Open Source Racecar

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Open Source Racecar

Post by Tim.Wright » Sat Sep 29, 2012 9:34 pm

For anyone who are interested I will do a bit of reporting on a project car I'm designing for myself.

Image

Since my FSAE days Ive been pretty keen on making something for myself. So a couple of years ago I started putting a few ideas together in my head. Now I have the beginnings of a CAD model.

The intention is put my learnings of vehicle dynamics of the last few years into practice. The expectation is that this will take a few more years itself to realise.

I will post as much technical information as I can as opposed to just pretty pictures. I'm happy to discuss technical issues that pop up too. Some healthy technical discussions on this will no doubt help me and be at least "entertaining" for people to read.

Some high level targets/specs/goals;
Mass: 900kg including driver
Mass dist: 39% front
Wheelbase: 2350mm
Track Front: 1600mm
Rear Track: 1600mm
Body Style: Closed wheels, closed cockpit
Engine: 3.2 or 3.0L Alfa Romeo V6 (approx 200PS)
Primary Structure: 4130 welded steel spaceframe
Suspension: Double wishbone front and rear, with adj antiroll bar. Undecided on pull/push-rod or direct acting.

You can probably guess from the specs that I'm not building a supercar. I'm targeting light weight, modest power with the focus on suspension and handling, because that is what most warms my noodle.

There are no promises of frequent updates regarding the design etc etc, but I will at least partake in any technical discussions that anyone start. Sometimes I go months without touching this.

Enjoy

Tim
Not the engineer at Force India

mep
34
User avatar
Joined: Sat Oct 11, 2003 2:48 pm
Location: Germany

Re: Open Source Racecar

Post by mep » Sat Sep 29, 2012 9:56 pm

Looks promising.
What are your plans for the gearbox?

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sat Sep 29, 2012 10:07 pm

This is the main reason I have chosen a FWD engine, so the gear box and diff is together in one nice unit. I'm not an engine man and I don't have the desire (or the money) to stuff about with rear mounted gearboxes, and adpter plates etc etc.

So I chose the biggest FWD engine I could find. The Alfa GTV and GT engine is also quite a beautiful looking engine too;
Image

I will buy an alfa, or at least the engine some time next year so I can start modelling it up bacuse I cant design anything at the rear until I know the size of the engine.

Tim
Not the engineer at Force India

spacer
8
Joined: Sun Nov 01, 2009 7:51 pm

Re: Open Source Racecar

Post by spacer » Sat Sep 29, 2012 10:12 pm

Big as in displacement/power or big as in reliable yet racing-pedigree v6?

You could take a look into the Honda K-series 2.4L unit. Recent engine, lots of potential and very easy to get to about 300BHP. But most important, probably lighter and easier to package. But then again, I don't know how much money the alfa unit will set you back...

Love these kind of things btw, been doing something similar in solidworks for myself. What cad system are you using?

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sat Sep 29, 2012 10:22 pm

Im using CATIA. Ive been using it at work for the last 4-5 years. I moved from Solidworks which I learnt in Aus to Catia which is used more in Europe where I live now.

Im somewhat open to the poweplant to be honest. When I meant big, I meant most power. I want the engine to be as plug and play as possible. I would rather not stuff about with non standard ECUs, tuning etc in the first instance.

I will look into the Honda, thanks for that suggestion. Im looking for something ideally with more than 200PS in the stock form and a few years old at least so it will be easier to find a complete engine second hand.. Any other suggestions are welcome.

Tim
Not the engineer at Force India

fausto cedros
0
User avatar
Joined: Sat Jan 30, 2010 9:22 am
Location: Brindisi, Italy

Re: Open Source Racecar

Post by fausto cedros » Sat Sep 29, 2012 10:51 pm

i'd say honda k20A. lightweight and powerful, but delivering btw 5500-8300 rpm. using it in my cn2
"Adding power makes you faster on the straights. Subtracting weight makes you faster everywhere" Anthony Bruce Colin Chapman

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sat Sep 29, 2012 10:58 pm

The first thing I did was set some targets based on cars already existing. Then I used this to set high level parameters such as wheelbase, track, mass and CG location. I found info on 4 cars with a bit of searching. Probably the closest to what Im trying to achieve is the Lotus (low weight, good handling).

Image

Finding info on the mass and distribution tool some searching, but was able to find something. auto-motor-und-sport.de has a massive database of cars they have tested which is very helpful.

First thing I did was set the CG and wheelbase. Not a lot of science in this. Just picked values in the range of the benchmarking cars. I chose a fairly rearward mass distribution which is typical of Porsche and Lotus. Now I have to try to stick to this target mass distribution as I design.

My parameters for wheelbase and track are similar to the KTM and Ariel Atom, though Lotus seems to have a higher wheelbase to track ratio to what I have chosen. So I'm thinking of reducing the track a little to bring it more in line with the lotus. Note, the crazy high speed cars have long wheelbases for more stability from yaw damping.

Image

Image

I've gone through and set a weight budget for all the major subsystems. This will be populated as the design progresses. Its a spreadsheet which calculates the total mass and CG location given masses of all the induvidual parts. High school maths but very useful.

Image
Not the engineer at Force India

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sat Sep 29, 2012 11:11 pm

Thanks for the suggestions on the engines. Im noting these down in my big excel sheet.

So as not to bore you all with tables and plots, I have been doing some CAD modelling too. Chassis is steel spaceframe and the windscreen I want to use is from an Ultima GTR. The screen is already E-marked and can be purchased alone because its a kit car.

Image

Image

Wheel and tyres have been sized too. I will explain the logic behind that another time.

Image

Have been playing around with a few bodywork ideas too

Image

Image
Not the engineer at Force India

mep
34
User avatar
Joined: Sat Oct 11, 2003 2:48 pm
Location: Germany

Re: Open Source Racecar

Post by mep » Sun Sep 30, 2012 12:29 am

Do you already know what tyres you are using and in best case do you have any data for them? Most likely not but you might be still able to base your weight distribution decisions on estimated grip conditions. This could be done based on tyre size or by doing some simple tests yourself.
Did you check by how much CoG height could by lowered by increasing wheelbase?
I would try hard to get CoG as low as possible.

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Sep 30, 2012 12:43 am

Tyres have been more or less chosen. I dont have any data except for trying to relate mass, distribution and engine power to tyre widths on already existing cars. I have a lot more charts on this which I will up load soon.

Tyres are;
Front: 195/45 R16
Rear: 245/40 R17

Either Michelin Pilot Sport 3 or Goodyear Eagle F1 are available in those sizes.

Yes, a low CG is important, but not as imporant as keeping the total weight to a minimum. At this stage of the CAD model its impossible to see how I can lower the CG by increasing the wheelbase. Generally a longer wheelbase will mean a heavier car.

Tim
Not the engineer at Force India

Jersey Tom
257
Joined: Mon May 29, 2006 7:49 pm
Location: Huntersville, NC

Re: Open Source Racecar

Post by Jersey Tom » Sun Sep 30, 2012 1:11 am

Are you concerned about the potential for excessive low-G oversteer with such a rearward weight distribution? What would the contingency be in that case? More tire size split?
Grip is a four letter word. All opinions are my own and not those of current or previous employers.

Nando
12
Joined: Sat Mar 10, 2012 1:30 am

Re: Open Source Racecar

Post by Nando » Sun Sep 30, 2012 2:29 am

Considering the relatively low weight of the car maybe a hayabusa engine with a turbo could do the trick?

Not sure how that would compare size-wise to a civic type r engine or alfa V6 but the fun factor once you get going should be pretty amazing.

Bigger bore, 1,6L 11,000rpm and some 300+ horses with a weight of 150lbs/68kg's.

http://www.h1v8.com/f/BUSA_layout_drawing.PDF
"Il Phenomeno" - The one they fear the most!

"2% of the world's population own 50% of the world's wealth."

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Sep 30, 2012 11:23 am

Jersey Tom wrote:Are you concerned about the potential for excessive low-G oversteer with such a rearward weight distribution? What would the contingency be in that case? More tire size split?
Thanks for chiming in Tom, to be honest I am not concerned about low-G understeer because I'm completely ignorant to it. So I understand better, when you say low G are you specifically referring to the linear range of the tyres, or to low steering angles? For example, if the tyres are quite linear to 0.6G, do you speak of the range 0-0.6G, or is there some effect which causes a change in balance at very low levels of lateral acceleration say 0-0.3? Its something I haven't seen before.

I'm familiar with characteristics which see the car's balance at a constant value in the linear range of the tyres, and then terminating in understeer at the limit (for road cars).

The mass distribution is rearward, but not out of the range of other vehicles. Since its impossible to come across tyre data for the tyres I want, I have done an investigation on tyre sizes using mass distribution information I have found from Auto Motor und Sport. For example here I've made two plots. the left one is relating tyre width distribution to mass distribution. The right one is showing the delta of tyre width dist to the mass dist. Here there is quite a distinct trend. The location of my car in these plots is the red point, and you can see, its very close to the regression line. Not very scientific, but the best I can do given the information available.

Image

On the left plot, I have seen a general, loose, trend which suggests a change of mass distribution of 1% requires a change in tyre width distribution of 0.168%. The right plot shows someething different. If it the difference of tyre_width_dist - mass_dist. Here there is a very strong relationship, but I dont really understand why. This is saying that at about 45%F mass dist, your tyre width distribution is also 45%F. As you move to a more rearward distribution, you obviously move to a more rearward tyre width distribution, but the front tyre increases more in terms of the relationship to the mass dist.

So its saying that the more rearward you mass distribution is, the manufacturers are "over-tyring" the front axle in comparison to the mass distribution. Though, I dont read this as meaning oversteer though. Im 95% sure all of these vehicles are understeering in the linear and limit regions, but it seems there is a requirement for more front tyre as you shift the mass backwards.

Interesting, and I dont have an explanation for it.

So JT, to answer your question better, I don't expect I will be struggling with oversteer given this mass dist and tyres size choice. In both of these plots, I am operating in the same region of Lotus and Porsche, both of whom use about 38-39% of mass distribution, and tyre their cars with a width distribution of +5-6% of that.

Another factor in the tyre width is the engine power. Without going into detail of engine maps, gear ratios etc etc etc, you can already pull out a trend which suggests (unsurprisingly) more power requires more rear tyre. In fact, I have seen that every 10PS of power requires 1.8mm of extra rear tyre width.

Image

So in my case, an estimation of 220PS tells me I need approx 245mm rear tyres.

So where does this leave me? With front and rear tyres sizes that may or may not be ok. But at least I can move forward now.

By following what the manufacturers are doing I know Im at least not going to end up with massive handling problems due to the tyres. Im also pretty sure that, barring anything stupid happening in the suspension, the car will have an understeering tendancy because this is how all manufacturers are designing their road cars these days.

I'm also aware that following the manufacturers I'm not guaranteed a brilliant hanlding car. Tyre selection is only one (important) part of the mind f***, I mean jigsaw.

Tim
Not the engineer at Force India

Tim.Wright
435
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Sep 30, 2012 11:27 am

Nando wrote:Considering the relatively low weight of the car maybe a hayabusa engine with a turbo could do the trick?

Not sure how that would compare size-wise to a civic type r engine or alfa V6 but the fun factor once you get going should be pretty amazing.

Bigger bore, 1,6L 11,000rpm and some 300+ horses with a weight of 150lbs/68kg's.

http://www.h1v8.com/f/BUSA_layout_drawing.PDF
The H1V8 is pure porn, but also massively expensive. I have ruled out bike engines because they need to be highly strung to get the power output I want. I'm only after about 220PS, nothing special but this would require a fair bit of tuning on a bike engine which I really can't be bothered to do.

The car engine for me, while being heavier, is a safer, cheaper option with less development work required on my side.

Tim
Not the engineer at Force India

marcush.
268
Joined: Tue Mar 09, 2004 3:55 pm

Re: Open Source Racecar

Post by marcush. » Sun Sep 30, 2012 6:28 pm

Tim.Wright wrote:Tyres have been more or less chosen. I dont have any data except for trying to relate mass, distribution and engine power to tyre widths on already existing cars. I have a lot more charts on this which I will up load soon.

Tyres are;
Front: 195/45 R16
Rear: 245/40 R17

Either Michelin Pilot Sport 3 or Goodyear Eagle F1 are available in those sizes.

Yes, a low CG is important, but not as imporant as keeping the total weight to a minimum. At this stage of the CAD model its impossible to see how I can lower the CG by increasing the wheelbase. Generally a longer wheelbase will mean a heavier car.

Tim

did you consider the Kumho Sports tyre range ?
I did use those for my Elise projects -you get a range of compounds the price is low and for trackday and road they are very useful ..