Presented by Nine Lives Racing
1 2 3 4
sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/9/20 3:05 p.m.

Alright, like last night, pic... then word salad edit...


steve, feel free to correct anything I get wrong-ish here wink

 

word salad  (eeek!  math!):

Alright, so there's this thing that usually gets summarized wrong with describing how wings work, called  Bernoulli's Principle.  The important part of that here, is that minus "system losses" an enclosed system has a constant flow rate that depends on the initial area and the initial inlet area.  My above simplification excludes "mechanical flow exitation" (i.e. fans).  So, increase the area, reduce the speed.  Reduce the area, increase the speed.   One super important thing to keep in mind:  when speed goes up, pressure goes down;  and when speed goes down, pressure goes up.

Point the second:  High pressure will move to a zone of low pressure.  I have no concern about if it get "pulled there" or if the high pressure is "pushed into" the low... that's philosophy.  We're dealing with reality... and trying to use it to our benefit.  So, if we have a tunnel, and it has an inlet area at the front doing 60mph... and we scoot in to the back of the car and constrict it to... fit between a tire and an engine... and it reduces by 60%... then we have flow coming out the back end of the car at 100mph, and lower pressure.  One other thing to keep in mind, is that after a certain length, the flow will be 'oriented' along the "lengthwise axis of the tube" (see: "fully developed flow") and it's effectively homogenous other than skin friction due to the wall (iirc).

If the "tunnel" isn't a tunnel... but two tunnels... the flow into the second tunnel has to deal with non-uniform flow into the second tunnel, and our "delta" efficiency loss goes up.  Probably significantly.   Can we do that?  Yeah, sure!  Maybe it's the right thing?  Maybe not?  If we do open up the side, then we won't have the same 100mph air in the "center-ish" of the back of the car.

So, why might it be important to put low-pressure high-speed air at the back of the car?  Well, for a couple of reasons.  One reason is it "fills in" a bunch of the hole punched into the air by the body of the car.  The low pressure also helps convince air coming around the "exterior surfaces" of the car to "contract back" as if the car hadn't been there, instead of swirling around lazily.   It just so happens that an area of expanding area and slowed air exists just below the exit of this air.... it's the flow coming out of the diffuser.  Remember, high pressure wants to move to low pressure.  So, the low pressure coming out of the tunnel, right over top of the diffuser actually acts on the whole of the flow under the car kind of like a huge "EXIT THIS WAY" sign... and that means that the whole diffuser makes more downforce.  Because the air under the car, all the way back to the nose and the splitter are reacting to what's effectively a fan at the back of the car.

That's why you probably want a tunnel along the car for the autocross.  I think you'll make more downforce with the tunnel there... although we're probably going to have to figure out what the "minimum area" available is between the frame, the engine, the rear suspension, and the tire.

 

Why do I want a tunnel for the drags, and to have the sides blanked off?

Well, again... as Steve says in his recent "misconceptions" article... "it depends"... or as my favorite college professor (the late, great) Dr. William Mason so often said:  "Well, yes, you can do that."

Sure, we could open up the side.  But... keeping the original exterior shape would be the least drag.  And, funneling the flow from the "Ojai Duct" into the tunnels would mean (as I said in page 2) that the car's total cross-sectional area (Ax) is half what it was.  Plus, if you squint at my dotted line, and think about the inside wall of the tunnel, you've effectively created a "laminar flow" airfoil shape around the formal car's frame.  Which is one of the lowest drag shapes out there.

Now.  The caveat:  The Drag's heavily favor weight/power ratio over aerodynamic drag.  So, you need to do this in a way that doesn't add a ton of weight-fraction.

One benefit, though, it we get to keep the body... which is budget-neutral.

stafford1500
stafford1500 GRM+ Memberand Dork
12/9/20 3:26 p.m.

That was delicious, what dressing are you using on this word salad? Well summed up, and with equations to boot.

I will add that based on the current power output of this car, the cooling requirements are going to be smaller than a higher horsepower car (which is typically larger as well). If we follow this logic you start down the path set by the DeltaWIng designer (Ben Bowlby or Ben Blowby to some of his co-workers...). The end results is a car that can perform as well or better than larger higher power cars due to the thoughtfull design and implementation of the car. The whole less weight --> less load --> smaller parts --> means less weight circle continues. It gets even better if you can get the parts to do more than one thing (like using the sides of the formula car chassis as one side of the ducting/providing extra stiffness to the chassis/keeping rocks out of your face/etc.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/9/20 3:38 p.m.
stafford1500 said:

That was delicious, what dressing are you using on this word salad? Well summed up, and with equations to boot.

...start down the path set by the DeltaWIng designer (Ben Bowlby or Ben Blowby to some of his co-workers...) ...

It's air we're talking about.... and I reckon the stress of this year has hurt all of our waistlines...

so probably a "misting" of oil and vinegar  wink

 

also,  after all, it was Bowlby who was also behind the GTR tunnel concept that we're effectively aping here

nocones
nocones GRM+ Memberand UltraDork
12/9/20 3:42 p.m.

Just to clarify these aren't Renderings they are just what Sketchup looks like.   Those take seconds to make, it's just an export of what you see live while modeling.

 

These are (Bad) renders.  I'm still learning the rendering software.  These take a few minutes each and will get better and better progressively the longer I let them run.  These churned for about 3 minutes each.  I ran 13-15 passes through.  It will max at 1000.  Each one adds more and more detail to the render.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/9/20 3:47 p.m.
nocones said:

Just to clarify these aren't Renderings they are just what Sketchup looks like.   Those take seconds to make, it's just an export of what you see live while modeling.

I think you just won a free dinner from mazdeuce - Seth... since you just pedantic'd the forum pedantic

 

edit:
seeing the actual render though, gives me some hope... since I think his wheels are smaller that the model's.  unless he's moving up in size for next year?

hmmm, maybe not?

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/9/20 4:13 p.m.

hmm... we may also have a bit of an issue with the rear suspension... if we're going to get monster diffusers as drawn...

edit:
I think we might have been going about this "tunnels" and "no rules" thing wrong

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/9/20 4:44 p.m.

In reply to sleepyhead the buffalo :

I was considering moving the rear brake to the middle of the car. 

nocones
nocones GRM+ Memberand UltraDork
12/9/20 5:03 p.m.

In reply to Robbie (Forum Supporter) :

There is also 0 reason structurally that that blue white bar can't be above the axle.  If you move the brake to near the other bearing so everything lower the the axle is inside those two longitudinal rails you will have pretty good depth to the tunnels.

 

There is also that other thing ..

stafford1500
stafford1500 GRM+ Memberand Dork
12/9/20 5:17 p.m.

Things running across the diffuser space can be faired in to reduce the effects. In the vein of my previous post, those fairings can also be used to stiffen the diffuser structure.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/10/20 3:42 a.m.
nocones said:

In reply to Robbie (Forum Supporter) :

There is also 0 reason structurally that that blue white bar can't be above the axle.  If you move the brake to near the other bearing so everything lower the the axle is inside those two longitudinal rails you will have pretty good depth to the tunnels.

 

There is also that other thing ..


I agree with moving the caliper to near the right side bearing.  I'd probably put it so it's about the same distance off the frame as the chain pulley.  That way we could have the option to create a duct in the middle that's 'centered'.

also, I wouldn't swing the axle all the way up, vertical.  I'd suggest swinging it up so that it's in a 10deg line from the axle center.  That way we can create a void for the drive axle and the suspension axle that's along the diffuser plane.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/10/20 3:53 a.m.

I went a dug back in Robbie's thread a bit to the bare chassis, to try and emphasis a point Steve's been making... that might be getting lost a little bit.  While also bring up something I'm seeing.

What Steve's try to convey is that finding a way to panel the side of the frame will have two benefits.  One, it will help stiffen up the torsional resistance of the frame.  Also, that paneling can form the "inside" wall of any tunneling/tube/diffuser that runs down the car.  One piece, doing two jobs.

Also, this view helps me show why I'd suggest raising the axle up only 10degree in line with the drive axle.  The "no rules" thing means we can make a "double height" diffuser, with one extending forward up towards the footwell.  And the second starting somewhere just forward of the drive axle.

And that leaves some volume between this diffuser and the side of the car... in the area ahead/abolish of the rear wheel... to play around with placing two smaller radiators for the engine.


edit, and I still think there's room to run a headlight/Ojai-duct fed tunnel down the side of the car:

nocones
nocones GRM+ Memberand UberDork
5/8/21 10:10 a.m.

Since this has become the home of really really detailed Aerodynamic discussions for bad idea challenge cars I finally got V2.0 of the LMP360 Aero "designed".  It's still probably a bit more of LMP Aero Cosplay but based on the feedback from this discussion I have made a bunch of changes.  I also adjusted the sizes of everything to be accurate to the chassis/wheels from the real car.

Inspiration for this round came heavily from the Praga R1, the Ligier JSP3 and I've digested a fair amount of online analysis of the AM Valkyrie.  The rear wing is still a bit of a placeholder, and the intention is for the bottom of the end supports to have some twist to them to provide some infill behind the car.  I will just make whatever profile rear wing is recommended.

 

Front and rear fender design is heavily influenced by the Praga R1.   Front aero "Theory" is to have the hood extensions direct some air up over the chassis that would then flow around the greenhouse before being addressed by the rear wing.  Immediately bellow this is openings that woudl allow high velocity air as well as the air from the top of the splitter/front wing to flow down the side of the chassis and mostly into the radiator openings, with the rest being allowed to exit behind the front fender.  

The front fender has a inner wall behind the tires but otherwise is open.  This would hopefully help evacuate the air from under the front wing.  A large vent in the side of the back of the front fender hopefully helps evacuate air from around the wheel.  

Moving down the side of the car the sidepods are slender on bottom and the rear fender has a horizontal divide behind the side pod entrance.  Air is given the ability to either go inside the rear wheels and then infill the top of the rear difuser or be shed around the side of the car.  The air going through the radiators would be released into the engine bay and infill the center of the back of the car or exit out of the top of the rear decklid to help infill under the rear wing.   

The wing stancions would have a gap behind the fender and help direct air behind the car to potentially infill the back.  

Underbody would be two tunnels that split to feed the rear difusers.  The width at the front will be determeined by being ~120% of the area that can be achieved in front of the rear wheels where the tunnels bend back into the difusers.  This will be the most restricted area (from a chassis clearance standpoint).  If it makes sense to make a choke point further up the car that can be accomplished easily anywhere along the sides of the car.

 

These are the theories anyway.  I'm not particularly studied in Aerodynamics.  I understand the theories somewhat but I haven't sayed in a Holiday Inn for at least 2 years so most of this is just replicating the average of what I see amongst cars of similar potential performance.  The Praga R1 and Valkrie seemed to be a good jumping off points becuase unlike most of the LMP1/2/3 type cars there are not rules dictating many aspects of the way they deal with Aero. 

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
5/8/21 10:23 a.m.

In reply to nocones :

I love it. Blend the entire floor into the diffuser though. Instead of air going around the flat bottom, you want air going past it too.

I'm really liking this idea for the iconic flat "datsun" panel at the leading edge of the hood for fdat.

 

nocones
nocones GRM+ Memberand UberDork
5/8/21 10:39 a.m.

In reply to Robbie (Forum Supporter) :

Yes sorry the rear of the floor will taper up as much as it can inside the tunnel exits behind the flat floor under the chassis.  

On F-Dat you can do a lot of different stuff with the front depending on how much metal you are willing/want to cut away.  The idea of the Konica-Minolta part being the "Hood" and the front of the car without the hood looking basically like that is a pretty good one for you.   Look at modern CSR/DSR cars which are basically fenderd formula cars for ways to treat the air at the front.  You could theoretically have tunnels that run down the sides of the car between the inside of the fenders and the actual chassis.  Sort of like what Sleepyhead showed for down car tubes but have the top be open so your front profile would look basically like that ACURA.  I'll fire up Sketchup Later and see what it would look like.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
5/8/21 3:25 p.m.

so, this is on my radar.  I'll try to have more thorough marked-up pictures tomorrow.

first thought, though, is that you probably want to widen the center section of the raised portion of the front element (I'm going to try to stop using the word 'splitter' from now on.  I think use of that term is leading many people down the wrong path on their design.) so that it's goes out to the inside edge of the front tire.  If you look at the minolta acura, it's about the same there.  So of that air will go through your tunnels, and some of it will end up on the upper surface of your side skirts.  Because of that, I'll be offering some suggestion tweaks.

Along those lines... look closely at the DP.Acura, and you'll see that the splitter really is more of a "wing"/"flap" shape, and it has some thickness.  For your design, and the need to carry the loads into the space-frame, having some thickness will be important.  Keep in mind, that the 'front lower flap' is going to make maximum downforce somewhere 1-25% back from the front edge... depending on "section" shape/profile (i.e. the curved shape cut along the x-z axis; whereas planform is the view from above along x-y axis).

potentially there's not a lot of benefit, and actually quite a bit of negative to extending the rear flap (nee 'wing') endplate supports all the way down to the diffuser... especially when it gets so spindly.  It might be better to 'leverage' off the fender approximately even with the wheel hub height.

one of the hardest things I think you'll have is that most of your fenders, flaps and aerodynamic accoutrements are going to be 1) self-fabricated;  2) cantilevered off the main body shell/structure.  Yes, there'll be some benefit from the desire to "hide" the upper arms from the wind... and how you can kinda approach the construction like a box-kite or bi-plane structure.  But, that strength... with the necessity for lo-o-o-o-w cost, the desire for low weight... but also significant rigidity (so the body doesn't contact the ground due to jitter/jounce/flex/downforce)... that's a definitely going to require some... um... constructive thought

nocones
nocones GRM+ Memberand UberDork
5/10/21 10:13 a.m.

In reply to sleepyhead the buffalo :

So my intention was for the areas outside the tunnel inlet to be wings with cord sections.  Are you saying it would be neat to have a full width "wing" a little above the floor with the tunnel inlet under the car behind the wing?  Or should some center section of the wing feed directly to the tunnel but the bottom edge be consistent with the bottom of the wing?

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
5/10/21 11:11 a.m.

I think it becomes a balance thing. If you put a wing at the front and get the air up right away, you'll have potential for lots of front downforce. If instead you channel it to the underbody and diffuser, that same air energy can be used toward downforce at the rear (or in the middle, at very near the CoM of the car). 

But instead of having front wings and rear wings, right now I'm leaning toward maximizing the floor and trying to center it's effect near the CoM on F-dat. I want a bit of fore/aft adjustability built in at the front as mentioned by stafford1500 last fall. 

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
5/10/21 3:16 p.m.

Apologies, I had some stuff go sideways on me, and I haven't gotten back around to this.

How many people flew balsa gliders as a kid?  It's a wing, even though it's just a flat plate.  That's actually where the basic lift curve slope that's shown in most textbooks derives from.  Most 'splitters' at the grassroots level are effectively the equivalent of a balsa wing set at an angle... and the flow is getting squeezed between the ground and the edge of the balsa wing, which is both a turning and an area reduction, thus the speed increase and suction/downforce is created.

Here's a bad drawing of a balsa-wing/GRM.panel-splitter:    

The thing is, though, that's a sharp corner to get around, so there tends be a separation bubble on the bottom side of the splitter which reduces the amount of downforce that's created.  So, frequently stafford1500 suggested that you round over the bottom edge... which gives a 'panel' splitter a faint resemblance of a Clark Y airfoil.

Remember... "The Body Is The Wing"... and "everything else are flaps."  The more you can round over the 'leading edge' of your front splitter-flap-wing-element-thing, the more downforce you can create at a slower speed.  Just because you've built a wing, doesn't mean the exit *can't* dump into the tunnels.  It just means you/I/we need to do some thinking to make sure we don't shove too much air into them such that we 'choke' the system.

otoh, we can use some of that wing curvature to feed into what are currently being called "splitter diffusers"... but really, they just after cambered flap sections.  Those are dumping into the volume between the wheel and the chassis.  Some will go into the tunnels,  it a fair amount will go up over the top side of the side skirt.  They create a 'very nice exit' for some of the flow from the front of the front flap-wing-splitter-element... including from parts of it that might be closer to the middle.

Wow, this is a word salad.  We need more scribbling on 3D model .jpgs up in here.

edited to change a can to a 'can't', and bolded it.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
5/11/21 2:10 p.m.

so, as promised scribbles.  First from the front, with some tentative side views to hopefully illustrate what I'm trying to describe...

So, the green note here kinda flows into some mark-ups I've done on you underside image...

So, looking at this, it looks like the panel under the cockpit area is flat, and I guess the tunnels slowly curve 'down' near the same height compared to the raised entrance in the splitter center?  But those two heights are fairly similar, and also 'up' compared to the side skirts?  Considering cost/material/time constraints... I think you should consider simplifying the floor section in the area under the cockpit all the way to where the tires have cutouts.  I think it should just be a flat floor section, with some vertical/horizontal walls to create the lowered sideskirt sections.  You want the edge of the flat floor/tunnel as sharp and vertical as possible, to discourage the low pressure under the car escaping out the side.

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
5/13/21 10:03 a.m.

While we're on the topic of floors, here's two options for fdat.

The center frame section is very low, and is low most of the length of the car. Between 2-3 inches above the ground, and about 20 inches wide. So the questions are as follows: make the whole floor that low? Or do tunnels? Also, skirt placement inside or outside wheels?

Sorry for the low tech design software.

stafford1500
stafford1500 GRM+ Memberand Dork
5/13/21 10:26 a.m.

For Noncones: from Sleepyheads images/sketches/notes Try to keep the walls of the tunnels closer to vertical and the forward facing corners radiused  (on the forward edge of the center flat section under the cabin, for example), but keep the aft facing corners sharp. That will maximize the retention of the low pressure generated by the flow. Skirts/rockers should be as sharp as possible to create flow across them to a minimum. The more torturous the path the more distinct the pressure/flow across those edge will be (higher gradient).

For Robbie: Skirts should be as far outboard as you can get them. That will maximize the surface area exposed to the low pressure under the car and separate it from the higher pressures outside. You can build the tunnels into the floor (outboard of the driver location). There will need to be some sort of cross-section reduction near the bottom of the drivers seat to get the center of pressure to be centered in that region. This will create an oval-ish shaped flat section in the middle. That can be with lateral or vertical reduction in cross section. Whichever makes the most sense based on the rest of the chassis structure. Get some pictures of the car on the lift at some point and I can scribble up a basic outline. There was discussion of venting some of the rad flow thru 'contained' ducts to the back of the car. Those may play into the shape of the floor tunnels, maybe with a common wall thru the middle section.

Teh E36 M3
Teh E36 M3 SuperDork
5/13/21 10:33 a.m.

As to ingress/egress: 

Epicstream 

A lawn tractor version of this would guarantee victory in any setting. 

stafford1500
stafford1500 GRM+ Memberand Dork
5/13/21 11:26 a.m.

In reply to Robbie (Forum Supporter) :

Thinking about it further, your central tunnel could work pretty well too. You will have to expand the rear around the engine/suspension differently than your sketch, but it could be done. The entrance presents plenty of options to make sure you have enough capture area to get the low pressure at the throat (small cross section under the driver). Pontoons along the outer regions and capturing the tires is not a bad plan. Still want skirts or similar at the outside edges.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
5/13/21 12:44 p.m.

One of the reasons I'd advocate for a flat floor for nocones is that he doesn't have much 'body' outside the frame that he's building.  So, the floor, fenders, and accoutrements that will be constructed have to create their own rigidity and transfers their loads back to the central frame.

FDat, even with some of the structure removed, has the benefit of the original steel body to connect to and develop/re-develop structure and stiffness too.  So, that'll be easier by comparison... and you're effectively a year ahead of him.  So, I'm making some simplified/simplifying suggestions with those things in mind.  As well as the fact that I'd like to encourage the mindset for nocones that the aero might look one way for $challenge, and then might get updated as he starts doing TT.

 

Another thing I've been thinking about with nocones car, is that there might some benefit to having the flat floor under the seating area, then the diffuser beginning around where the engine is... would have the benefit of the diffuser downforce being located at/near the car's CG.  Thus, tweaks to it, and some of the effects that front/rear aero might have to increase flow through the diffuser will be 'neutral'.   That might be particularly helpful when he starts shaking the car down on-track... where the speeds/dynamic will be higher than AutoX/drags.

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
5/13/21 12:55 p.m.

I only drew 'half' of the lower picture, hope that's not confusing the issue. There would be a similar tunnel on the right side too, and yes I was thinking that the surface I labeled "tunnel area" would be both the top of the underbody tunnel and the bottom of the front-to-rear ducts. 

My main question though hinges around "is #2 worth the effort over #1"?

#1 would be easier to build, but maybe not by a whole lot. I just don't know if 2-3 inches of ground clearance is really enough to keep enough volume of air moving under the car. Between the insides of the front wheels, we're only talking about 40-45 inches wide. On the other hand, the "tunnels" would be able to be raised up a bit to allow more air through. If that makes sense. So say that using the tunnels I could almost double the minimum cross sectional area of the under car area as viewed from the front.

#1      .....[]___[].... 

vs 

#2     .....[]-_-[]....

1 2 3 4
Our Preferred Partners
tvJopEh92KfQgUDxgXFe6HDKYLzXreP3X225FlAaHpaeeHnudVfXPXJNlbuKFQ47