IRS LOADING DISPARITIES

[BillyShope]

Some 48 years ago, I was taught that, with an IRS, right-to-left load variations are unaffected during acceleration. But, I just realized that, if anti-squat is present, some of the reaction torque must be diverted to appear as vertical loading on the unsprung mass, resulting in a load disparity opposite to that which occurs in a beam axle. This would be a function of the relative compliance of the gear housing mounting and the suspension links, of course.

So, has anyone investigated this and, perhaps, established a relationship between load disparity and percent anti-squat? Is the disparity even significant? Obviously, if it was as bad as that which occurs with a beam axle, it would have been immediately noticed. So, how bad is it?

 

[BillyShope]

No, the effect would be to unload the right rear, just as with a beam axle. Must have had my glasses on upside down. Sorry.

 

[GregLocock]

The load transfer in a beam axle is due to t he need to react the propshaft torque. If the halfshafts+CV joints in an IRS are working correctly this is resolved in the diff bushes and is not seen by the wheels.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[NormPeterson]

This sounds closely related to the reason why some drag racers with certain rear suspension arrangements have been going to a huge rear sta-bar, to re-plant the RR tire to some extent using the engine torque reaction fed back through the chassis and rear suspension.

I've seen the above related topic elsewhere and presented the logic that the effect is going to be dependent on the relative roll stiffnesses of the front and rear suspensions and the torsional stiffness of the chassis, and is independent of anti-squat (up to the point where one front wheel lifts clear of the ground, anyway). I did not attempt any derivation or present any numbers.

Anti-squat needs to work from the longitudinal traction forces, which are dependent on the wheel loads, so that effect is a step downstream from the effect of engine torque reaction and the X-axis stiffnesses. A-S may also be of unequal percentages, left vs right, so the A-S share of wheel load won't necessarily be the same L vs R even if the instantaneous vertical wheel loads happened to be identical.

Norm

 

[GregLocock]

Whoops, now I see I think. Bear in mind that the diff bush torque is exactly opposite and equal to the engine/trans mount torque, in the steady state case.

Best example of this is in the ~1988 Camaro or MX5, where the diff is part of the engine structure. Then there is no external Tx, only a Ty.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[NormPeterson]

The torque arm on the 3rd and 4th gen Camaro primarily resists only the axle My, as it utilizes a sliding connection (within a bushing) at the transmission tailshaft. There's perhaps a little induced loading in the TA due to lateral displacement of the chassis pickup in roll and also as a consequence of axle rollsteer, but there's no Mx carried through the TA.

The MX5 may be different.

Norm

 

[BillyShope]

Ideally, the structure between the engine/transmission mounts and the differential housing could be considered perfectly rigid and there would be no effect on wheel loadings. When I first posted, I was "picturing" the reaction torque feeding back to the rear and I saw part of it being diverted into the trailing links. But, I then realized I was looking at it incorrectly, for there cannot be a net torque imbalance on the differential housing. So, I then looked for a place to absorb the torque from the differential housing mounts. Again, if the structure is perfectly rigid, the torque is simply taken at the engine/transmission mounts. But, if there is any deflection, loads will be transmitted throught the trailing links, resulting in wheel load variations. It now appears, however, that, since the compliance of the differential housing mounts does not enter into the picture, the load in the links would be small compared to that carried by the other structure, meaning similarly small wheel load variations, even with a large amount of anti-squat.

(Norm, it sounds like we've both been encouraging dragracers to increase the ratio of rear-to-front roll stiffness. A few of them are putting big bars on the rear, but it's like pulling teeth to get them to remove the one at the front! And, this next season, a few will be running higher rate springs at the right front.)

 

[Warpspeed]

The big bar at the back idea sounds good until one day you get yourself sideways at the start. The uncontrollable oversteer even after you back off might be somewhat embarrassing.

And if you get sideways half way down the track at a reasonably high speed, that could be even more interesting.

 

[BillyShope]

If you think that's bad, the Altered class cars are now allowed to run NO rear suspension with a stock front suspension. Let's face it: The competitive cars in any of the drag classes are a handful even in a straight line pass. They'll do whatever's necessary.

But, I would agree that there's a better way to achieve a more stable launch. Specifically, the driveshaft torque can be dynamically canceled by forces in an asymmetric link arrangement and the rear (or front)suspension springs don't even have to be involved. That's certainly a much "cleaner" way to do it!

 

[NormPeterson]

The big bar at the back idea sounds good until one day you get yourself sideways at the start. The uncontrollable oversteer even after you back off might be somewhat embarrassing.

Actually, the tendency from an immense rear sta-bar should be for a straighter launch as compared with an otherwise identical car and setup that does not use the big rear bar. There will be a smaller absolute amount of roll steer present (even though the rate of roll steer has probably gone up).


What's of somewhat greater concern is that some (many?) of these cars are street-driven, and that not all of those who have taken this approach understand that this is not a particularly street-friendly set-up under all likely conditions. A few that do recognize this and take it seriously (or have perhaps had a less than pleasant experience) have mentioned disabling the big rear bar for the street.

Billy - from time to time I actually do see some discussion regarding the removal of the front bar or the disconnection of at least one front sta-bar endlink for strip duty. But I have no idea how representative that is of the way the typical drag racer's car shows up in staging. And I'm a bit skeptical about how many of those that do either of those mods understand that they're tinkering with the front:rear roll stiffness distribution in addition to losing weight off the front or allowing the front to rise a bit more easily.

Norm

 

[Warpspeed]

I agree with you. The slingshot dragsters have a solidly mounted rear axle and seem to work o/k. But if they do start to go sideways, the result is never pretty.

It worries me that guys in the lower street classes might start fitting huge rear antiroll bars, and then try to drive to work every day in the same car.

 

[patprimmer]

It worries me that the guys that don't know any better do all sorts of things to their cars, then drive the car to work every day.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.