Boxster 986 Aerodynamics @180km/hr

[Nine8Six]

^^^ There... some new ideas for sweet t-shirts lollll

Originally Posted by rs10

Very cool thread!

I even I registered for the forum because of it, and I don’t even hava a 986 (yet).

If I may ask a silly question, is it correct that the software – or at least the 986 model, doesn’t take into account the air passing through the radiators, nor what happens in the wheel wells when the wheels are turning?

And in addition to showing what the air is doing, can your software estimate the drag or drag coefficient? And the lift over each axle? It would be quite interesting to see how close that is to reported figures. (Since lift figures are usually at 200 kmph, it would help to do it at that speed, though one can make a pretty good estimate based on the lift at 180.)

I’m sure someone on the forum can tell you the reported figures for the 986. I can at least tell you them for the 996.2: 24 kg of lift at the front, 6 at the rear. And yes, it would be great if you could put that nice 996 you showed in an earlier post into your virtual wind tunnel!

Please also add me to the list of people who’d love to see what a duck tail does. I’m interested in putting an immitation of the 997 Sport Classic on my car (a 996.2), and it’s one of the great mysteries of the Porsche forum universe what effect they have, if any.

Thanks!

I've seen your note and thanks! I'll get back to you once you've introduced yourself bud. Let us know where you are from, what you do! If student what uni ect etc. I'll be happy to get back on the subject and pass on all the Intel I got back to you

In a meantime I am doing a whole new kind of CFD Analysis. My wife (aka-therapist) suggested below:

[rs10]

Originally Posted by Nine8Six

I've seen your note and thanks! I'll get back to you once you've introduced yourself bud. Let us know where you are from, what you do! If student what uni ect etc. I'll be happy to get back on the subject and pass on all the Intel I got back to you

Sorry for the long radio silence! I got some double-nasty cold or flue, and with the wife sick too, I still have to take care of the kids, so not a lot of time for things Porche related :-(.

Anyway, I’m happy to say hi and a few words about myself. I’m new to this forum as I don’t actually have a 986, though I do have the good fortune of driving one a couple times a year. I’m not quite so new to Porsche forums though, and if you’ve seen my user name on another forum, it’s probably me. My PCar is a 996 Targa. And one of the reasons this thread really caught my attention is that I spend a fair amount of time driving it on the autobahn, so before messing with the aero on my car, I’d kind of like to have an idea what it is going to do.

And Nine8Six, thanks again for the very cool thread!

[Nine8Six]

Originally Posted by rs10

And Nine8Six, thanks again for the very cool thread!

No worries for the silence and hope you and your second-half get better soon. Apologies for forcing an introduction from you btw (forum rules, not mine!). So welcome to 986forum bud, great place and great people as you've already noticed I'm sure!

You are welcome for those 'genreric' air flow visuals! I plan on doing the 996 but will post using my RL account under the same alias. I'll shoot you a PM to let you know where to find those. As for messing-up with aero; I think it is fair to say that Porsche have carried out this work and have designed the car to safety limits already Any modification can be costly if not validated indeed. So best to stick to what the pro (Porsche) recommends loll

E.g. in another coupled-CFD-vs-structural analisys that I've carried, i.e. by exporting node-to-node pressure/force from one mesh to another, I was able to carry a static (down)force analisys using the nastran SOL101 solver for each independent wheels (using a pressure plate/sensor under each wheel). This is how I've noticed that my rear-end was in-fact lifting at speed greater than 80km. Ran this at different velocity increments to then acknowledge that 120km/h is pretty much the 'safety limit' for my current set-up. Go Go eBay.com!!!

Additionally, and to make my simulation even more complex (bringing it to its knees) I've ran the same CFD but using shifting winds by offsetting the car in an angular X/Y position (15degree) to get a whole new set of data fields. Once ran transient I've noticed force/damping waves in the plots. The bounce was quite pronounced in the X plane (left/right dancing) but nearly invisible in the Z (up/down). Can't imagine what it would feel like cornering my car at 160km/h at my International circuit here (some fast corners on our SIC). Hope the software engineeres of the package I'm using got it all wrong LOL

Fascinating model/solvers; all part of the training that I have to go through. Once we've done enough and graduated this course, and clearly understand the limitiations of the package we have here, we'll start to offer this as a corporate PLM service.

I'd be pleased to help other automotive artists/designers and get "validation" carried out for their concepts. I do have access to pretty much any of the Porsche car's 'solid' models from a 3rd party source, and/or can easily reverse-engineer their design for that matter.

Lift & drag, independantly set down-force points, air-flow, compressed air sources, validating structural fatigue on mounts (e.g wing brakets, mounting points, etc). Just to name a few... We are now geared up to carry those analisys if needed

[rs10]

Originally Posted by Nine8Six

E.g. in another coupled-CFD-vs-structural analisys that I've carried, i.e. by exporting node-to-node pressure/force from one mesh to another, I was able to carry a static (down)force analisys using the nastran SOL101 solver for each independent wheels (using a pressure plate/sensor under each wheel). This is how I've noticed that my rear-end was in-fact lifting at speed greater than 80km. Ran this at different velocity increments to then acknowledge that 120km/h is pretty much the 'safety limit' for my current set-up. Go Go eBay.com!!!

Additionally, and to make my simulation even more complex (bringing it to its knees) I've ran the same CFD but using shifting winds by offsetting the car in an angular X/Y position (15degree) to get a whole new set of data fields. Once ran transient I've noticed force/damping waves in the plots. The bounce was quite pronounced in the X plane (left/right dancing) but nearly invisible in the Z (up/down). Can't imagine what it would feel like cornering my car at 160km/h at my International circuit here (some fast corners on our SIC). Hope the software engineeres of the package I'm using got it all wrong LOL

Sounds like you have some amazing tools at your disposal! Not that I can follow every detail of course, and I’m particularly puzzled about one thing.

I understand you are doing virtual aero modeling/testing using computational fluid dynamics. But you also mentioned putting a pressure plate/sensor under each wheel. You also have access to a windtunnel? Or … ?

Also, the 15 degree test at 160kmph you mentioned sounds like driving 160 with a 40+ kmph sidewind. That’s rather quick, so no wonder you got a left/right bounce! But I wouldn’t worry about experiencing anything like that when cornering at 160, as what is happening in the air should be quite differernt, no? For instance, suppose that on a 160 kmph corner you cross a line on the track perpendicular to your direction of travel. If your car is cornering neutrally, then the air that passes exactly over the center of your car when the front wheels cross the line will also pass exactly over the center of your car when the rear wheels cross it (Well, ignoring how the air might get pushed to the side or forwards by the car). That’s quite different from a side wind. (And also rather different from going straight. As the front and rear of the car pass the same line, they will be further to the outside. And as the middle passes it, it will be further to the inside.)

Also, even if what happens in the air were the same as with a side wind, 15 degrees is quite a lot. Suppose it was exactly like a side wind (with I can only imagine happening if you have four wheel steering such that the car stays pointing in the same direction, but it moves 15 degrees to the right or left). The wheels would have to be turned at least 15 degrees for that to happen, so your steering wheel would have to be turned, I don’t know, 240+ degrees? I would definitely worry about that at 160!

[Nine8Six]

Originally Posted by rs10

Sounds like you have some amazing tools at your disposal! Not that I can follow every detail of course, and I’m particularly puzzled about one thing.

I understand you are doing virtual aero modeling/testing using computational fluid dynamics. But you also mentioned putting a pressure plate/sensor under each wheel. You also have access to a windtunnel? Or … ?

Also, the 15 degree test at 160kmph you mentioned sounds like driving 160 with a 40+ kmph sidewind. That’s rather quick, so no wonder you got a left/right bounce! But I wouldn’t worry about experiencing anything like that when cornering at 160, as what is happening in the air should be quite differernt, no? For instance, suppose that on a 160 kmph corner you cross a line on the track perpendicular to your direction of travel. If your car is cornering neutrally, then the air that passes exactly over the center of your car when the front wheels cross the line will also pass exactly over the center of your car when the rear wheels cross it (Well, ignoring how the air might get pushed to the side forwards by the car). That’s quite different from a side wind. (And also rather different from going straight. As the front and rear of the car pass the same line, they will be further to the outside. And as the middle passes it, it will be further to the inside.)

Also, even if what happens in the air were the same as with a side wind, 15 degrees is quite a lot. Suppose it was exactly like a side wind (with I can only imagine happening if you have four wheel steering such that the car stays pointing in the same direction, but it moves 15 degrees to the right or left). The wheels would have to be turned at least 15 degrees for that to happen, so your steering wheel would have to be turned, I don’t know, 240+ degrees? I would definitely worry about that at 160!

Admire your knowledge; Clearly there are two kind of people. Some that are clueless and need CFD tools to aquire understanding and knowledge, and some cool others who can compute nearly all of this stuff out of the head. Fascinating thread indeed!

No 'physical' wind tunnel here, you asked.

That virtual wind tunnel/CFD data (results) was exported into spreadsheets then called back into a totally separate structural analisys. That cfd data can be re-used for e.g. excitations or forced displacements if you prefer. Post structural analisys is meant at solving mostly displacement, force and strains/stress problems on the car componets.

So... set evironment; gravity, humidity, temp, and you can predict the behaviour of lets say; your suspension's preload/compression, damping, the force generated on any brackets, your mirrors' mouting points... stuff like that. You can also predict response dynamics (modal) such as vibrations onto body panels (accoustics) and some other more adv nonlinear problems eg. elasticity, buckling, laminates/ply; how much is my 5mm thick cheap fiberglass front bumper is bending at 200miles an hour (seriously lol).

All started from a 180km/hrs beaten up air fluid data mapped node-by-nodes onto a Boxster car! My training folder is now at 30+ gigabytes of raw *.dat text files bud

Learning the fun way! Chk back in two years... I'll be the official 986Forum Formula 1 Team Aerodymics - adviser

[rs10]

Originally Posted by Nine8Six

You are welcome for those 'genreric' air flow visuals! I plan on doing the 996 but will post using my RL account under the same alias. I'll shoot you a PM to let you know where to find those. As for messing-up with aero; I think it is fair to say that Porsche have carried out this work and have designed the car to safety limits already Any modification can be costly if not validated indeed. So best to stick to what the pro (Porsche) recommends loll

If you’re also planning on doing something 997 related, well, I’ve done my homework on the Sport Classic and I have a suggestion for you.

The Sport Classic is the only 996 or 997 with a duck tail – on other cars it wasn’t even an option. As far as I can tell, the closest equivalents to the 997.2 Sport Classic are the 997.2 C4 GTS and the 997.2 C4S with the X51 powerkit. All three are wide bodies, and will have larger air intakes in the front bumper than normal Carreras. (The next closest match is the 997.2 C4S without the powerkit, though not sure if the bumper, radiators, etc. are the same.)

The C4 GTS has a Cd of 0.30, versus the Sport Classic’s 0.32. It’s easy to blame the duck tail, but it could also be the double bubble roof or the aftermarket looking side skirts. And I’m not sure the bumper is the same either. It would be very interesting to see what your software shows about both drag (where the difference is) and lift.

By the way, the GTS was available with and without a GT3 style wing (which was probably an option on C4Ss as well). Comparing eiher version to the Sport Classic would be hugely interesting. And if you will be modelling enough cars to do all three rear ends (normal, duck tailed, and winged), then you’d have all the rear aero answers!

Anyway, much though I’d love to see the aero analysis of my car, as you’ve probably guessed, I’d give my vote to the Sport Classic. And I’m betting I’m not the only 996 (or 997) owner who would do the same.

[Robert986]

I guess I´m not the only one who isn´t too pleased with the visuals of the original wing. From a visual standpoint I think this one is the best I´ve seen (my personal preference): http://pictures.topspeed.com/IMG/crop/201012/first-generation-por-4_1600x0w.jpg

Even if I still would like to modify it, I think it has the right thing going for it. Anyway, since I´d except the Porsche guys to have made their homework I´d hesitate to make any changes without some fluid-dynamic calculations to back it up. Great to see this initiative from Nine8Six, I´m hoping for some more conclusions regarding the rear wing vs aftermarket stuff.. My dream product would actually be a carbon fibre rear hatch with included (good looking) wing with a mechanical raise-function like the one on the Turbo.. Unfortunately I don´t think I have the ability to produce it myself, at least not the time..

Again all credits to Nine8Six, lifting the forum to another level