Full mapping of Legacy Suspension Points?

[BAC5.2]

Has anyone done this and can save me some time?

 

My intent was, sometime this week or early next week, break out the tape measure and a notebook and map out all of the suspension points.

 

I'd then take those points and build a virtual model of the suspension in 3D (Solidworks). That way I can get a predictable model of camber and toe-change, as well as calculate boring numbers that mean things only to vehicle dynamics guys like me.

 

The ultimate goal is to identify the ideal ride-height, as well as other things that I should change (such as body spacers), for when I do my Endura-Tech coilover install at the end of this month.

 

One question, aside from the first one I have, is if anyone knows the amount of shock stroke a stock OBXT has? I could estimate it, but it would be nice to know.

 

Thanks!

[teK--]

Subscribed. I look forward to seeing the results of what you are planning to do.

 

I have read that the rear multi linkage design increases toe in and negative camber as the suspension compresses. The front, well being a Macpherson strut behaves like any other design since the 80s or whenever it was introduced.

[BAC5.2]

I suppose the results will only directly reflect what happens on an Outback, but I suppose the general idea will carry-over to the Legacy. I have body-spacing mounts to take into account, along with the longer struts and whatever other differences an Outback has between a standard Legacy sedan/wagon.

[ifbiker]

I did this but would say my numbers are not good enough to share. If you have a level floor, inclinometer and laser you can do a pretty good job.

[grimsleeper]

This is definitely my kind of project. I do 3d modeling/rendering/simulation for Sikorsky helicopters and have thought about doing this in my free time, although the reality is when I get home at night it's the last thing I want to be doing:lol:

 

You'll be able to do the modeling in Solidworks but in order to get an accurate physics simulation you'll likely need to bring the model into a more specialized app - something like:

http://www.lmsintl.com/simulation/vehicle-motion or

http://www.mitchellsoftware.com/prod01.htm

 

Very interested to see the process & results! If there's anything I can help with let me know.

[BAC5.2]

Thanks Grim, I'll be sure to remember that. I have loads of modeling experience, designing gearboxes and CV joints and such, so this really shouldn't be tooooo difficult.

 

I've got more than a few books regarding vehicle dynamics, so I think I can get Solidworks to show me the data I want (albeit, it may not be as "pretty" as Mitchell or others).

 

Worst case, I can upload everything into Adams, and do analysis with that.

 

I don't have a laser, nor an inclinometer. I DO have a level floor, a few yard sticks, a note book, some sharpies, and a tape measure. That should get me close enough.

 

I might start building the model today. I've got some other CAD work I need to do, so I might take a break and do a quick rendering of the wheels. I want to build a reference base, and I want to adjust things so I can type in tire/wheel sizes and have the whole model respond. The FSM lists track (center to center) and wheelbase (hub-to-hub), and I know wheel dimensions, so I should be in good shape!

[grimsleeper]

I think the key to getting decent results from this will be dependant on the simulation of weight forces (and their location) on each corner. In reality this is going to be a dynamic number but I think an approximate static weight should net the results you're looking for. Maybe this data could be derived from someone who's had their suspension corner balanced?

 

I know Solidworks has a decent rigid dynamics sim but how does it handle flex resistance, i.e. from bushings, springs etc.?

[BAC5.2]

Solidworks CAN account for compliance through material selection, but I'm not really interested in going that far.

 

I want to simply build a representative model of the suspension. Then things like ride-height and alignment can be optimized for best overall performance. Getting into dynamic events, I'd probably just use Adams, and I'm not overly interested in that.

 

I mainly want to see geometry changes with respect to wheel travel, and see what changes things like wheel width and offset have on other aspects (such as roll center, scrub radius, etc.) I'm not overly concerned with dynamic impact of bushings and spring deflection.

 

First and foremost, I'm doing this as a hobbyist, and to ensure that putting coilovers on an otherwise stock Outback isn't going to jack up geometry.

[grimsleeper]

First and foremost, I'm doing this as a hobbyist, and to ensure that putting coilovers on an otherwise stock Outback isn't going to jack up geometry.

 

I don't think there's any question that it will;) especially if you're considering a significant drop - which you may or may not have a choice in the matter depending on what CO you go with.

 

still in for the results, though.

[BAC5.2]

I think it'll be OK. And, part of the reason I'm doing this, is to find out which spacers NEED to come out, to retain optimized geometry.

 

I'm looking at a drop of right around 2" flat using Endura-Tech twin-tubes with custom, wagon friendly, spring rates.

[BAC5.2]

So I am at an impass.

 

I had initially thought that I would model each corner individually, and simply leave it as sketches. Since all I really care about is geometry, that should suffice.

 

But then I began thinking about it, and it would be pretty cool to model each suspension component as close to accurate as possible. Then, simply build an assembly of each one.

 

Doing it this way would certainly make things more "visible" and probably be easier to visualize and read. It would also allow me to model different wheels/offsets, and simply interchange them. I could also put all 4 corners in a single assembly and organize key measurements to be labeled appropriately. The down side is, it would take longer.

 

I was going to "add-in" a sketch of the body. That would give me a rough estimate of visual ride height, which could be useful.

 

What do you guys think? Any input at all?

[BAC5.2]

I think I might just model each component. Not necessarily for correct visuals, but at least correct necessary geometry. So instead of that curved lower rear link, it'll just be a straight bar. But the distances should be the same.

 

That way, things will be a little easier to visualize and locate. I can just use floating points to set the different pickups, and I can keep things lined up correctly. I think it's the best bet, especially since it'll be visual. I can even make animations, watch the interaction of all tires at once with various different body motions.

 

I think it'll probably take me a solid week. I better get cracking since I expect to install my coilovers on the 25th.

[scoobydoobie]

subscribe to see Phil work his magic

[BAC5.2]

I'll use this thread as my working log and try to post updates as the project progresses.

[BAC5.2]

I drew up an Outback wheel really quickly. Took maybe 20 minutes. I'm not really happy with it, but it is technically correct and the important dimensions are correct (offset, diameter, width).

 

I'll probably change it, but I figured I should get started.

 

http://i267.photobucket.com/albums/ii287/bac52/SWWheel.jpg

[thx247]

On an alignment rack I added weight to the rear of a stock 05 OB and watched rear toe increase a couple hundredths, negative camber increased a couple tenths.

 

Doesn't account for driving force though, but its data.