29 replies to this topic

[NeilR]

What were the differences between single seat cars and group C sportscars in tunnel design. Also was there an ideal angle for a tunnel or an expansion ratio? Finding data is really quite hard and I cannot find any detail on the reynard inverter etc, which is about the most modern application.

[NeilR]

No one? perhaps my question was a bit unresolved so I'll try again.
1. The oft quoted angle of 7 degrees is given for attached flows to the walls of a venturi and seems to have been applied to the venturi of a ground effect tunnel. This angle is increased with aerodynamic contribution from a rear wing - which apparently is a good thing (AGT). However a ground effect tunnel is not a proper venturi in many respects with one wall being a moving plane. did F1 cars and sportscars approach this in a different manner or was the solution essentially the same?
2. If a competition vehicle has a speed range of 60-160kph is there any point in pursuing ground effect?
3. The CAD drawing of the inverter seems to have the tunnel starting very early and a very small throat area indeed, yet the suggestion in Katz is that the throat area is in fact the most productive underfloor surface. Would the reynard treatment reduce pitch potential sensitivity? Or would there be another reason behind their decision?
http://picasaweb.goo...073941975531954

[DOF_power]

No one? perhaps my question was a bit unresolved so I'll try again.
1. The oft quoted angle of 7 degrees is given for attached flows to the walls of a venturi and seems to have been applied to the venturi of a ground effect tunnel. This angle is increased with aerodynamic contribution from a rear wing - which apparently is a good thing (AGT). However a ground effect tunnel is not a proper venturi in many respects with one wall being a moving plane. did F1 cars and sportscars approach this in a different manner or was the solution essentially the same?
2. If a competition vehicle has a speed range of 60-160kph is there any point in pursuing ground effect?
3. The CAD drawing of the inverter seems to have the tunnel starting very early and a very small throat area indeed, yet the suggestion in Katz is that the throat area is in fact the most productive underfloor surface. Would the reynard treatment reduce pitch potential sensitivity? Or would there be another reason behind their decision?
http://picasaweb.goo...073941975531954

1] I don't know the angles, but the solutions where different.
In F1 GE was banned after 1982, witch is really when GE kicked in in prototype racing.
F1 cars usually generated most of their downforce via the underbody undertrays (even tried to get rif of the wings), while SP went beyond that by using the rear wing as you mentioned, and ended up producing a lot more downforce. F1 cars also initially used sliding skirts initially, not rigid ones.



2] Even old VW bettles where modified to become GE cars.
I don't know how effective they were though.


Links that might help.

http://prototyp.org/...eators_002.html
http://forums.autosp...w...=98037&st=0

[J. Edlund]

What were the differences between single seat cars and group C sportscars in tunnel design. Also was there an ideal angle for a tunnel or an expansion ratio? Finding data is really quite hard and I cannot find any detail on the reynard inverter etc, which is about the most modern application.

Sportscars ran with flat floors and venturi tunnels limited in size by the regulations. F1 cars ended up with something more like an inverted wing which ran the length of the sidepods. The Inverter is using a similar solution.

Some of the later sportscars did produce a lot of downforce, but they were also very draggy.

[Greg Locock]

The moving plane thing is a bit of a furphy, if the plane is moving at airpseed then the venturi is just a mirror image of the tunnel.

I'm assuming you've got Katz? fig 6-39 shows an optimum diffuser angle of 4 degrees, for Cd, tho for lift the more the better. Again 6-40 says the more the better.

[gruntguru]

2. If a competition vehicle has a speed range of 60-160kph is there any point in pursuing ground effect?

Yes. Even at 60 kph, a lightweight car with effective GE acting on a substantial floor area will obtain significant downforce.

[NeilR]

It is more accurate to say I had Katz...twice. Which just goes to show that I need to be more careful who I lend books too.
I've tried contacting Katz to no avail. Since regulations have moved away from tunnels it is difficult to find people who have worked with them in the past 10 years...hence my interest in the inverter.
For the application (Formula Libre hillclimb) drag is not much of an issue, particularly as 40% of the course is down hill:
Issue is that in direction there are 3 blind crests followed by off camber corners, another direction there is 4 blind crests and a significant dip that gets cars light. 1300cc cars are using WOT for approximately 30% of the track. So more downforce and more grip is potentially the way to go faster, hence questions on tunnels.

Edited by NeilR, 09 November 2009 - 00:09.

[Greg Locock]

It is more accurate to say I had Katz...twice. Which just goes to show that I need to be more careful who I lend books too.
I've tried contacting Katz to no avail. Since regulations have moved away from tunnels it is difficult to find people who have worked with them in the past 10 years...hence my interest in the inverter.
For the application (Formula Libre hillclimb) drag is not much of an issue, particularly as 40% of the course is down hill:
Issue is that in direction there are 3 blind crests followed by off camber corners, another direction there is 4 blind crests and a significant dip that gets cars light. 1300cc cars are using WOT for approximately 30% of the track. So more downforce and more grip is potentially the way to go faster, hence questions on tunnels.

Quick look suggests that 0.25 Cp should be attainable. So at 100 kph that is 30 m/s, stagnation pressure is 1/2*1.2*30^2=500 N m-2. So if you've got 3 square metres that is 3*500*.25=400N, or 40 kg. Hmm, not as much as I'd have thought. Given that the behemoth weighs a ton that's only worth 0.05g. On the other hand it is free.

Edited by Greg Locock, 09 November 2009 - 01:59.

[NeilR]

behemoth!...for a moment I thought you meant me. Yes the project car will have something in addition to gross weight to hold it on to terra firma.
However I was serious when I said Formula Libre. I was speaking to the Monash FSAE team and they took their car to the track and were 2sec off the record. I am surprised that 40kg is all.

[Greg Locock]

behemoth!...for a moment I thought you meant me. Yes the project car will have something in addition to gross weight to hold it on to terra firma.
However I was serious when I said Formula Libre. I was speaking to the Monash FSAE team and they took their car to the track and were 2sec off the record. I am surprised that 40kg is all.

Well, it all depends on how effective you can make the venturi. Some of the other designs were more like Cp=0.6 When I wrote behemoth I was thinking of your car (and wondering how the hell you'd fitted sidepods)

[Fat Boy]

This picture gives you an idea of the tunnel size of the old GTP cars



Basically, on what I've seen, you can approach tunnel design with a liberal dose of More's Theorem. More's Better.

[DaveW]

Basically, on what I've seen, you can approach tunnel design with a liberal dose of More's Theorem. More's Better.

Not my subject, but be wary of More's Theorem (with apologies to FB). Think expansion ratio (as the D/F builds, the chassis is pulled down & the venturi expansion ratio increases). At some point (airspeed) the flow will separate & D/F will reduce significantly. The result is normally called porpoising. The Lotus 80 porpoised (I recall) at <100 mph when it first ran (on relatively soft springs). It never raced because of that, but it did lead to the '88. On the other hand the '79 was brilliant. Tunnel airflow can be controlled by running soft springs & bump rubbers (the former to satisfy a minimum static ride height regulation, the latter being the "running" springs), but bump rubber characteristics & packer gaps are likely to be critical for stability. Might also want to think about the tunnel profile, because this will determine the centre of pressure.

[NeilR]

Well Greg, I went to the hill climb championships and had a moment of madness. I looked at the single seater cars made from pipe cleaners with 1100-1350cc engines and I thought...'a slower one of those would be fun to share with friends and a good, cheap way to learn some skills'. Then I met the FSAE guys, who curiously are using the same wing method I am since they read about it in the magazine. Anyway a mindless moment on ebay and I typed in 'indycar' and lo and behold a plethora of parts was before me. Since I can resist anything but temptation (apologies to O Wilde for the misquote) I ended up with a NOS Lola steering rack and a set of four double adjustable WP dampers for less than a pair of tyres (r-spec 185/60/14 if you must know). I contacted the engineer in the Indy team concerned and they said they have a box of 20 they want to get rid of. So if I can find WP seals etc then they will all go as in the mean time I have found two other interested people who would like something similar and we are discussing a barter arrangement. Fortunately one fellow runs an engineering business, so I will not have to make a chassis. The cars (if this goes ahead) will run a 600-750cc bike engine, Gemini front uprights and formula ford tyres (second hand from local teams they are $50 a set) and wings (foam/FG) to have a bit of fun. Such a car could be double or triple entered in local hill climbs and should be a hoot to drive.
But no rush I have plenty of other things to do...so just stockpiling parts.
The tunnel question comes partly from this, but also from the observation that all the FL cars were running a flat bottom despite tunnels being legal. I took the opportunity to ask Malcolm Oastler a couple of questions when he was free from running his special. The tunnel efficiency issue with sealing is well known, but then Tony Southgate had a unique solution at Jaguar with a high mounted LCA that bent over the tunnel. So with these thoughts and with Reynards inverter I thought I'd investigate them.
Yes DaveW CoP issues are legion, though I think the low speed may reduce the likelihood of it arising?

Edited by NeilR, 09 November 2009 - 08:40.

[Tony Matthews]

Not my subject, but be wary of More's Theorem (with apologies to FB).

I think the McLaren M28 suffered from too much 'More's'.

[cheapracer]

and a significant dip that gets cars light.

eh?

If you look through the online SCCA rulebook you can find the tub specs for the F Alantics from memory.

[NeilR]

eh?

If you look through the online SCCA rulebook you can find the tub specs for the F Alantics from memory.

Dip in the road/track. If you watch the video you will see the spot. Now called 'OH ****!' corner after the current Australian champion wrote of a car there...cars to not steer or change direction with wheels off the ground and landing with brakes on is a natural reaction but not helpful.

[Fat Boy]

Not my subject, but be wary of More's Theorem (with apologies to FB). Think expansion ratio (as the D/F builds, the chassis is pulled down & the venturi expansion ratio increases). At some point (airspeed) the flow will separate & D/F will reduce significantly. The result is normally called porpoising.

Completely true. I was overstating things a little. Never take me too serious....

There aren't too many modern racecars with tunnels to look at. The Lola and Reynard ChampCars were very limited by the rules on what they could do, but the tunnels were still very strong. The Panoz was much less limited and the tunnels for it were significantly larger, even though they stopped at the rear axle centerline. Things like the late 80's-early 90's GTP cars are amazing to me. The shear volume of those tunnels is intimidating.

For a small car, I think you'll run out of real estate before you stick too much tunnel on it. 4* is certainly not even close to getting you into trouble. You'd probably have to be 3, maybe 4, times that before even considering a 'too big' problem. To stick some number on it, at 12*, if your diffuser is 1.5 meters long, then you're at a 31cm tunnel exit height, which at first glance isn't excessive. It's only slightly above axle centerline. In fact, you might need to go higher or lower so you don't have to deal with the tunnel and the axle crashing into one another. Just another packaging issue to deal with. If you think there's a separation problem, stick a 10mm gurney on the exit. Tunnel gurneys can be very powerful.

[NeilR]

Well the rules allow a maximum of 1000mm past axle centreline for wing and tunnel. Given a 20.5 x 7 Hoosier slick, the axle line height of tunnel will be approx 140mm. Wings are free in No, so a two element wing could be placed down near the tunnel exit.

[McGuire]

2. If a competition vehicle has a speed range of 60-160kph is there any point in pursuing ground effect?

In theory, perhaps. But in practice, for that speed range, given finite time and money the resources would be more effectively spent elsewhere, I would think.

That said, the biggest tunnels I have ever seen on a serious race car were on a hillclimber, Rod Millen's Pikes Peak Toyota. A foot and a half square at the exits. Lee Dykstra did the aero. The course is around 12.5 miles long and the record is just over ten minutes, sooo....

[Fat Boy]

Lee Dykstra did the aero.

Meh, what does he know?

[McGuire]

Meh, what does he know?

Well, he took the job. What can you say.

An interesting aspect of Pikes Peak: the altitude of the course is 9400 ft at the start and 14,000+ feet at the summit, so the air density is not so much to write home about. Also, the road surface was gravel all the way up when the truck was built, though about half of it had since been paved the last time I looked. Not too many GE cars at Pikes Peak over the time I was paying attention. The more customary setup on the unlimited cars was giant wings like a Sopwith Camel.

[jeremy durward]

Well the rules allow a maximum of 1000mm past axle centreline for wing and tunnel. Given a 20.5 x 7 Hoosier slick, the axle line height of tunnel will be approx 140mm. Wings are free in No, so a two element wing could be placed down near the tunnel exit.

i assume you are going to run the car in formula libre? if so there is no rule in the cams manual for how far back you can go. there is a rule for formula 4000/3/ford etc but not for libre, the only rule in there for them is not to go over 900mm and a cockpit opening dimension rule

so basiacally go nuts with how far you want the tunnels to go

[Fat Boy]

Meh, what does he know?

This was tongue in cheek. Lee knows his aero. His Mazda GTP car, which IMO was a really nice looking car, had some pretty massive tunnels as well. I think you can be pretty aggressive with tunnel volume. As far as the speed range, the really important thing to know is what is your cornering speeds. If they're 120kph or more, then tunnel aero (any aero, for that matter) is going to really help. Below 80 kph, it's probably easier to rely on wings. It really is amazing at how wings come into play even at low speeds, though. It goes to show how we play around with just a small amounts of grip here or there to balance things out.

[jeremy durward]

Well, he took the job. What can you say.

An interesting aspect of Pikes Peak: the altitude of the course is 9400 ft at the start and 14,000+ feet at the summit, so the air density is not so much to write home about. Also, the road surface was gravel all the way up when the truck was built, though about half of it had since been paved the last time I looked. Not too many GE cars at Pikes Peak over the time I was paying attention. The more customary setup on the unlimited cars was giant wings like a Sopwith Camel.

Personally I’d look more to the English hillclimb cars for inspiration than Pikes Peak. And they run tunnels on their cars, in fact I’m told that when the British champion came out to Aus to compete he said that underbody aero work gets some really good gains over what you can get out of the wings.

And at the end of the day, will it hurt to try it out?

[cheapracer]

Dip in the road/track - with wheels off the ground

eh?

[Greg Locock]

eh?

Nah 'Straya, not Canadia.

[NeilR]

Thanks for the clarification Jeremy, I was not sure after reading the CAMS manual. So the bigger the better! I think the challenge will be to make them stiff and light.
I take the point about wings too.

[jeremy durward]

Thanks for the clarification Jeremy, I was not sure after reading the CAMS manual. So the bigger the better! I think the challenge will be to make them stiff and light.
I take the point about wings too.

well thats the way i read it anyway, light is the thing isn't it, like i said it can't hurt... but on the other hand it will if its heavy

1. For each automobile (1st, 2nd, 3rd, 4th, 5th and 6th Categories) coachwork shall be deemed to include all
external parts of the automobile which extend above the highest point of either the front or rear complete
wheels (with tyres) with the exception of units definitely associated with the functioning of the engine or
transmission and the roll bar.
Any specific part of the automobile which has an aerodynamic influence on the stability of the automobile
must be mounted on the entirely sprung part of the car and shall be firmly fixed whilst the automobile is in
motion.
Neither the roll bar nor any of the units associated with the functioning of the engine or transmission shall
have an aerodynamic effect by creating vertical thrust.
All external projections swinging in a horizontal plane shall have a minimum radius of 15mm. The leading
edge of any aerofoil fixed to the front of the automobile shall not be sharp.
Switches for battery isolation and fire fighting equipment may project beyond the coachwork without
infringing regulations.
3. 1ST CATEGORY
Each automobile shall comply with the following requirements:
3.1 No element of coachwork may exceed in height a horizontal plane situated at 900mm above the ground.
Neither the roll bar nor any of the units associated with the functioning of the engine shall be included.
Measurements are to be taken with the driver on board.

3.2 Each automobile of a type registered at 1 January, 1975, but constructed after 1 July, 1975; and each
automobile of a type not registered at 1 January, 1975, but constructed after 1 January, 1976; and each
automobile registered at 1 January, 1975, but which have subsequently been substantially varied; must all
comply with the following requirements:

Table for Art. 3.2
F4000 F2 F/Ford
Maximum width ahead of front wheels 1,500mm 1,500mm 950mm
Maximum width ahead of front wheels, above height of 1,100mm 1,100mm 950mm
wheel rims
Maximum width between front and rear wheels + 1,300mm 1,100mm +200mm 1,300mm
deformable
Maximum width behind rear wheels 1,100mm 1,100mm 1,100mm
Maximum front overhang 1,000mm
Maximum rear overhang (from centre of wheel/axle) 800mm 1,000mm

3.3 Each wheel shall be external to the coachwork.

3.4 Unless otherwise specified in technical regulations the coachwork opening giving access to the cockpit must
be at least 600mm long; and 450mm wide, maintained over 300mm from the rearward point of the seat
backrest towards the front. It must be able to be entered or left without it being necessary to open a door or
remove a panel. Sitting at his steering wheel the driver must be facing forwards. Moreover, the cockpit must be
so conceived that the maximum time necessary for the driver to get out does not exceed five seconds

[cheapracer]

Nah 'Straya, not Canadia.

I'm still tring to understand how to get 'air' in a dip........

[NeilR]

I'm still tring to understand how to get 'air' in a dip........

Well for everthing, including sex, there is a start, a middle and an end...of course for some these things happen so fast or indeed not at all that they don't notice the detail...so let me explain for the hard of thinking. There is a flat bit of road, the road starts to dip, the car gets light as the car enters the dip, the car gets heavy at the bottom of the dip and the car gets light again as it leaves said dip.

Edited by NeilR, 12 November 2009 - 05:43.