Post rigs

[Mystery Steve]

haha drivers are lying drivers are a burden..
sorry guys ,you are catching the cow at the wrong end.
It should read:
Drivers are the most tuneable device on any racecar.Your input to them will help or worsen the performance of the whole package.
If you are aproaching the situation with something like:
look here my data says this nut here should make the whole shitbox 1tenth faster and the driver comes back to you telling you the shitbox turned into something worse ..you are in the middle of a problem not asociated with engineering but more what is needed to make your hero tick..
I really and honestly think the approach of :hey i know the numbers ,if it don´t work the driver is a girl and I will not trust his words will never ever lead to success.at least not with this guy.Or not with you .It`s your choice.

As for everything you try to simulate ,the story is never complete as you may not
be able to even see all variables affecting your sim.So maybe on certain sims you are very close to getting the correct answer but maybe tomorrow you suddenly start to simulate in an area were all your calculation is overcast by a factor you are not taking into account and from one second to the other your results will go from correlation to useless crap .that is why the actual verifying of the sims in real world is so important.Of course you can verify during race weekends but then you are back to the early days in Motorsport ,empirical ways of approach.With a bunch of expirenced guys this will of course also have its place.

My point I think Markus. We can all guess but let's use the word 'Validate' as it should be used and not as an excuse for bad guesses.

What? Validate means to confirm, or verify, which is how we are using the word... and that is what marcush is saying. Do you have an actual stance on this, or are you just stirring the pot here?

[ringo]

Interesting thread, I am learning a lot here; any rigs out there with motor driven rollers for the wheel posts?
Th data could be more accurate to reality if previous data on vehicle speed and brake and throttle position on certain points on track could be mapped to motor speed control for the wheel posts; the posts having rollers in contact with the wheels similar to a dyno.
Having the rollers heated could also help; after all tyre rigidity and pressure vary with temperature, which also affects the spring rate of the tyre and hence the displacements and damping of the suspension.

[autogyro]

"What? Validate means to confirm, or verify, which is how we are using the word... and that is what marcush is saying. Do you have an actual stance on this, or are you just stirring the pot here?"

If it were possible to fully validate all the factors, there could only be one result.
The car would win.
Any other level of validation must include a percentage of guess work.
How then can you tell the difference?
If the pot is never stirred then we would still be racing steam cars on wooden wheels.

[Mystery Steve]

If it were possible to fully validate all the factors, there could only be one result.
The car would win.

Validation does not mean perfection. Yes, there is a percentage of guess work. And the more you can validate the simulation data against real world data, the more value it holds.

And even if you could "fully validate all the factors" and have a "perfect" vehicle simulation, it would by no means automatically make for a winning car. The "perfect" vehicle cannot exist because there is a trade-off of variables, and increasing performance in one aspect usually means a decrease in another area of performance. That is why at the end of the day, it is the responsibility of the engineers to develop the tools that are most useful to them and to use them aggressively with an appropriate level of discretion while being mindful of their limitations.

[Mystery Steve]

Without going in to specific details of our test programs and sequences, there's not much more I can be say than the above post really.

Out of curiousity, are there any variables that F1 teams use post rigs to examine that I didn't mention? I talked about pitch/heave control because they were in the paper. Is roll control examined? I'd imagine that might be trickier to nail down mathematically, but I'd have to think about it more to be sure.

And is the damper matrix method used on playbacks of data from track tests? I wouldn't expect you to reveal any company secrets, but maybe you can speak in generalities?

[Belatti]

I will end my posting on this thread here, there doesn't seem to be any direction.
Regards

Its a pity all this happens when Im on a racing weekend (when I usually dont read nor post) and some archaic mind ruins a perfectly technical thread with their "in the past everything used to be better" approach.

Go now please autoguru and win an F1 championship without CDF, FEA and a 7 post rig. Make sure your design has the correct amount of helixes driven by your magic gearbox.

@F1_eng: I would like to give the direction deserved to this thread but first Id have to sit down, read some more (thanks for those papers Steve ), learn a little bit more about Matlab[/b] (I started learning it with simulations -clasical 1/4 car model- and also validating my calculated natural frequencies with damper sensors data using FFT) and actually go with a car to a rented 4 post rig.

And thats motive behind me starting this thread: Ill go this summer with one car to a post rig. My teams engineering group is formed by.... ME! -a 27y.o. mech Eng that have been working professionally in motorsports for about 6 months

Right now my focus is on learning the strategies for calculation used in the rig. Thanks to many formers I have learn that there different approaches, like using track recorded data or simply introducing inputs.
Im looking to re-learn the math behind all these, too .





PS: sorry for the bad mood of this post, but I cant get people that doesnt like "x" and keeps bashing "x" in 5 different threads shooing the really interesting guys from the forum.

[autogyro]

Sorry Belatti but you really should open your mind further.
F1-eng says he can find no direction on this thread.
With his experience and knowledge, perhaps he should supply such direction for the benefit of those less fortunate.

I am not 'knocking' the use of rigs or anything else.
Just warning about accepting it all as gospel.

So far the 'American 8 point rig shown has been discounted by some and on the F1 rigs, the jury is out.

I was looking forward to more sensible discussion from the 'very' clever guys posting on here.

[Jersey Tom]

It's not that the 8-post is discounted. Just a different application. Having read up on it, here's the facts.

8-post rig: A 7-post with an additional frame loader for warp effects. In theory you only need 7 posts to do all your work, and on an open-wheeler with a very stiff carbon tub, that works just great. On a more compliant frame, having the 4th load point lets you put some twist on the frame essentially. Has nothing to do with aero.

Needed in some race series, not needed in others.

[autogyro]

It's not that the 8-post is discounted. Just a different application. Having read up on it, here's the facts.

8-post rig: A 7-post with an additional frame loader for warp effects. In theory you only need 7 posts to do all your work, and on an open-wheeler with a very stiff carbon tub, that works just great. On a more compliant frame, having the 4th load point lets you put some twist on the frame essentially. Has nothing to do with aero.

Needed in some race series, not needed in others.

So these big American oval circuit cars don't create any downforce (or updraft!) effect on the suspension and tyres then?

[xxChrisxx]

"What? Validate means to confirm, or verify, which is how we are using the word... and that is what marcush is saying. Do you have an actual stance on this, or are you just stirring the pot here?"

If it were possible to fully validate all the factors, there could only be one result.
The car would win.
Any other level of validation must include a percentage of guess work.
How then can you tell the difference?
If the pot is never stirred then we would still be racing steam cars on wooden wheels.

I really don't get how you CANT understand what validation means. You are doing it here and you have done it on the CFD thread. All it means is checking yours and the computers sums with reality to make sure they are correct.



It's possible to have your sums right and validated but a --- design. That means you are a bad engineer.

It's also possible that you dont valide the sums, and design to incorrect input data. This does not mean you are a bad engineer, jus tthat you were an idiot for not validating.

The third possiblity is that you fluke a good design on unvalidated data, and are hailed as a hero. This means you are a lucky sob.

PLEASE NOTE: These sums ARE NOT TRIVIAL. The answers have to be found experimentally.



I'll give you an example:
We come up with a computer model for how high a ball bounces after dropping.

We first VERIFY the model. This means you make sure it works correctly, and doesnt give a blantantly stupid answer. So if it says the ball bounces higher than you dropped it we know something is wrong.

After its verified to work. We then VALIDATE. This would involove dropping a real ball on to a real floow and measuring the results.

We then compare data. If the ball in the computer model bounces too low compared to the real ball, we know that some variables need tweaking to ge the model correct. We can then estimate model accuracy over a range, and come up with a domain that the model is valid for.



The above... is science.
Correct design decisions are an art.
Engineering is a beautiful mix of the two.

[Jersey Tom]

It's not that the 8-post is discounted. Just a different application. Having read up on it, here's the facts.

8-post rig: A 7-post with an additional frame loader for warp effects. In theory you only need 7 posts to do all your work, and on an open-wheeler with a very stiff carbon tub, that works just great. On a more compliant frame, having the 4th load point lets you put some twist on the frame essentially. Has nothing to do with aero.

Needed in some race series, not needed in others.

So these big American oval circuit cars don't create any downforce (or updraft!) effect on the suspension and tyres then?

Stock cars do indeed create downforce.

The 8th connection is not for aero loading. Just isn't.

3 points define a plane. With a 7-post you use 4 posts for the tires (one each) leaving 3 for the sprung mass. With those 3 alone you can add any amount of "downforce", lateral or longitudinal load transfer you want. By moving the whole chassis purely in a vertical direction you can adjust the amount of total vertical load on the tires, and then adjust the front-to-rear split with pitch.

[autogyro]

Sorry but now someone is saying that aero effects are taken into account.
Which is it fellas?
Of course three points can be used to load the suspension and tyres.
Four is a bonus and needed on a not so rigid and heavy chassis.
Aero or not aero?

[Mystery Steve]

Sorry but now someone is saying that aero effects are taken into account.
Which is it fellas?
Of course three points can be used to load the suspension and tyres.
Four is a bonus and needed on a not so rigid and heavy chassis.
Aero or not aero?

Yes, aero is taken into account. They determine from the speeds on the track how fast it was going and thus how much aero loading should be applied.

[marcush.]

but then you got an aero map which is heavily influenced by dynamic ride height ,so the loads especially for flat bottom cars could and will change dramatically in certain areas of the map ,also effects like porpoising will surely be hard to get a handle on ,as the ride height changes and so does the aeroforce effectively the actuator has to act like a spring there, right?

[Jersey Tom]

Sorry but now someone is saying that aero effects are taken into account.
Which is it fellas?
Of course three points can be used to load the suspension and tyres.
Four is a bonus and needed on a not so rigid and heavy chassis.
Aero or not aero?

I don't recall anyone saying that aero wasn't taken into account. It is. That's just not what the 8th post is for.