Buying this Mustang and building the tube chassis for my Falcon has sent me down a rabbit hole of trying to understand suspension geometry, dynamics and kinematics. In trying to gain a better understanding of why the Mustang handles the way it does I made a quick sketch which shows the difference in roll center and roll axis with and without a Panhard bar. The MM site does little to help someone like me understand why...maybe because nobody else cares, haha. I've never been a 'just tell me what parts to buy' kind of guy, though, I know plenty of people who are, and they seem much happier .
And, as much as I want to share what I think I've figured out, I'm hoping someone might have some corrections to make to my understanding if it is indeed wrong.
Anyway, it seems most people's understanding or explanation of why and how the Panhard bar improves the handling is "it makes the car feel more precise" or "it keeps the axle centered under the car better". All fine and true, but I'm more curious than that. Why?
Using a plumb bob last year I plotted all the connection points of my rear suspension and put them into CAD to see if it would give me some clues. [I'll be doing this with the front suspension as well in the next couple of weeks]
You can see without the Panhard bar, the roll center is around 18" high, and the roll axis of the rear axle is about 4.7º which would seem to be the cause of the roll steer, as the outside tire would move forward and the inside tire moves rearward...in addition to the axle drifting side to side. On my car, which is lowered, you can also see how the upward inclination of the RLCA's (body to axle) affects anti-squat and shortens the lift point/SVSA (side-view swing-arm).
If the roll center height defines the length of the moment arm, then I think of it like the amount of leverage that the center of gravity has to roll the body of the car around the axles. Without the Panhard bar, the rear roll center is 18" and the COG is approximately 20". So, aside from the rear suspension being a big ol' bag of BIND, the COG only has a ~2" lever to try and force it into roll compliance. Right? So, once the available leverage has squashed the springs and the rubber bushings as much as it can, it goes solid, and the rear stiffness goes to infinity. But the interesting thing is that it doesn't get loose like popular knowledge would suggest the tighter end of the car would do. And I suspect the stored energy in the springs and bound up rear suspension is part of the cause of the 'snap' oversteer.
At the same time, the front of the car, especially when it's lowered, has a much longer moment arm (more leverage) and a bit more weight on it.
Once the rear suspension is sufficiently bound up, the front suspension absorbs the weight transfer. Since the front springs are absorbing the weight transfer, a softer setup is employed to lessen understeer. Which leads to excessive body roll which leads to tons of static camber. The rear suspension really is the crux of the issue. The bandaid is to stiffen the rear sway bars to try and induce some oversteer to balance the car out. But ideally, taking grip from one end of the car to fix the other is not the best solution. I'd rather add grip to the front to balance the car.
The roll center when a Panhard bar or Watts link is employed is the center of the bar or the center of the bell crank on the watts. So, add the Panhard bar, and the roll center is now ~9". The COG now has an 11" moment arm. Now the COG has a lot more leverage to roll the car around the rear axle. Now, the rear suspension takes up some of the roll moment of the car when you go into a corner reducing some of the work the front axle has to do and bingo bingo, understeer is reduced. And now you can start to stiffen things up, reduce body roll, and maintain better wheel alignment in cornering.
Here's the part I'm less sure of- On a three link, the roll center and the convergence point of the RLCA's defines the roll axis. I do not know for sure and cannot find a solid answer to whether this is true or not when the triangulated 4 link is still installed. If yes, then the rear axle's roll axis is greatly affected, causing it to go about -1º, which would cause a bit of roll counter-steer. Which would explain the reduced roll steer.
The last thing to mention is Roll Axis Inclination. Which is the imaginary line that intersects the front and rear roll centers and is the imaginary axis the the car rolls around. This photo represents a good starting point for roll center heights on a RWD, Non-Aero race car.
[Image borrowed from https://suspensionsecrets.co.uk/roll-centre-and-roll-moment/]
So, if you consider the stock mustang has about a 90% rear roll center height and it's essentially below ground in the front once you lower it on the stock k-member...you can see how it's backwards of the recommended starting point and by a lot.
Anyway, once I get the front suspension plotted and input into CAD I'll be able to show you guys what the roll axis inclination looks like without the Panhard bar, with the Panhard bar and with the corrected front roll center.
Brad
And, as much as I want to share what I think I've figured out, I'm hoping someone might have some corrections to make to my understanding if it is indeed wrong.
Anyway, it seems most people's understanding or explanation of why and how the Panhard bar improves the handling is "it makes the car feel more precise" or "it keeps the axle centered under the car better". All fine and true, but I'm more curious than that. Why?
Using a plumb bob last year I plotted all the connection points of my rear suspension and put them into CAD to see if it would give me some clues. [I'll be doing this with the front suspension as well in the next couple of weeks]
You can see without the Panhard bar, the roll center is around 18" high, and the roll axis of the rear axle is about 4.7º which would seem to be the cause of the roll steer, as the outside tire would move forward and the inside tire moves rearward...in addition to the axle drifting side to side. On my car, which is lowered, you can also see how the upward inclination of the RLCA's (body to axle) affects anti-squat and shortens the lift point/SVSA (side-view swing-arm).
If the roll center height defines the length of the moment arm, then I think of it like the amount of leverage that the center of gravity has to roll the body of the car around the axles. Without the Panhard bar, the rear roll center is 18" and the COG is approximately 20". So, aside from the rear suspension being a big ol' bag of BIND, the COG only has a ~2" lever to try and force it into roll compliance. Right? So, once the available leverage has squashed the springs and the rubber bushings as much as it can, it goes solid, and the rear stiffness goes to infinity. But the interesting thing is that it doesn't get loose like popular knowledge would suggest the tighter end of the car would do. And I suspect the stored energy in the springs and bound up rear suspension is part of the cause of the 'snap' oversteer.
At the same time, the front of the car, especially when it's lowered, has a much longer moment arm (more leverage) and a bit more weight on it.
Once the rear suspension is sufficiently bound up, the front suspension absorbs the weight transfer. Since the front springs are absorbing the weight transfer, a softer setup is employed to lessen understeer. Which leads to excessive body roll which leads to tons of static camber. The rear suspension really is the crux of the issue. The bandaid is to stiffen the rear sway bars to try and induce some oversteer to balance the car out. But ideally, taking grip from one end of the car to fix the other is not the best solution. I'd rather add grip to the front to balance the car.
The roll center when a Panhard bar or Watts link is employed is the center of the bar or the center of the bell crank on the watts. So, add the Panhard bar, and the roll center is now ~9". The COG now has an 11" moment arm. Now the COG has a lot more leverage to roll the car around the rear axle. Now, the rear suspension takes up some of the roll moment of the car when you go into a corner reducing some of the work the front axle has to do and bingo bingo, understeer is reduced. And now you can start to stiffen things up, reduce body roll, and maintain better wheel alignment in cornering.
Here's the part I'm less sure of- On a three link, the roll center and the convergence point of the RLCA's defines the roll axis. I do not know for sure and cannot find a solid answer to whether this is true or not when the triangulated 4 link is still installed. If yes, then the rear axle's roll axis is greatly affected, causing it to go about -1º, which would cause a bit of roll counter-steer. Which would explain the reduced roll steer.
The last thing to mention is Roll Axis Inclination. Which is the imaginary line that intersects the front and rear roll centers and is the imaginary axis the the car rolls around. This photo represents a good starting point for roll center heights on a RWD, Non-Aero race car.
[Image borrowed from https://suspensionsecrets.co.uk/roll-centre-and-roll-moment/]
So, if you consider the stock mustang has about a 90% rear roll center height and it's essentially below ground in the front once you lower it on the stock k-member...you can see how it's backwards of the recommended starting point and by a lot.
Anyway, once I get the front suspension plotted and input into CAD I'll be able to show you guys what the roll axis inclination looks like without the Panhard bar, with the Panhard bar and with the corrected front roll center.
Brad
Last edited: