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.
rossnzwpi
0
Joined: Sat Dec 24, 2016 1:33 am

Re: Open Source Racecar

Post by rossnzwpi » Sat Dec 24, 2016 3:00 am

excuse the quick posts - but I need 5 before I can send messages :)

rossnzwpi
0
Joined: Sat Dec 24, 2016 1:33 am

Re: Open Source Racecar

Post by rossnzwpi » Sat Dec 24, 2016 3:02 am

Your project aligns with my interest in developing a sort of Locost mid engine sports car (suitable for road and occasional track day use). Like yours it would use the transverse Alfa V6 and transaxle. It would also have a multi-tubular chassis. I'm certainly not an engineer though so would be looking to adapt an existing sports car suspension system or find open source plans.
Cheers
Ross

Tim.Wright
444
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Jan 22, 2017 5:44 pm

rossnzwpi wrote:Hi Tim, I've joined the forum to follow your open source project. Have you continued to develop it?
cheers
Ross (in NZ)
Hi Ross, sorry I let this reply slip.

In short yes I'm still working on it but it's moving very slowly due to heavy work commitments.

I'm wanting to do a few simulation experiments modifying the suspension kinematics to solve a few curiosities I have regarding things like roll centre placement, ackermann and other similar intellectual masturbations. I'm hoping to finish the roll centre study in the next couple of months which I will post here once it's done (I'm actually preparing a paper for this research work).

If you make a similar thread regarding your project I'd be happy to follow it. How much of the alfa engine are you going to use? Just the main stuff like block, crank and internals? Or also all the auxiliaries? That engine is DAMN heavy. I'm using partly for novely value (the busso engines sound amazing and make decent power) and partly because I scored the engine + gearbox for free. If I had to make an unemotional logical choice I'd take something like a Honda S2000 engine which makes more power and weighs waaaaaay less.

But I figured I'm a logical man during business hours at work. Outside of that I'm quite happy to mechanically compromise a vehicle and it's handling dynamics in order to have a cool sounding engine because fùckit.

Tim
Not the engineer at Force India

etusch
92
User avatar
Joined: Sun Feb 22, 2009 10:09 pm
Location: Turkey

Re: Open Source Racecar

Post by etusch » Sat Mar 04, 2017 12:49 pm

spacer wrote: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?
Agree. You must choose 2.4 or 2.0 type R engine or something like that. Because for example 2.4 Lt engine also near to 200 Hp but more Light weight. I think this engine will give you better performance because of weight advantage.

garyjpaterson
21
User avatar
Joined: Tue Oct 25, 2016 11:59 am

Re: Open Source Racecar

Post by garyjpaterson » Thu Jun 01, 2017 4:31 pm

Just found this thread, skimmed through it looking at the pretty pictures, then been so impressed I had to go back and read the whole lot from the start..!

Whilst I find the whole thing incredibly interesting, I can't pretend a lot (most) of it doesn't go straight over my head - it does! Huge respect to you and the many other guys who are hugely knowledgeable in these fields, massively impressive.

So thanks for this thread - really enjoyable to read, educational too!

Just wondering though, are those line drawings on the previous page from 2015 the most up to date? I'm tempted to have a stab at the styling (if I find some time).
And if I was able to come up with something worthwhile, I'd also be keen to implement it into a driving simulator as a bit of fun, complete with accurately represented physics (though the interconnected suspension is beyond consumer driving sims...). Would be cool to see it in motion virtually anyway :)

Tim.Wright
444
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Thu Jun 01, 2017 5:42 pm

garyjpaterson wrote:
Thu Jun 01, 2017 4:31 pm
And if I was able to come up with something worthwhile, I'd also be keen to implement it into a driving simulator as a bit of fun, complete with accurately represented physics (though the interconnected suspension is beyond consumer driving sims...). Would be cool to see it in motion virtually anyway :)
This would be interesting - I work with a couple of driving simulators and have been meaning to setup a model but I've had such little time available in the last year. I'm particularly interested in seeing if I can come up with a way to manage the transient instabilities which are inherent in vehicles with so much rear weight bias. Then I've got a shopping list of things I want to investigate from roll centres to ackermann. I've actually written a parametric K&C curve generator and want to use it to do some subjective driveability studies in the simulator.

What kind of simulator do you use? What vehicle model is behind it?

Honestly I've done very little work on the whole project in the last 1-2 years due to work commitments. I'm hoping to pick it up again during this year - perhaps after summer if work slows down a bit.

I think I changed something drastically in the frame around 6 months ago so I don't think the line drawings are accurate. I will see if I can make something more updated. Someone at work made a shooting brake concept and it looks awesome so I'm quite set on that idea. Should give me storage space for beer in the back which you otherwise wouldn't have for a coupè.
Not the engineer at Force India

garyjpaterson
21
User avatar
Joined: Tue Oct 25, 2016 11:59 am

Re: Open Source Racecar

Post by garyjpaterson » Fri Jun 02, 2017 10:54 am

Tim.Wright wrote:
Thu Jun 01, 2017 5:42 pm
This would be interesting - I work with a couple of driving simulators and have been meaning to setup a model but I've had such little time available in the last year. I'm particularly interested in seeing if I can come up with a way to manage the transient instabilities which are inherent in vehicles with so much rear weight bias. Then I've got a shopping list of things I want to investigate from roll centres to ackermann. I've actually written a parametric K&C curve generator and want to use it to do some subjective driveability studies in the simulator.

What kind of simulator do you use? What vehicle model is behind it?

Honestly I've done very little work on the whole project in the last 1-2 years due to work commitments. I'm hoping to pick it up again during this year - perhaps after summer if work slows down a bit.

I think I changed something drastically in the frame around 6 months ago so I don't think the line drawings are accurate. I will see if I can make something more updated. Someone at work made a shooting brake concept and it looks awesome so I'm quite set on that idea. Should give me storage space for beer in the back which you otherwise wouldn't have for a coupè.
Nothing special, probably nothing compared to what you're used to - Just Assetto Corsa. I use rF2 aswell but I don't have any experience creating anything with it. In AC its very simple, the only problems I usually have is just due to the lack of real data, which obviously isn't a problem in your case.
Its great fun experimenting with the suspension geometry and seeing/feeling the effects, though my experiments are mostly based in trial error, your would no doubt have some real theory behind it!

I'm just about to start modelling and implementing a FSAE car to AC in the next few weeks, looking forward to starting a project with proper data available from the start.

And yes, shooting brakes are universally awesome, I think that's a given :)

AngusF1
7
Joined: Sun Aug 13, 2017 9:54 am

Re: Open Source Racecar

Post by AngusF1 » Sat Nov 03, 2018 1:53 am

Hi Tim, were you able to make any further progress on this project? I've read the thread a few times with interest over the last couple of years.

Some thoughts on the weight distribution:
This is caused by two factors. One, your transverse engine is practically between the rear wheels, and two, the seat positions are very far rearward - this is enabled by the transverse mounted engine and lack of large fuel tank. Why not either increase the car width, allowing you to push the driver forward so the pedal box is in between the front wheels, or increase the wheelbase by lengthening the passenger compartment and moving the seats forward by the same amount? I know normally increasing the wheelbase forward would send the distribution rearwards, but if you move the pilot along with it the net effect could be more frontwards. A third option could be to move the rear wheels further back, but I guess you're stuck because then you'd have to move the engine backwards, too. How much angle could you put on the driveshafts? If they can endure some angle, you could move the rear wheels back as much as possible. I think some combination of all the above would solve your problem.

Some thoughts on front tyre width:
You remark that front tyres of rear-engined sports cars are proportionally larger than an allocation purely based on weight distribution would suggest. Yes, I've noticed the same thing. I think the reason is that allocating purely by weight distribution % assumes that a weight of zero implies zero tyre width (ie, the minimum tyre size is zero). In reality, for anything approaching a normal car there is a minimum tyre width of something between 100 to 150mm.

To illustrate, consider a car with a 40-60 weight distribution and 245 width rears. Allocating purely according to weight distribution would result in fronts of only 163mm (never seen in reality for a car like that), but sizing assuming a minimum width of 150mm results in [maths: (245 - 150)*40/60 + 150)] a width of 213mm - possibly a bit wide but more reasonable. 205 might be a suitable tyre in reality, which actually falls out perfectly from assuming a minimum width of 125mm [245-125)*40/60 + 125]. That isn't a very physical explanation, but I think something like that is occuring. Probably something to do with the minimum feasible widths of radial tyres.

hexen
0
Joined: Sat Oct 26, 2019 8:17 am

Re: Open Source Racecar

Post by hexen » Sat Oct 26, 2019 8:55 am

Hello,

I see you made a CAD engine model from the Alfa V6, I have draw the front of my car, and I'm looking for the engine. I would like know if it's possible to have it.

Regards

Tim.Wright
444
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Nov 03, 2019 6:32 pm

I've put the engine model (with original FWD gearbox) in STP format here:
https://1drv.ms/u/s!AioEiFs0jfZSgWqTBy_ ... v?e=xacYy1

The alternator and AC compressor are not in the production locations but they are the production parts as I measured them.
Not the engineer at Force India

Tim.Wright
444
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sun Nov 03, 2019 6:46 pm

AngusF1 wrote:
Sat Nov 03, 2018 1:53 am
Hi Tim, were you able to make any further progress on this project? I've read the thread a few times with interest over the last couple of years.
No real progress recently. I'm without access to CAD and MBS software at the moment which restricts the ability to do things properly.
AngusF1 wrote:
Sat Nov 03, 2018 1:53 am
Some thoughts on the weight distribution:
This is caused by two factors. One, your transverse engine is practically between the rear wheels, and two, the seat positions are very far rearward - this is enabled by the transverse mounted engine and lack of large fuel tank. Why not either increase the car width, allowing you to push the driver forward so the pedal box is in between the front wheels, or increase the wheelbase by lengthening the passenger compartment and moving the seats forward by the same amount? I know normally increasing the wheelbase forward would send the distribution rearwards, but if you move the pilot along with it the net effect could be more frontwards. A third option could be to move the rear wheels further back, but I guess you're stuck because then you'd have to move the engine backwards, too. How much angle could you put on the driveshafts? If they can endure some angle, you could move the rear wheels back as much as possible. I think some combination of all the above would solve your problem.
Angles on the driveshafts are at their max that I'm comfortable with. A lot of their angle is eaten up already in the static alignment because I'm trying to keep the drivetrain as low as possible and the wheels as far back as possible.
AngusF1 wrote:
Sat Nov 03, 2018 1:53 am
Some thoughts on front tyre width:
You remark that front tyres of rear-engined sports cars are proportionally larger than an allocation purely based on weight distribution would suggest. Yes, I've noticed the same thing. I think the reason is that allocating purely by weight distribution % assumes that a weight of zero implies zero tyre width (ie, the minimum tyre size is zero). In reality, for anything approaching a normal car there is a minimum tyre width of something between 100 to 150mm.
That's seems reasonable but I've developed another theory over the years. More rearward weight distribution causes bigger delays between steering input and LatAcc/Yawrate response. When you have a rear-heavy vehicle, fitting larger front tyres it helps reduce these delays so that they will remain under the driver's perception thrshold of a "slow" handling response. Of course this comes at the cost of reduced understeer/stability.
Not the engineer at Force India

hexen
0
Joined: Sat Oct 26, 2019 8:17 am

Re: Open Source Racecar

Post by hexen » Sun Nov 03, 2019 8:13 pm

Tim.Wright wrote:
Sun Nov 03, 2019 6:32 pm
I've put the engine model (with original FWD gearbox) in STP format here:
https://1drv.ms/u/s!AioEiFs0jfZSgWqTBy_ ... v?e=xacYy1

The alternator and AC compressor are not in the production locations but they are the production parts as I measured them.
Thanks you

maunde
1
User avatar
Joined: Sat Nov 02, 2019 11:36 am

Re: Open Source Racecar

Post by maunde » Sat Nov 09, 2019 5:43 am

Hello Tim, I have recently joined the forum and have loved reading this thread. I am a mechanical engineering graduate from NZ and am wanting to pursue a career in aerodynamics - I have been modelling vehicles and aero simulation for some time now.

I would love the opportunity to model your car in CFD, would you be willing to collaborate on this? I can provide some simple aero mapping in order to find the Aerodynamic centre.

I have a question regarding your concepts for aero. If you have a very rearward COG, it is true that you would want a COP (aero - centre) slightly behind the COG. However, would you not want an element of downforce on your front wheels to reduce understeer? Albeit without pushing the COP forward of the COG in order to maintain yaw stability.
A kiwi looking to fly like McLaren.

Tim.Wright
444
User avatar
Joined: Fri Feb 13, 2009 5:29 am

Re: Open Source Racecar

Post by Tim.Wright » Sat Nov 09, 2019 12:41 pm

Hi Maunde,

Apart from the fact that I have done very little on the project for a couple of years now, I had anyway bailed on the idea of using aero as a fundamental part of the car for a few reasons:
  • Firstly, contrary to the topic title, the car would be primarily a road vehicle not a pure race car. So aero devices are going to be doing nothing but decreasing my fuel efficiency for 90% of it's life.
  • Secondly, for road based race cars, the amount of downforce that you can produce as a ratio to the vehicle weight (which is the ultimate number that determines the effectiveness of the aerodynamics) is very low such that at typical road car speeds it will be practically useless.
  • Lastly, given time and budget constraints for such a project, it's not really possible to do be doing the required wind tunnel or full scale aero tests to validate the CFD results. Such iterations would take years to setup and feed back into the design
I think for a vehicle like this (>60% weight on the rear axle), a rear wing will definately be required to increase the rear cornering stiffness at high speed for for stability. The transverse engine unfortunately doesn't leave enough room for a decent diffuser.

BTW, I'm not a fan of speaking about a single COP in the context of vehicle performance. This is because from the chassis point of view, the aero forces are not unique to a single point but act along a line which depending on the point alont the line of action, could be either in front or behind the COP.

A lot of people think that the stability from the aero having it's "COP behind the CG" is due to the "pulling" or "stabilising" effect that the COP has w.r.t. the vehicle CG. In my opinion the mechanism at play is much simpler in that a rearward biased downforce increases the vertical force and hence the cornering stiffness of the rear axle which increases understeer, static and dynamic stability.
Not the engineer at Force India

maunde
1
User avatar
Joined: Sat Nov 02, 2019 11:36 am

Re: Open Source Racecar

Post by maunde » Sat Nov 09, 2019 1:12 pm

Hey Tim, that all seems to make sense to me, however, given that you will already be having a large proportion of understeer due to your COG, would you not want to reduce this at low speeds?

All valid points as far as road-car practicality from what I understand.

May I ask what your experience has been like working in Europe and moving there from Aussie?
A kiwi looking to fly like McLaren.