A new free vehicle dynamics resource - Dan's Vehicle Dynamics Corner

[ChassisSim]

Hey Guys,

For those of you using ChassisSim the following is worth a look,

http://www.chassissim.com/blog...-lap-time-simulation

It's about how to dial in grip and bump scale factors manually. It's a really good guide to what to adjust and when to adjust it. It also walks you through the mechanics of how to do it.

I realise this tutorial is a bit long in the tooth and has been pushed to the side by the advances we have made with auto grip and the advances in bump profiling. That being said if you have a circuit that is difficult to model this tutorial will be of great help and there are some good insights you can learn from.

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[ChassisSim]

Hey Guys,

I just posted this article on my blog,

http://www.chassissim.com/blog...a-to-refine-aeromaps

It's about how to refine aeromaps from pitch data from the racecar.

I realise this isn't ultra applicable for a FSAE car but its still a useful resource to have. I know some of you might be considering jobs in motor racing and this is a good thing to have in your back pocket.

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[ChassisSim]

Hey Guys,

For the ChassisSim users in the FSAE community I've just posted this video on how to consolidate your circuit models,

http://www.chassissim.com/blog...-master-circuit-file

It's about how to use the Master circuit file which we have found to be a great tool.

I also realise that I've posted this before but this is a great guide to getting going with lap time simulation,

http://www.chassissim.com/blog...onnecting-the-dots-2

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[ChassisSim]

Hey Guys,

I've just posted this article on how to use virtual shaker rigs,

http://www.chassissim.com/blog...lbox-tips-and-tricks

This has proven to be a very powerful tool in the ChassisSim community. In particular I think it is something the FSAE community could get quite a bit out.

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[ChassisSim]

Hey Guys,

Over the last couple of weeks I've had a few people in the ChassisSim community asking me questions about the order in which they should dial in bump and grip scale factors when creating their circuit models.

I've seen the same theme's recurring and this tutorial addresses this,

http://www.chassissim.com/blog...ating-circuit-models

Think of it as the follow up to my post from a couple of weeks a go on the mechanics of dialling in bump and grip scale factors.

For those of you who are ChassisSim users you'll find this very useful.

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[BillCobb]

I'm obliged to tell you that your simple model equations are missig the 4 tire aligning moments. The net effect is a very large rigid body yaw moment as well as a likely understeering effect from steering compliance.

The net effect is the predicted total vehicle understeer will be far different than actual and the predicted system (steering) gain will be too high.

You can spend a lot of money buying or renting MTS or Michigan Scientific wheel force transducers to validate these conclusions. The parametric models provided with nonlinear tire data MUST include the MZ functions. MX functions can be important, too, if the car reacts badly to aligning torque camber inducements.

Since a 'race' car has low, but measureable total vehicle understeer (SAE definition as the difference between the front and rear axle sideslip gradients), the tire MZ terms are a very large fraction of this value. Getting that right will alter any simulated track or test results, as you can imagine...

[ChassisSim]

Hey Guys,

Just fixed the link for the latest video tutorial. I hit the publish too soon! Sorry about that.

Bill - with regards to the concern that you raised, self aligning torque primarily makes its presence felt with regards to steering feed back. In particular it's really important for driver in the loop simulation.

However with regards to vehicle performance prediction it's impact is minor. Hence why I didn't include it. This allows the reader to get straight to what really counts. This isn't just based on hearsay. The ChassisSim community covers categories as diverse as GP2, F3, V8 Supercars, DTM, Sports prototypes, GT and the FSAE community to name a few. These cars are very heavily instrumented and all of them have outstanding correlation. Bottom line if this was an issue I would have heard about it a very long time a go.

However don't take my word for it, do some numbers on it. Take a FSAE car weighing 250kg in a 1g turn with a 50/50 weight distribution and a 1.7m wheelbase. Also assume the peak trailing arm of the self aligning torque is say 50mm. Calculate the moment arm from the front axle to the c.g and do the same for the self aligning torque. The numbers are very revealing.

Enjoy Guys

Danny Nowlan
Director
ChassisSim Technologies

[Z]

Of the above two posts (Bill's and Danny's), I feel Bill's is closer to the truth, but I don't know by how much.

So, for any FSAE-VDer's who might be interested...

Under/oversteer can be thought of in terms of "static margin". Roughly speaking, this is the plan-view-longitudinal-distance between the car's CG and the (roughly lateral) Line-of-Action of the resultant force vector from all four wheels (Edit! -> ), when this force vector is that which gives neutral steer (<- EndEdit). If the CG is in front of neutral-steer-force-LoA, then understeer. If CG behind LoA, then oversteer. If CG right on LoA, then neutral steer. (MoreEdit. See Tim's post below for better definition...)

Anyway, for a reasonably handling car the static-margin is only a few inches/cm. This is a similar distance to the "Pneumatic Trail" of typical tyres. And since Mz of a tyre can be thought of as the Fy force multiplied by the PT distance (ie. Mz = Fy x PT = Fy force acting PT behind tyre centre), then ignoring Mz can give an error that has similar magnitude to the static-margin.

BUT! Mz is typically zero at zero slip-angle, rises to its peak at about half peak-Fy-slip-angle, then drops back to about zero at peak-Fy-slip-angle (roughly speaking). This is because the pneumatic trail starts at maximum at zero slip-angle and ~zero Fy (so Mz = 0 x PT = 0), and PT then drops to ~zero at peak-Fy (so Mz = Peak-Fy x 0 = 0), but inbetween, at ~half-peak-slip-angle, there is some Fy and some PT, so some Mz. So the above paragraph makes sense at low and half-peak cornering forces when the PT is of similar magnitude to the static margin. But at peak cornering forces the Mz effect may be negligible, because PT = ~0???

I haven't seen enough of the curves of real tyres to know just how big Mz is at typical peak-Fys. My guess is that the Mz curves move about quite a bit depending on all the other factors (longitudinal forces, camber, tyre pressure, etc...).

So, for the numbers, it is over to you guys, Bill and Danny.....

Z

[ChassisSim]

Hey Guys,

The number's speaks for themselves,

*Moment arm from tyre to c.g 800mm - 1500mm depending on the car.
*Self aligning torque moment arm in the tyres are in the order of 20mm - 50mm. Also this bleeds off as we hit peak slip angle.

Self aligning torque can be visualised as a result of the lateral force applying a moment arm about the centre of the contact patch. Consequently when you do the numbers for the static margin it's effectively swamped by the moment arm of the axle to the c.g. Also this moment arm bleeds off as we hit peak slip angle.

Consequently in terms of car handling and static margins it's a mild second order effect. However it makes its presence felt in the torque the driver feels through the steering wheel. This is where the good drivers earn their money.

However guys - work through the example I presented above in my previous post. To really ram home the point calculate the typical contact patch length for a given load, tyre pressure and area. The conclusions are pretty obvious.

Enjoy

Danny Nowlan
Director
ChassisSim Technologies

[Z]

Danny,

The large moments about the CG from the front and rear tyre forces largely cancel out (hopefully!), leaving only the small moment of the resultant force from all four tyres multiplied by <STRIKE>the static margin</STRIKE> its offset from the CG. ( <- Edit.)

If the effect of pneumatic trail (or Mz) moves all four tyre forces backwards by your quoted "20mm - 50mm", then that is a significant first-order effect, because that distance is similar to the static margin.

So that leaves the question of how small does Mz/pneumatic-trail become as Fy reaches peak? Bill??

Z

PS. The decreasing pneumatic trail (or Mz) as Fy approaches peak moves the handling balance towards more oversteer, so it is worth keeping in mind...