How to maximise downforce on flat floor+diffuser on race car?

[hsg]

How to maximise downforce on flat floor+ rear diffuser on race car(not open wheel car)?

[Silent Storm]

What's your max tunnel height and width?
Floor length?
What's your ride height envelope?
What speed range and track type is this for?
Car type?

[hsg]

What's your max tunnel height and width?
Floor length?
What's your ride height envelope?
What speed range and track type is this for?
Car type?

modified Toyota mr2

[Silent Storm]

I can't figure out what you have in mind, but I'll tell you what I'd do given your three rules.

1) Tunnel sizing & diffuser geometry
2) Edge sealing
3) Wheel wake management

Let's start with flat floor length...
Roughly 60% of your 3.8m, remaining 1.5m for your diffuser as that gives you plenty of ramp.

With floor clearance at 100 mm and a target trailing edge height of 200 mm, your diffuser angle is...
α=arctan(200-100)/1500
That gives you 6.7, which is safely under 7°

If you need even more expansion, split that 100 mm rise into two 50 mm steps with a shallow strake mid height. That delays stall and lets you push closer to a 1:10 expansion ratio.

Edge sealing...
Vertical fences at the floor to diffuser join, about 80 mm tall, generate strong vortices to seal the underfloor. You can add rake so front is at 90 mm and rear at 110 mm.

Wheel wake...
Triangular plates mounted just ahead of the diffuser throat about 50 mm tall and angled 30°outboard will steer the turbulent tire wake around the diffuser inlet. Properly pitched, their vortices run parallel to the ramp and actually re energize the boundary layer rather than trip it.

911 skirts...
Rubber skirts wear out, plastic is fragile. Production homologation rules, ground clearance rules (minimum gc for production car at kerb weight) and NVH targets would ban devices that touch the ground. So deflectors are a practical choice.

[hsg]

Let's start with flat floor length...
Roughly 60% of your 3.8m, remaining 1.5m for your diffuser as that gives you plenty of ramp.

With floor clearance at 100 mm and a target trailing edge height of 200 mm, your diffuser angle is...
α=arctan(200-100)/1500
That gives you 6.7, which is safely under 7°

Diffuser cant be so long, because at rear is engine/gearbox, it can start from rear axle downstream.

[Silent Storm]

If not 1.5 m, then I hope you have atleast 1 m to 1.2 m...
To keep your diffuser angle under 7° you need to drop your rise to 160-170 mm at the lip or break the ramp in two stages, a shallow 4-5° stage upto say around 140 mm then a second 6-7° stage to 200 mm. Each stage less than 7° so flow stays attached even in shorter length.

If you are going for vertical fences then case 1, place it right at flat floor to diffuser transition to pin your under floor vortex.

Regarding 30° flow separation... you should get 30° in stages, each <10°

Plug those into your CFD loop, tweak vane angles by a few degrees, and you’ll land on the optimum balance of area, angle, and vortex strength.

[hsg]

If not 1.5 m, then I hope you have atleast 1 m to 1.2 m...
To keep your diffuser angle under 7° you need to drop your rise to 160-170 mm at the lip or break the ramp in two stages, a shallow 4-5° stage upto say around 140 mm then a second 6-7° stage to 200 mm. Each stage less than 7° so flow stays attached even in shorter length.

If you are going for vertical fences then case 1, place it right at flat floor to diffuser transition to pin your under floor vortex.

Regarding 30° flow separation... you should get 30° in stages, each <10°

Plug those into your CFD loop, tweak vane angles by a few degrees, and you’ll land on the optimum balance of area, angle, and vortex strength.

Case 1 has vertical fences behind front wheel wake, they stay parallel to inner flow, how they will produce vortices?

[Silent Storm]

If not 1.5 m, then I hope you have atleast 1 m to 1.2 m...
To keep your diffuser angle under 7° you need to drop your rise to 160-170 mm at the lip or break the ramp in two stages, a shallow 4-5° stage upto say around 140 mm then a second 6-7° stage to 200 mm. Each stage less than 7° so flow stays attached even in shorter length.

If you are going for vertical fences then case 1, place it right at flat floor to diffuser transition to pin your under floor vortex.

Regarding 30° flow separation... you should get 30° in stages, each <10°

Plug those into your CFD loop, tweak vane angles by a few degrees, and you’ll land on the optimum balance of area, angle, and vortex strength.

Lets try to first solve skirts/fences location and angle to flow.
Case 1 has vertical fences behind front wheel wake, they stay parallel to inner flow, how they will produce vortices?
Also if fences is behind the wheel, front wheel wake will enter in the floor, that is not good
Why fences at case 1 better then case 2 and 3?

Even straight, parallel fences generate counter rotating longitudinal vortices due to the pressure differential (Δp) between the low pressure underfloor and the higher pressure ambient air. These vortices form at the fence edges and travel downstream, acting as virtual skirts that inhibit high pressure air from leaking into the diffuser.

Any abrupt change i.e. the fence’s top and bottom corners, trips the boundary layer into a shear sheet. That sheet rolls up into a vortex even if the plate is flush and parallel. The under floor cavity sits at, say, 70 kPa, while outside is 101 kPa. That Δ𝑝 forces flow through the tiniest opening and around the fence tips, energizing your vortices.

Sure, some of the wake can sneak under the fence if you place it too far downstream. You can add a turning vane or shift the fence upstream. Like I said plug these in your cfd loop and don't aim for perfection.

Let's look at pros and cons of each case...

Case 1 -
Pros - Max underfloor area and simple to package
Cons - Wheel wake

Case 2 -
Pros - Less wheel wake interaction
Cons - Narrow venturi loses total downforce

Case 3 -
Pros - Slightly wider capture area
Cons - Outer edges stall vortices breakdown leading to massive separation and leakage

Basically, case 2 loses absolute floor area as your plenum is narrower, so peak suction falls. Case 3 tries to flare too aggressively, but that big outward angle stalls the fence’s own flow and kills your vortex strength. Case 1 hits the sweet spot.

[hsg]

Basically, case 2 loses absolute floor area as your plenum is narrower, so peak suction falls. Case 3 tries to flare too aggressively, but that big outward angle stalls the fence’s own flow and kills your vortex strength. Case 1 hits the sweet spot.

?



You think this fences design(below photo) at flat part of floor is not good?
It seems they tray to blocked some tire wake.

[Silent Storm]

Are you serious about this project or just curious?

Are you running a front splitter?

Wheel arch vents or louvres can bleed off some high pressure pockets. Turning vanes at wheel trailing edge angled outwards rolls the wake away from your diffuser.

And to the picture... There’s no absolute good or bad here, it depends on the intended flow management and use case. I’ll review it properly once I’m off my phone.

[hsg]

Are you serious about this project or just curious?

Are you running a front splitter?

Wheel arch vents or louvres can bleed off some high pressure pockets. Turning vanes at wheel trailing edge angled outwards rolls the wake away from your diffuser.

And to the picture... There’s no absolute good or bad here, it depends on the intended flow management and use case. I’ll review it properly once I’m off my phone.

Serious.

[Silent Storm]

Are you serious about this project or just curious?

Are you running a front splitter?

Wheel arch vents or louvres can bleed off some high pressure pockets. Turning vanes at wheel trailing edge angled outwards rolls the wake away from your diffuser.

And to the picture... There’s no absolute good or bad here, it depends on the intended flow management and use case. I’ll review it properly once I’m off my phone.

Serious.
They are loosing much of floor area with design.
It will be the best to see how real aero cars solve this problems, like VW IDR from Peaks Peek, must be car without rules, because you never know if something is done because of rules or this is best solution.
But there is not undertray photos of this exotic cars.


F1 has used flat floors for decades, but of course venturi floors are much more efficient at creating downforce.

Just a quick note... Originally you mentioned working under certain rules (flat floor + diffuser, no sliding skirts, etc.), but now we’re drifting into “no rules” prototype territory.

That car lives in a realm of essentially unlimited CFD, real world testing and wind tunnel budgets, bespoke chassis packaging and structural freedom... Nothing like a modified MR 2 with a budget and a gearbox sump in the way.

F1 “flat floors” were flat only because the regulations forbid venturi tunnels, not because they’re more effective. In a zero rule world, you’d always choose a true venturi... It simply outperforms a flat plate every time. But unless you’ve got equivalent CFD time, tunnel hours and aero team, trying to copy the ID.R’s deep tunnels is unrealistic. Copy paste doesn't work here.

My advice is focus on proven compromises you can implement in your MR 2 package. Maximize the contiguous floor area you’ve got, keep your diffuser ramps in two sub 7° stages, use robust 80 mm edge fences (and wheel arch vents/turning vanes) to manage wakes, and iterate with what little CFD or track testing you have. That way you get most of the venturi benefit without chasing a prototype you can’t feasibly replicate.

[hsg]

@Silent Storm
yes I talk about no rules car ,only because they use optimal solution, of course if they know what are doing. My only rule is flat floor + diffuser, not allowed venturi

Indeed my question is "simple" make fences on edge of floor(1.) or inside to try avoid tire wake...

[hsg]

tire wake

[Silent Storm]

@Silent Storm
yes I talk about no rules car ,only because they use optimal solution, of course if they know what are doing. My only rule is flat floor + diffuser, not allowed venturi

Indeed my question is "simple" make fences on edge of floor(1.) or inside to try avoid tire wake...

If your main goal is to avoid front tyre wake entering under the floor, fences placed inside (inboard) of the floor edge are the better choice. Placing fences slightly inboard lets you catch and manage the dirty tyre wake before it spills into the low pressure floor area.

Position the fences roughly along the trailing plane of the front wheel (or just slightly downstream), and angle them a few degrees outboard if needed to better guide the wake away from the diffuser throat.

On a side note... You need to lower your car to make all of this work better, 100 mm is not enough, 40-60 mm is needed.