Pitch, roll, yaw M.O.I

[HenningO]

I'm wondering if anyone out there has measured the full car (with driver) mass moment of inertia? Or maybe anyone with a very complete CAD assembly might have been able to get it out of the software?

Pitch and roll values are of most interest!

It can obviously be estimated, but hard numbers are always more fun!

[HenningO]

I'm wondering if anyone out there has measured the full car (with driver) mass moment of inertia? Or maybe anyone with a very complete CAD assembly might have been able to get it out of the software?

Pitch and roll values are of most interest!

It can obviously be estimated, but hard numbers are always more fun!

[Erich Ohlde]

yes, we built a swing rig and tested the yaw and roll MOI's. I don't remember what the numbers were but it was pretty simple to do.

[flavorPacket]

lol

[Mike Cook]

I have tested to acquire the value you ask for. I'm not going to give them to you but I will tell you this: We made a simple spread sheet and measured the weight and distance from the CG of all the main parts on the car and calculated yaw inertia this way. It was very close to the measured value. I'll be honest, I don't think many teams use these values other than to compare one car to the next. Also I don't think roll and pitch inertia are a big deal.

[HenningO]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mike Cook:
It was very close to the measured value. ... Also I don't think roll and pitch inertia are a big deal. </div></BLOCKQUOTE>

Good to know, thanks!

I would disagree to some extent that the roll and pitch inertia aren't a big deal. Im currently doing my thesis work for a company delivering components for the motorsport industry, they say, and after simulating our cars in their software, the inertial properties can have a quite big impact on car balance. Of course there are other aspects of the car that rightly should be adressed before this one but still... http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Kurt Bilinski]

Maybe not a big deal (a very subjective term anyway) in a FSAE car, but it is for larger cars with heavier parts. Probably a good idea to find what it is and see if it can be improved, if for no other reason than practice for life after college.

[Ben Beacock]

In a paper I just finished, I found that decreasing yaw inertia actually increased the lap time slightly. Compared to some other parameters it really didn't have much effect.

[Kurt Bilinski]

FWIW, I tried keeping PMOI as low as possible in my mid-engine Mini. At the Streets of Willow Springs, there's a fast left-right-left, and it's absolutely amazing how fast the car changes directions. Well worth it in my book, though to be fair, there's a point of decreasing returns as the parts you're anguishing over get really small and light... at what point is "enough"?

[Mike Cook]

Ben, I'm not sure why you would not expect the lap times to drop as your yaw inertia decreased? I certainly think decreasing yaw inertia will help these cars a lot but eventually you get to a point where its about as low as you can get it and it probably isn't worth fighting to minimize it 1 more percent.

Henning, I would say that roll and pitch inertia can be a big deal on many other cars, especially production based race cars, but our cars already have signifcantly low pitch inertia and roll inertia values, and small pitch/roll angles so it probably isnt a big deal. Or as you put it, there is lots more to be addressed first http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Christopher Catto]

Ben. please explain in short:

what simulation software/algorithm
what vehicle type (especially mass, wheelbase and track)
basic effect of decreased yaw inertia in cornering (if oscillations or loss of momentum etc)

just brief. no need to make your paper available http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif

[Ben Beacock]

&gt;what simulation software/algorithm

-IPG CarMaker,
-pacejka tire data from the TTC,
-IPG Kinematics for the suspension model
-IPG driver retrained for each iteration
-a monte carlo simulation for parameter sensitivity (2000 iterations)

&gt; what vehicle type (especially mass, wheelbase and track)

250kg vehicle, 90kg driver, 1640mm wheelbase, 1250mm front track, 1150mm rear track.

&gt;basic effect of decreased yaw inertia in cornering (if oscillations or loss of momentum etc)

I ran a full autocross sim, as well as individual tests on 75m of slalom and a hairpin turn.
The slalom and hairpin time both decreased along with decreasing yaw inertia, 0.0021 and 0.00033s/kg.m^2 respectively, but the autocross increased (barely) by 0.0048s/kg.m^2 over a 60s lap. As a comparison, the mass sensitivity was around 0.031s/kg
As to why.. well I haven't really gone that far. I've been meaning to check the slip angles to check for spikes but I'm busy with other data right now.

[HenningO]

I didn't necessarily mean that they should be minimized, even though Ben simulations shows that could decrease laptimes. I was more referring to having the numbers avaliable for use in suspension tuning.

[C.Zinke]

0.031s/kg ist quite a lot. This would mean, that Lions or Stuttgart schould have a disadvantage of 1.55sek per lap in comparision to TU-Graz. I bet their car is not that bad, that the can not show this advantage.

[Ben Beacock]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by C.Zinke:
0.031s/kg ist quite a lot. This would mean, that Lions or Stuttgart schould have a disadvantage of 1.55sek per lap in comparision to TU-Graz. I bet their car is not that bad, that the can not show this advantage. </div></BLOCKQUOTE>

This could be explained by 3 things:
a) the simulation does not account for tire temperatures, with an advantage to the heavier car

b) the simulation assumes the same engine output, so a (generally) heavy 4cyl turbo vs a NA single

c) the mass parameter was tested with a standard deviation of 25kg, so it is more accurate with small changes in mass vs large changes