Internal lift and down-force

[ringo]

Do teams consider the lift that internal channels create?
On the other side of the coin, are internal channels purposely designed to increase down-force?

I think it would be interesting to look at this issue. It's possible a team is losing a lot of down-force on the car because there is a duct running from the roll hoop all the way down to the gear box.
The lift is caused by the reaction force of the air turning down into the car.
While down-force would be added by the opposite.

A few examples of cars that avoid steep downward ducting:



no high openings in sight. KERS and oil is cooled through the sidepods, no internal upward reaction.


a few that have downward ducts:


There is an upward reaction in both these high cooling solutions.

[Jersey Tom]

Team engineers are very well aware of internal and external aerodynamics. Is there some reaction of redirecting air? Sure.

Maybe a good example... of when one thing in isolation doesn't mean much, because it's much more important how everything else works around it.

[marekk]

They'll lose som downforce for sure, but not a lot. What's expected mass flow for the channels - a few kg/s ? (we already know, that this big one to the engine is about 0,4 kg/s)
Much more is lost IMO on rear wing, due to disturbance of the flow. It's a little dragy as well.
The only reason they'll make it, is because they have to. There's not enough space in R31's and Macca's sidepods.

[MeowMix]

Yes Marek, but its that 0.25% better performance or downforce that allows you to qualify on pole and win. Even if you qualify half a second ahead of someone in a 1:25 lap is a 0.5% in times.... even an extra 10 or 20 N of downforce will matter.

Back to the actual question however. I reckon the only solution to the problem is to use air from lower down in the car, but that would disrupt and take away air from the diffusers etc. Also, remember that they still would need a crash / roll structure to protect the driver's head creating just as much drag. It makes sense then (at least to me) to have the intake on top, even if it does create a small bit of lift. This allows as much air to get to the diffuser as possible, the extra downforce created by the diffuser would likely be greater than any lift created I reckon.

[Jersey Tom]

Yes Marek, but its that 0.25% better performance or downforce that allows you to qualify on pole and win. Even if you qualify half a second ahead of someone in a 1:25 lap is a 0.5% in times.... even an extra 10 or 20 N of downforce will matter.

Except when the 0.25% better performance you get by some different radiator arrangement costs you 1% in some other aerodynamic performance, or 100% reduction in performance when your engine overheats.

BIG PICTURE, guys. CANNOT just look at individual pieces and parts in isolation all the time.

[MeowMix]

Yes Marek, but its that 0.25% better performance or downforce that allows you to qualify on pole and win. Even if you qualify half a second ahead of someone in a 1:25 lap is a 0.5% in times.... even an extra 10 or 20 N of downforce will matter.

Except when the 0.25% better performance you get by some different radiator arrangement costs you 1% in some other aerodynamic performance, or 100% reduction in performance when your engine overheats.

BIG PICTURE, guys. CANNOT just look at individual pieces and parts in isolation all the time.

of course, but all im saying is that it cant be discounted if theres another solution, and that im sure that the teams look at it and try to reduce it as much as possible. if you read the rest of my response you would have realized I get the point of a compromise! See my reasoning for keep it there in the first place?

[Twaddle]

Don't they level off again so that they air exits parallel to the intake (and usually the free stream)? That makes them net neutral except for effects due to changes in the pitch of the car.

Edited for clarity.

[Sayshina]

Don't they level off again so that they air exits parallel to the intake (and usually the free stream)? That makes them net neutral except for effects due to changes in the pitch of the car.

Edited for clarity.

They're all designed to be filled to capacity at fairly low ground speeds, if that's what you mean. However, they all do change the direction of at least some of the flow they see, and Ringo is right they will all cause some lift because of that. But lots of parts on a car cause lift, including the cockpit.

Ringo, the Beneton springs to mind offhand. It had the airbox inlets just above the sidepods. Whenever you see alternatives tried for a few years and then abandoned, assuming no rule changes forcing said behavior, it's reasonably safe to assume the abandoned method to be inferior in some way. This isn't VHS vs. Betamax where people intentionally choose an inferior product because of marketing.

The higher up you place the airbox inlet the cleaner the flow it sees is likely to be. Cleaner better flow means you can get the mass you require with a smaller inlet, which you would expect to mean less drag. Placing it directly over the inlet trumpets would seem to be nice and elegant.

As for the McClaren KERS cooling solution, Gary Anderson has said he thinks it would allow better flow to the wing, presumably by shaving off the boundary layer.

Someone, possibly Mike Gascoyne, said a few years ago during the chimney and shark fin era that standard practice was to constantly reduce area in the sidepod after the radiator to gain a little thrust from the cooling flow. And we all know Audi recently made an LMP car that was all about flow through the car. So it's safe to assume designers are VERY interested in that area. Yet I can't recal anyone ever commenting on the need to prevent lift from shifting mass flow downward. There have been lots of aero elements on cars in the last couple decades that have intentionally directed flow downward and took the lift in order to condition the flow to a downstream part that allowed a net downforce gain.

As for putting the airbox inlet down low, have you never noticed how much trash they dig out of the radiator inlets at your average stop?

[riff_raff]

I would imagine that with the design of engine airbox and radiator inlet ducts, the more important issues are smoothly diffusing the flow to build pressure and creating a uniform pressure distribution at the engine inlets or radiator core face. Both airbox and radiator ducts also require the flow to change direction at the engine inlets or core face. I can't imagine the internal duct geometries could be altered sufficiently to affect downforce without impacting the other flow requirements.

With radiator duct design, there are many compromises to be made between things like core thickness, face area, inclination of the core to incoming airflow, duct length, flow losses, mass flow, etc. For a given heat rejection rate, a heat exchanger can have a small frontal area with a thick core, or a large frontal area with a thin core. The smaller frontal area core would allow the radiator to be more easily positioned normal to the airflow, thus simplifying the ducting. But it would also have a higher pressure drop across the core. The larger area core would need to be inclined to the incoming airflow, and would need a more complex duct to smoothly turn the airflow as it enters the core. But the thinner core would have less pressure drop.

Most current F1 cars seem to be using thin, large area cores. So that must be a better compromise. To get an idea of the radiator inlet duct geometry requirements, compare the duct inlet area to that at the face of the core. The change in cross section is quite large and must occur over a relatively short length, without producing excess turbulence.

[Twaddle]

Don't they level off again so that they air exits parallel to the intake (and usually the free stream)? That makes them net neutral except for effects due to changes in the pitch of the car.

Edited for clarity.

They're all designed to be filled to capacity at fairly low ground speeds, if that's what you mean. However, they all do change the direction of at least some of the flow they see, and Ringo is right they will all cause some lift because of that. But lots of parts on a car cause lift, including the cockpit.

Correct me if I'm wrong, but I think you're missing my point slightly. My point isn't that they don't change the direction of flow (they clearly do), but that they do so twice in equal and opposite directions. This only applies if the ducts are shaped somewhat like a stretched Z, like this ¯\_ if that's easier to visualise.

[ringo]

I would imagine that with the design of engine airbox and radiator inlet ducts, the more important issues are smoothly diffusing the flow to build pressure and creating a uniform pressure distribution at the engine inlets or radiator core face. Both airbox and radiator ducts also require the flow to change direction at the engine inlets or core face. I can't imagine the internal duct geometries could be altered sufficiently to affect downforce without impacting the other flow requirements.

With radiator duct design, there are many compromises to be made between things like core thickness, face area, inclination of the core to incoming airflow, duct length, flow losses, mass flow, etc. For a given heat rejection rate, a heat exchanger can have a small frontal area with a thick core, or a large frontal area with a thin core. The smaller frontal area core would allow the radiator to be more easily positioned normal to the airflow, thus simplifying the ducting. But it would also have a higher pressure drop across the core. The larger area core would need to be inclined to the incoming airflow, and would need a more complex duct to smoothly turn the airflow as it enters the core. But the thinner core would have less pressure drop.

Most current F1 cars seem to be using thin, large area cores. So that must be a better compromise. To get an idea of the radiator inlet duct geometry requirements, compare the duct inlet area to that at the face of the core. The change in cross section is quite large and must occur over a relatively short length, without producing excess turbulence.

Yes, but how about the gearbox cooler and the KERS coolers? We are seeing differing locations of the inlets. For instance the Renault and Mclaren inlets on level with the roll hoop. The should expect some lift with this option.

I don't think there is much that can be done with the radiator cores as you say. Maybe they could be slanted backwards, but i am sure F1 teams tried that and had other issues with the rest of the car. So the radiator ducting probably has little variation.
Though running the flow up the sides of the engine instead of through the side pod seems to be preferred option this year.

[riff_raff]

Offhand, it would seem that putting cooling ducts along either side of the airbox would not be efficient. The duct length is long, and one inlet side or the other gets disrupted airflow under yaw.

But F1 aero guys know what they're doing, so I'm sure it works. Aero design is all about getting the best compromise between lift & drag or thrust & drag. A good heat exchanger installation can have no net drag penalty, mostly due to the large amount of energy transferred to the cooling airflow.

Electric KERS would likely require liquid cooling of the motor permanent magnet materials and power switching electronics, due to their high power densities. But I would imagine engine oil could be used for that, so no additional cooler would be needed.

The trans lube oil would not need much cooling because there is not much efficiency loss or heat generated in the gearbox. Gearbox losses are likely less than about 3% of BHP at most. And if the trans housing is metal, there is lots of wetted surface area to transfer heat convectively (passively). Gear oil can also be safely used at much higher temperatures than engine oil, so the trans cooler core matrix can be smaller in size.

Engine coolant radiator installations are the most critical by far, since they potentially have the most energy to contribute. 15 or 20 percent of the fuel's energy content passes through the radiators, versus 30 or 35 percent through the rear wheels.

[Sayshina]

Don't they level off again so that they air exits parallel to the intake (and usually the free stream)? That makes them net neutral except for effects due to changes in the pitch of the car.

Edited for clarity.

They're all designed to be filled to capacity at fairly low ground speeds, if that's what you mean. However, they all do change the direction of at least some of the flow they see, and Ringo is right they will all cause some lift because of that. But lots of parts on a car cause lift, including the cockpit.

Correct me if I'm wrong, but I think you're missing my point slightly. My point isn't that they don't change the direction of flow (they clearly do), but that they do so twice in equal and opposite directions. This only applies if the ducts are shaped somewhat like a stretched Z, like this ¯\_ if that's easier to visualise.

Well, I was thinking specifically about the airbox path when I wrote the above, which does not see a "Z" shape to its flow. It is directed downward, then gets very complex inside the airbox. We know this because the airboxes are baffled. The flow then gets directed downward again into the inlet trumpets.

The flow shape of the sidepods is likely to look like 4 or 5 "Z" shapes strung out. Alternative cooling paths are probably almost as complex. It's unlikely any of these are downforce neutral, because 0 is a dificult number to achieve. It's also probably more likely they create a net lift than any downforce, simply because you would have other higher priorities.

Riff raff, I'm not so sure about placing additional loads on your engine oil. In this heavily restricted and homologated environment even a tiny miscalculation could wipe out your entire season.

[ringo]

This topic was actually pretty interesting. There was an image with the internal flow of the redbull car. Can someone post it again?

And to bring up riffraffs points:

Electric KERS would likely require liquid cooling of the motor permanent magnet materials and power switching electronics, due to their high power densities. But I would imagine engine oil could be used for that, so no additional cooler would be needed.

The trans lube oil would not need much cooling because there is not much efficiency loss or heat generated in the gearbox. Gearbox losses are likely less than about 3% of BHP at most. And if the trans housing is metal, there is lots of wetted surface area to transfer heat convectively (passively). Gear oil can also be safely used at much higher temperatures than engine oil, so the trans cooler core matrix can be smaller in size.

Engine coolant radiator installations are the most critical by far, since they potentially have the most energy to contribute. 15 or 20 percent of the fuel's energy content passes through the radiators, versus 30 or 35 percent through the rear wheels.

Riff Raff, can you pin point which elements of the KERS system require cooling.
The motor and batteries are the most obvious.

It would be good if the heat energy can be quantified as a percentage of the power output.

-We have the radiator at 20-30% of bhp
-gearbox at 3%
-KERS motor at ~15% ? of motor output.
-batteries at ?? % of motor output or discharge.
-KERS components at ??% of motor output or discharge/recharge.

[tommylommykins]

I also still don't see this one?

Every Action has an Equal and Opposite reaction implies that unless the exhaust air is moving in some particular direction up or down, you're not going to get lift or downforce at all? Regardless of lift (and ignoring pressure differences and heat differences), whatever happens internally (baffles, passing through the inside of the engineblock, etc.) if the air doesn't change direction on exit, then there's no overall effect..?