motormechanic

iTrader: (1)

best places for an inlet for cooling the brakes is up front. If you take a look at Plucas' model you'll see there's high pressure where the air hits the bumper.

emilio700

iTrader: (3)

Not usually. The air that slows in front of the nose gradually speeds up again (regains energy) as it moves to the sides and over the hood. Right under the headlights say, the air is moving a quite a tangent to the centerline/drirection of travel. Thus a forward facing duct right under the headlight has a bad angle of incidence to the airflow. Sorta like whistling across a bottle. If you can run yarn on the nose and get pics, you'll see this. In that case a NACA duct aimed towards the centerline might work better.

We locate our ducts as close to the car centerline as possible for this reason. They can also be located in the bumper opening for the rad. The air is all but stopped here but is significantly higher pressure than the backside of the rotor. In this example the differential moves air down the duct tube for you.

mx5-kiwi

I have just made v1 of a nose undertray/flat bottom and need to add extensions out to the wheels from the initial middle section.

Unfrotunately I overestimated the required wheel and tyre clearance and need to add extensions on each side..

But, how close should I get to the wheels/tyres? ( on the sides i.e by the lower control arms) and in front of the tyre, do people extend the tray deeper than the wheel arch, towards the tyre...?

As close as possible or it doesn't matter too much in this turbulent area.....?

Some intersting results on my first drive with this v1 middle section....yet to be decided conclusively but the gearbox tunnel seems to be considerably cooler (as compared with no engine undertray where the gear box tunnel was almost to hot to touch...56+ deg C)

My slight front crank seal oil leak appears to be blowing out the back and not dripping straight down the front of the engine like it used too.....(inside the tray is spotless despite a 60km test drive whereas I have oils spots all over my drive from the persistant leak previousley.....)

So, while I haven't had time to prove either point conclusively it would appear the flat tray has raised the air flow or suction out of the engine bay considerably...is this likely to that extent?

And is the reduced tunnel temp likely or do you guys think I am probably imagining it?

mr_hyde

iTrader: (4)

AFAIK, nobody has made conclusive before/after temperatures of transmission and differential fluid temps with undertrays and while the tunnel temp from the cockpit isn't a perfect metric for that, a cooler tunnel certainly can't be a bad thing.

The general assumption is that the undertray would trap engine and exhaust heat so it is very interesting if you feel it is actually sucking the heat out. Do you have an extraction hood or any other topside vents in front of the firewall? At a glance, they would appear to lower the pressure/velocity/volume of air rushing through the tunnel even though they help the heat exchangers.

Maybe a functional approach when coolant temps isn't the absolute priority (i.e. no 100*f ambient racing, WHP<250, etc.) is to have high pressure at the front, medium pressure in the engine bay and low pressure where the tunnel exits past the diff.

mx5-kiwi

Just for the record, I do run FM bonnet louvres and the undertray on this car is just to the normal mounts behind the steering rack.

Without any front tray the tunnel temps were so bad I had ordered some reasonably expensive heat mat for under the car inside the tunnel but now it looks like I may not need it.....

That would be good. Although all bets are off if I move from 10 to 14lbs boost I guess....

Anyway, I love this thread, great info. Just thought I would contribute my tidbit.

Shahab

iTrader: (2)

I'm glad Paul posted the results here. I provided Paul with the model and the simplified wheels are the dimensions of my 225 rs3's mounted on 15x9 6uls if anyone was curious. Also as Paul mentioned the underbody is flat to simplify the cfd.

One main thing I wanted to achieve with these cfd runs was to compare options 4-6. Obviously these options are what most people in the miata community are running.

A few months ago there was a thread where someone explicitly asked what was better option 4 or 5. When I suggested that 4 (my current car) would make more downforce, someone suggested that wasn't true because option 5 reduced lift and the net change would be more downforce than option 4. Obviously these results show that is not the case and prove me right, huzzah!


But seriously I'm glad we were able to get this model together and quantify some things in regards to miata aerodynamics. I would also like to echo Emilio's comments in regards to the curse of the internet. Couldn't have put it better myself.

triple88a

iTrader: (2)

I don't know it just makes no sense to me how #4 can have less drag than #1.

Shahab

iTrader: (2)

Moving tires are a huge source of drag. #1 has them in free air, #4 has them covered.

danny2747

No induced drag, or only induced drag? Your CFD model seems to be calculating Base drag. What about parasite and rotating wheel drag? I venture to say that it is quite significant on these little cars. And what about drag during side slip? As in 4-12% lateral angle of attack for maximum tire grip?

I love this topic and the analysis which is why it peaked my curiosity? Does any other forum go to this level of detail to analyze and improve their car?

danny2747

For most MT dot net lurkers, hydrodynamics trumps aerodynamics (sex-wise) and incompressible fluids mark their passion. I for one, find as much joy in compressible fluids as in incompressible.

Shahab

iTrader: (2)

Are you asking about the effects of yaw? I'm sure if you pay Paul enough he will run it at whatever yaw angle you are curious about.

What do you mean by this?

gohammer

Great article. I think drag is also reduced because the splitter/air dam forces more air to go over the car and to the side of the car reducing airflow under the irregular disruptive and turbulent bottom of the Miata. If you compare 5 and 6 you can see how the splitter helps in this way.
Something I didn't notice any one talking about is the use of a Gurney Flap at the trailing edge of the air dam to scavenge even more air from the area in front of the tire, it will increase drag a little but it's really worth it for downforce gain.
If reducing CD was the goal just fair the wheels, make the bottom smooth, and allow the air to flow smoothly to the bottom by rounding the radius of the bottom leading edge. (this also reduces total frontal area)

1993ka24det

I'm doing my front end aero and since we are on the subject. I was looking around on the net to see what kind of canards there are for the NA. I remember seeing the Brightning, but their pictures suck(Brightning Front Canards For Miata MX5 MX-5 89-97 JDM Roadster : REV9 Autosport). Does anyone know of a better pic? I'm going to make my own out of pre-preg and searching for ideas.

triple88a

iTrader: (2)

Gotta do some solid rims too.

plucas

Do what? I know what induced drag is but do you? What is base drag? Not to be an *** but it sounds like you are throwing words around without knowing what they are.

Induced drag is drag that occurs or induced when a moving object redirects the airflow moving around it. It is usually referring to wings... It is inversely proportional to velocity.

Base drag is.... I have never heard of it

Parasitic drag is pressure drag and friction drag and increases proportionally to the square of the velocity.

Rotating wheel drag? I understand this...but what do you mean?

The drag that is calculated in my analysis is total drag.
Total Drag = Friction Drag + Pressure Drag + Induced Drag

Are you talking about yaw for the last part? I really do not understand.

Mobius

iTrader: (2)

Two of his questions as I understood them were:

a) what about cases where airflow is not directly head on, but is yawed horizontally at an angle of between say 4 to 12 degrees. I'm not entirely positive as to the value of this. Having airflow yawed at a static angle is a poor approximation of simulating the car during a turn.

b) rotating wheel drag - he wants to know if your model is assuming stationary wheels, or if some function is being used to approximate the different drag created by wheels that are rotating. I believe you said earlier that your model uses a moving floor, correct?

Shahab

iTrader: (2)

This was mentioned already.

plucas

a) The car or airflow (depending on how you look at it) is not yawed. I have no interest in this either unless somebody wants to pay for cfd work (currently busy in paid projects so no time for freebies except for shahab ). Yawing the airfow/car was the way to simulate turns and still is for windtunnels. However now with cfd, you can actually flow the air in a radius to simulate a corner situation. The big guys in motorsports are doing this (not really useful in this level of motorsports imo).

Not directed towards you as you were just clarifying somebody else's post.

This

Dot3

iTrader: (1)

What do you guys think about the kazespec splitter mounting kit?
KazeSpec Engineering FULL Aero Kits!!!! - Page 11 - ClubRoadster.net

motormechanic

iTrader: (1)

Looks pretty nice.