For strut bars up top I have the preload as tension down here I think I have compression is this correct? I put in bars in all three locations many years ago....and don't recall what I did

The lower bar prevents outward movement of the uprights, so the bar above is under tension.

Can you explain your reasoning behind this more? It's the opposite of what I'd expect geometrically given the location at the bottom of the suspension. When cornering hard the major force is from the tire on the outside of the turn pushing to the centerline of the car, which would tend to compress the bar. Wouldn't this compression cause any bar pre-loaded with tension to just go slack in a turn, and then do nothing at all? Pre-loading with compression should be necessary to prevent suspension deflection in turns.

It makes sense to me to apply tension pre-load to something at the top of the strut, and the opposite to the bottom of the strut, to counteract chassis flex in cornering, which is a twisting force on the entire suspension assembly (clockwise on the right, counter-clockwise on the left of the car).

Find a diagram of the car's resuspension set up; the tops of the uprights are tied together by the car's chassis, while the bottoms are not. It isn't the chassis that is flexing, it is the uprights themselves that are moving. Under hard turn in, the inside strut is pulled outwards while the outside upright is pushed inwards because neither is really firmly attached to anything substantial. Tying one to the other helps reduce the deflection impact on both struts, and slightly preloading the tie bar (read tightening) increases that load sharing and reduces deflection, and the resulting tire scrubbing, for both uprights.

https://www.planet-9.com/attachments...ion-jpg.53899/

Ideally, you would run a triangulation bar upwards on a 45 degree angle from the bottom of the uprights toward the centerline of the car, where they would connect to a bar running across from one side of the car to the other, which would prevent movement in any direction; only problem is that the triangulation bars with have to run through the transmission, which happens to be in the way. So going across underneath is the next best alternative.

If we imagine your described ideal dual diagonal bar setup, the outside strut diagonal will be compressed in a turn, and the inside diagonal bar would be under tension...

So if we are using a single bar due to spacial limitations, having it under tension would seem to better control deflection on the inside, and having it under compression would seem to better control deflection on the outside. (edit: note that these are in direct conflict, so we have to decide which side is more important to stabilize!)

Now, consider that a Boxster basically unloads the inside rear wheel at maximum cornering, so all of the force is really on the outside wheel. Given this situation, wouldn't a compression pre-load be what you want - by prioritizing the correct geometry for the more heavily loaded side?

edit: Also, since the forces are higher on the outside and we we would expect the deflection here to be much greater, in a turn the bottom of the suspension sides will get closer together (compress together), because only the outside suspension has enough force to really move a lot. This suggests that a bar pre-loaded with tension would actually exasperate the suspension flex issue, causing further increased flex!

Only problem is it doesn't seem do that, preloading slightly tight tends to settle the car more; preloading outwards doesn't have the same effect. We have had customers play with these things a lot (give a track rat something adjustable, and they will play with it), and the general consensus is that running the tie bar slightly tight at rest gives you are more settled and predictable corner exit. Granted, a lot of that is seat of the pants, but drivers that feel more confident based upon how the car feels to them tend to be faster.

It's hard to argue with real world experience, but that is really surprising and confusing. I wonder what is really happening here?

Perhaps rather than preventing the cornering deflection that normally occurs in a stock car without the brace, having the brace causes it to always be "fully deflected," even in a straight line, so you don't get the abrupt transition/shift that makes the car feel less stable.

I'd also expect the factory alignment to anticipate the shift caused by the cornering load... so if compressive loading stabilizes the suspension better, perhaps you'd only get the benefit by also altering the suspension alignment such that it's no longer pre-compensated as heavily for this shift/distortion?

You have to remember that some people are running the nonadjustable solid bar, and they say they can feel the difference on the street, just like the adjustable bar cars, and where the suspension loads are considerably less.

As for factory alignment expectations, I would seriously doubt they factored in what happens when someone with 500 hours of track seat time and R compound tires hit the apex. The factory only concerns themselves with the more pedestrian, everyday driver loads.

My comments were in the context of considering adding a bar to my own car (completely stock, street use).

Wont argue with experience, however it surprises me. Nevertheless can we agree that the forces are higher on the outmost side, and that they mainly affect toe? Why I see it this way is that the subframe obviously have two "legs" the forward one holding the LCA is already tied together, this takes the main cornering force, the rear leg holds the toe link so...