Lightweight AWD Sportscar

designerguy

10 mW
Joined
Jan 8, 2018
Messages
23
I’ve just started on the design and development of my own personal project car. I have been a long time reader of these forums, constantly inspired by all the collaborative development that happens.

Some background:

By trade I am an automotive exterior designer. I work at a one of the big 3. Last year I sold my personal car as I hadn’t used it in over a year, I had been walking to work. A week later I was told my job was moving and suddenly my commute got quite a bit longer. I started looking into buying another car, being a designer I wanted something special (read: AM V8 Vantage, Lotus Elise, Lotus Evora.) It’s winter here currently in Michigan, and dealers aren’t too keen on me test driving their sports cars in the snow. That’s when my brilliant (stupid) idea came to design and build my own car.

So I decided to jump head first off of the development and financial cliff.

The Idea:

600kg
180kw
Individual wheel motors with belt reduction
Low Cd (~0.2)
Low frontal area (~1.5 m^2)
Smallish battery pack with high charge rate (19kw LTO)

My bad influences:

Palatov Motorsports
Formula SAE
larsb motor remake
Every Lebowski controller thread
Farfle/Miles various motor projects

So far I have bought the batteries and motor stators to help characterise the rest of the car around it. I contacted Alexey from Revolt who sent 360mm of RV 160 stators. That will make 4 90mm stacks, one for each motor. This should be good for ~130 Nm per motor, with a 4:1 reduction ratio that’s 520Nm per wheel (1.2 g acceleration!)

Things to do from easiest to hardest:

Exterior
Chassis and suspension
Motor mechanical
Motor electrical
Motor Controllers
High amp BMS
High power battery charger
...
...
Make it all work

I will be making threads for the motor and controller seperately.
 
Will you be trying for regen or eabs braking, with that belt reduction?

Any concepts for the body yet, beyond the mentioned models?

Have you already worked out the registration / insurance path hurdles?
 
Registration is pretty easy in Michigan. If I follow basic FMVSS guidelines I should have no problem. Insurance will be another story but I cross that bridge when I get to it.

I do plan on using regen braking with the belt drive. LTO have an insane charge rate (8C) which translates to 152KW under braking, in other words close to 1g would be possible just with regen. I’m not sure how close I will get to that number, but I would like to get as close as possible.

No real sketches or models for the body design yet, that’s the part I’m most comfortable with, and least worried about. I’ll end up changing it right up until I cut the molds anyways :lol:
 
Got the batteries and stators in tonight. Made a quick model of the center battery box/ frame and ran a torsional moment through it. It gets about 10k lb-ft/deg just with just the center frame and no passenger cell!
 

Attachments

  • rivetedaluminumskin.JPG
    rivetedaluminumskin.JPG
    127.5 KB · Views: 4,441
Collin,

i'm looking forward to following this thread ! Most guys wouldn't even know how to start a body design / fab and you say that is the easiest part .

with those LTO batteries you will be able to charge that car from dead to full in 7.5 minutes ! your going to love that . what pack voltage are you going with ? have a large charger coming in next week , came with some LTO cells i purchased .Its big , i will check its rating .
mike
 
I'm currently working on getting some more accurate CFD software to test ideas before I show the whole design. I thought I would leave this teaser here for you, though.
 

Attachments

  • Image002.jpg
    Image002.jpg
    23.6 KB · Views: 4,441
  • Image003.jpg
    Image003.jpg
    26.1 KB · Views: 4,441
  • Image001.jpg
    Image001.jpg
    26.3 KB · Views: 4,441
Nice work! :D I always thought one of the best from the past to take inspiration from is the stratos.
Lancia-Stratos-rally-car_06-990x603.jpg


https://3bv8x43y68hc448rg43goku7yq-wpengine.netdna-ssl.com/media/2012/04/Lancia-Stratos-rally-car_06-990x603.jpg
 
i have the CFD software working a little better now. I have been running non-stop simulations, iterating to get a lower Cd. So far I'm at 0.20 for the Cd and 1.54 m^2 frontal area. That gives a drag area of 0.307 m^2; ~the same drag area as a bicyclist with drops! That's about 7kw at 30m/s (highway speed) including rolling resistance which means that a range of over 120 miles should be possible.

Some big contributors to this were the decision to use BMW i3 wheels and tires, and to move away from a flat underbody to a shaped underbody similar to Ecomarathon cars and Solar Challenge cars.
 
speedmd said:
Nice work! :D I always thought one of the best from the past to take inspiration from is the stratos.
Lancia-Stratos-rally-car_06-990x603.jpg


https://3bv8x43y68hc448rg43goku7yq-wpengine.netdna-ssl.com/media/2012/04/Lancia-Stratos-rally-car_06-990x603.jpg

Watch for the Geneva show next month.
The Stratos is being relaunched with a Ferrari drivetrain.
http://www.caradvice.com.au/621491/lancia-stratos-return-geneva/
 
The Stratos is pretty lovely.....

Here is the latest iteration. The Cd now stands at 0.181. It would be great if these numbers were accurate! 4.5 KW aero-load at 30m/s!

Now that the aero package is coming together I need to start on the motors to make sure I don't lose too much in iron losses at speed :D
 

Attachments

  • simulation004.JPG
    simulation004.JPG
    221.3 KB · Views: 5,332
wow, 0.18 is crazy low cd, I drive a mk1 insight and that has some of the best oem cd for a mass produced vehicle.

BTW like the fact you are interested in Lebowski threads, things should move along for the bobc Lebowski project now. Nice renderings btw.
 
I’ll be needing to make 8 controllers :lol:

I’m hoping to develop a nice compact 25kW powerstage for the SMD controller for insane ebike and light motorcycle use. Once again I’m learning as I go with all of this :lol:

To put the drag into perspective the first gen insight has a CdA of 0.56m^2This has a CdA of 0.28m^2.
 
like I said, I thought it was pretty insane cd and thus cda vals

8 controllers eh?

I am planning to make a few boards each evening once the kid and wife are tucked up in bed. Also got plans to build an opensource pick & place using my anet a8 so that should help with some of the tedium. Glad to have another Lebowski convert on board with lofty plans. Let me know what you need via pm and I shall endeavour to deliver. I wonder whether it will get Arlin's CRX a run for it money when its done :shock:
 
For the controller I’m looking at 3 MTI85W100GC similar to what you had posted in one of the threads. It would be nice to be able to buy MTI145WX100GD so I could actually make decent power though.
 
GOOD NEWS!

Stator losses are only about 50W per motor at 30m/s! (At least per the simulation) I suspect it will be a bit higher since the laminations are joined together via a stamped tab, shorting the stack. There will also be stator, bearing, windage, drivetrain, and controller losses to take into account. I can only hope to keeps them below 1kw together to get the total power at 30m/s under 7kW. (Assuming 1.5kW rolling resistance)
 
designerguy said:
GOOD NEWS!

Stator losses are only about 50W per motor at 30m/s! (At least per the simulation) I suspect it will be a bit higher since the laminations are joined together via a stamped tab, shorting the stack. There will also be stator, bearing, windage, drivetrain, and controller losses to take into account. I can only hope to keeps them below 1kw together to get the total power at 30m/s under 7kW. (Assuming 1.5kW rolling resistance)

If the lams are staked together at a location that has little to no flux density change as the motor operates, it will cause little to no added eddy/core loss to have them electrically connected.

You may find soon enough that putting less mass and bulk in the reduction stage and more mass and bulk into copper and iron in your stator makes an overall higher performance vehicle, especially so with respect to drag times.
 
liveforphysics said:
If the lams are staked together at a location that has little to no flux density change as the motor operates, it will cause little to no added eddy/core loss to have them electrically connected.

You may find soon enough that putting less mass and bulk in the reduction stage and more mass and bulk into copper and iron in your stator makes an overall higher performance vehicle, especially so with respect to drag times.

Absolutely. Eventually if budget permits I would like to move to a direct drive motor. The biggest issue was just finding stator laminations (YASA, EVO, and EMRAX would be a significant portion of my budget) as it stands I’m just using a single stage reduction to the wheels. Each motor is good for >140Nm, so even at 4:1 its >1g acceleration.

The stator is staked through the center of the teeth. I’m not sure how bad this will be for eddy currents...
 
I played with narrowing the tracks, rear track more so than front. Drag area is now 0.22 m^2. That's a decrease of 21% from the previous test. I want to play with how much I can reduce drag for maximum range.
 

Attachments

  • simulation006.JPG
    simulation006.JPG
    199.3 KB · Views: 5,116
Narrowing the rear axle length will affect your centre of gravity and make the car wobbly in straight line travel. That can be dealt with but it will be costly. The simplest solution is to keep the car on equal footing, front and back.

If you were an automotive exterior designer, surely you would have heard from your structural engineering buddies that the best location for the GC is in the dead centre of the car.

I also noticed you want the motors in the wheels. That too will spread your GC towards the outside of the car. Not a smart idea. Do not be fooled by solar cars, they do not care about the GC, only about efficiency. Motors in wheels will make steering rather heavy, which will require a top of the line hydraulic/electric steering. And you will still feel it. Plus, you intend to add gears to the wheels/motors... I suggest you call one of structural engineers, if you know any, who worked in the same automotive companies as you. Run your ideas by them and go with their advice.
 
Inducter said:
the best location for the GC is in the dead centre of the car.

I assume you mean in terms of front-back/left-right and not vertically?

There have been a lot of successful cars that haven't had perfect weight distribution. I wouldn't get too bogged down - it's important to be able to decide what is "good enough" to avoid the tyranny of perfection and never completing a design (or making in infeasible expensive or complicated).

It's hard to believe that heavy wheels would make a noticeable difference to steering weight as the acceleration of the steering is quite low, but this is only intuition. Individual wheel motors could still be used but inboard with a simple driveshaft.

For actual in-wheel motors with gearing there is a Formula Student car that utilised this approach. I can't think of any other car that does off the top of my head.
 
CG perfectly centered will most likely end up with a loose handling car. It will most likely be more effected by cross winds. Best to err with a bit forward bias on the mass IMO. Agree, weight generally centrally located and as low as practicable, will give good general results. Battery layout can lend itself to some great options to dial this in. You could move more mass to the extremities making it more stable with regard to direction changes or inboard to quicken this. Dial it in for wide open desert racing or tight twisty rallies just by moving some battery modules.
 
Back
Top