Push Rod Flex

[gato azul]

sorry it's a bit OT, but one could look at the "Archer paradox" to see some similar oscillation in a bow arrow (straight rod) during launch/initial flight.

a high speed video can be seen here

BTW:
I would disagree with the observation that this is the unloaded (inner pushrod) oscillating. From the video it can be seen
that the car uses the outer curb exiting a right hand corner, same as the Red Bull. ( you can see the apex curb at the beginning
of the clip in the background)
Therefore I think that we are talking about a fully loaded pushrod. The load will fluctuate due to the curb (rumble strip) shape,
and I guess the inerter will (try to) oppose the deflection of the spring/damper unit, letting the tyre deal with the "brunt" of the
initial impact(s).

[DaveW]

... and I guess the inerter will (try to) oppose the deflection of the spring/damper unit, letting the tyre deal with the "brunt" of the initial impact(s).

An inerter, if one is used, would probably be acting across the rockers (heave & pitch). This is likely to complicate the response to what is essentially a single-sided input.

[riff_raff]

.....If the push rod is completely symmetrical, and the ends are perfectly pinned, then the push rod should not bend, not until the Euler compression load is exceeded, anyway......

If the "bending" is due to a structural modal response, it may occur even if there is not net compressive force being applied. This "S-wave" bending can occur in the pushrod even in a completely unconstrained (free-body) state.

As for the other comments regarding the structural vibration modes of composite parts, I can't give provide a good answer, since I'm not an expert in this subject. All I can offer is that I believe composite structures exhibit non-linear properties in this regard, so it would seem to be quite difficult to predict these modes by analysis.

slider

[ringo]

I'm late to the party. But the flex in a pushrod, should it be so much times minsucule compared to the suspension movement or stiffness?
I don't think it will show considerable buckling unless the acceleration is considerable enough the rod compresses before the suspension even can dream of moving. And that's a long shot.

[marcush.]

.....If the push rod is completely symmetrical, and the ends are perfectly pinned, then the push rod should not bend, not until the Euler compression load is exceeded, anyway......

If the "bending" is due to a structural modal response, it may occur even if there is not net compressive force being applied. This "S-wave" bending can occur in the pushrod even in a completely unconstrained (free-body) state.

As for the other comments regarding the structural vibration modes of composite parts, I can't give provide a good answer, since I'm not an expert in this subject. All I can offer is that I believe composite structures exhibit non-linear properties in this regard, so it would seem to be quite difficult to predict these modes by analysis.

slider

I´d go even one step further the unloaded/unconstrained part should have an easier time to show bending with any visible amplitude when excited .As soon as you constrain both ends there is not much in terms of amplitude available as the the part cannot change its length .
The interesting bit is why would the pushrod flex and bend instead of the dampers dooing their job ? this hints at a rediculously stiff initial damper response (transmissibility)it reacts as if it were a fixed rod with ultimate stiffness.

[Jersey Tom]

The interesting bit is why would the pushrod flex and bend instead of the dampers dooing their job ? this hints at a rediculously stiff initial damper response (transmissibility)it reacts as if it were a fixed rod with ultimate stiffness.

The wheel rates of a high downforce open wheel car are quite high to begin with. I still don't think this has anything to do with the dampers or inerters or whatever (though they certainly will add to higher dynamic loads). The pushrod doesn't care where the load comes from; you exceed a load, it buckles.

You could take that car and put it on a pull-down or K&C rig and start gradually applying vertical load to the suspension. It will ever so slowly travel through its operating range. Then you hit a crossover point, and the pushrod buckles (or some component fails).

Here's how I see the series of events:
1. The design group is in charge of making lighter components.
2. Designers get a dynamic load trace of expected front pushrod loads from a multibody simulation. They see there is a max load of 'X'
3. Designers draft a pushrod with some factor of safety of 1.05 * X (or 1.10 * X or whatever)
4. For whatever reason, on track that FOS is exceeded. I could be because...
(a) There's some variation in manufacturing where the component was built out of tolerance, or the tolerance was too broad
(b) Just by virtue of the fact you will never have a 100% accurate prediction of track performance, loads were higher than the FOS afforded
(c) An incorrect design criteria was specified.. for example designing for max stress and not fully capturing the potential for buckling

The fact that the pushrod buckled is inconsequential IMO. It's just one type of failure mode (in this case a stiffness failure rather than an overload failure and a shattered part). Fundamentally no different than Red Bull blowing out front suspensions a couple years ago and having to tell their drivers to take it easier on the kerbs. You're trying to come as close as possible to a minimum practical safety factor to get weight out of the car... sometimes you go a little too far.

[olefud]

Strad’s slow mo of Button’s episode shows the tires,front and rear, in what appears to be Whirl Mode resonance. This can be pretty violent. Most of the components that I thought were markedly moving only appear to be doing so –the tire is the big mover. If the push rod is lightly loaded, it, the brake duct etc cold be fluttering in response to the tire gyrations.

[DaveW]

Sadly, nobody has confirmed (or otherwise) my attempt to estimate of the time expansion ratio of the movie clip.

The interesting fact about the strut deflection is that it occurs at a frequency that is roughly double that of the kerbing input. I have struggled to find an explanation for that. On reflection, I think that the "buckling" reference I found could well contain an important clue, under the heading "Flutter instability", although it doesn't look very likely at first sight.

I ask you to imagine that the two beams of the "Zeigler column" are the simplest lumped parameter approximation of the push rod, and the outer ball joint is represented by the point "C" in the diagram. The outer ball joint must be assumed to have, what is called "dry friction", as I suggested (kind of) in a previous post. Now, with that in mind, you might want to watch the explanation contained in this video, also referenced in the "Flutter instability" section.

If you find that convincing, then the solution would be to reduce friction in the outer ball joint. Alternatively, an FE model might be used to confirm the instability(*), and to investigate other changes to improve the "stability margin" of the strut. If I were to do that, I would very much like to know the actual frequency of the instability.

Overall, I think that JT's conclusion was correct but, perhaps, the reasons are rather more complex than would appear at first sight.

(*)Edit. Incidentally, I would include the upright in the model, & also the power steering system (I guess it would be hydraulic).

[Tommy Cookers]

there is an old saying "if it can vibrate, it will !"
(apparently unknown to the perpetrators of the Millenium bridge, Shuttle booster rockets, and F1 pushrods ?)

people are taught simplified samples from the full suite of possibilities that can be forced upon engineering design
that's ok if we remember that, and give ourselves a bit of a margin wherever it seems reasonable
sooner or later it will save a lot of trouble
(it worked for me ...... 99% of the time !)

BTW are these pushrods behaving as if made from an isotropic material, or do they have 'tailored elasticity' ?

[DaveW]

BTW are these pushrods behaving as if made from an isotropic material, or do they have 'tailored elasticity' ?

I can't imagine designing a push rod with "tailored elasticity" - but then, I'm not an aerodynamicist....

[Tommy Cookers]

a carbon-fibre driveshaft is tailored to favourable strength and stiffness for its job as a driveshaft, this sacrifices strength and stiffness in bending etc
is behaves as of made from a material with non-isotropic properties

are we sure that no such adverse side-effect is possible here ?

[gixxer_drew]

BTW are these pushrods behaving as if made from an isotropic material, or do they have 'tailored elasticity' ?

I can't imagine designing a push rod with "tailored elasticity" - but then, I'm not an aerodynamicist....

Just an opinion from one aerodynamicist, that has got to be the very last place I would ever want to look to get some benefit. I just cant think of any characteristic that I would want I couldnt get easier and safer from somewhere else. CYA is often job 1 on a big team. The things that can go wrong there are not stuff you want to write your memoirs about, especially with a struggling/new car. My $.02.

[aussiegman]

Just an opinion from one aerodynamicist, that has got to be the very last place I would ever want to look to get some benefit. I just cant think of any characteristic that I would want I couldnt get easier and safer from somewhere else. CYA is often job 1 on a big team. The things that can go wrong there are not stuff you want to write your memoirs about, especially with a struggling/new car. My $.02.

I have to agree with this 100%, you would not want it try and introduce flex or deflection into these load bearing/stressed members. As pointed out, a failure at high load/speed would not be something you you want to put your name to and obtaining consistent results would be difficult with resources best spent elsewhere.

This is one I obtained from an ex-F1 car and have used as a wing support and they are specifically made (to my eye at least) to resist bending and/or flex.

[aussiegman]

a carbon-fibre driveshaft is tailored to favourable strength and stiffness for its job as a driveshaft, this sacrifices strength and stiffness in bending etc
is behaves as of made from a material with non-isotropic properties

are we sure that no such adverse side-effect is possible here ?

Carbon driveshafts are typically made from continuous filiment wound tube, transmitting torque from fiber to fiber allowing a torque tube that is lighter than steel or aluminum with approx three times torsional strength of steel equivilent.

The improved torsional spring rate allows the shaft to absorb high moments of torque and then release stored energy (stored as tension in the tube) for smoother power delivery and reduced shock loadings on the drivetrain and tyre.

These characteristics can be varied by changing the windings, filament angles, thickness of the tube and/or overall diameter. As such the non-isotropic nature of the filament winding in the tube is a benefit not a detractor from performance.

Additionally, carbon and carbon-kevlar/spectra etc shafts provide a safety factor in so much as under stress is will "shatter' and form small parts and not steel and aluminum shrapnel,

[Tommy Cookers]

a carbon-fibre driveshaft is tailored to favourable strength and stiffness for its job as a driveshaft, this sacrifices strength and stiffness in bending etc
is behaves as of made from a uniform material with non-isotropic properties

the cfc driveshaft has intentionally much of the fibre aligned around 45deg to give the greatest strength in torsion
(if it was designed for best bending strength much of the fibre would be aligned longitudinally)
relative to its torsional strength the driveshafts bending strength is poor due to the 45deg alignment (reasonably)
similarly with stiffness properties
looks like tailored elasticity to me (nothing to do with aero)

I just wondered whether the pushrod could have inadequate bending stifness for a similar reason
(just a suggestion)