How can full size cars be tested in so small A2 wind tunnel?

[g-force_addict]

Weren't cars or models supposed to occupy only a small fraction of the wind tunnel cross sectional area?
Yet in the A2 wind tunnel
http://www.a2wt.com/

cars use about 1/3 of the tunnel width and height


How do they avoid interference from the tunnel walls?
I read somewhere that they adjust the tunnel ceiling height or shape to the shape of the vehicle.

[n_anirudh]

a) There will be higher blockage and this will affect Cd Cl values
b) Most tunnels have an adjustable roof which can be changed to accomodate larger vehicles.
c) maybe they were testing internal aero in this particular case?

[Tommy Cookers]

a) There will be higher blockage and this will affect Cd Cl values

the F1 Jaguar stalled its rear wing on track at an AoA that the wind tunnel said was just fine (ask Eddie Irvine)
this was because the tunnel was too small for the model, ie just too small

naval aircraft operate unusually close to the stall in carrier approaches
public domain sources could show which ones have had problems consistent with deficiencies in tunnel size and Re number
the wind tunnel is a design tool whose cost, value and availability attributes may be competing with those of other design tools

[Jersey Tom]

How? Roll the car in, close the door, turn on the fans.

Doesn't mean that it's making good data, or that it can [correctly] answer every question you throw at it.

[gixxer_drew]

With a static floor and boundary suction, everything is a generalization anyway. You can still go faster from what you learn. What level are you developing at?

You designing front diffusers or cooling systems? Big difference in what you need to go faster at the end of the day. Taking an axe to your aero or a scalpel?

[Pierce89]

A2 works because they mostly use it for stock cars and they have developed fudge factors. I worked for a small Dirt Late Model team and we made good gains in A2. They also do some bicycle stuff there. My team had to wait when we arrived then out walks Josh Hayes so they apparently do superbikes too.

[g-force_addict]

a) There will be higher blockage and this will affect Cd Cl values

the F1 Jaguar stalled its rear wing on track at an AoA that the wind tunnel said was just fine (ask Eddie Irvine)
this was because the tunnel was too small for the model, ie just too small

naval aircraft operate unusually close to the stall in carrier approaches
public domain sources could show which ones have had problems consistent with deficiencies in tunnel size and Re number
the wind tunnel is a design tool whose cost, value and availability attributes may be competing with those of other design tools

Did CFD say the wing would not stall as well?
I guess it did as CFD usually precedes wind tunnel testing.

How close walls have to be to affect model aerodynamics?
Do circuit walls and tire barriers affect aero as well? I guess they do so the question is what is the distance threshold. Specially interesting in Monaco.

With a static floor and boundary suction, everything is a generalization anyway. You can still go faster from what you learn. What level are you developing at?

You designing front diffusers or cooling systems? Big difference in what you need to go faster at the end of the day. Taking an axe to your aero or a scalpel?

Did you mean with a static floor and WITHOUT boundary layer suction?
So it looks like boundary layer suction is also good for the wind tunnel walls and ceiling, not only for the floor?

[P.S.]

Usually you should not exceed 5% blockage in the test section. Otherwise it will effect the simulation. But this rule leads to a very large wind tunnel with large costs. So there is a pretty big interest in decreasing the wind tunnel size.

Lots of wind tunnels for cars are build with a blockage of 10% still getting decent simulation results.
F1 wind tunnels are designed large enough staying inside the 5% limit to avoid any risk.

Probably the idea behind the A2 is to adapt the tunnel walls in the same way the air would become displaced in free stream. Some smart guys have shown in the 80s or so, that it is possible to get the same aerodynamic results increasing the blockage up to 20% with this technic. But you need to adapt the walls for every car apart if you want to be very correct.

If they don´t adapt the tunnel walls for each car properly, the error is probably not that big. I would say, especially if you have a small budget, this tunnel can be a nice choice testing full scale stuff.

[gixxer_drew]

Usually you should not exceed 5% blockage in the test section. Otherwise it will effect the simulation. But this rule leads to a very large wind tunnel with large costs. So there is a pretty big interest in decreasing the wind tunnel size.

Lots of wind tunnels for cars are build with a blockage of 10% still getting decent simulation results.
F1 wind tunnels are designed large enough staying inside the 5% limit to avoid any risk.

Probably the idea behind the A2 is to adapt the tunnel walls in the same way the air would become displaced in free stream. Some smart guys have shown in the 80s or so, that it is possible to get the same aerodynamic results increasing the blockage up to 20% with this technic. But you need to adapt the walls for every car apart if you want to be very correct.

If they don´t adapt the tunnel walls for each car properly, the error is probably not that big. I would say, especially if you have a small budget, this tunnel can be a nice choice testing full scale stuff.

I think that rule of thumb comes from aircraft stuff for high downforce ground vehicles I think its really defined by the upwash. Sometimes you need a huge vertical upwash area like what was being discussed in the monaco tunnel thread. The sidepods can be tricky as well with requiring a lot of width to the tunnel as well and it all depends on the designs in play. I think this is where most teams get into trouble with the tunnels. Everything is riding some engineering compromise and the design changes but the tunnel does not...

It is unfortunate to me that the cost and complexity of tunnels has gotten to where people can no longer design their own tunnels. Much understanding is lost between designer and users especially when selling has to go on to make the deal happen.

[wesley123]

a) There will be higher blockage and this will affect Cd Cl values

the F1 Jaguar stalled its rear wing on track at an AoA that the wind tunnel said was just fine (ask Eddie Irvine)
this was because the tunnel was too small for the model, ie just too small

Did CFD say the wing would not stall as well?
I guess it did as CFD usually precedes wind tunnel testing.

Jaguar was 2004, with Irvine it was 2002. There was CFD, but it wasn't employed at such a scale as it is now.

[thisisatest]

the compromises are definitely there. that's why they built the larger tunnel next to it.
someone brought up cost. the a2 is around $400/hr. i believe the main one is $1200.
specialized bicycles recently completed their own wind tunnel...
http://www.specialized.com/us/en/news/latest-news/15948

[riff_raff]

According to the A2 website, the tunnel is only capable of producing airflow velocity of 85mph.

[g-force_addict]

According to the A2 website, the tunnel is only capable of producing airflow velocity of 85mph.

Would the close placed walls make it worse at a higher speed?

[riff_raff]

The results from wind tunnel tests are usually assessed on a qualitative basis. First, a baseline is established for the test configuration. And then a series of test cases are run with a specific variable changed for each run. Then the results for each test run are compared to the baseline to determine if performance is better or worse. If the testing only seeks qualitative changes between test cases under similar test conditions, then there may not be an issue with the tunnel used in the example noted.