at 4:27 pm
Hi Vasco,
It was the trailer idea, which I found about here on your site that gave me the idea for a bolt on. My idea is to use a similar set up as the trailer, but instead of an actual trailer, to use a “snap-on” (I know I said bolt-on, but I really meant a quick connect) third axle.
I think a third axle set up would be better for drive-ability. This would mean placing the actual EV motor in the rear and the trunk space in the front. This way you could connect and disconnect the ICE by pushing it onto the read end, and have some quick connect clamps that lock in place. In order to disconnect you’d have to be in park with both engines off and then use a control in the dash. You know safety features so you don’t disconnect while flying down the highway.
On your wiring, I think you’ll find Litz wire to be an interesting topic. I’m wanting to build three stators in a model size and build a model rotor to see what kind of power I get from them. Model 1 will be coreless; model 2 will be with an Iron core, and model 3 will be with Litz wire. Litz wire though is a lot more expensive, I think, and harder to get.
Meanwhile, I have my first half of production magnets now and will be looking for appropriate discs today and picking up a spool of wire. It’ll be enough to do a coreless production and a cored production stator. Now all I need to do is settle on a design for a cored stator. Not an easy task. I have a few ideas, but I think that will be my biggest challenge. To produce a viable cost efficient stator core for an axial motor. I think that is the main reason you see so many coreless axial stators. How do you build a layered stator that won’t fly apart?
I understand wanting 90% efficient. I want to go even higher. I’ve seen claims of 98%. I’m not sure I buy that claim though, except in maybe empty space. My main concern is the magnet bending the steel plate when under load.
My magnets are high-temp Nd magnets making a 8″ o.d. wheel 1/4″ thick (16 magnets 4″i.d.). These babies are powerful magnets. They attach to each other with more than 50lb of force. They can’t be pulled apart by hand, at least not by me. You have to slide them apart, and you’d better be ready for the twisting torque (I’d guess about 15-20lb) on your hand when you do. Even 2″ apart they exert consider force of attraction. I can’t wait to get these on a disc!
As you can see, I have more of an experimentalist approach.
I’m off! I can’t wait for your next installment. Once I get things going, I’ll start putting up details.