fe1rx adds full aero

[135]

Originally Posted by fe1rx

In order to maximize cooling effectiveness, I am sealing up the inlet air leaks around the radiator and intercooler.
...

Amazing attention to detail.

[fe1rx]

Originally Posted by tracer bullet

When you mention -12 and -22 sizes, is that a reference to AN tube sizes? If so, I didn't see a 22 option, only 20 and 24.

I am using Aeroquip AE102/624 firesleeve. The dash number is the ID in 1/16" increments. Nothing really sacred about the sizes. Any two nesting sizes would probably work. I chose -12 and -22 because I have quite a bit of those sizes on hand that is surplus from an old project. The stuff you have is always preferred to the stuff you have to buy

[davidwarren]

great work. Looks phenomenal

[chris_flies]

Your attention to detail and thoughtfulness about everything is inspiring. It would be so cool if this could be marketed and sold to the wider community as a mostly bolt-on affair, but I'll settle for great pictures and detailed explanations if/when I do something like this.

[rac]

nice work. i agree with the previous comments about the splitter angle.

downforce measurements through the revisions?

it would be good to verify revision improvements. if you have a long smooth and straight road and dont need to hit the brakes or turn suddenly you could measure shock travel at different speeds. or for ease of reading fabricate 4x soft consumable wear markers at each corner and measure how far they wore down at speed 1, 2, 3 etc. you intimately know your vehicle spring rates so should be able to work out the rest.... just throwing out some ideas. i love your charts so....

[fe1rx]

Originally Posted by rac

nice work. i agree with the previous comments about the splitter angle.

downforce measurements through the revisions?

it would be good to verify revision improvements. if you have a long smooth and straight road and dont need to hit the brakes or turn suddenly you could measure shock travel at different speeds. or for ease of reading fabricate 4x soft consumable wear markers at each corner and measure how far they wore down at speed 1, 2, 3 etc. you intimately know your vehicle spring rates so should be able to work out the rest.... just throwing out some ideas. i love your charts so....

Even with shock potentiometers (which I don't have) and smooth roads, apparently it is quite difficult to separate the signal from the noise in terms of suspension compression in response to aero loads. Using a tell-tale indicator of maximum suspension travel would end up just showing you the peak deflection, where you are really looking for the "steady" state value.

In any case, rather than trying to determine my actual downforce, I will judge the mod in terms of effect on lap times primarily, with a closer look at the data for things like apex speeds, specifically at the Mosport GP track.

[fe1rx]

My splitter stays were attached to the bumper impact beam via eyebolts into rivnuts, per an earlier image. I considered the combination acceptable, but only just, so at V3 I have replaced the rivnuts with welded in bushings and the rivnuts with stainless steel straps.



I have also added an additional pair of outboard stays, to better support the outer ends of the splitter. A thin nylon washer is installed between the straps and the bumper cover, so that the straps don’t wear through the cover.



The stay cables are fabricated from 1/16” 7x 7 stainless “aircraft” cable with a PVC coating. The ends are finished using Nicopress sleeves and AN100 thimbles. The inner and middle cables are almost the same overall length, differing only by 0.05”, so I have colour coded the middles to differentiate them. The outboard cables are much longer so easily differentiated from the others, but build tolerances in the splitter result in the ideal length of the two outer cables being different LH and RH by 0.09”, so these two cables are also colour coded (port side red) to avoid head-scratching down the road.



The cables connect top and bottom using AN115 shackles, with clevis pins . The upper clevis pins are secured with cotter pins and the lower ones with cowl latch safety pins. The safety pins are new at V3 and they really make life easier when removing and installing the splitter.



My car was built with headlight washers, which don’t really provide any useful function for a track car. As the bumper cover comes off with some regularity, I have left the washer covers off for a while now. I would prefer to seal up the holes to avoid errant airflow, so I decided to remove the washers entirely. All washer functions are on a single fuse (48, behind the glovebox), so disabling the headlight washer by pulling the fuse would also disable the windshield washer. Instead, I removed the headlight washer pump and all the associated plumbing, and fabricated a nylon plug to seal up the pump grommet.



The splitter diffusers have been revised at V3, with a curved ramp. I have no idea how far up the ramp the flow will remain attached, but I will check this out with flow vis when I get the car moving. Here is the V2 diffuser:



Here is the V3 diffuser:



The splitter diffusers attach into helicoils in the plastic splitter, to minimize the chance of stripping the threaded holes and to eliminate the bother of hex nuts. As a general philosophy, lap joints are completely avoided in the undertray, so the splitter has been rabbeted to allow the diffusers to remain flush.



Previously I relied on the existing undertray (installed above the splitter) to manage cooling air flow, but these are now removed. The splitter aft attach bracket has been replaced by a new one that provides the desired ground clearance, and that also now extends the full width of the undertray. This serves to seal the passage under the front subframe. I want to make sure that any air that moves past the engine and transmission actually does some cooling, rather than bypassing underneath the subframe.



I have fitted silicone baffles to seal the wheel well liners to the splitter diffusers. These allow the splitter to move upward in response to ground strikes, without applying shock loads to the liners.



Last year’s Delrin wear pads have almost worn down to the screw heads, so I will start this season with new ones. Although the part is a bit too pretty for its intended purpose, it adapts a part I have on hand from a completely unrelated product/project, so was the easiest solution for me.



That wraps up the splitter changes at V3.

[fe1rx]

The centre section extends from the back between the forward and aft jack pads. As this section has little means of support, I have used bonded honeycomb panel material in the outboard sections, as this provides appropriate stiffness with no additional bracing. The centre panel is 0.063” aluminum. The panels are sealed to the underbody to constrain all cooling air flow to the drive shaft tunnel. Top view:



Bottom view:



On the left side, removal of the plastic undertray elements exposes the fuel and oil lines, so these have been boxed in with aluminum.





On the right side a baffle has been installed to separate the transmission side of the passage from the exhaust side. This should minimize heat transferred from the exhaust to the transmission.



A baffle has been installed under the transmission prevent the air from passing “unused” under the transmission. Silicone engine baffle material has been used to allow a close fit and to prevent wear between adjacent parts.



To provide additional support to the sheet metal centre panel, a bracket has been installed on the transmission cross brace.



The bonded panels edges have been closed out with formed aluminum “Z” sections. These were bonded in place using a Hysol structural adhesive. Clamping during cure was achieved by vacuum bagging.



The tunnel seal is accomplished by means of a formed aluminum section, to which a silicone baffle seal has been attached with pop rivets. This section also serves to splice the side panels to the centre panel with a butt joint. Here, instead of rivnuts, I installed self-clinching “Pem” nuts. These are more difficult to install than a rivnut, requiring a rivet squeezer and access to both sides of the part, but they have the benefit of being flush on the bottom side. They are also lower profile, allowing the use of shorter screws. Normally the tunnel seal sections will stay attached to the outer panels, even when those panels are themselves removed.



The geometry of the seal provides a good seal between the undertray and the bottom of the vehicle.



The result is an isolated tunnel that constrains the cooling air within the drive shaft / exhaust tunnel.



The combined weight of the centre section panels, cross members, baffles and splicing/sealing parts is approximately 34 lbs.

The panels provide a smooth, flat, rigid aerodynamic surface.

[135]

I was going to say amazing attention to detail but noticed that was my comment last time.

So I'll say it again... amazing attention to detail!

Everything I've been wanting to do to my car - and more.

[asbrr]

Truly is lovely work.

fe1rx I could help notice what look to be stock transmission mounts in one of your photos - are those upgraded M3 pieces or the originals? If original don't you get the downshift lockout under heavy braking and downshift actions?

I was getting that until I upgraded to the Rogue Engineering solid rubber pieces which improved things. But I think engine mounts play as big a factor and those are still stock for me.

[Traf]

Hi f3rix, given the aero u added, i'm assuming this will increase your cornering G's a fair bit. Couldn't you have done all these pannels with core foam and carbon fiber tho ?
Anyway, are you going to do something about oil starvation due to cornering ?

[fe1rx]

Originally Posted by asbrr

Truly is lovely work.

fe1rx I could help notice what look to be stock transmission mounts in one of your photos - are those upgraded M3 pieces or the originals? If original don't you get the downshift lockout under heavy braking and downshift actions?

I was getting that until I upgraded to the Rogue Engineering solid rubber pieces which improved things. But I think engine mounts play as big a factor and those are still stock for me.

Sharp eye asbrr . They are original. I should investigate the mod.

[fe1rx]

Originally Posted by Traf

Hi f3rix, given the aero u added, i'm assuming this will increase your cornering G's a fair bit. Couldn't you have done all these pannels with core foam and carbon fiber tho ?
Anyway, are you going to do something about oil starvation due to cornering ?

We'll see. My expectations are relatively modest - maybe 0.1 g on fast corners. The track I am interested in has several approximately 160 km/h corners, so the issue will be downforce at about that speed.

Aluminum is relatively easy and quick to work with for a prototype setup. If it works well, I might look for some weight savings on a later version.

Depending on results I may consider an oil pan baffle.

[AndyW]

What thickness HPDE did you use for your diffuser?

[fe1rx]

Originally Posted by AndyW

What thickness HPDE did you use for your diffuser?

The splitter and the diffuser strakes are both 3/8" thick HDPE.

[fe1rx]

The forward section extends from the back of the splitter to the forward jack pads, and it represents the aerodynamic throat of the underbody.



Normally engine cooling air exits the engine bay into the front wheel wells and below the car from the openings in the plastic undertray. Having provided hood louvers, I want to eliminate the cooling airflow into the wheel wells, and to ensure that any engine cooling air that passes under the car, exits above the diffuser into the low-pressure area behind the car. To accomplish the required sealing, sheet metal sealing panels have been installed to seal around the front subframe.







This section also incorporates a converging exit section to merge the wheel well air with the air traveling under the car.



An unknown, but significant amount of engine heat is rejected directly from the crankcase to the air passing though the engine bay. To maximize the effectiveness of this cooling mechanism, the potential air path under the front subframe has been sealed by a series of baffles the ensure that all the air passing through the engine bay passes close to the engine.

The forward undertray incorporates a NACA duct (same design as for the differential) to introduce additional cooling air in the vicinity of the transmission.



To retain the ability to jack the car at the forward subframe, a low-profile jack pad has been installed in the place of the existing plastic jack fitting on the surface of the panel.





Our power steering reservoirs are notorious for venting a mess into the engine compartment. By pressing a short length of 3/16” stainless tubing into the vent hole, and connecting this to a length of rubber hose, I have previously vented the reservoir into the LH wheel well. That keeps the engine bay clean but the LH suspension parts get fogged with oil and attracts dirt. To improve this further, I have run a hose down through a grommet in the centre section undertray. This hose splices in the wheel well, so the undertray can be removed without disturbing the hoses.





The vent hose stays attached to the forward panel. This panel weighs approximately 10 lbs. and is constructed from 0.063” aluminum sheet.



This completes the underbody. As is apparent from the reflections from the underbody panels, the shape is well controlled with only minor surface irregularities.



This gets us to the point where the car was ready for initial trials.

[asbrr]

Great update & progress!

What if any thoughts have you put towards brake cooling? I know you have that fantastic AP racing custom setup and maybe this is a moot point, but I can't remember if you've done any duct work and venting from the wheel well for the brakes? (May be in another thread you did, can't recall...)

[fe1rx]

Originally Posted by asbrr

Great update & progress!

What if any thoughts have you put towards brake cooling? I know you have that fantastic AP racing custom setup and maybe this is a moot point, but I can't remember if you've done any duct work and venting from the wheel well for the brakes? (May be in another thread you did, can't recall...)

Thanks!

No brake ducts per se, but scoops that extend inboard of the tire:

https://www.1addicts.com/forums/show...04&postcount=1

Incidentally, these scoops caused me to add 10 mm Delrin spacers to the steering rack, which reduces full lock travel. My forward undertray takes advantage of this reduced lock. It would have to be narrower to avoid a rub, if I didn't have that feature installed.

I don't have a problem with the calipers getting too hot. Peak caliper temperatures do not exceed 300°F based on temperature labels I installed on them. The rotors might be happier with some more cooling, but I have never faded my pads (DTC-60) either.

[fe1rx]

Heat management has been my biggest concern, and the initial shakedown (street, not track) test did reveal some problems in this regard.

First, the exhaust runs hotter now that it is wrapped and fully enclosed. As a result thermal expansion is increased enough that the muffler contacted the tow hitch and body when hot. I have addressed this by reworking the axle-back, removing 5/8” from its length.



I have installed non-reversible heat labels on the differential and transmission to monitor their maximum temperature condition. Based on my assumption that anything under 250°F is acceptable, I chose a label that ranges from 250°F to 320°F. The differential labels are visible from the rear of the car with a flashlight, or through the jacking access hole.



The transmission label is only visible when the mid center panel is removed.



Street driving was not sufficient to get a response from any of these labels.

I did get some bad smells and smoke from some heat damage to the RH fiber under body cladding. Just the edges were exposed to the radiant heat from the exhaust, but this was enough to cause a problem in a couple of spots. This material is self-extinguishing and not really a fire hazard, but it does decompose when overheated.



To address this, I have removed the original RH underbody cladding completely, installing aluminum sheets where necessary to provide a sealing surface for the silicone tunnel seals.





The LH side was not heat affected but a small amount of the cladding was still exposed.



I was able to reduce the exposure by simply tucking the fibre under the heat shield





I have also painted the undersides of the forward and mid panels flat black where they see radiant heating from the exhaust.



Rather than reflecting any radiant heat, these panels will now absorb more of it. I am happy to let them dissipate it under the car.

Another quick test revealed no more bad smells, and no problem with the exhaust impinging on the tow hitch.

I am planning on installing thermocouples on the differential and transmission, and into the forward and middle air spaces so that I can monitor these temperatures in real time, but that will be after the car’s first track test.

[fe1rx]

Before I take the car to the track I want to confirm that the air is flowing over the car as I intended. In particular, I am interested in:

1) how the hood vents flow.
2) whether the splitter fences (aka tire blockers) and rocker panel treatments generate a counter-rotating pair of self-supporting vortices, which is my intent. My intent is indicated in the image below:



3) how cleanly the wing flies
4) how cleanly the air exits the rear diffuser

To answer these questions, I have tufted the RH side of the car and taken some phone camera video of the car at 100 km/h. The images that follow are 5 successive frames for a forward quartering and aft quartering view. Multiple frames reveal if the tuft direction represents a steady flow or a turbulent one.

FRONT QUARTER:







Observations I find interesting from the front quarter:

Above the splitter at the base of the air dam, the flow is strongly outboard. This is the flow I am intending to utilize with my fence to generate a counterclockwise (RH side, looking aft) vortex across the middle of the front wheel. As an aside, Andrew Brilliant has made great use of this flow with his Infinity Wings.

Flow exiting the hood vents is good. A wicker at the leading edge might improve the flow further.

Behind the centre vent and at the base of the windshield, there is a recirculation bubble. I suspect this is due to the flow from the vent and is not present in the baseline. Flow reattaches over the windshield and remains well attached over the roof.

Flow over the side of the car at wheel height is turbulent but sufficiently orderly that it may be indicative of the splitter fence vortex I intend. Or maybe this is wishful thinking. My intent is that this vortex assist in extracting air from the front wheel well, improving brake cooling.

The rocker treatment appears to generate the clockwise (RH side, looking aft) vortex that I intended. The purpose of this vortex is to inhibit underbody inflow between the wheels.

REAR QUARTER







The rear quarter view reveals that the flow remains fully attached over the rear glass and trunk lid. This is really good aero design by BMW. A long tuft attached to the top of the shark fin confirms how smoothly the air flows over the top of the car. The wing is flying in very clean air.

I installed tufts on the trailing edge of the wing, and these are trailing aft nicely demonstrating the Kutta condition present in an unstalled wing.

Similarly tufts installed at the trailing edge of the rear diffuser trail aft in the centre tunnel but demonstrate some vorticity in the outer tunnel. This is not surprising due to the divergent vanes in the outer tunnel.

I am interested in examining the splitter fence flow further, and also to look at how the flow changes with the removal of the fence. That experiment will have to wait for after my first track day.

I also want to look at under-car flow using flow vis fluid. Specifically, I would like to know how much inflow there is from the sides of the car, and how well the flow stays attached in the splitter diffusers and in the rear diffuser. I think that will be very interesting, but again a track day beckons first.

My wing is set at the incidence I settled on for V2. The car feels good so I will leave it there initially.

As for the track, my first concern is thermal management.

[Dmak]

Always love reading your post, fe1rx!! Cant even image the time spend for all this r&d.

I also have hood vents at pretty much the same locations as yours. Its nice to see real data proofing it works(even tho, i know it helps as soon as i have mine installed)

I also notice large gap at both sides of the intercooler while i was doing my install and made 2 side pieces to block the air leak.

[spxxx]

Have you considered trimming the radiator shroud behind the grills to allow for more direct airflow to the radiator?

I've also been working on sealing airflow around my intercooler, oil coolers and radiator but it seems there's an opportunity to increase airflow to the radiator with larger openings behind the kidney grills.


Btw... Love your car, such attention to detail. The flat bottom is great!