Suspension strain gauge loads vs. time

[riff_raff]

I agree that when it comes to choosing what types of instrumentation give the best bang-for-the-buck, race engineers are good judges. But purely from a technical standpoint, an accelerometer can only measure accelerations produced about its local frame of reference. Strain gauges can only measure relative strain produced at the local surface it is mounted to. And displacement sensors (LVDTs, prox sensors, rotary pots, etc.) can only measure relative linear/angular displacements.

Thus, how would one assess the true impact of dampener and spring changes based solely on the acceleration and displacement data from just two points within the suspension linkage? The suspension and chassis are basically a free-body system, which means the forces, accelerations, and strains produced within any suspension members are dependent upon the response of the rest of the chassis.

[speedsense]

Because measuring acceleration is not going to give you push rod force

Accelerometers on the uprights and the chassis, work out the positional. It has a level of accuracy, but so does everything else we talked about here. Cheap enough to do to your family car since I dont think anyone is going to post up their team's strain gauge data.

Position won't give you a reliable measure of strain or force on the spring though. Consider the case where you have steady cornering on a smooth road (turn 8 at Turkey). The vertical hub accelerometer will be reading practically zero but you have massive loads from cornering and vertical aero force going through the spring. Its these constant "DC" loads which accelerometers miss.

Hub accelerometers are mainly useful when trying to recreate bump profiles on a circuit. Its then quite a convuluted path to get from accelerometer reading to spring load (or push rod strain). The only way I can think to do it via "calculation" would need good knowledge of the dampers dynamics and an iterative approach.

We do use Front/Rear Accelerometers over the axle plane of the suspension (inboard on the chassis center), mostly for US/OS navigation and calculation of Yaw (even with Yaw sensor). There are uses of accelerometers at the uprights, but mostly for accelerations and shock work. The presence of upright acceleration is used more frequently in off road racing but again for accelerations of the wheel. These are normally in conjunction with either load measurement through load cells (built into shock tops for instance) or strain gauging pushrods.
Also to add, the steering tie-rods are also strain gauged, for reasons of measuring under steer.

A couple of questions speedsense;
Why do you use front and rear accelerometers if you have a yawrate sensor?
What does the tierod strain gauge have to do with understeer?

Tim

Because a yaw rate only shows an acceleration of rotation. You don't know, without further investigation, if the rotation is driver induced, car induced or track induced. With the added accelerometer of front and rear, it helps further define the hardest aspect to analyze, under steer. The grip level (front to rear and cross) is more defined with the added accelerometer, when combined with other sensors, such as steering. Over steer is an obvious analyst, under steer is subjective, not in engineering, but were the driver is concerned. It differs between drivers. For instance a driver with a gokart background views understeer differently than a driver with no background in karting. The slightest hint of understeer is bad to a an ex-karter and might not be understeer to someone else.
I can safely say that you can put drivers in two categories, those who can rotate a car and make do what he/she wants, eg: driving beyond under steer with induced rotation, effectively killing it or reduction of it, for example. And the other type, a driver who needs the car to rotate for them. These are two entirely different setup approaches. As an analyst, you have to know the difference. Absolute number finding or number crunching does not work here.
When one accelerometer is used, you have an "overall" grip level in combination graphs, typically steering angle and Lat G. Having front and rear Lat G, the "grip level" is more defined to front and rear when subjected to the same type comparison, and puts the analyst closer/focused on what the driver feels/needs depending on their driver technique.

The tie rod strain gauge, is a "trick" I learned running banked ovals. Banked ovals create a few strange phenomenon's in drivers. The lateral G is combined with vertical g. In such a way the road racing driver "feels" G force on a road course from between his hips. On banked ovals the same cornering force is now being felt from his left shoulder going through his right hip, on a left turning high bank oval. The driver can no longer judge the grip level of the car in a normal way. He must now rely on his steering wheel, more exactly his forearms. Heaviness in his forearms, means understeer about to happen. Lightness in his forearms means oversteer. This is because, on ovals, loads are vertical and lateral G, so roll increases on one end of the car create problems/ handling issues due to load and typically that end of the car with the most roll will let go first. Very different than road courses. Hence, the strains allow you to see the loads your driver is feeling at the wheel.
Adding tie rod strains on road race cars, further illustrates/helps define in the analyst of under steer through load the driver feels at the wheel. Reduced load in the tie rod at certain points in a corner are loss of grip. It's just one more point of measurement of U/S.
In engineering terms and in some data programs, we can define under steer on a car. But in driver terms can you? If I could strain my driver's hands and feet, I would.
In my mind, knowing everything, every decision, every thought process, every point of recall (data systems are well beyond his recall) and combining that with a balanced car suited to him, leads more often than not, to a winning car.

As always in MHO

[speedsense]

I agree that when it comes to choosing what types of instrumentation give the best bang-for-the-buck, race engineers are good judges. But purely from a technical standpoint, an accelerometer can only measure accelerations produced about its local frame of reference. Strain gauges can only measure relative strain produced at the local surface it is mounted to. And displacement sensors (LVDTs, prox sensors, rotary pots, etc.) can only measure relative linear/angular displacements.

Thus, how would one assess the true impact of dampener and spring changes based solely on the acceleration and displacement data from just two points within the suspension linkage? The suspension and chassis are basically a free-body system, which means the forces, accelerations, and strains produced within any suspension members are dependent upon the response of the rest of the chassis.

We don't solely base it on those. All movements/inputs of the driver and just about every sensor on the car is part it. You don't change shock settings/re-valve, springs, ARB's, or just about anything based solely on suspension sensors and/or strain gauges, at least if you want a happy driver that is . The driver, the car and the track reading are all part of the analyzing process IMHO.

[DaveW]

Because a yaw rate only shows an acceleration of rotation.

Apologies, speedsense, but that is not quite correct. Yaw rate is the rate of change in yaw angle. Hence if two lateral accelerometers are in place mounted on the centreline, but a distance apart, yaw rate is the integral of the difference between the accelerometers divided the distance between them. If that makes sense.....

Other than that, I thought your post was very interesting, I will study with care....

[Tim.Wright]

I agree there.. Basically with two accelerometrs you don't need a yaw sensor. Likewise with one accelerometer and a yaw sensor you can calculate the acceleration at the front and rear axle (which is the normal method I use)

Regarding the strain gauges, what you are measuring is effectively the steering effort. This is linked to the front axle grip but NOT to underster/oversteer. To measure OS/US you need to compare some measurement of the front axle to the same measurement of the rear. The force you measure is affected by the tyre's aligning torque characteristics, the caster and caster trail of the suspension and the scrub and scrub radius of the suspension. All of these change with different setups and from car to car.

Also, typically, the steering effort will start reducing but the front tyre force is still slightly increasing. A good suspension will make the difference between the peak front axle force and the onset of the steering torque drop-off very close so the driver feels the front axle limit better.

The best way to characterise the balance is to calculate the elastic steer anglewhich is the difference between the actual steer angle and the theorectical angle you would need if the slip angles at the front and rear tyres are zero.

[Tommy Cookers]

You don't know, without further investigation, if the rotation is driver induced, car induced or track induced. With the added accelerometer of front and rear, it helps further define the hardest aspect to analyze, under steer. The grip level (front to rear and cross) is more defined with the added accelerometer, when combined with other sensors, such as steering. Over steer is an obvious analyst, under steer is subjective, not in engineering, but were the driver is concerned. It differs between drivers. For instance a driver with a gokart background views understeer differently than a driver with no background in karting. The slightest hint of understeer is bad to a an ex-karter and might not be understeer to someone else.
I can safely say that you can put drivers in two categories, those who can rotate a car and make do what he/she wants, eg: driving beyond under steer with induced rotation, effectively killing it or reduction of it, for example. And the other type, a driver who needs the car to rotate for them. These are two entirely different setup approaches. As an analyst, you have to know the difference. Absolute number finding or number crunching does not work here.
When one accelerometer is used, you have an "overall" grip level in combination graphs, typically steering angle and Lat G. Having front and rear Lat G, the "grip level" is more defined to front and rear when subjected to the same type comparison, and puts the analyst closer/focused on what the driver feels/needs depending on their driver technique.
Hence, the strains allow you to see the loads your driver is feeling at the wheel.
In engineering terms and in some data programs, we can define under steer on a car. But in driver terms can you? If I could strain my driver's hands and feet, I would.

super informative !, ss
just thinking out loud here ......
for 50 years race car cornering has been optimised by (mainly and still currently) ARB adjustment, ie what we call balance
this requires the chassis to be much less compliant torsionally than the torsional compliance equivalent from the suspension and tyres

because karts have no diff, their wheel loads must be unbalanced ie the inside rear must have only a small load
the chassis is designed to be very compliant torsionally (combined with the non-compliance of the non-suspension) to do this

drivers typically progress via their karting ability and then by their ability to un-learn this for car racing ?
and I thought after 25 years of real data acquisition the best way of driving would be apparent !
bringing the standard of most of the field (usefully for their employers) closer to the top few

[speedsense]

You don't know, without further investigation, if the rotation is driver induced, car induced or track induced. With the added accelerometer of front and rear, it helps further define the hardest aspect to analyze, under steer. The grip level (front to rear and cross) is more defined with the added accelerometer, when combined with other sensors, such as steering. Over steer is an obvious analyst, under steer is subjective, not in engineering, but were the driver is concerned. It differs between drivers. For instance a driver with a gokart background views understeer differently than a driver with no background in karting. The slightest hint of understeer is bad to a an ex-karter and might not be understeer to someone else.
I can safely say that you can put drivers in two categories, those who can rotate a car and make do what he/she wants, eg: driving beyond under steer with induced rotation, effectively killing it or reduction of it, for example. And the other type, a driver who needs the car to rotate for them. These are two entirely different setup approaches. As an analyst, you have to know the difference. Absolute number finding or number crunching does not work here.
When one accelerometer is used, you have an "overall" grip level in combination graphs, typically steering angle and Lat G. Having front and rear Lat G, the "grip level" is more defined to front and rear when subjected to the same type comparison, and puts the analyst closer/focused on what the driver feels/needs depending on their driver technique.
Hence, the strains allow you to see the loads your driver is feeling at the wheel.
In engineering terms and in some data programs, we can define under steer on a car. But in driver terms can you? If I could strain my driver's hands and feet, I would.

super informative !, ss
just thinking out loud here ......
for 50 years race car cornering has been optimised by (mainly and still currently) ARB adjustment, ie what we call balance
this requires the chassis to be much less compliant torsionally than the torsional compliance equivalent from the suspension and tyres

because karts have no diff, their wheel loads must be unbalanced ie the inside rear must have only a small load
the chassis is designed to be very compliant torsionally (combined with the non-compliance of the non-suspension) to do this

drivers typically progress via their karting ability and then by their ability to un-learn this for car racing ?
and I thought after 25 years of real data acquisition the best way of driving would be apparent !
bringing the standard of most of the field (usefully for their employers) closer to the top few

Not at all!! In a kart, any understeer is slowing the kart. It is a tendency that the go-karter's mentality has when they get to cars and tend to favor a bit of O/S as it makes the car's attitude more familiar to them. Though IMHO, a small amount of U/S isn't any slower than the same small amount of O/S. It is in kart but not in a car.
Also the roll aspect is the other mentality. A Karter will feel impeding roll on one end of a car, and "think" that end of the car is about to let go, as it would in a kart. This can be "cured" by disconnecting the rear bar, making him push the car into over-steer, rather than have him presume the car is "going to go into oversteer". Thus defining for him the difference of roll in a kart vs roll in a race car. IMHO

[speedsense]

Because a yaw rate only shows an acceleration of rotation.

Apologies, speedsense, but that is not quite correct. Yaw rate is the rate of change in yaw angle. Hence if two lateral accelerometers are in place mounted on the centreline, but a distance apart, yaw rate is the integral of the difference between the accelerometers divided the distance between them. If that makes sense.....

Other than that, I thought your post was very interesting, I will study with care....

Sorry short hand thinking, should read "acceleration rate of rotation". I would add that a single 2 axis accelerometer, and an "added" steering signal multiplied by derivative (showing accelerations of inputs at the steering wheel) can be a "backwoods" yaw sensor when you don't have one on the car. Also it's a good tuning tool as well, as when accelerations of the steering slow down, the car is getting better handling/balance wise.
Most systems that would be in the Math channels, Derv (steering signal- calibrated at the steering wheel, IE: 180,90,0,-90, -180 in degrees rather than caster plate degrees) Also calibration for the Lat G and steering angle in the same direction. For me that's not "SAE convention" where right hand turns are positive numbers but rather for me they are negative.
This is only for reasons of ease as I come from the ERA prior to track maps (was personally involved in the track map's creation). When graphed this way, left turns are positive, if you were to walk across your computer screen from the left side of the plot facing the right side, left is left and right is right. SAE convention does not function this way.

[speedsense]

I agree there.. Basically with two accelerometrs you don't need a yaw sensor. Likewise with one accelerometer and a yaw sensor you can calculate the acceleration at the front and rear axle (which is the normal method I use)

Regarding the strain gauges, what you are measuring is effectively the steering effort. This is linked to the front axle grip but NOT to underster/oversteer. To measure OS/US you need to compare some measurement of the front axle to the same measurement of the rear. The force you measure is affected by the tyre's aligning torque characteristics, the caster and caster trail of the suspension and the scrub and scrub radius of the suspension. All of these change with different setups and from car to car.

Also, typically, the steering effort will start reducing but the front tyre force is still slightly increasing. A good suspension will make the difference between the peak front axle force and the onset of the steering torque drop-off very close so the driver feels the front axle limit better.

The best way to characterise the balance is to calculate the elastic steer anglewhich is the difference between the actual steer angle and the theorectical angle you would need if the slip angles at the front and rear tyres are zero.

I don't agree, it's the "best way". A great tool, it is, but I have seen very talented drivers take a ill handling car, bend it to his will and make that graph look neutral and balanced when the car wasn't.
Combined with other graphs and handling signal layouts rather than completely relying on that one graph is best. IMHO

[Tim.Wright]

For me it's always told the same story that the drivers are telling.

Ok, perhaps if the car is completely out of the zone it might not be reliable. Though in that case you don't need the data to tell you that, the driver will let you know.

I would add that a single 2 axis accelerometer, and an "added" steering signal multiplied by derivative (showing accelerations of inputs at the steering wheel) can be a "backwoods" yaw sensor when you don't have one on the car.

An easier one: Yawrate = LatAcc/Velocity

[speedsense]

For me it's always told the same story that the drivers are telling.

Ok, perhaps if the car is completely out of the zone it might not be reliable. Though in that case you don't need the data to tell you that, the driver will let you know.

I would add that a single 2 axis accelerometer, and an "added" steering signal multiplied by derivative (showing accelerations of inputs at the steering wheel) can be a "backwoods" yaw sensor when you don't have one on the car.

An easier one: Yawrate = LatAcc/Velocity

Ok, perhaps if the car is completely out of the zone it might not be reliable.

Tim, I'm not in implying it's not reliable, just citing cases where it can be fooled, therefore it's not the "best" route to me. It's a great tool to have, but as with all analysis, there's more than one way to get accurate results.

Though in that case you don't need the data to tell you that, the driver will let you know.

One of the experiences I had with this was at Sebring, turn 1. Both drivers of the car are top factory drivers. The driver that made this graph look perfect, had zero issue with the car in T1. The second driver however had a different style and approach to turn 1, thought the car had too much under steer (for him) on entry, yet he was faster out of the turn 1. and the graph clearly showed the cars issues. Other graphs, than the handling graph, pinpointed the issue with car, Driver A's ability to drive around it and we found a comprise in balance that suited both. And made both faster in that turn and resolved other issues we had elsewhere. The handling graph, was important in the analysis, but in equal importance, so were the other graphs in pinpointing the issue. IMHO

An easier one: Yawrate = LatAcc/Velocity

Yes that is simpler, however, for me, in determining yaw and chassis setup direction an analysis must include in the results, whether the driver induced rotation (on purpose) or the car induced it (on it's own) or the track caused it. Can also be a combination of varying degrees of each. Yaw amounts, as an absolute number alone does not answer how or what or why. So even with an existing yaw sensors, other sensors, like steering, G, throttle, etc. need to be considered either in raw or math induced form. IMHO

[Jersey Tom]

Over steer is an obvious analyst, under steer is subjective

Eh.. can't say I particularly agree with that assessment. Tight/free, understeer/oversteer (from driver comments) are both fundamentally subjective measures [for reasons below]. It's up to the engineer or analyst to have a methodology able to objectively quantify both ends of the spectrum. To only have one side of the "neutral" point be obvious by objective measurement says to me that the method is not robust.

For instance a driver with a gokart background views understeer differently than a driver with no background in karting. The slightest hint of understeer is bad to a an ex-karter and might not be understeer to someone else.

Same can be said of "oversteer" (itself an imprecise word). Going to the extreme of short track dirt oval racing, a driver can put the car in what would otherwise be an obvious oversteer state... but yet it's quite calm, comfortable, and controlled. How a dirt racer vs. a kart or road course guy deems a car "oversteer" or "loose" or "free" is extremely subjective... even if there are standard definitions of some of those terms (the practicality of which is itself an individual assessment varying among people).

[Greg Locock]

Here's some understeers, oversteers are generally opposite

Start with Olley, an understeering car steers down the crown of the road.

I drive a bit faster round a constant radius skidpan, and I need more steering wheel angle

At a very high speed ignoring aero and tires) the car becomes less responsive (oversteer car becomes unstable)

the nose of the car points outside the tangent to the circle it is driving on

Applying throttle pushes the nose out

I drive a bit faster with the same lock and the radius of turn increases

...and so it goes. All of those are measurable.

[speedsense]

Over steer is an obvious analyst, under steer is subjective

Eh.. can't say I particularly agree with that assessment. Tight/free, understeer/oversteer (from driver comments) are both fundamentally subjective measures [for reasons below]. It's up to the engineer or analyst to have a methodology able to objectively quantify both ends of the spectrum. To only have one side of the "neutral" point be obvious by objective measurement says to me that the method is not robust.

For instance a driver with a gokart background views understeer differently than a driver with no background in karting. The slightest hint of understeer is bad to a an ex-karter and might not be understeer to someone else.

Same can be said of "oversteer" (itself an imprecise word). Going to the extreme of short track dirt oval racing, a driver can put the car in what would otherwise be an obvious oversteer state... but yet it's quite calm, comfortable, and controlled. How a dirt racer vs. a kart or road course guy deems a car "oversteer" or "loose" or "free" is extremely subjective... even if there are standard definitions of some of those terms (the practicality of which is itself an individual assessment varying among people).

In data, spotting an oversteer condition is far easier to identify than an under-steer condition. I've met many a driver who is comfortable with some amount of over-steer, and that in road racing.
Judging under steer is not as simple of a matter as it is with O/S. It takes a fair amount of analysis experience to spot a slightly under steering car, especially with a driver who is comfortable with it and may in fact desire it over O/S, and may fail to mention it due to his comfort level with it.
Funny you brought up dirt cars, I spent a half season with the Outlaws. Had to run four wheel (before GPS) speeds, even with 4 wheel speeds, spots of the track were missing reliable speed. Wheels in the air, spinning wheels. You would average out the speeds and you still missed a lot of the cornering speeds. Just wild stuff... IMHO

[Tommy Cookers]

Mr Olley came up with the understeer and oversteer concept related to steady-state limiting cornering on a test circle ?

this caused cars to become close enough to a balance for the determined and skilled driver to achieve in the right situation either what he calls understeer or what he calls oversteer
if this was recognised as driver induced understeer and driver induced oversteer then both drivers and engineers could command a 10% pay raise, and car journalism would for once be correct

there is only one quickest way round a track (and it shouldn't look understeery or oversteery) ?
hopefully these days it is being found, via the data
if teammates have different routes to the same laptime, there must be a better route than either has found ?