[mfucci]

With some friends, I am building a vehicle to beat the current Guinness Book solar-powered speed record.

I am looking for advices on high efficiency electric engines that we can use.

Our vehicle is a train, 30 meters long, covered with solar panels providing about 15kW of power and we are hoping to reach 65 mph (driving on actual train tracks).

I am hoping to find an electric engine + controller with an efficiency above 92% @15kW power input, the gearing will be adjusted to have the best rpm @65 mph.

I looked at the engines of those manufacturers: NuGen Mobility, NetGain and HPEVS (Am I missing some?).

Would AC-35 or AC-50 from HPEVS be potential candidates?
I am ideally looking for an efficiency map if someone has done dyno measurement on those.
For instance, the engine graphed here is too powerful for our application, since 21kW are required to reach 92% efficiency and 65kW for 95+% efficiency.
But 15kW would bring in the 92+% zone for this Prius engine.

A lot of electric engine are boosting 95% efficiency. However it is rarely clear if it is just the engine or this includes the controller loss, this top efficiency is only reached at very specific tpm / torque.

Does anyone as an experience of an electric engine that actually as 95% efficiency for your vehicle?

Any other advice would be welcome!

Thanks!
- Marco

 

[Kevin Sharpe]

Our vehicle is a train, 30 meters long, covered with solar panels providing about 15kW of power and we are hoping to reach 65 mph (driving on actual train tracks).

What's your target weight for the finished vehicle? Do you have a time limit to reach top speed? Is this a 15kWp solar array? Are you charging a battery/capacitor or expecting to use the solar electricity without storage?

 

[mfucci]

What's your target weight for the finished vehicle?

The total weight should be around 3 tons (6,000 pounds).
We may have to add more weight, since more weight is actually desirable for a locomotive (more weight = more traction).

Do you have a time limit to reach top speed?

I received Guinness World record rules and there is no time limit nor max distance to reach top speed. However the record has to be attempted in both direction in less than an hour (and the average or the min speed reach is the record).

Is this a 15kWp solar array?

Yes, we are planning to use commercial solar panel.
The surface is 30m x 3m = 90m^2 of solar panels.
Building our own custom panels out of cells would just take too long...

Are you charging a battery/capacitor or expecting to use the solar electricity without storage

?
No storage, the solar energy should be used directly.

 

[Kevin Sharpe]

The total weight should be around 3 tons (6,000 pounds)..

The Tesla Model S weighs just under 2.5 US tons and iirc requires ~15kW to cruise at 60MPH. Obviously the Tesla is extremely aerodynamic and even if you make your train as 'slippery' you'll need more than 15kW to break the speed record.

I received Guinness World record rules and there is no time limit nor max distance to reach top speed. However the record has to be attempted in both direction in less than an hour (and the average or the min speed reach is the record).

Can you turn the train around or must it run backwards at full speed?

Yes, we are planning to use commercial solar panel.
The surface is 30m x 3m = 90m^2 of solar panels.

I estimate your solar array could produce ~18kWp at noon in the Bay Area if you're using cells with 20% efficiency (see SF Solar Map here).

No storage, the solar energy should be used directly.

I guess that might be possible if you are only running for an hour around noon on a clear day.


Have you considered fitting a Tesla Model S with track wheels and towing the solar array? I'm not sure whether this would qualify as a 'train' but it might meet your aerodynamic requirements

 

[Duncan]

Hi
The idea of using the rails to get a very low rolling friction is a good one

I suggest you aim for a much lower weight - 600 Kg rather than 3000 Kg
A lightweight frame as optimised as possible and as aerodynamic as possible

You wants some rails that are at the best possible orientation to the sun - and then design your vehicle around that

The most difficult aero will be around the wheels and the bottom - as th wheels don't have to steer the body and wheels must be as close as possible

All of these are much much more important than motor efficiency

Also it is motor efficiency at full speed that is important - a lower efficiency at lower speed is OK as you won't need as much power at lower speeds

 

[brian_]

The total weight should be around 3 tons (6,000 pounds).
We may have to add more weight, since more weight is actually desirable for a locomotive (more weight = more traction).

That only makes sense if the locomotive is pulling something... and this is not.

 

[brian_]

No storage, the solar energy should be used directly.

This makes sense for this type of challenge: with storage, it would be difficult to prove that at top speed the solar panel was providing all of the power being used.

Without storage, acceleration is going to be slow, since the solar panel output becomes the peak power limit, rather than just the average power limit. On the other hand, there's no big rush as long as you have enough track, so this might not be a problem.

 

[brian_]

The total weight should be around 3 tons (6,000 pounds).
...
Yes, we are planning to use commercial solar panel.
The surface is 30m x 3m = 90m^2 of solar panels.

Why has a specific size already been chosen? Is this the result of some design optimization (balancing solar power and drag), or just an arbitrary guess?

A 30 metre long array implies a 30 metre long vehicle... which seems too long for a single unit, and a multi-unit vehicle would add substantial complication.

 

[brian_]

The Tesla Model S weighs just under 2.5 US tons and iirc requires ~15kW to cruise at 60MPH. Obviously the Tesla is extremely aerodynamic and even if you make your train as 'slippery' you'll need more than 15kW to break the speed record.

Maybe not...

That Tesla runs on tires, not steel wheels on steel track; the rail vehicle will have substantially lower rolling drag.

Aerodynamics are also interesting. The rail vehicle should have much lower frontal area than a large sedan, and the length-to-area relationship is favourable, but the sheer bulk of this 30-metre-long thing will be an aerodynamic challenge in anything but zero crosswind and straight track.

 

[mfucci]

Thanks for the interest on this project!

I will try to answer everyone.

now this obviously shows your lack of basic knowledge for mechanical engineering. more weight = more traction?!?!? i guess it's technically true, but then, what would you be doing with the traction that you get? how are you going to utilize the traction to your advantage? have you thought through everything??

Here is the wikipedia article on train traction.
Steel on steel is slippery. If there is not enough weight on motorized wheels, 15kW will create too much torque, even close to top speed.
In our current design, only one axle out of four is motorized with about one ton of weight on it. We will probably use sanding or add weight on the motorized wheels to increase traction.

More importantly, you intend to propel your 30m, 3 ton train to 65mph using only 15kW?? seriously? You probably did your engineering weight/aerodynamic calculations and came up with that number (at least I hope you did), but what about acceleration? how long are you gonna take to get to 65mph with 15kW?do you have like 5 minutes of time and track to spare to go there?

As long as we don't take more than 10 minutes to reach top speed, we should enough time to do two attempts under one hour and we will have about 10 kilometers of tracks. Sure, we would have prefer to accelerate faster so we are stil trying to not be stupid on the weight...

Anyway, most motors are about 85-95% efficient, and controllers are about 95-97% efficient, so unless you're running at the absolute peak efficiency all the time (which is not possible in the real-world), you won't be making your 92% efficiency. And this doesn't include the efficiency of the batteries, drivetrain etc. I hope you did your homework on those too.

Only efficiency at to top speed matter (as Duncan pointed out), so yes I hope to be in the 92% efficiency range... If you have some experience on a specific engine, I would be happy to know which efficiency you reach under some specific conditions.

ok, so this probably answers my question about you doing engineering weight/aerodynamic calculations. I think you probably did it, that's why you said no storage and solar energy will be used directly. You need to figure why you need storage, which will then answer the part on why using 15kW to go 65 mph is not the best of ideas.

We have no choice. To qualify for this record, no energy storage is allowed.

The Tesla Model S weighs just under 2.5 US tons and iirc requires ~15kW to cruise at 60MPH. Obviously the Tesla is extremely aerodynamic and even if you make your train as 'slippery' you'll need more than 15kW to break the speed record.

I am interested in knowing where did you find this info (but seems right!).
I chatted with a Telsa aerodynamic engineer about this project and here is what he said: "If you don't have styling design constraints, you could get the Cd of a train down to 0.15 I would guess."
Most cars are not designed to be aerodynamic, aesthetic and handling characteristic are more important. Here is a car with a 0.15 Cd.

Per this page, Telsa S has a Cd of 0.24, slightly better than a Prius.
We are shooting for Cd of 0.15. Even if we don't reach it, we should be able to compensate by make the train longer since parallel drag is much lower than front drag and longer will give us more power.
An a bicycle powered by a man can reach 83mph with less than 1kW.

Can you turn the train around or must it run backwards at full speed?

Either are fine, however aerodynamic is not symmetric, so we are planning to turn around the train. There are railroad loops that would flip around a train in our area like this one. However, we are not planning on being able to drive on curves since the axles will be fixed and far apart (no bogies, so it can only drive on straight tracks). And it is difficult to secure those specific tracks. We will use cranes to lift and turn the train.

I estimate your solar array could produce ~18kWp at noon in the Bay Area if you're using cells with 20% efficiency (see SF Solar Map here).

I have a similar number: 16kW
Here is my estimation:
Solar energy @90deg/clear sky: 1025W/m2
Solar panels: 56 x Sunpower X22-360 22.2% efficiency at 25C (the best commercial panels on the market), 1.63m2 each
Efficiency at 35C: 19.3% (0.29% lost per degree above 25C)
Attenuation due to dust / humidity: 7% (20% humidity, 50km visibility)
Sun altitude 75deg (May / Jun): 3.5% attenuation

Have you considered fitting a Tesla Model S with track wheels and towing the solar array? I'm not sure whether this would qualify as a 'train' but it might meet your aerodynamic requirements

We actually considered that! Car wheel distance and train wheel distance are pretty close, so it is possible to have car running a train track with few tweaks!
However, the Telsa S didn't fit in our budget

The idea of using the rails to get a very low rolling friction is a good one

Yes, rails give advantages. Rolling friction is one (although only 9% of the energy is lost in rolling friction, per my calculation), but also carrying heavy weights (not need to optimize weight with custom solar panels) and no problem for a long structure.

I suggest you aim for a much lower weight - 600 Kg rather than 3000 Kg
A lightweight frame as optimised as possible and as aerodynamic as possible

It is tricky to reduce significantly the weight: the solar panels alone will weight 1000 Kg and we will need steel beams to have a 10 meters spans between the axles... And at the same time, we are very limit in traction...

You wants some rails that are at the best possible orientation to the sun - and then design your vehicle around that

This was one of the ideas, but not to maximize power.
In May / June, sun altitude is around 75 deg in California, which only reduces the power received on an horizontal surface by 3.5%.
However, solar panels are very sensitive to the temperature: efficiency drops with height. So putting the solar panels at an angle would have allowed us to attempt the record much earlier in the year, when the air is cooler which would have cooled down the solar panels to keep a better efficiency.
We ruled out this idea, since this adds more constraints to the design and does seem to be needed to beat the record...

.

The most difficult aero will be around the wheels and the bottom - as th wheels don't have to steer the body and wheels must be as close as possible

Right, here are some examples of aerodynamic train designs.

All of these are much much more important than motor efficiency

Sure! But I am the one in charge of finding the right motor, so even a few % of efficiency will reduce top speed by a few mph !

That only makes sense if the locomotive is pulling something... and this is not.

Actually, it is! To qualify as train, this should have at least two vehicles.
Here is the train definition I found online: "a series of railroad cars moved as a unit by a locomotive or by integral motors."
We are planning on a 20 meters locomotive pulling a 10 meter car.
The locomotive by itself can attempt the solar-powered vehicle speed record (one driver is enough to qualify as a vehicle), but two vehicles are required to be a train. There is no solar-powered train speed record so we are sure to get this one regardless our speed!

Why has a specific size already been chosen? Is this the result of some design optimization (balancing solar power and drag), or just an arbitrary guess?

I made a spreadsheet that computes max speed with length as an input.
Above 20 meters, extra length only adds a few extra mph.
We decide to build a 20 meters motorized car and a 10 meter car, attached together to be a train. The 20 meters car by itself should be able to beat the record, but won't qualify as a train. The two cars couple together may have additional drag, so may be slower than only the 20 meters car but will be a train.

Aerodynamics are also interesting. The rail vehicle should have much lower frontal area than a large sedan, and the length-to-area relationship is favourable, but the sheer bulk of this 30-metre-long thing will be an aerodynamic challenge in anything but zero crosswind and straight track.

You are correct, the front drag is much lower than the drag created by skin friction of the air flow parallel to the body of the train.
Fortunately, flat surfaces are much easier to build than a fancy aerodynamic frontal area!

 

[Bird of Prey]

What are you gonna do about the motor drive system ? If you don't have the money they will make up every reason in the book for not helping you . Just look at me and the AC motors I wanted to use . Did you see if a train company would sponsor you after a test run ? Could you use capacitors to keep the power up ? Not for a battery but a boost ? Hope it works .

 

[brian_]

I assume with responses by the original poster to most of the comments, no followup for over two months, and not even a login for almost two months, that this idea has been abandoned.

What are you gonna do about the motor drive system ? If you don't have the money they will make up every reason in the book for not helping you .

Who are "they"? There would be little motivation for any motor manufacturer to put resources into this project, just as it doesn't make sense for manufacturers to support special requests for single-vehicle projects not leading to sales of significant quantity.

I think it is reasonable to assume that some sort of regular production motor or motors will be used.

Did you see if a train company would sponsor you after a test run ?

After a test run? Isn't support from a sponsor needed to develop and build the vehicle, which must happen largely before the first possible test run?

Could you use capacitors to keep the power up ? Not for a battery but a boost ?

That would still be storage.

 

[Bird of Prey]

1. Sevcon an Kelly both have programs to run motors without the encoder . but none of there distributors will help . And yes , I called more then 5 , and 2 sold Sevcon an Kelly .
2.You should show a working model or a good looking model if your not G.E , or Chip Foose or Amtrack .
3. There are capacitors in all controllers and amplifiers .
But that may all be a mute point . He could be getting money , parts , and a place to work out of . I didn't come on-line when I was pulling my motor and wiring my car . And doing overtime to pay for parts kicks my on-line .

 

[brian_]

3. There are capacitors in all controllers and amplifiers .

True, and there's no way to avoid that, but the energy storage would not be significant and whatever the rules for this record attempt, there is presumably some provision for this issue. Some capacitance in the controller is different from a separate and substantial energy buffer. "No energy storage" would rationally mean no battery or big capacitor in the panel to motor power path.