How Stiff is stiff enough upright hub assembly ?

[Mudit]

FEA Results for braking + cornering are showing 0.1 mm deflection in upright and 0.13 mm deflection in hub. This cumulatively results in a camber change of 0.13 deg which is ~10% of the camber change calculated using kinematic analysis.

I am currently using Al-6061 for both the components. Are this results suggesting that upright hub assembly is stiff enough ? If not, What should be the target values ?

[penna]

That is up to you.

How much camber gain from deflection do you have to get to significantly affect the grip the tyre gets?

What will increase the performance of your car more, reduced unsprung mass or increased stiffness?

Is the stiffness and fit of the rest of the suspension assembly good enough that increasing the stiffness in your hub/ upright will noticeably increase the stiffness of the system?

asking yourself these questions might help you decide

[Goost]

I think he is starting a legitimate discussion, it's an open ended design decision and he has already put a bit into it if he has actually done the FEA...
Mudit, in 2012 we designed with a target of less than 0.1 degree deflection (due to compliance) in all cases, like what you are doing it seems. Ended up when it was actually built the physical system was only 72% the stiffness we had calculated. We were able to modify the components before Nebraska, ended up more rigid than the design specs actually, but did run the compliant system at FSAEM. Anyway, I think the typical engineer way is usually to neglect effects that are an order of magnitude away from the control value (e.g. chassis spring rate 10 time the suspension rates). Which, as it sounds like you are doing, was pretty much the choice to target <0.1 degrees.
Just to say from experience though; when you actually make it and have it connected through the arms to the chassis, the deflection probably won't be what that FEA is showing. As far as whether the 'order of magnitude' rule applies here, I would be curious about what other teams think. I'm going to maintain it isn't quite right - who's to say it isn't an order of magnitude in base 14 that we're supposed to shoot for? http://fsae.com/groupee_common/emoticons/icon_wink.gif

[penna]

Sorry what I was trying to get at is it depends on your system. If you tyres are not very camber sensitive then weight might be more important than stiffness, and if you get alot larger amount of deflection from other components it is probably worth adding the extra weight there instead as you will see larger gains.

[Mudit]

Thanks Penna and Goost.

I calculated stiffness of the entire assembly to be close to 5 times the suspension stiffness. I think I will make the assembly a little more stiffer as Penna suggested. Weight is second priority for me right now.

But numbers are not drastically off. Right ?

[Zac]

don't neglect the influence of compliance stacking in your analysis, particularly how your bearing configuration will impact installed stiffness. I've seen noticable diffence between a 10" and 13" wheel package.

As measured on a full K&C rig, an installed stiffness for a top level car is .15-.375 deg/kN Lateral Camber Compliance, .2-.3 deg/kN Lateral Steer Compliance (rear toe compliance will be lower than the front).

[Claude Rouelle]

I calculated stiffness of the entire assembly to be close to 5 times the suspension stiffness.


Then your car compliance is worse than a passenger car with bushings...... Good luck for driver feedback and car response to throttle, brake and steering input...

[Kealan O Carroll]

Originally posted by Zac:
don't neglect the influence of compliance stacking in your analysis, particularly how your bearing configuration will impact installed stiffness. I've seen noticable diffence between a 10" and 13" wheel package.

As measured on a full K&C rig, an installed stiffness for a top level car is .15-.375 deg/kN Lateral Camber Compliance, .2-.3 deg/kN Lateral Steer Compliance (rear toe compliance will be lower than the front).

Is there any way to calculate the compliance in off-the-shelf bearings or is physical measurement the only way to get an accurate enough result ?

What I'm getting at is if we were to use these figures as targets in an FEA simulation, are there a baseline set of compliances for certain bearing types ?

[Claude Rouelle]

Is there any way to calculate the compliance in off-the-shelf bearings or is physical measurement the only way to get an accurate enough result ?

What I'm getting at is if we were to use these figures as targets in an FEA simulation, are there a baseline set of compliances for certain bearing types ?


Catalog info?

[EPMPaul]

Never seen it in catalog data at least not on the SKF data other than in qualitative terms. however you might want to look into a book by tedric harris called "Rolling Bearing Analysis". Already had a look at it. I believe there were compliance calculations but it's a little far to be entirely sure.
It'd definitely give you a good idea and/or references to do said calculation

[Francis Gagné]

Google lead me to this catalog with some numbers, I have no idea how good they are but it is a starting point for at least an order of magnitude.

http://www.nachi-fujikoshi.co....web/pdf/B1031E-5.pdf (http://www.nachi-fujikoshi.co.jp/eng/web/pdf/B1031E-5.pdf)

There is also a stiffness calculation guide from Schaeffler on there website:

http://medias.ina.com/medias/e.../tg_hr*ST4_102146955 (http://medias.ina.com/medias/en!hp.tg.cat/tg_hr*ST4_102146955)

That should get you started!

[Simon Dingle]

It's worth remembering that your bearing stiffness will be highly affected by the operating temperature of the bearings and uprights.

Saw an FSAE car recently where, straight after running while everything was still hot, you could move the wheel by hand and see the play in the wheel bearings with the naked eye.

[Claude Rouelle]

Yep. Preload on bearings should take into account thermal expansion. Want another examples of temperature influence? Chassis torsion stiffness and differential preload. Never have calculated any of the two but have been part of Chassis K&C measurements on test bench at different temperatures in temperature controlled room. Not negligible! As far as diff preload, measure it with a torque wrench with one driven wheel of the ground when the car is at ambient temperature. Make a few laps measure again; you could loose as much as 50 %.

[kkung]

don't neglect the influence of compliance stacking in your analysis, particularly how your bearing configuration will impact installed stiffness. I've seen noticable diffence between a 10" and 13" wheel package.

As measured on a full K&C rig, an installed stiffness for a top level car is .15-.375 deg/kN Lateral Camber Compliance, .2-.3 deg/kN Lateral Steer Compliance (rear toe compliance will be lower than the front).

Hi, I am curious to know where this data came from and if it is available publicly. We're trying to spec stiffness targets for our 2014 car..

[Zac C]

Hi, I am curious to know where this data came from and if it is available publicly. We're trying to spec stiffness targets for our 2014 car..

The full data set is not publicly available. Those numbers also only represent what 'good' teams typically achieve, not an optimal target to shoot for.

[kkung]

The full data set is not publicly available. Those numbers also only represent what 'good' teams typically achieve, not an optimal target to shoot for.

I see. Can you at least tell me which car it was?

[Edward M. Kasprzak]

I see. Can you at least tell me which car it was?

Zac has measured a bunch of Formula SAE/Student cars on a K&C machine, so his comments don't come from just one car.

[kkung]

Zac has measured a bunch of Formula SAE/Student cars on a K&C machine, so his comments don't come from just one car.

Thanks for the clarification!