Testing rig capabilites

How do they test the stabilisers on a testing rig. Same with anti dive devices and such, say on a F1 rig.

Must be hard to simulate dive and roll. Its not like you just tilt the thing on its sides, is it?

How do they test the stabilisers on a testing rig. Same with anti dive devices and such, say on a F1 rig.

Must be hard to simulate dive and roll. Its not like you just tilt the thing on its sides, is it?

http://en.wikipedia....i/7_post_shaker (appalling article by the way)

or a K&C rig.

http://en.wikipedia....i/7_post_shaker (appalling article by the way)

From the Wikipedia link.
The 4 post shaker is commonly used by vehicle manufacturers to determine if their vehicles will handle specific road inputs. Manufacturers do not normally use a 7 post rig for road cars because these vehicles are not normally subject to the same aerodynamic effects as a race car operating at high speeds.

Is this true Greg? A 4 post rig will tell you a lot, but what about steady-state roll and pitch? These still occur in the absence of aero and surely manufacturers would like to add those to the mix?

From the Wikipedia link.
The 4 post shaker is commonly used by vehicle manufacturers to determine if their vehicles will handle specific road inputs. Manufacturers do not normally use a 7 post rig for road cars because these vehicles are not normally subject to the same aerodynamic effects as a race car operating at high speeds.

Is this true Greg? A 4 post rig will tell you a lot, but what about steady-state roll and pitch? These still occur in the absence of aero and surely manufacturers would like to add those to the mix?

No. 4 posters aren't used for anything much to do with durability or suspension design. We use them primarily for S&R and some diagnostics for NVH. The reason is that the tire behaves very differently on the rig to reality, so it doesn't really generate the right loads. For durability vertical forces are not the only ones of interest, so for that we use a road load simulator which drives each wheel hub in 3 or more DoF, taking the tire out of the equation.

We have had race cars on the 4 poster but they spent most of their time obsessing about wheelhop, see my fourth sentence, and didn't learn much they were prepared to talk about.

No. 4 posters aren't used for anything much to do with durability or suspension design. We use them primarily for S&R and some diagnostics for NVH. The reason is that the tire behaves very differently on the rig to reality, so it doesn't really generate the right loads. For durability vertical forces are not the only ones of interest, so for that we use a road load simulator which drives each wheel hub in 3 or more DoF, taking the tire out of the equation.

We have had race cars on the 4 poster but they spent most of their time obsessing about wheelhop, see my fourth sentence, and didn't learn much they were prepared to talk about.

Much of that is true.... & yet is misleading, in my view. I certainly agree that the Wikipedia article is factually incorrect, for the most part.

I have used "your" rig at Geelong to help understand & set up an OzV8. It is true that "wheel hop" (the rear hub mode) is something of an issue with that specific vehicle because the live rear axle weighs north of 150 kg & the control tyres are are not a good match for the vehicle. The tyres drive set-up towards low rate springs, but controlling the rear hub mode properly imposes a lower limit on rear spring stiffness. I would not call that "obsessing" about wheel hop, just understanding the hardware better. I have also used my rig in Thetford to help set up OzV8's. This option has now been "banned" by the series largely, I suspect, because other teams have protested. They probably wouldn't have protested if rig testing added no value. In the last year it was allowed, "my" customers filled the first three places in the series - not entirely down to what I do, of course, but because they are capable teams, & rig testing helps them to understand their vehicles better (I would argue).

Oops, other commitments call. I may return later.

Edited by DaveW, 16 May 2010 - 08:33.

Greg/Dave W. is this article beterthan teh Wiki one?

http://www.circletra...mics/index.html

Also if you are looking at very high frequency motions would you not need an eight post rig so as to be sure that any lack of torsional resistance in the chassis is not affecting the results?

Or do you swap over the postion of the two posts and the single post of the loading posts and then re run the test to see if the answers change at all?

Greg/Dave W. is this article beterthan teh Wiki one?

http://www.circletra...mics/index.html

Also if you are looking at very high frequency motions would you not need an eight post rig so as to be sure that any lack of torsional resistance in the chassis is not affecting the results?

Or do you swap over the postion of the two posts and the single post of the loading posts and then re run the test to see if the answers change at all?

Terry Satchell is a knowledgeable & experienced engineer who, last time I looked, helped Ford NASCAR teams with rig tests. The referenced article is fine, but very non-technical.

Back to Greg's post & philosophising for a moment, it is possible to pour scorn on all types of testing, whether it be "hardware in the loop" or mathematical modelling. Each has its strengths & weaknesses, but all can, if interpreted & applied properly, be used to improve the product being tested, regardless of whether the objective is performance, stability, "ride" or durability.

Thus, for example, neither a 4 nor a 7/8 post rig can load vehicles laterally, but including real tyres, with all that implies, does generally add value to a test (in my view). Again, neither can predict reliably the effect of vehicle & suspension changes on race vehicle lap times, although either can yield on-track performance improvements, provided the information generated is used wisely. A "spindle loaded" test can add lateral & longitudinal forces, but tyre information is lost, & it still will not represent accurately the effect of some components of inertia force. It is, I think, for that reason that a spindle loaded durability test can & does yield both false negatives & false positives (failures are experienced during a test that are not found in actual vehicle trials & vice versa). A cynic might say, with justification, that vehicle mathematical models simply "repackage" what is known or assumed about a vehicle. They add value, nonetheless, particularly for situations that cannot be represented accurately by hardware tests, or for conditions that might be hazardous for hardware tests.

Back to the original question. No rig or mathematical model test is capable, by itself, of achieving a complete, optimal, "on-track" set-up, although rig tests can often achieve improved track performance directly. "Hardware in the loop" tests can confirm that many of the intended design objectives have been achieved (or not), and can also expose some unintended characteristics. Mathematical model tests, augmented &/or validated with information from hardware tests, can also make important contributions. At the end of the day, however, the principal function of such work for a race vehicle is to help a competent race engineer and driver achieve a set-up that will tick most "performance" boxes and, probably more importantly, can be used to guide a race engineer to implement changes that will correct (or at least improve) a deficiency without compromising other aspects of performance. Track testing is essential to achieve close to optimal performance for a race vehicle, but it is not a reliable alternative to rig testing &/or mathematical modelling. In short, the cockpit is the place to identify that a problem exists, but it is not a good place from which to identify the source/cause of a problem.

Edited by DaveW, 16 May 2010 - 21:01.

Is someone willing to update the Wiki article?

If we rewrote the Wikipedia entry we wouldn't have it to sneer at. Try this:

http://www.theoryinp...01-0804v001.pdf


In the OE world, 4-post machines far outnumber 7-post and for all kinds of uses -- NVH, component durability, etc -- beyond spring and damper. 7 post rigs were more of a motorsports development driven by aero and based on the 4 post machine expertise. The number of rams is a number, depends how many data channels you can stand in light of the scheduled 24 hour day/7 day week while many installations are complicated with semianechoic chambers etc and/or specialized instrumentation. Some of the 7-post NASCAR tables are actually 8-posters and there is at least one 11-post machine in F1. However, the automakers employ one, two, etc post machines for specialized development incl. powertrain, half car and quarter car work. And there are hub-loaded tables, too. Most impressive shaker rig I ever saw is a 6-post rig designed for Class 8 trucks. Now there is a seismic machine bed. When it was under construction some nice men in black suits stopped by to inquire what they were making in there.

If someone were to ask me, not that anyone did, the 7-post shaker is overhyped in motorsports at this moment. This is not intended as a knock on the potential of shaker rigs whatsoever but more on the biz if anything. There is always a trick of the week and this is that week. But in my view, many race programs headed that way would be better off starting with K&C and even pulldown rigs, for some reasons alluded to above. Will be of more benefit initially IMO and will also make the shaker time more productive when they get there. Just my $00.02, opinions vary.

A good reference, McG.

I agree that a K&C test will provide much useful quasi-static information about a vehicle. I'm not so convinced about "dynamic" K&C tests, however. Incidentally, I also think that a day (or more) spent track testing can, for a race vehicle, be a valuable precursor to a 4/7 post rig test.

The fundamental issue with a 4 poster or 7 poster is that the tire behaves substantially differently when it is rolling compared with when it is being pummelled up and down standing on a plate. Roughly speaking on average you might be talking about 30% difference in rate.

Over the years I have seen (and tried) mods suggested to get better correlation, such as lower tire pressures, or using a ridged plate instead of a flat one. If you enter the road profile into the rig, and measure the hub accelerations, and compare those hub accelerations with the ones you measured on the road, you discover that even for different but similar tires on the same car the 'adjustment' that needs to be made is different. Attempts to modify the drive file to account for this for specific tires haven't really worked, and in Tony Wright's opinion the game is not worth the candle.

I quite like putting cars on shakers in general, but I think the tests strength is in understanding what goes on from the hub onwards, rather than the wheel/road interface.

So far as I am concerned the only way to get a good handle on wheelhop is to put 80 g of out of balance on one rim and then drive the car through its speed range. This test, while realistic, throws up one of the other problems - coupling with the wheel recession mode. That shouldn't apply on a race car but I bet it does sometimes. The final linear range difficulty I am aware of is coupling between wheels, but that is best handled analytically.

Of course if you are trying to control wheelhop off a kerb, which is the most likely condition, then none of the above is going to be definitive due to the non linear and time based response of the tire.

The fundamental issue with a 4 poster or 7 poster is that the tire behaves substantially differently when it is rolling compared with when it is being pummelled up and down standing on a plate. Roughly speaking on average you might be talking about 30% difference in rate.

Tyres (tires) are indeed annoyingly complex.

Vertical stiffness & the ability to "wrap around" encountered obstacles are two relatively independent properties of a tyre. The first is, with care, measurable on a multi-post rig. The second is not, simply because it is not exercised in any meaningful way by a rig. If a rig is to be used to reproduce vehicle responses to an actual road surface, then the fidelity of the result will suffer because, whilst the wheel platforms can represent the "magnitude" of a road surface input correctly, they cannot represent the "shape" of the road surface input. Simple logic suggests that the effective "shape" of an input will depend upon input wavelength, vehicle speed and tyre diameter. Hence I believe your quoted "30% difference" is more a reflection of the way observations have been interpreted, than it is a description of an intrinsic tyre (or rig) property.

Attempts to modify the drive file to account for this for specific tires haven't really worked, and in Tony Wright's opinion the game is not worth the candle.

I quite like putting cars on shakers in general, but I think the tests strength is in understanding what goes on from the hub onwards, rather than the wheel/road interface.

You have certainly made a case against using a rig for "track simulations", & it is a case that I (&, I suspect the F1 community at large) would endorse. I'm not sure who Tony Wright is but I do take issue with his conclusion, however. I also take issue with your last statement quoted above. You imply (I presume) that a tyre and the suspension it supports can be separated logically. That is a common assumption, but it is certainly not correct. A tyre, primarily by virtue of its vertical stiffness property, is an integral part of the suspension of both road and race vehicles, and the overall performance of a vehicle suspension cannot be assessed independently of the tyres fitted to the vehicle. I'm sure you are as aware of that fact as I am. If it is true for road vehicles, it is even more so for race vehicles, for which the ratio of spring/tyre stiffness can exceed unity, especially when aerodynamicists rule.

My point is that, whilst multi-post rig tests might not be capable of executing accurate "track simulations", they are invaluable tools for assessing overall suspension performance, including tyres (track inputs are not the only disturbances that a suspension must deal with). In fact, "optimizing" the suspension of a race vehicle is very much an exercise in matching the vehicle to its tyres. Alternatively, of course, it is possible to use a rig to select tyres that integrate well with a vehicle. One race tyre manufacturer uses "my" (the Multimatic) rig at Thetford for that specific purpose.

Rig tests are also useful for identifying some potential vehicle issues, and providing estimates of suspension & vehicle properties that are useful for validating & augmenting mathematical models. A group within Multimatic actively supports race teams using Adams models, & they use results extracted from rig tests of the target vehicle to help to generate their models. F1 teams also use rig tests for that purpose, & some teams do, certainly, hire my rig to help with that activity.

I also help race teams directly, from series ranging from FFord through to F1. Apart from matching dampers to the vehicle & its tyres, we also explore the effects on response of tyre pressure, temperature & camber, damper & spring changes, damper temperature, & the side effects of suspension geometry & ballast changes. All, it might be argued, aimed at improving an understanding of the likely effects of changes they might encounter or contemplate at a track.

The fact that many of my customers are fairly regular visitors is at variance with Tony Wright's opinion that "the game is not worth the candle". That is an example of shooting the messenger because he doesn't like (or understand) the message.

Edited by DaveW, 17 May 2010 - 09:17.

A lot of this is way above my level of knowledge but that just makes it more interesting!

One simple question can a 4/7 post rig be used to determine initial suspension "stiction " and if so would that be one of the first tests you do - to get that non -linear effect of of the way analytically - or do all the maths just work it out for you as you go along?

One simple question can a 4/7 post rig be used to determine initial suspension "stiction " and if so would that be one of the first tests you do - to get that non -linear effect of of the way analytically - or do all the maths just work it out for you as you go along?

The simple answer to your question is yes....

However, friction/stiction is a complex phenomenon that some have modelled as a single, time-dependent, process. The model has merit, because two sliding surfaces that are separated by an oil film when moving would be expected to have a low friction. Allow the two surfaces to remain static, however, and a pressure at the interface can extrude the lubricating oil molecules, leaving a time-varying surface-surface contact, that will require a time-varying force to resume relative movement.

The model would suggest that estimating "stiction" & removing that from subsequent results might not be a valid strategy. Hence, in our case, we "let the maths work it out", as you mentioned.

A good reference, McG.

I agree that a K&C test will provide much useful quasi-static information about a vehicle. I'm not so convinced about "dynamic" K&C tests, however. Incidentally, I also think that a day (or more) spent track testing can, for a race vehicle, be a valuable precursor to a 4/7 post rig test.

How would you advise a client to best use that track session to optimize their shaker rig time?

How would you advise a client to best use that track session to optimize their shaker rig time?

I wouldn't presume....

Playing your question straight, most race engineers have the knowledge & experience to "rough out" an acceptable vehicle & suspension set-up (better than I could, anyway). What is likely to happen during a track test will depend upon the preferences of the engineer, his driver and conflicting vehicle requirements. Bear in mind that a good set-up compromise will involve many variables, only a few of which are exposed by (visible during) a rig test. However, after a day's track set-up work, when most (or some, at least) of the "invisible" variables will have been addressed, the changes they made (& why) to the "visible" variables will often be helpful when suggesting sensible directions they might explore using, as a background, the "view" provided by analysing each rig test run.

The strategy helps me to avoid (hopefully) many potential pitfalls. It will (also hopefully) engage the race engineer sufficiently for him to take "ownership" of suggestions and give him an insight into the probable effects of different set-up changes. The last of these is probably the most enduring benefit of carrying out a rig test.

Well i guess you can make a testing rig to do anything. i mean if you want dive you could just create attachment points on the chassie and use actuators to force it down in front and up in the back etc.

When the teams simulate a track. Is the track readings from sensors measuring to the ground and such or is it from wheel movement.

Using wheel movement and tire pressure sensors to generate track surface seems like a troublesome route.

Edited by MatsNorway, 17 May 2010 - 19:32.

Can someone explain for me, the operation of a seven post hoist? To my simple way of thinking, you shouldn't need 7 actuators (I assume they are all single DOF). 4 actuators (on the wheels) plus 3 attachment points to locate the chassis would enable simulation of all combinations of Z axis deflections would it not?

Edited by gruntguru, 17 May 2010 - 23:56.

Can someone explain for me, the operation of a seven post hoist? To my simple way of thinking, you shouldn't need 7 actuators (I assume they are all single DOF). 4 actuators (on the wheels) plus 3 attachment points to locate the chassis would enable simulation of all combinations of Z axis deflections would it not?

I posted an attempted explanation elsewhere a while ago. Here follows a copy of that. I hope it helps.

The hardware required for track simulations comprises four hydraulic actuators supporting wheel platforms (as for a four post rig), usually exciting a vehicle through its tyres (wheel actuators), augmented by three (or four) additional hydraulic actuators attached to the vehicle "sprung mass" to apply controlled loads to the vehicle (force actuators). The wheel actuators operate under "displacement control", whilst the force actuators, unsurprisingly, operate under "force control".

A track simulation is implemented by recording a lap, and then exciting an identical vehicle installed on the rig so as to minimize the differences between a selected set of track measurements and the identical set of rig measurements. The procedure is (roughly) as follows:

1. Create "drive files" for the force actuators to simulate deterministic loads applied to the vehicle. These are rolling moments caused by lateral acceleration, pitching moments caused by longitudinal acceleration, driven wheel torque reaction & aerodynamic pressure (drag), and vertical loads caused by aerodynamic pressure. Other forces act on a real vehicle, of course, but these cannot normally be simulated with confidence using a tyred vehicle on a rig.

2. It is then assumed that differences between selected track measurements and the equivalent rig measurements when the vehicle is driven "deterministically" are the result of vertical inputs to the tyres caused by track irregularities. A complex iterative process is followed to arrive at four wheel actuator drive files that minimize differences between the selected track and rig measurements. The wheel actuator drive files are assumed to represent track inputs encountered by the tyres as the vehicle executes a lap.

In theory, at least, the calculated aerodynamic components of the load drive files can then be replaced by "coupled" components (the aero forces being computed in real time using the aero map, airspeed time history and rig-based ride heights). Vehicle response should not, of course, be changed by this unless or until the vehicle set-up is changed.

The simulation can now be used to explore the effect of set-up changes on vehicle response.

It is worth reviewing what has been created by putting together the hardware-in-the-loop "track simulation".

First, the simulation will not be exact, because all the forces applied to the vehicle on track have not been represented. Second, any errors in track measurements and vehicle parameters used to compile Step 1 (c.g. position, aero map, etc.) will "feed through" to corrupt the computed track input drive files (this will include any vehicle differences, of course). Third, the simulation does not extend as far as a stop watch, so a "cost function" is required to assess the likely effect of set-up changes. Fourth the "driving line" around a track is invariant. Fifth, the poor signal-noise ratio of a track simulation implies that it is unlikely to yield good estimates of some parameters compared with more idealized rig tests. Lastly (I think) it important to be aware of hardware limitations: force actuators (specifically) connect the sprung mass directly to a virtual earth and, unless actuator control is precise, they are likely to be the most efficient dampers attached to the vehicle.

Nevertheless, "track simulations" are likely to provide a good approximation to real suspension movement (positions and velocities) around a circuit, and for that reason alone, they can be worthwhile.

As has already been stated, the only true simulation is a vehicle negotiating a track and the primary Performance Index is a stop watch. Anything else is an approximation that might purport to "optimize" a vehicle but, unless the "optimized" set-up can put the vehicle on pole consistently, its main function is actually to help to understand better the vehicle and the problems encountered by its driver, and a test or simulation technique that fails in this respect, either directly or indirectly, must be of questionable value.

Edited by DaveW, 18 May 2010 - 05:32.

Thanks Dave, I'm starting to see the picture.

http://www.youtube.c...amp;feature=sub

Interesting..... Looks like half a K&C rig without vehicle tie-down structure. Didn't look much like a "track simulation", so I guess they were running pseudo-random inputs with no applied down-force (at least I could see no evidence of down-force actuators).

Look closely at 0.33-35. tires compress.

looks like some wires in the center at 0.08.

those might be capable of pulling the car down.

i first believed they also could emulate roll but would that not be inacurate as the angle of the wires chance? it would also pull the car sideways.

Edited by MatsNorway, 13 July 2010 - 17:06.

There's a lot of funny ideas about test rigs out there. I just had a vehicle dynamics engineer trying to tell me that he couldn't tell the difference between a softer steering column and a change in steering ratio on a K&C rig. I must find out which company he's working for.

Apologies, Mats, I can see what I suspect are transducer cables attached to the chassis, but no obvious down-force cables or links.

Incidentally, the K&C "half" comprises the hexapods under the front wheels & the rather clunky structures (not attached) for measuring wheel rim positions. I suppose structure would be added to locate the chassis when the rig was used in the K&C role. I find it odd that they combined the K&C & vertical dynamics functions into a single rig. Can only be that they were short on real estate (or the hardware salesman was overly enthusiastic).

The use of hexapods is interesting because, whilst they can control the wheel platforms in 6 degrees of freedom, the transformation from (required) platform position to (required) actuator positions is not explicit (requires an iterative solution) & it is, therefore, not easy to achieve accurate dynamic positioning in real time. I guess MTS would take their usual approach to dynamic problems & use RPC to generate drive files "off line". I shudder to think about those 12 UJ's per corner being hammered mercilessly during dynamic tests.

I wonder where the wheel platform load cells are? For the K&C role & think I would want a load cell in series with each hexapod actuator, but that would not be the case for the vertical dynamics role.

Also interesting is the use of plastic sheet (bin liners?) as poor man's slip plates to prevent tyres bonding themselves to the wheel platforms.

p.s. I guess that the horizontal cables placed longitudinally along the centreline of the vehicle are intended to stop the vehicle rolling off the rig. Now that is a good idea....

Edited by DaveW, 14 July 2010 - 08:43.

I have no clue about testing riggs. All i know is that if i were given the task to make the spec for a machine like this it would be more sophisticated than that.

I do like the front wheel "hexapods" tho.. they seems to be capable of being tilted in real time for simulating curb riding. Right?

I also like that there seems too be camber lift on the Renault.

In that "tape" around the tire i would not be surprised if they had some load cell/pressure sensor in there. Is it piso electric tape its called?

Are they able to add the rotation mass from the tire after the result?

I have no clue about testing riggs. All i know is that if i were given the task to make the spec for a machine like this it would be more sophisticated than that.

Actually, sophistication (such as it is) occurs not so much in the rig, but in the way it is used, in the software used to analyze measurements, and in the way the results are used.

I do like the front wheel "hexapods" tho.. they seems to be capable of being tilted in real time for simulating curb riding. Right?

A simple, single, actuator (like the one shown supporting the rear wheels of the Renault) is (or should be) capable of simulating kerb strikes (for kerbs that an F1 vehicle is likely to encounter, anyway), & will probably be better at it than a hexapod.

In that "tape" around the tire i would not be surprised if they had some load cell/pressure sensor in there. Is it piso electric tape its called?

I could be wrong about the function of the "bin liners", but I don't think so.

Are they able to add the rotation mass from the tire after the result?

If they are wise, & are interested in contact patch loads, then they would certainly make the required corrections, either in real time or post run (having said that, one manufacturer of multi-post rigs famously made the correction with an inverted sign). Not quite sure what you mean by "rotational mass, however.

Most teams incorporate information obtained from K&C tests, multi-post rig tests, wind tunnel tests, various component tests and tyre models into full vehicle simulations, which they use to examine in detail various aspects of on-track performance more realistically than they could ever achieve with a hardware-in-the-loop test, however complex that might be. Increasingly, such models are used in full driver-in-the-loop real time simulations. In that context, tyre models are probably the weakest link in the whole house of cards at present.

tyre models are probably the weakest link in the whole house of cards at present.

A pity, seeing the tyre is the most performance critical component on the whole car.

A pity, seeing the tyre is the most performance critical component on the whole car.

I agree with you, although I'm not entirely sure that aerodynamicists would.

Whether or not it might be possible to improve tyre models experimentally is debatable, but the debate would be pointless because F1 teams are not permitted to characterize their tyres. Curiously, given their hyper-regulated universe, F1 teams are, apparently, permitted to spend many millions of dollars on driving simulators.

I agree with you, although I'm not entirely sure that aerodynamicists would.

Yes they do tend to have tunnel vision.

Given the choice of 10% extra tyre grip or 10% extra downforce is a no-brainer however.

Whether or not it might be possible to improve tyre models experimentally is debatable, but the debate would be pointless because F1 teams are not permitted to characterize their tyres.

Does that mean they cannot rig test their tyres? A bit pointless IMHO as I'm sure they could get reasonably close by analysing track data.

Yes they do tend to have tunnel vision.

If that was an intended pun - nice one!

A simple, single, actuator (like the one shown supporting the rear wheels of the Renault) is (or should be) capable of simulating kerb strikes (for kerbs that an F1 vehicle is likely to encounter, anyway), & will probably be better at it than a hexapod.

only problem with non angled kerb riding i see is that you get a longer arm as the tire might only be in contact on the outer side of the tire.

They might be able to add realistic data on top of everything but if you do to many corrections your back to simulations.

i meant the centrifugal forces.

I see a lot of ifs and buts here and therefore i think it should had things like the wheels spinning during testing and tilting hexapods for curb riding.

i think it should had things like the wheels spinning during testing

Why? Gyroscopic forces only apply to tilting of the axis of rotation (ie camber change) - a fairly minor aspect when suspension operates.

Why? Gyroscopic forces only apply to tilting of the axis of rotation (ie camber change) - a fairly minor aspect when suspension operates.

No longer a big deal unless you are running enormous camber gains. In the days of beam axle front suspensions it cause steering shimmy, you'll find that older dynamics books mention it.

Given the choice of 10% extra tyre grip or 10% extra downforce is a no-brainer however.

I couldn't argue with the statement as such. However, similar statements are made to justify aero-dominated suspension set-up strategies. They do often incorporate alleviating features, but F1 could (in my opinion), and other race series do, use strategies that are rather rather more grey than your simple black & white statement would suggest.

Does that mean they cannot rig test their tyres? A bit pointless IMHO as I'm sure they could get reasonably close by analysing track data.

Extract copied from the 2010 F1 Sporting Regulations:

25.5 Testing of tyres :
a)Tyres supplied to any competitor at any time may not be used on any rig or vehicle (other than an F1 car on an F1 approved track, at the exclusion of any kind of road simulator), either Team owned or rented, providing measurements of forces and/or moments produced by a rotating full size F1 tyre, other than uniquely vertical forces, tyre rolling resistance and aerodynamic drag.
b) Tyres may be used on a test rig providing forces control and monitoring by F1 rim manufacturers for the sole purpose of proof testing their products.

Using track data to compile tyre models is problematic for a variety of reasons. For example, "grip" varies with both tyre & track states, both of which are difficult to quantify, contact patch loads cannot be measured (directly), & accurate tyre slip angles cannot be measured (I think, although it has been attempted - some claim successfully).

Using track data to compile tyre models is problematic for a variety of reasons. For example, "grip" varies with both tyre & track states, both of which are difficult to quantify, contact patch loads cannot be measured (directly), & accurate tyre slip angles cannot be measured (I think, although it has been attempted - some claim successfully).

Ultimately you aren't really interetsed in test rig type datat, if you can only measure slip angles with a correvit etc then you are measuring the vehicle slip angle (etc) and can at least work out wht the tires are doing relative to vehicle.

In the linear range I overlay measured and predicted average axle slip angles to within say 0.1 deg/g, but of course that is not limit handling where slip angles and tire behavior are much more difficult. Also that is on one piece of road, in a known condition, on a known vehicle, so I already know a great deal about the relatiosnhip between Flattrac and the test results.

Ultimately you aren't really interetsed in test rig type datat, if you can only measure slip angles with a correvit etc then you are measuring the vehicle slip angle (etc) and can at least work out wht the tires are doing relative to vehicle.

In the linear range I overlay measured and predicted average axle slip angles to within say 0.1 deg/g, but of course that is not limit handling where slip angles and tire behavior are much more difficult. Also that is on one piece of road, in a known condition, on a known vehicle, so I already know a great deal about the relatiosnhip between Flattrac and the test results.

The better known the road, vehicle etc, the better your chances of inferring the type of tyre performance data you would get from a tyre test rig, which is of course a highly controlled environment itself.

Would not a skid pan be the best place to emulate test rig results?