Rear diffuser?
 

Has anyone on here made a rear diffuser for any of their vehicles?

Any tests done?

Since I’m doing a smooth under tray, I’m going to obviously fabricate a rear diffuser. Seems like it’s all benefits and no downsides

So far my inspiration is making one that’s like on the vision eqxx by Mercedes

Yes.

Yes.

I haven't done one myself, the VW Beetle body doesn't lend itself.

I have squirreled away some examples:

Start big and whittle down
https://ecomodder.com/forum/member-f...80-silver4.jpg

If the outer fences are curved, you have a start on the rear wheel spats.

https://ecomodder.com/forum/member-f...bonnevette.jpg

That was the closest I got to that car. The fences start at the axle line and flare to the outside face of the tire.

Wonder why they don’t use any strakes or fins. Was going to try and make mine look like that and connect to the rear wheel skirts

Hey Phase! Strakes are only beneficial when they prevent vortices from forming because of other elements. Otherwise they deflect air that doesn't need to be deflected, i.e. they cause drag. It's easier and better to just have a well shaped smooth surface and let the air determine it's own least energy path. This is especially true if you are making mods upstream, as the optimal strake angles depend on what's happening along the entire length of the car.

If you are putting on rear wheels skirts, it might be worth having a strake on each side, as an extension of the wheel well. Then have the outside skirt and the inside strake meet, making a boat tail for the wheel. That's what all the solar car winners / serious aero cars do. (They even go a step further and make a whole fairing for each wheel.)

I always like the skegs on the 1961 Oldsmobile:

https://cdn-0.barnfinds.com/wp-conte...-3-630x390.jpg
https://cdn-0.barnfinds.com/wp-conte...-3-630x390.jpg

Im a 57 chevy fan myself.

Has anybody ever run very long tufts to see exactly what is going on back there in the turbulent section?

Like mentioned above, arbitrarily adding strakes can be harmful to aero. They should only be added where they're actually needed. And to add strakes without disrupting airflow, you'll need CFD and/or wind tunnel testing.

i figured id just skip the strakes and go for the eqxx style diffuser

Sounds like a plan. Also, you need to be mindful of the departure angle. If it's too aggressive, the air stream can detach and create a bunch of turbulence (i.e. drag). If you could do some tuft testing, that should help.

im going to keep it fairly flat and a mild angle like the eqxx also. pretty much just extending the existing underbody shape out another foot or so

The underbelly angle must be kept very shallow to retain attachment (2 or 3 degrees if I recall correctly), otherwise turbulence results.

The diffuser fences/strakes allows a roughly doubling of inclination angle. The 10 degree angle I drew was considered undo-able.

Going off memory of sharing a coffee with a professional aerodynamicist about five years ago that had work experience with several automakers.

The fences work because air attaches to the vertical surface much the same way as the near horizontal surfaces, I suspect as parasitic drag hence people's words of caution about assuming a rear diffuser will always lower drag.

These are of course generalisations and not necessarily applicable to every car type/groud clearance.

"im going to keep it fairly flat and a mild angle like the eqxx also. pretty much just extending the existing underbody shape out another foot or so"

That is rarely a bad thing.

Generally, high diffuser angles (greater than 18 degrees) cause seperation and vortices and increase drag. Low diffuser angles (less than 12 degrees) allow attached flow and can reduce drag.

What that exact optimum (for low drag) angle is on your specific car? Who knows, likely somewhere between 0-12 degrees.

If you are talking about just extending out the diffuser a long way, past the end of the car then there is a paper on this. In their tests, on their car, they saw around 9 degrees was the optimum angle. Attachment 32470Attachment 32471 Paper name: Drag and lift reduction of a 3D bluff body using flaps

www.researchgate.net/publication/226210472_Drag_and_lift_reduction_of_a_3D_bluff_bo dy_using_flaps

Interesting data aquisition via olive oil.

The way the rear of my car is shaped, it looks like my diffuser will have to get more narrow the farther out it goes. In theory, shouldn’t that also benefit since a more narrow path increases flow and thus should spit out even faster flowing air into the wake area? I’m just going off what I know from sluice boxes when gold panning in water lol

Any success with the sluice box? Or do you keep that sort of knowledge held close?

Depending on tuft testing, the long skirts should have some taper in plan. It should transition from the rectangular body section to an increasing radius. (pace Morelli) A skeg at 45&deg wouldn't add side area in crosswind conditions. I imagine something like the Cadillac Biarritz turned upside down, at an angle.
https://bestcarmagz.net/sites/defaul...ar_3-4_Web.jpg
https://bestcarmagz.net/sites/defaul...ar_3-4_Web.jpg

That's not optimal. Ideally, the diffuser would be a diverging nozzle, i.e. bigger at the back.

http://https://upload.wikimedia.org/...A_1.6_Rear.jpgWell the way the back bumper is shaped on the corners and how I’m “ supposed to integrate the flares coming out of the rear wheel skirts), I don’t see how that’s possible

See. Tapers in

Convergent on the outside, divergent on the inside.

is it even worth building the diffuser from the rear wheel spats though? theres going to be a lot of turbulance under body behind the rear wheels anyways with detached airflow

whats the point of trying to reattach the airflow for another inch, just for it to become separated after the diffuser? obviously between the wheels is the most important part

The Bonnevette in Permalink #2 had a duct wall that started at or before the axle line. It walls off the underbody flow from the wheelwell turbulence.

I think the outer wall of a rear wheel spat is optional for the reason you state.

so pretty much i could taper in the rear wheel spat to a straight diffuser, while blocking in the wheel well section from messing with the underbody, then expand that area out and wider from the inner section of the flat underbody/diffuser. dont know if i would both boat tailing behind the rear wheel wells, or maybe just do a mild boatd tail behind the wheel to slighly guide the edge air at the side ends of the diffuser

diffuser skirt wheel fairing
 

The diffuser defines all the other architecture.
Once the diffuser is defined, it will reflect on the wheel skirt design.
The diffuser can close the gap of the wheel well, up to the inner face of the rear tires.
The wheel fairings would only be behind the wheels.
And these wheel fairings would create fences on the outer extremities of the diffuser.

Yeahs that’s what I’m going for. Main diffuser air between the wheel wells, and then bring in the trailing edges of the wheel skirts like a mini boat tail that also acts as some fences/strakes. I saw one of the newer Priuses doing something similar actually

http://https://www.hilandtoyota.com/...ar-Angle_o.jpg

Obviously I’d have deeper boat tailing behind the rear wheels to create the fences. It would look like a pinching shape behind the wheels once I have the rear wheel skirts over. Tesla also uses mini rail guards next the inside wheel wells to block off some turbulence and guide the air out to its diffuser

I can't find the original Lotus Exige wind tunnel testing page I read a lot of years ago but this is a pdf of another? (or maybe the same one): https://www.reverie.ltd.uk/Downloads...st_Session.pdf

Notice that the extended strakes to seal off the wheels drops drag by a massive amount.

Saw this yesterday. Now that’s a rear diffuser…

I scanned through there, (:confused:), do you have a page or image number? Front or rear wheels?

Nonactive video links in the PDF when I viewed it, would you by any chance know where they are online?

They claim large differences in drag, I'm just not as impressed as the authors I guess.

number
 

It's @ #9, at te bottom of the first 'RESULTS' table.
Adding the outboard diffuser vanes dropped drag, from Cd 0.555, to Cd 0.496, a delta- 0.059, 59-counts, a 10.63% reduction.:)
Polishing a turd.

https://youtu.be/Mi5NgRyAqt4

https://youtu.be/Zq2KjcNE50c

two videos showing a flat floor isnt necessery. it helps a little and is better than a dirty floor, but the diffuser helps all cars

youtubes
 

I don't know anything about this person.
I don't know anything about the CFD he uses.
The models he depicts do not represent any 'reality' I've ever experienced.
Perhaps, when I have nothing else to do, I can spend some time vetting him, but until then , I find his conclusions highly dubious.
There's some 'context' associated with his presentation, however there's no transparency provided in its content.

Those are two different people doing two different tests

Bottom line is yes, a flat floor is best, but you don’t need a flat floor for a rear diffuser to work

The first video is someone I'm familiar with. He does this stuff (aerodynamics engineering) for a living. So there's a pretty good chance he actually knows wtf he's talking about.

The second video, I'm not familiar with. So I can't comment.

don't need
 

It flies in the face of all previously published foundational research. And I don't understand the rush to dismiss what, up to now has been considered 'settled science.'
So, like something the late Carl Sagan might say, it would require a remarkable argument.

chance
 

When I can, I'll try and dig up a portfolio on the guy.
Some CFD I've seen, like Air Shaper, produces quanta which doesn't resonate within this universe. Orders of magnitude.

https://www.ioniqforum.com/attachmen...zed-jpg.13113/

https://www.ioniqforum.com/attachmen...ed2-jpg.13121/

i mean its not as perfect as an EV, but other than the rear section, it seems like its pretty streamlined for an underside

DrivAer model
 

Here's a link to Munich's model:
https://www.researchgate.net/profile...r-OpenFOAM.pdf
For the notchback smooth underbody model, CFD was 4.26% different than wind tunnel.
Same car, but with detailed underbody, the CFD was 11.18% different than the wind tunnel model.

'smooth' vs 'dirty'
 

I went to the Munich, Germany site.
I reverse-engineered an estimated frontal area.
I reverse-engineered the VERSUS ENGINEERING test velocity.
Using Munich's drag coefficients for the wind tunnel tested-model, I reverse-engineered for VERSUS's velocity.
With their own numbers, I calculated that the 'smooth-bellied' notchback with diffuser would have 39.1 fewer counts than the 'dirty-bellied' notchback with diffuser, making 13.82% lower drag.
With only the smooth belly and diffuser, the IONIQ might hit it's target of Cd 0.20.

Why’s the Ioniq ev got the same or higher CD then if that has a smooth flat under body and a mini rear diffuser? That’s what I don’t get

I’m pretty worried about cooling when trying to cover the underbody mufflers and exhaust pipes