Post rigs

[Jersey Tom]

A lot of people, even professionally, are very naive as to the capabilities of computer simulation.

[F1_eng]

how do you come to this conclusion?

[Giblet]

Are any of the teams using an 8 post rig yet?

the 8th post is to load individual points of the car (ie one wheel).

You already control each of the individual corner loads with 3 DOF's (heave, roll, pitch)...

Unless you mean adding banking and warp stiffness effects?

That is exactly what I mean. An eight post rig. Like this one. To simulate aero loads or mechanical loads.

[youtube]http://www.youtube.com/watch?v=hqyUwCTIGQg[/youtube]

Are they using them in F1 yet, or is there any need?

[F1_eng]

yes

[Jersey Tom]

Are any of the teams using an 8 post rig yet?

the 8th post is to load individual points of the car (ie one wheel).

You already control each of the individual corner loads with 3 DOF's (heave, roll, pitch)...

Unless you mean adding banking and warp stiffness effects?

That is exactly what I mean. An eight post rig. Like this one. To simulate aero loads or mechanical loads.

[youtube]http://www.youtube.com/watch?v=hqyUwCTIGQg[/youtube]

Are they using them in F1 yet, or is there any need?

I guess I'm still not following what the 8th degree of freedom is exactly.

On a 7-post, each tire has one linear, vertical DOF. The rest of the chassis has a total of 3 more; heave, pitch, and roll. With that heave (vertical motion of chassis) DOF, you can add as much vertical "aero" load as you want.

What's #8?

[autogyro]

I think you will find that this American rig has four suspension/wheel actuators and four 'pull down' aero actuators and no roll or pitch actuators as such.
This is probably because of the great big heavy American thing sitting on it.
Moving that about any more would hardly be accurate IMO.

If this is the best that can be done, then I am not surprised that there is insufficient general data to establish a decent computer data base for future work.
If the same level of testing is common in wind tunnels, then it confirms the fragmented and ('convenient' for some) current methods of vehicle testing.

All it is doing is saving some track time by second guessing from track data.
I would sooner use a good test driver and work up the car.
American stock cars are practically the same anyway, so why bother with nerds?

[Giblet]

Maybe the video was a bad example, with the pads. Pads would work the same would the not for a wind tunnel with no rolling road?

I thought the 8th post would allow one corner of the car to be loaded (as opposed to one wheel, or the whole car from the top middle) to simulate cornering loads and aero loads.

I am just trying to understand as I read an article in Racetech about how new (rad:uncommon) 8 post rigs are, and how few wind tunnels have them but I am having trouble finding the article.

I was not trying to infer that NASCAR is weak.

[Giblet]

F1_eng

Thanks for the clear yes.

Where could I look to read about 8 posts in F1. Maybe an article needs to be produced.

Pleas don't respond:

"google"

[autogyro]

Maybe the video was a bad example, with the pads. Pads would work the same would the not for a wind tunnel with no rolling road?

I thought the 8th post would allow one corner of the car to be loaded (as opposed to one wheel, or the whole car from the top middle) to simulate cornering loads and aero loads.

I am just trying to understand as I read an article in Racetech about how new (rad:uncommon) 8 post rigs are, and how few wind tunnels have them but I am having trouble finding the article.

I was not trying to infer that NASCAR is weak.

Pads are fine for the very limited use such a rig as this has.
Even a rolling road surface in a wind tunnel plus the new F1 8 point rigs cannot simulate cornering tyre slip and the effects on roll,mass transfer and air flow this has.
How do you simulate all the factors in a controlled slide and from it work out where the car goes out of control and how suddenly?
Do that for any speed and add in braking distances and you might be getting somewhere.
Until then testing rules.

[Giblet]

I came up with this idea a few years ago.

http://www.halfbakery.com/idea/Rolling_ ... 1183663712

hmmm halfbakery should be right up your alley autogyro.

Rolling Road Indoor Go-Kart
I hope its not already thought of.

(+4) [vote for,
against]

This device would tie in to existing PC racing sims.
The track portion would be a rolling road (coveyer belt wooo) a few inches longer than the wheelbase of the go-kart corner to corner, able to rotate 360 degrees,and tilt 15 or so degrees.
The go-kart portion would have a ring around the rolling track portion, and the go-kart is rigidly mounted to this ring, which can also rotate and tilt the same amount.
The screen in front is the race sim you are tied into.
When you drop your clutch and launch your go-kart, wheelspin will be detected and the road will roll towards you, increasing speed as the tires stop burning out, as your car rushes ahead on the screen.
As you brake for the corner, the rolling road begins to slow. Fighting against the adhesion of the tires, a small puff of smoke comes off just as you give it too much brake while turning in.
When the wheel is turned for the corner, both the track and the kart rotate and pitch, with the angle of the car slightly steeper then the track going into the corner. Sliding could be produced by too much angle vs speed on the rolling road and kart, while showing you drifiting to the outside of the corner on the screen.
Most speeds and track configurations could be accounted for, while using a real cart to simulate real racing. Bumps could be simulated as well, either under the track or the whole machine bouncing up and down.
— Giblet, Jun 30 2007

[autogyro]

I came up with this idea a few years ago.

http://www.halfbakery.com/idea/Rolling_ ... 1183663712

hmmm halfbakery should be right up your alley autogyro.

Rolling Road Indoor Go-Kart
I hope its not already thought of.

(+4) [vote for,
against]

This device would tie in to existing PC racing sims.
The track portion would be a rolling road (coveyer belt wooo) a few inches longer than the wheelbase of the go-kart corner to corner, able to rotate 360 degrees,and tilt 15 or so degrees.
The go-kart portion would have a ring around the rolling track portion, and the go-kart is rigidly mounted to this ring, which can also rotate and tilt the same amount.
The screen in front is the race sim you are tied into.
When you drop your clutch and launch your go-kart, wheelspin will be detected and the road will roll towards you, increasing speed as the tires stop burning out, as your car rushes ahead on the screen.
As you brake for the corner, the rolling road begins to slow. Fighting against the adhesion of the tires, a small puff of smoke comes off just as you give it too much brake while turning in.
When the wheel is turned for the corner, both the track and the kart rotate and pitch, with the angle of the car slightly steeper then the track going into the corner. Sliding could be produced by too much angle vs speed on the rolling road and kart, while showing you drifiting to the outside of the corner on the screen.
Most speeds and track configurations could be accounted for, while using a real cart to simulate real racing. Bumps could be simulated as well, either under the track or the whole machine bouncing up and down.
— Giblet, Jun 30 2007

Very funny Giblet.
I did much the same when I built a three axis cockpit for one of my flight simulators years ago when the P2 chip ruled.
You are at least ten years out of date.
All I see today is more and more complex equipment for guessing.
Years ago we actually did things.

[Jersey Tom]

I think you will find that this American rig has four suspension/wheel actuators and four 'pull down' aero actuators and no roll or pitch actuators as such.
This is probably because of the great big heavy American thing sitting on it.
Moving that about any more would hardly be accurate IMO.

If this is the best that can be done, then I am not surprised that there is insufficient general data to establish a decent computer data base for future work.
If the same level of testing is common in wind tunnels, then it confirms the fragmented and ('convenient' for some) current methods of vehicle testing.

All it is doing is saving some track time by second guessing from track data.
I would sooner use a good test driver and work up the car.
American stock cars are practically the same anyway, so why bother with nerds?

You don't need specific roll and pitch actuators. That's what I'm trying to get at. 3 points define a plane, so with 3 vertical actuators on the sprung mass you can independently control any combination of roll, pitch, and heave. There's no need for a 4th.

I have to say though, in addition to being a little annoying, you're being really naive as to the capabilities and application of real world testing.. and what's possible and appropriate for computer testing.

And no. American stock cars are nowhere near the same.

[noname]

www.instron.com/wa/library/StreamFile.aspx?doc=403

Opel's test rig with rolling "road" under the wheels

[autogyro]

I think you will find that this American rig has four suspension/wheel actuators and four 'pull down' aero actuators and no roll or pitch actuators as such.
This is probably because of the great big heavy American thing sitting on it.
Moving that about any more would hardly be accurate IMO.

If this is the best that can be done, then I am not surprised that there is insufficient general data to establish a decent computer data base for future work.
If the same level of testing is common in wind tunnels, then it confirms the fragmented and ('convenient' for some) current methods of vehicle testing.

All it is doing is saving some track time by second guessing from track data.
I would sooner use a good test driver and work up the car.
American stock cars are practically the same anyway, so why bother with nerds?

You don't need specific roll and pitch actuators. That's what I'm trying to get at. 3 points define a plane, so with 3 vertical actuators on the sprung mass you can independently control any combination of roll, pitch, and heave. There's no need for a 4th.

I have to say though, in addition to being a little annoying, you're being really naive as to the capabilities and application of real world testing.. and what's possible and appropriate for computer testing.

And no. American stock cars are nowhere near the same.

If American stock cars are not exactly the same, they are as near as dammit.
The tighter the regulations the more the same they become. It is going the same way in F1.

It is true that you only need three actuators for roll,pitch and heave but the four corner connections on the rig shown are used primarily to simulate down force.
In theory these four could also be used for roll,pitch and heave but what about lateral slip?
You would need a rolling (conveyor) road surface that could be rotated laterally and continually moved sideways relative to the car to do that.

Not impressed, sorry.

[F1_eng]

I'm afraid I don't know where you could read about 8 posts in F1 Giblet, most F1 teams use 7 posts.

I haven't watched the video, no offence but most of the American race car stuff annoys me so I won't watch it, agricultural springs to mind in most cases.

Why would you want to try and simulate slip on a test rig like these? It has no bearing on what is trying to be achieved through the rigs. A lot of the work we under-take on the rigs is related to ride, things like hitting curbs and the like where component installation stiffness and damping properties are observed as well as different component responses transiently.

Why on earth would we buy this type of equipment for show? No-one really knows that we have an 8 post rig, so its no type of marketing ploy.
We can simply replay a whole track on the rig, and compare how the car behaviour on track varies to the rig. We are not blessed anymore with endless track testing plus drives will lie very often, we need to tell them which set-up will be quickest very often, not just listen to their feed-back.

If you can't imagine what you could do with a rig like this and how it benefits the performance of a racing car, perhaps you should start looking in to dynamic behaviour of systems and not look for flaws in the test rigs.