Jump to content
LegacyGT.com

4th gen aero


boxkita

Recommended Posts

sent you a pm.

 

At one point, I was a fair-to-middling programmer of obscure languages. Never learned python, but for this, I'm willing to learn.

 

The 2d seems a bit off to me. The front nose actually has a bit of hook under the front bumper, while the rear is really flat without the knob sticking out.

 

Can you share the 2d model you created? Or it in a special cad software only format?

 

The model is in a Siemens NX .prt file - not sure if other CAD packages could open that. I could save it out as one of the open source formats, but would lose all the parametric aspects of it so we'd have to handle any changes with cuts/additions to the baseline geometry.

Link to comment
Share on other sites

smoothing rear window to make it more flat, like audi does with their wagons? Madrig was willing to make a full tailgate if I could get the design to him.

 

Figuring out how to cure the suction effect at the rear? The dragon of drag :)

 

Mounting a wing in the middle of the curl off the rear? Supposedly that's why f1 wings sit where they do... to gather air from over & under the car and redirect.

 

Sounds like a few single point runs to me!

 

I don't think there's any curing to the suction short of adding a massive faring, but I bet we could reduce the effect a bit.

 

Would mounting the wing like you said be any different than a waist spoiler? One of my shopmates has a racebrick FXT, had a conversation with him a while back about waist spoilers. He mentioned that they'll completely eliminate the snow/dust fouling that we've been talking about on wagon hatches, which to me seems to imply they'd have a pretty big effect on the recirculations back there.

Link to comment
Share on other sites

This is awesome. I am definitely interested in making the rear underside more flat near the subframe and exhaust. I know some Python if you need any help. It has been about 3 years since I have done anything intricate in Python but I am sure I could dust off a little. I am getting married in a week so I will be useless for the next two (honeymoon)...then I am back to serious wagon work!

 

Sounds like you just volunteered to make a wrapper/GUI for the response surface we're talking about :D

 

I am interested to see a marginally longer wing as well (and maybe stupid long wing). Makes me think of the tail end of drag cars, how they extend the trunk lid with sharp fencing on the edge

 

It shall be done.

Link to comment
Share on other sites

Sounds like a few single point runs to me!

 

I don't think there's any curing to the suction short of adding a massive faring, but I bet we could reduce the effect a bit.

 

Would mounting the wing like you said be any different than a waist spoiler? One of my shopmates has a racebrick FXT, had a conversation with him a while back about waist spoilers. He mentioned that they'll completely eliminate the snow/dust fouling that we've been talking about on wagon hatches, which to me seems to imply they'd have a pretty big effect on the recirculations back there.

 

found a post on nasioc on the waist spoiler - https://forums.nasioc.com/forums/showpost.php?p=788541&postcount=19

 

My original reason for removing the machV wing was to mount a waist level wing. From all the pics, putting in the area where air starts to curl back above the bumper is probably right location.

Link to comment
Share on other sites

ultimately I'm interested in achieving negative lift, ie downforce. Drag can be overcome, but lift loses races.

 

What happens if the air dam touches the ground? In this pic, air still gets by. Making a seal is possible.

attachment.php?attachmentid=287255&d=1601217606

 

Update: not what you expect, according to Neal Cassidy. Boundary effect is unsettled close to the ground so being closer hurts more than helps.

Edited by boxkita
Link to comment
Share on other sites

Before I exceeded Gunnar's patience with his freshly waxed wagon, we had determined adding vertical extensions at the trailing edge of the rear window helped airflow off the spoiler. Someone in one of the fb groups made a wagon spoiler with side pieces. It looked like a step in a better direction, however, I dropped off fb before any pieces shipped so no idea how it really worked.

 

Interesting. I recently saw a picture newer SUV or wagon where the hatch spoiler looked like it wrapped down around the sides of the rear glass. The parts that wrapped down around the the sides of the glass weren't jutting too far out, probably 2inches or less, and tapered as they went down. It looked weird, but not terrible. I'll see if I can dig up which car it was.

 

 

Aren't we in a situation dealing with one of the same principles that people believe to give an airplane wing lift? The air running over the top of the wing from the incident point at the front has to travel a longer distance than the air along the bottom of the wing, which causes the air to lift up into the pocket. Can you calculate total body length (including curves going up the bumper, up over the windshield, along the roof) and compare it to the essentially straight line length along the bottom of the car? If you reduce this top/bottom length difference (via a very long front splitter or even a long rear diffuser), do the numbers change? I know these are unrealistic, but I'm wondering if it'll provide any insight. Maybe some of your tested scenarios already have done this.

 

The front splitter might not help since top and bottom "wing" length are both increased with that...

 

Either way... data are cool.

Link to comment
Share on other sites

Under car lift/downforce is driven by the venturi effect to create high speed/low pressure air under the car, whereas a traditional wing achieves this high speed/low pressure air on the underside of the wing by forcing the air to travel a longer distance as you described.

 

A long splitter/rear diffuser reduce lift by increasing the Venturi effect (generally). The front splitter also pairs with a vertical air dam to increase the pressure differential and the rear diffuser if position correctly with the rear wing can have beneficial interaction to reduce separation, as Boxkita mentioned earlier.

 

Some of the 80s F1 cars, Group C and Le Mans prototypes did essentially have wing shaped side pods/tunnels, combining the wing shape and Venturi effect, but that’s going beyond what we’re looking at with a road vehicle.

Edited by Neal Cassady

2006 Legacy GT Wagon - Regamasters, Evo X Recaros, STI 6MT, Brembo 6-Pots (in progress)

@regashiii

Link to comment
Share on other sites

Because we don't have nearly enough parameters to model, I was chatting with one of the master techs today. He suggested active aero using spring steel, which technically isn't illegal. :)

 

Regards the model, there's a python library for the software Awfulwaffle is using. Theoretically, other cad can import the model. The concern is loss of fidelity of the model during import/export. Haven't any thoughts on how to resolve it, then again parameter driven design is news to me.

Edited by boxkita
remove unproven ramblings
Link to comment
Share on other sites

thinking something like this with the diffuser. basically a dual plane rear spoiler.

blue smudge is supported away from the body on an arm. The black smudge is the lower plane is attached to the bumper or bottom edge of the hatch. And Removing the oem spoiler for a rounded cap?

 

crappy design conversion courtesy of https://www.freeconvert.com/jpg-converter/download , so I can find it again. :)

3046D6AB-18CA-4618-87A3-175A1DF48306.thumb.jpg.db12dd427473b0095f8a70338eeb7d07.jpg

Edited by boxkita
Link to comment
Share on other sites

  • 2 weeks later...
I can't decide whether that's the result of some very thorough 3D CFD/wind tunnel work, or someone with a bucket of leftover carbon fiber bits and a hot glue gun looking at pictures of aero setups and trying to duplicate...

 

 

Haha I stared at it a long time. According to the article it's from wind tunnel testing.

 

By article I mean press release. I read about it in road and track I think. Not auto blog.

 

 

Sent from my iPhone using Tapatalk

Link to comment
Share on other sites

Because we don't have nearly enough parameters to model, I was chatting with one of the master techs today. He suggested active aero using spring steel, which technically isn't illegal. :)

 

Regards the model, there's a python library for the software Awfulwaffle is using. Theoretically, other cad can import the model. The concern is loss of fidelity of the model during import/export. Haven't any thoughts on how to resolve it, then again parameter driven design is news to me.

 

The fluid-structure interaction would be an interesting can of worms. I'm sure we could do it, but I'd almost rather try it manually at first (ie calculate load, tweak model based on some simplified calcs, run again, repeat until force balances out). Not sure how quickly that procedure would converge though

Link to comment
Share on other sites

thinking something like this with the diffuser. basically a dual plane rear spoiler.

blue smudge is supported away from the body on an arm. The black smudge is the lower plane is attached to the bumper or bottom edge of the hatch. And Removing the oem spoiler for a rounded cap?

 

crappy design conversion courtesy of https://www.freeconvert.com/jpg-converter/download , so I can find it again. :)

 

Added to the docket! Single point run before digging into parametric stuff.

Edited by awfulwaffle
Link to comment
Share on other sites

The fluid-structure interaction would be an interesting can of worms. I'm sure we could do it, but I'd almost rather try it manually at first (ie calculate load, tweak model based on some simplified calcs, run again, repeat until force balances out). Not sure how quickly that procedure would converge though

 

I found the object model for your drawing system and then realized I had no way to test it. I'm sure your bosses would be delighted to get random code from the internet to execute on your system. :-)

 

I'm good with manual approach. What can I do to help? Guessing crayon on iphone doesn't translate well?

Link to comment
Share on other sites

 

the new wrc rules allow for aero mods. not quite gruppe B, but more than has been allowed in past seasons.

 

Lots of neat aero tech on the car. Especially with changing movement of tire smoke down the body instead of up in the air. Note the exit of the spins, where tire smoke stays almost flat exiting the rear.

 

Sht.Gator's sti uses a similar trunk mounted cooling system. idk if he has done the door mods, but it's very supercar method of routing air to the coolers

Edited by boxkita
Link to comment
Share on other sites

  • 2 weeks later...
Flipped far back enough into one of my desk references at work, found this.

 

Thanks for looking it up.

 

I read this as front air dam needs to be 1/3-1/2 distance of bumper to ground. Angle of rear window needs to be roughly 34 degrees from horizontal. Rear spoiler at waist height needs to be roughly same as Porsche Turbo or 930, as the top spoiler doesn't help much. Underside of car needs to be very smooth.

 

Still boggles that an air dam matched with a smooth underbody doesn't create a lifting surface.

 

Rough measure of wagon rear window is 35 degrees from horizontal. Go figure, its about ideal

Link to comment
Share on other sites

No problem, I happened to notice a picture of a car when looking at the section on drag.

 

Those takeaways make sense to me too, with one tweak - I'd think that the rear spoiler needs to be placed in a region of flow that's dynamically similar to that where the Porsche spoiler lives. Seems to me like it'd be somewhere between where we originally wanted the waist spoiler, and where the top spoiler would be. Thoughts?

 

Only way I can picture the dam + flat underbody doesn't induce lift is if you incur enough total pressure loss and reduce static pressure recovery due to the presence of the dam...

Link to comment
Share on other sites

2nd paragraph: You have mentioned running a parameter driven run. Any chance of doing a run with spoiler location moved from bumper to roofline spoiler?

 

I wonder if a bi-plane spoiler would work? Looking at Porsche tails, the spoiler is at the top of wheel fenders, more or less. The 959 skips the spoiler and just has a wing, tho.

 

The test runs you've already done suggest targeting the waist area will have highest success. I'm coloring the design, but the coming off the rear windows changes from laminar to vortexes at the tailgate line. If the back wash from the roof spoiler and the bumper were intercepted by a wing?

 

The pdf's parameters have no allowances for a roof spoiler. If the area under that spoiler was filled in, then what?

 

3rd paragraph: I've read it four times, still dont grok it. Translation?

Link to comment
Share on other sites

Yeah I think so. I'll motivate myself to do a parametric model w/ just spoiler location to start - should be a good trial run for getting things to play nice between all the modules, before we dive into a model with a bunch of stuff moving around.

 

I wonder if one of the effects of the roof spoiler is to have a sort of variable impact on the airflow down the back of the car at different speeds, per this effect: https://en.wikipedia.org/wiki/Coand%C4%83_effect

 

^^That's a 6:30 AM just-got-to-my-desk-and-ingesting-my-caffeine-thought, so take it with a grain of salt.

 

3rd paragraph (sorry if I say stuff you already know):

 

 

So, fundamentally lift is the sum of the force caused by a static pressure difference across a body in the direction of lift, plus the force imparted by momentum transfer from the fluid on any portion of the body that is not parallel to the lift direction and is also in contact with moving fluid that's not entirely tangent to said surface.

 

Pressure can be considered in two components. Static pressure is that which exists in the fluid due to random motion of the molecules in the reference frame (moving with) the fluid under consideration. Dynamic pressure is the equivalent pressure associated with the energy of bulk motion of the fluid (ie flowing together in one direction). The sum of these is called the total (aka stagnation) pressure, which is what the static pressure would be if all of the dynamic pressure was converted to static with zero losses.

 

Any time a fluid is forced through a restricted area, there is an exchange of static pressure for dynamic, as the fluid accelerates. There is also some finite loss of energy/total pressure due to friction/viscosity, etc. Any time a fluid passes from a smaller flow area to a larger one, there is an inverse exchange - dynamic pressure is converted to static, with some conversion efficiency (static pressure recovery coefficient) based on how gradually the fluid is expanded. I think of it as - dynamic pressure is scrubbed during acceleration (really, at all times to some extent), and the ideal static pressure is not realized during sudden deceleration since some of the dynamic energy is converted to useless energy modes instead of those that contribute to static pressure.

 

The thought is, the dam might serve a twofold purpose - impose a penalty on the total pressure (total pressure loss), and then also have really crappy static pressure recovery from this already-reduced total pressure since the fluid is immediately dumped into the much larger area underneath the car aft of the dam. These two combined would lead to a reduced static pressure under the car than in the case where there is no dam at all.

 

The pdf text puts emphasis on crossflows and effects of other vehicles on the road, so this is most likely only part of the picture. Also trying to think through what impact ground effect might have as Neal Cassady mentioned earlier in this thread.

 

All them wordz above seem to disagree with the difference in calculated lift in the runs with and without the dam. Gonna poke around the solutions a bit to see if I can make it make sense in my head.

Edited by awfulwaffle
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

Terms of Use