Benefit of Pilette suspension design

[Jersey Tom]

This design has a goal, unknown at this time to us, that required the use of a strut system inside the wheels. These struts could well be an engineering or materials problem that was never solved correctly. That does not negate the value of the underlying goal.

Underlying goal of a suspension... I separate into 3 items:
1. Control of lateral (or more properly yaw) dynamics
2. Body attitude control
3. Control of vertical dynamics

They're all intertwined but in my mind those are the general goals. These are unchanging, and chassis engineers merely have ideas for different approaches to meet those goals to an appropriate level for the given platform and end user / consumer.

The realist in me takes a look at this - a one-off F3 design from almost 20 years ago - and is pretty comfortable in writing it off as a failed approach and/or inappropriate goal spec (the relative importance of items 1, 2, and 3 to each other). We can stare at this thing all we want, but bear in mind there's nothing inherently there to say that any of this design process was a good idea. Don't get your hopes up too much. Also, to be fair, nothing inherently there to say it's junk other than the end result (which does speak strongly to me).

It seems what you are proposing is that we crawl into the mind of this chassis engineer and attempt to figure out what they were thinking when they came up with this design. I have a hard enough time understanding what's going on in my mind, much less someone else's. But doesn't mean we can't dive into it.

As far as the actual chassis is concerned, there's not much magic to Item 1 on a vehicle as linear as this. Quasi-steady state distribution of tire loads, inclination and steer angles is relatively straight forward to analyze within the constraints of a double a-arm suspension. Tires dominate this area more than anything anyway.

I'd go so far as to say that there isn't an awful lot of magic in the coupling of Item 1 and Item 2, particularly the geometric anti's with jacking coefficients. They are what they are, can achieve what you want by moving "roll centers," adjusting anti-dive rates, etc.

That leaves Item 3, which for me is harder to "eyeball" mostly in that I don't have much working experiencing. By changing the distribution of masses (sprung vs unsprung) as well as how the various force elements (corner springs, ride springs, dampers, bars, inerters if you have them) act you can change the dynamic response of the platform with regard to heave, pitch, and roll inputs. This is by far DaveW's realm more than it is mine.

Now that we know this though, if we circle back is this design really that unique in a fundamental sense? The pushrod ARB thing really fundamentally different than a conventional T-bar. Just acts directly rather than through a bell-crank. The only unique thing is what's going on with the chassis-side mounts of those pullrods and whether they are fixed or not. Hard to see. One would have to think so, though... to fix that wheel-side UCA point and let the shorty (shawty) coil spring become the main ride spring. I suppose that would give you the potential for some different kinematics, but the travel of that spring is so damn short it might not amount to anything. Hell maybe the whole thing just goes into coil bind at high speed, like landing on a bump-stop.

So what effect does fixing the wheel-side UCA point and putting the main ride spring within the wheel have? Who knows. With a given overall ride rate and sprung mass, I'd guess your primary ride frequency is no different. It's an easy damn thing to tune anyway. Some second order ride dynamics effect by moving weight to-and-from the body and wheel assembly? At the end of the day this --- really does boil down somewhat simply. Couple springs and couple masses at each corner, and some amount of direct linking between them (ARB).

Could very well be just an overly complicated approach to the problem, with too much time spent chasing something low on the "bang for your buck" list. In a series like F3, chasing second and third order things probably isn't the best use of resources.

[Onch]

... Hell maybe the whole thing just goes into coil bind at high speed, like landing on a bump-stop.
...

There has been mention of the Lotus 88 earlier in the thread, isn't this the F3 incarnation of the concept?
The 'knuckle springs' go coilbound at reasonably low speeds and from that point on aero loads are fed directly to the tyres.
There is then a conventional suspension activated through the pullrods, and only the ARB design is a bit exotic with separate pushrods (impossible to package the ARB at the bottom of the tub?).

[scarbs]

I've pinged a message off to Martin Ogilvie who was the designer, lets see if he responds.

[Jersey Tom]

... Hell maybe the whole thing just goes into coil bind at high speed, like landing on a bump-stop.
...

There has been mention of the Lotus 88 earlier in the thread, isn't this the F3 incarnation of the concept?
The 'knuckle springs' go coilbound at reasonably low speeds and from that point on aero loads are fed directly to the tyres.
There is then a conventional suspension activated through the pullrods, and only the ARB design is a bit exotic with separate pushrods (impossible to package the ARB at the bottom of the tub?).

It's easy enough to do that with a conventional setup. Adjust your packers on the dampers such that you come down and arrest the further travel of the car by sitting on bump stops.

...not that that is always the best approach.

[hardingfv32]

It's easy enough to do that with a conventional setup. Adjust your packers on the dampers such that you come down and arrest the further travel of the car by sitting on bump stops.

This pretty much sums up the thread. Why go to the trouble of doing this exotic suspension when everything, based on our observations, can be accomplished in a much simpler manner?

My vote is that it is all about packaging for better aero.

Scarbs, demonstrate your genius and voice opinion on the reason for the design BEFORE you get the answer from the designer.

Brian

[Pierce89]

... Hell maybe the whole thing just goes into coil bind at high speed, like landing on a bump-stop.
...

There has been mention of the Lotus 88 earlier in the thread, isn't this the F3 incarnation of the concept?
The 'knuckle springs' go coilbound at reasonably low speeds and from that point on aero loads are fed directly to the tyres.
There is then a conventional suspension activated through the pullrods, and only the ARB design is a bit exotic with separate pushrods (impossible to package the ARB at the bottom of the tub?).

In the Lotus 88 the body which fed aero loads into the uprights was seperate from the chassis that used the conventional suspension. I don't think this f3 car is that extreme. I believe they just moved the suspension into the uprights for aero packaging(narrower chassis ?)

[DaveW]

My vote is that it is all about packaging for better aero.

I believe they just moved the suspension into the uprights for aero packaging(narrower chassis ?)

If that is the case, then (forgive me) the project might stand as a case study for educating vehicle aerodynamicists on the importance of compromise.

At its first race at Hockenheim Jeremy Dufour walked out from the team in disgust when it was eight seconds off the pace. At the end of the year the Pilette name vanished from the scene

[ubrben]

I think if you look back to that era you had a vast array of mechanical design concepts. In the days of minimal 3D CAD and simulation you had designers who could throw these concepts at cars without any evidence that they were better. Generally they seem to have been able to do this on reputation.

What you see is a visual manifestation of one persons idea of how to make a car fast. In reality most of these ignore the reality that a simpler concept, but well integrated, is generally more successful.

For a good discussion about a systematic approach to design (as opposed to the individual artisan approach exemplified by cars like the Pilette) see here:

http://fsae.com/eve/forums/a/tpc/f/1256 ... /217101453

How you approach the design process is more important than the individual design concept.

Ben

[ubrben]

Incidentally the modern equivalent of the artisan designer sterotype is the complete obsession about being "data-driven" from simulation tools, whilst completely ignoring whether the result is a drivable race car.

I've spent too much time with laptop wielding rocket scientists insisting a given rake is "optimum" for downforce generation who completely ignore the driver, who is quietly (and sometimes not so quietly) trying to articulate that the aforementioned rake has reduced directional stability on braking to the extent he can't drive into the corner.

This is generally followed by much head scratching about why the lap-sim predicted times are nowhere near being achieved. The next step after that is to blame the tyres, engine, track condition, etc....

This scapegoat bingo from aero or control systems people is a big part of why Honda, Toyota and BMW are no-longer in Formula 1.

In the age of CFD, FEA, etc, the myth that there's an equation or formula for everything is a massive problem.

Ben

[hardingfv32]

I think if you look back to that era you had a vast array of mechanical design concepts. In the days of minimal 3D CAD and simulation you had designers who could throw these concepts at cars without any evidence that they were better. Generally they seem to have been able to do this on reputation.

What you see is a visual manifestation of one persons idea of how to make a car fast. In reality most of these ignore the reality that a simpler concept, but well integrated, is generally more successful.

Ben

In the case of the Pilette, I am going to assume that the main goal was better aero. The suspension design would only be required to be of equal performance to a standard design. For reasons we discussed, there are some very big challenges in making this suspension design function at an equal performance level. This type of suspension had never been done before. How is the designer to know that the challenges could not be overcome. Isn't this usually the case with designs that push systems to the next level?

This could be a one man design project and we do not know that he did not if fact use a systematic approach. A systematic approach does not guaranty success does it?

A simpler concept is not always the answer. Simple solutions do not always satisfy the
requirements of complex problems.

In this case we are judging Martin Ogilvie's design without all the facts. About all we know for sure is that the car was slow. If that is a very good, but simple criteria.

Brian

[ubrben]

How is the aero better? It has two links in the airstream - a push and pullrod in addition to the wishbones. How is that better than just picking one or the other and having the springs, dampers and ARBs inboard.

Ben

[hardingfv32]

Would a slimmer body be a valid aero benefit? Seems to be valid in F1.

Also, remember I was using better aero as an premise for the design. I am drawing the conclusion that the overall design makes sense IF you accept the aero premise.

Brian

[Jersey Tom]

Would a slimmer body be a valid aero benefit? Seems to be valid in F1.

Also, remember I was using better aero as an premise for the design. I am drawing the conclusion that the overall design makes sense IF you accept the aero premise.

Brian

You are making the assumption that this design allowed for a slimmer body, in addition to slim being better. I may give you the latter, but perhaps not the former. Would have been conceivable to package the ride springs and dampers inside the existing monocoque structure as had been done in pro open wheel previous to this design.

To be honest Brian it comes across that you are not being objective in this "analysis." Continually seems like you are latched onto the idea that just because it was done, this must have been a good idea to pursue and we're just not seeing the reason why.

Awfully big assumption.

[hardingfv32]

[
To be honest Brian it comes across that you are not being objective in this "analysis." Continually seems like you are latched onto the idea that just because it was done, this must have been a good idea to pursue and we're just not seeing the reason why.

Awfully big assumption.

To be very precise it would be my premise that:

A) Martin Ogilvie was/is an accomplished designer.
B) Count Van der Stratten was an accomplished car owner who knew a "reasonable" design when he say it.

It would be my conclusion that "they" thought it was a good idea and yes, we have not determined why "they" thought is was worth pursuing.

Is this not a valid assumption?

Brian

[Jersey Tom]

[
To be honest Brian it comes across that you are not being objective in this "analysis." Continually seems like you are latched onto the idea that just because it was done, this must have been a good idea to pursue and we're just not seeing the reason why.

Awfully big assumption.

To be very precise it would be my premise that:

A) Martin Ogilvie was/is an accomplished designer.
B) Count Van der Stratten was an accomplished car owner who knew a "reasonable" design when he say it.

It would be my conclusion that "they" thought it was a good idea and yes, we have not determined why "they" thought is was worth pursuing.

Is this not a valid assumption?

Brian

Many things seem like a good/reasonable idea at the time. Doesn't mean that they are. I don't care who the person is or what their history entails. Not everything that Ross Brawn or Adrian Newey comes up with is going to be gold either. You have an idea for something, you try it, maybe it works maybe it doesn't.

Given that this type of topology didn't go out breaking any speed records and wasn't used subsequently, I'm going to make the assumption that it wasn't the best design solution.

I think trying to get into someone else's head and figure out their thought process from years ago is often an exercise in futility.