shiners780

There has been a lot of discussion lately regarding rear coilovers, both in addition to torsion bars and in place of torsion bars. I have seen reference to people keeping the torsion bars in, but having them indexed to a neutral setting when the car is at ride height. Their theory, as I understand it, is that the car's weight is fully supported by the coilover allowing for a somewhat 'softer' ride. Then when needed (spirited driving) the torsion bars aid in added spring action.

I don't understand this theory.

When a rear wheel is pushed up (in essence what is happening when a car leans on that corner) it doesn't matter if the torsion bar had a zero rate on it or 200 pounds of force being applied to it when it was static. What does matter is that either way, a force is now being applied to both the coilover spring AND the torsion bar. So whether the torsion bar has a static rate (wheel or actual...doesn't matter for the sake of this discussion) of zero or 200 pounds, the effect of the torsion bar on the car's handling will be exactly the same.

Whether the car is being driven on the street or track, gently or aggressively doesn't matter. Any force applied to the wheel (a bump, turning, braking, accelerating, etc.) will act upon both the torsion bar and the coilover spring. Any suspension movement moves both the torsion bar and the coilover spring. So if the force is great enough to compress the spring, the force causing the suspension movement will also twist the torsion bar. Whether or not the torsion bar was twisted before the force was applied to the suspension is irrelevant. So how does a "neutral torsion bar setting at ride height" help handling?

Maybe I am not understanding the whole "neutral torsion bar setting" thing. So if someone could please explain it, I'd greatly appreciate it.

sawood12

This is how my KW's are set up. Not sure of the benefits myself, it just seemed natural to me. I wouldn't say that the rear of my car is soft in anyway shape or form and the springs are rated at 250lb and the TB's are 25mm M030 TB's (no idea what their equivalent lb rating is). How would you set the car up otherwise? If you set your TB's as usual what height would you set your coilovers to? If they are set too short they are going to be fighting against the TB's and want to pull the wheel up into the arch, if they are set too long they will be trying to raise the rear of the car, if they are set to match the ride height then you have the reverse situation i.e. loaded TB and neutral coilover. I suppose it depends which spring you want to be your primary spring and which is to be your helper. I need to take my car in for a set up check soon so i'll ask why they have set up the car like they have.

2bridges

One purpose some keep the torsion bars is design flaws of current conversions.

Current coilover conversions result in the entire rear suspension relying one bolt per side. This is not itself an issue. The real issue is there is not a "throughbolt design" that is supported on each side of the lower coilover mounting points to share the load and eliminate leverage forces. Many here have reffered to this as single shear design.

Also many coilovers require using a reduced size stud. So on top of a single shear design you are forced into using a bolt size near half originally designed just for a shock.

Tons of racers do it, tons of street cars do it, and many will tell you how many years/race events etc they have run without issue. I know of 5 race cars in this configuration that have never experienced a failure.

Regardless I find this a fundementally flawed design for the back of any rear drive vehicle. This is why my upgrade retains torsion bar rear suspension.

__________________________________________________________

More info - specific to initial post:
torsion bar forces are not linear
the more you twist, the more force it takes to twist

xsboost90

mine are set w/o tension, and perhaps actually some negative load on the rear suspension, as i had to pull the rear arms down to attach the coilovers at the bottom.

shiners780

With the coilovers supporting the weight of the car and the torsion bars at a neutral setting, there are no helper springs and neither component would be considered a helper. I'm not sure what you are referring to when you mentioned primary and helper springs.

And for the sake of this discussion, let's assume the springs in question are not progressive springs. That just complicates things even more.

2bridges

uhhhhhhhh except torsion bars are progressive.

shiners780

Ah, you did throw something into your post relevant to the topic...

That was a question I had about torsion bars. If it takes 200 ft/lbs of force to twist a TB 1*, does it take more than an additional 200 ft/lbs of force to twist the TB an additional 1*?

However, even if that's true, wouldn't the rate of change of the TB be the same whether it was preloaded or not? If TB's are progressive at 0* twist and progressive at 1* of twist and progressive at 2* of twist, why does it matter where they're set at ride height? As long as the rate of change of the TB is constant, wouldn't it make no difference whether or not they are preloaded at static ride height?

2bridges

no - the rate of change is not the same. the further you twist the faster the rate increases.
as such where they are set - will be key in how you height adjust the coilover..... this is where many pick which spring is "primary" and load it for most of ride height. Then use the other spring barely loaded as a "secondary"

shiners780

I shouldn't have brought up 'rate of change' because I think it's completely irrelevant to this discussion now that I think about it.

I am trying to understand what benefit is derived from indexing a torsion bar at a neutral setting at static ride height when being used in conjunction with a coilover. So far no one has stated what that benefit is.

1. To me, it seems like an awful lot of time and effort to realize no gain. In fact, I think it has several pitfalls and could likely cause odd handling issues. What happens if you decide to change your ride height, or install different rate springs on the coilovers, or cornerbalance the car? In order to maintain your "neutral" TB setting, you would have to reindex the TBs, if you could even figure out what that setting would have to be.

2. Any force applied to a rear corner of the car (cornering force, acceleration) engage both the TB and the spring on the coilover. Whether the TB is set at a static rate of zero or 200# preload, a force is still being applied to it as soon as that corner of the car begins to compress, just as the force is also being applied to the coilover spring. So what is the benefit of having the TB resist that force beginning with zero preload as opposed to 200# of preload?

3. Any time a given rear corner of the car lifts (inside during hard cornering), your torsion bar (having been set at zero preload at static ride height) quickly changes from unloaded to a negative load. What effect does a negative load on a TB do? It creates a lifting force on the inside wheel, which is less than ideal.

I don't know, it seems pretty whacked to me, but maybe I am just not understanding the concept.

sawood12

OK I just double checked with the specialist who installed my KWv3. They didn't actually set up the TB's as neutral as I thought before. They set the ride height with the TB's as usual then installed the coilovers as helpers. This is the way they set up their street cars due to the mounting points on the trailing arm not really being intended to carry the full weight of the car (i know this is controvertial and there are plenty of people out there running no TB's on the road with no probs). For track use they might run the cars without TB's or set the TB's neutral as helpers. Apparently there is no right or wrong way, it's how you like to set the car up for what you want out of it. I guess at least you don't want the two springs to be fighting against each-other.

Van

Just to clarify for people: both normal coil over springs and torsion bars are LINEAR springs -- this means they have one "k" value (i.e. spring rate). For a coil over, it's lbs per inch -- if the spring is 400 lbs/in, then it takes 400 pounds to compress it 1 inch. It needs another 400 lbs to compress it the second inch - which would be 800 lbs total. To compress it 2.5 inches would take 1000 lbs.


If you take these numbers, along with the inches compressed, and put it into a line graph, you'll get a straight line (hence the term "linear").

Torsion bars are the same way -- it takes a certain amount of force to rotate (twist) the bar 1 degree. It takes twice the amount of force to twist it 2 degrees. Just like the coil over example above, this is a linear spring.

There is such a thing as a progressive spring -- this is made by CHANGING the wire DIAMETER on the spring. I.e. the top of the spring has a wire diameter of 0.375" and the bottom of the spring has a wire diameter of 0.25". The smaller diameter (not having as much material) will deflect first. Once it has compressed to the point where the force equals the other rate, then the larger diameter will start to deflect.

Here's an example: Our spring has a dual rate of 200 and 300 lbs per inch. After the spring is compressed 1", it feels a force of 200 lbs. At 1.5", the spring is feeling a force of 300 lbs. At 2", it's feeling a force of 450 lbs (the 300 lbs from the first 1.5" of deflection plus .5" at 300 lbs/in). And, at 3 inches of compression, the spring is feeling 750 lbs of force. If you plot these numbers on a graph, you will find two lines, with different slopes -- one slope for the first rate and one slope for the 2nd rate. (NOTE- these are slightly theoretical numbers to illustrate the example - the smaller diameter part of the spring will also compress more until the coils touch eachother)

If one is able to actually have the torsion bar at the neutral position while the car is at static ride height and all the weight is supported by a coil over helper, then the compression rate below ride height would be torsion bar rate + spring rate. And the compression rate ABOVE ride height would be the spring rate - torsion bar rate.

Usually, the only time the rear is above ride height is under breaking when the weight has transfered forward. Under cornering, if you had no sway bar, the inside wheel would drop below ride height (making the car ABOVE ride height). However, with a rear sway bar, it will work to limit the amount that inside wheel whill move down. If your sway bar rate is greater than your 'spring rate - torsion bar rate', then that inside wheel would never drop below the ride height level while cornering.

Is this setup better than just having the springs and torsion bars working together at one rate? I don't know. For all I know, it could assist in a quicker weight transition from braking to being ballanced before a turn (due to the lesser rate being compressed faster). But on the other hand, in motion, the car is almost never "at static ride height" -- unless you're coasting down a level stretch of raod, in which case, it may not matter WHAT springs you have!

However, if you like the feel of it, by all means, set the car up that way. You're going to be the one driving it.

Chris White

If you are just concerned with compression / extension of the ‘entire’ rear it is not a big deal (other than the fact that the compression rate is different than the extension rate – could cause interesting weight transfer to the front under braking due to rear ‘jacking’). It gets interesting when you get the car to lean (with a neutral TB setting). In theory you will increase the outside wheel spring rate (Coil + TB) while decreasing the inside wheel spring rate (Coil – TB). Then you tie the two together with a sway bar….that makes for a complex system to tune!!!
Just pull out the TBs and get on with it, it makes ride height and corner balancing much easier!

Van

+1 Chris!

shiners780

I've been TB-free for a year and a half, and I don't miss them at all.

I have seen 'neutral set TB' mentioned several times here on the 'list, but had never seen any logical reason for having such a setup. Due to my never ending quest for knowledge I figured I'd start a thread to see if we could figure out exactly why people do this.

edit: oh, and thank you Van for the clarifications.

Skip Wolfe

This has been debated over the years and the - Porsche designed it that was so it must be the best" - is one of the favorite arguments that drives me nuts. The only valid argment relates to the strength of the mounts - upper and lower - and the rigidity of the pivot point. All of the issues can be addressed by upgraded bushings, hardware, etc. Other than the structural points there is no reason to use the bars and as Chris stated, only complicates the equation with essentially two springs of different rates working in conjunction. The KISS theory should be applied here - suspensions are hard enough to optimally tune in standard form without factoring in multiple spring interactions.