Printable Version of Topic

Posted by: Dave-O Oct 22 2006, 02:09 PM

Hi guys/gals,
Now that the AX season is officially done up here in the north, it's time to start thinking about winter "improvements". At the beginning of this season I set my alignment/ride height/corner balance and just left it as is all year, focusing on driving. I'm contemplating trying to drop the front of the car down another inch in an attempt to transfer a little more of the weight up there. Right now I have about 2.5" of suspension travel in the front and don't want to sacrifice any of that.

In order to lower the car I would have to chop the struts and put shorten inserts in. To justify this, I wanted to verify that the camber curve would still be acceptable. I made a quick excell sheet graphing the camber gain as a function of A-arm deflection. Hopefully this will be useful to other people as well.

As of now, at rest the A-arms are at about +3 degrees. On the bumpstops they are at about -10. By lowering the car an inch they would start at about -3 degrees and assuming I cut the struts 1.5 inches on the bump stops would be at about -19.

I took some quick measurments on the car but they are by no means extremely precise.

Here is the graph...


Attached image(s)

Posted by: Dave-O Oct 22 2006, 02:11 PM

Based on the graph I think I will still do the shortened struts as I could use some new struts anyway. I may not go down a whole inch but we will see when I get the car on the scales next spring.

here is the excel sheet...let me know if it works for you.




Attached File(s)
 camber.xls ( 27.5k ) Number of downloads: 92

Posted by: Aaron Cox Oct 22 2006, 05:40 PM

what about raising spindles?

A arms now point down farther starting out further positive at the same ride height, and giving you a bit more travel....

Posted by: Dave-O Oct 22 2006, 06:17 PM

Step 0, read the rule book. Altered spindle location is not allowed in my class.

Posted by: Aaron Cox Oct 22 2006, 06:52 PM

but you can shorten the housings? wow...

cool prject tho!

Posted by: groot Oct 23 2006, 06:39 AM

Hey, Dave,

I love to see people using science to make cars faster!!! Good work.

For reference a certain vehicle (no specifics due to proprietary information restrictions) with a Macstrut front suspension that I have actual Kinematics and Compliance data on has a fairly linear camber curve of .168 degrees/cm around 0 (+/- 2.5 cm) on one side and .160 on the other. One of the limitations of the macstrut is the lack of ample camber gain.... ideally, you'd like it to be .4 degrees/cm or so... depending on many things.

More often than not most people measure camber gain by degrees per distance of wheel travel.... but I understand why you compared it to LCA angle to horizontal.

Posted by: brant Oct 23 2006, 08:46 AM

Aaron,
alot of rules say exactly this.
this is the reason I don't have raised spindles either, but instead have chopped housings.

Moving a suspension point is not legal.. Technically the wheel spindle is a suspension point.

however the use of any shock choosen by the owner is legal.

brant

Posted by: Dave-O Oct 23 2006, 09:53 PM

I certainly have a "certain vehicle"'s suspension design in mind for a future project. If you take my excel sheet and add 10" to the "a" and "c" components the graph gets much much nicer!

Class rules aside...What do you think of using a later model porsche strut with a larger strut angle and moving the upper mount inboard? I haven't done any measurments on these struts but assuming you could move the top inboard by 2.5 inches. Here is what the graph would look like. With longer A-arms it would be more linear.


Attached image(s)

Posted by: groot Oct 24 2006, 05:53 AM

That's an interesting idea (using the later model strut), which I hadn't considered. I didn't even know that Porsche changed that angle. Seems like a reasonable way to get better camber gain.... but I have to think on it a bit.

The "certain vehicle" I was referring to is currently in production, but it's camber curves are fairly representative of the front macphearson strut suspension.

But, if you were thinking about my race car (thanks), please know that I optimized for roll center migration. I certainly weighed roll center migration higher than both camber gain (since I actually decreased it) and bump steer (I raised the rack, but my toe curve still got slightly worse than stock). But, I did that because I like to use suspension travel and think huge stablizer bars aren't the proper way to deal with roll. In my opinion this makes the car easier to drive.

Posted by: groot Oct 24 2006, 06:34 AM

So, I've thought a bit more about the steeper angle on the strut.

If one makes this change without moving point 3 and point 4 on the LCA (where the lower control arm meets the body....maybe this is the "a" and "c" you referred to). There will be an increase in camber gain, increase in side load on the strut (resulting in more stiction), greater change in the toe curve (curve is less linear), slight (negligible) change to roll center... I think that's most of it. Well, not quite, there's kingin angle and scrub radius, too, but those are relatively minor contributors to this situation.

Now, if you move your steering rack you can center your toe curve to minimize the effect of the increased non-linearity. You can move points 3 and 4 inboard to settle your roll center migration down at the expense of some lost camber gain.

Seems like a reasonable thing to try.

Posted by: ottox914 Oct 24 2006, 06:35 AM

Cool stuff, Dave. Was this little exercise part of your application for the masters program for ME's? See you at the club meeting tomorrow, we can talk about that future "project" and see what taladaga mike has for us...

Posted by: Dave-O Oct 24 2006, 04:41 PM

the "a" value is the length of the LCA. the "c" value is the horizontal distance between the LCA mounting points and the upper strut mount. For simplicity, in my tables I just took the distances from the centerline of both mounting points of the LCA. In the graph above I simply decreased the c value by 2.5 inches to simulate moving the upper strut mount inboard by this amount.

I have not done any actual measurements on the late-model struts as unfortunately I don't have a 996 or a boxter sitting around to take measurements off of. I have heard that they have a larger strut angle but I really don't have any real data to back that up. The 2.5" difference is simply a guess.

It seems that with a larger strut angle and with something similar to your longarm suspension design, you could get a slightly more aggressive camber curve while keeping it relatively linear.

I'm entirely a newbie when it comes to suspension design. I had not accounted for toe-changes. I could potentially make another graph to look at this.

The other benefits of using a later model strut is that it is already set up for coil springs. Another concern we had was with the overall width of the vehicle (very important in autocross). Potentially with a larger strut angle we could keep the wheels further inboard. Finally, it also seems like a cool thing to do!

Later I will take some time to change the independent variable on the graphs to cm of suspension travel to make it more relative.

Here is a graph of the camber curve while moving the upper strut mount inboard by 2.5 inches and using 21 inch control arms.




Attached image(s)

Posted by: Dave-O Oct 24 2006, 04:53 PM

I didn't mention any of this craziness on my application, which is probably why I got admitted! They still think I'm a mild mannered biochemist who is interested in mechanical systems.

Posted by: J P Stein Oct 24 2006, 05:05 PM

Kevin:
Take a look at Bills Buckster Build pics.
There's a good pic of the front strut from which you can discern what they did to maintain the proper angle on the strut (to keep the side loads under control) while getting the front hub outboard.

I'm already tearing up Bilstein inserts with side loads....doan need to increase that angle.

Posted by: jhadler Oct 24 2006, 05:19 PM

David,

Very interested in what you come up with, the Street Prepared rules regarding struts are pretty much the same as in Street Touring, so that could be of use to folks in other classes as well!!

-Josh2

Posted by: Brad Roberts Oct 25 2006, 03:23 PM

For experienced drivers? Yes. For newbies? No. People who have ran go karts with no suspension travel can jump into anything with super high spring rates and small anti roll bars and drive the wheels off of it. Somebody who hasnt driven a stiffly sprung car and gets into one typically feels uneasy.

Everyone is different

Now back on topic!

How does this help me go faster? I already know not to turn the wheel a whole lot or the car slows down. I actually teach people this for AutoX. I make them push a car.. then push a car with the wheel turned.

The numbers are great. I understand everything being said, but bottom line, how does it help us? If a Boxster strut can be used (easily actually) why wouldnt I just run a Boxster?


B

Posted by: jhadler Oct 25 2006, 04:10 PM

'cause a boxter costs more????

But really, if a Boxter strut can significantly improve the camber curve of the front, that would be way cool. In my class I'm allowed _any_ strut, so long as I don't alter pick-up points...

-Josh2

Posted by: Dave-O Oct 25 2006, 04:37 PM

Josh, the talk of the boxter strut is for an XP 914 that ottox914 and I have been daydreaming about. The problem with the boxter strut on a stockish car is that the top mounting point will have to be moved inboard by a significant ammount. For our cars in Street Touring/Street Prepared rules we are stuck with the first graph that I posted. Since we can't change the A-arm and we can't change the upper strut mount, it doesn't matter what strut we run, that is the camber curve we will get based on those parameters.

Brad, as to why not just run a boxter, I had conisdered it. But we are thinking (dreaming) about building a car for a specific SCCA AX class. To reach the minimum weight in the class the boxter would require a lot of sawsall action. Also the initial cost is higher. Finally, we want to run a WRX STi drivetrain...sticking with the 914 chassis just makes more sense.

Why do these crazy mods to a 914? If we are going to spend a bunch of money building this car? Why go half-assed? Like you said, if the strut can be used easily. We also will get the boxter brakes, hubs, and coil-spring set-up. Seems like win/win/win.

This all just talk right now, I won't have any money to sink into such a project till I finish gradschool. But it seems like a much better idea to figure these things out before I fire up the credit card and the welder!

Posted by: Brad Roberts Oct 25 2006, 05:03 PM

I have a 993RSR hub/carrier and a Boxster hub/carrier (both with struts) Tell me what to do and what to measure. I'll let you know what I find.

GT2 Porsches and 993RSR's use a "raised" wheel bearing. They lower these cars by raising the bearing in the carrier compared to a stock 986/996 carrier. The lower control arms are identical in a GT3 street/GT3 Cup/GT3RSR/GT2.

My friend Jason in San Jose has been buying complete suspension setup from wrecking yards for under $800.



B

Posted by: J P Stein Oct 25 2006, 07:22 PM

OK, Brad, lay the Boxster hub/carrier up into the 914 strut area and tell me what needs to to altered (oher than the ball joint) to make it fit. I can't run my big wheels/tires in front due to scrub radius problems. The cure is a more room twixt the
hub & strut. I don't mind altering the upper strut mount within reason.

Posted by: groot Oct 26 2006, 05:46 AM

For the record, I don't run really high spring rates (currently 400 #/in in the front). I do agree stiffly sprung cars are difficult to drive....

You're absolutely right, everyone is different

Posted by: Brad Roberts Oct 26 2006, 12:13 PM

Ok. This weekend I will place a 914 strut (minus spindle) into a Boxster and 993 front hub for comparison purposes. I'll find out what it is going to take to make a Boxster upright work in our tubs. This has always interested me since I first laid my hands on the Boxster upright and found out it uses a 914 rear wheel bearing!!

One of the guy's I'm working with right now happens to have several 914 struts stripped of their spindles.

400#'s is pretty stout for a 4cyl car that weighs 1800lbs!! (I think class weight used to be 1850 or something)



B

Posted by: Dave-O Oct 26 2006, 04:55 PM

Wow Brad, that would be awesome! I think the only thing we really need to know is the difference in angles between the hub and strut itself. This will determine where the top mount will have to be placed (assuming you use the same location for the ball joint relative to the tub. If you get the measurements I could do the trig. to figure the angle and determine how much the mounting point will have to be moved. (lay a long straight edge on the hub and measure the distance between the straight edge and the strut at two points, (e.g. 3" at one point and 4.5" at a point 10" up the strut). I hope I didn't make that confusing...

The other measurement would be the length of the top of the strut to the ball joint when fully compressed. Since it would make sense to use all the aftermarket boxter dampeners, we would have to know how much (if any) the upper mounting points would have to be raised. On our AX project it wouldn't matter much if we had to cut holes in the hood for the mount, but some people may not be as utilitarian as I am!

Thanks again!

Posted by: Brad Roberts Oct 27 2006, 06:28 PM

First observation:

914 stock strut tube measures 52mm at the fat part
50mm's at the narrow lower section where the spindle rides.

Stock Boxster strut where it enters the wheel carrier = 51.50mm's

The stock 914 4 cyl 4 lug strut tube is a NO brainer for the Boxster wheel carrier. One small piece of metal welded to the outside and the wheel carrier will pinch down on a 914 strut tube with no issues. I can even show you how to raise and lower ride height based on the position of the welded tab. The tab allows the tube to be locked down beside the "pinch" method the hub uses. If we use the 914 or 911 strut tube.. then we dont have to fool with ball joints.

I'll shoot pics soon.


B

Posted by: groot Oct 28 2006, 09:54 AM

You guys might be on to something here.....

Brad, with the 400# springs, my ride frequency in the front is only 2.5 Hz, (my motion ratio is different than other 914s)... which is a lot lower than most racers run their production based cars... and lower than most sports racers, too. Wheel rate is ~280 #/in.

At this weight (1870 race weight), I think of my 914 somewhere between a production-based car and a sports racer with no aero.

Posted by: Dave-O Oct 28 2006, 03:11 PM

I didn't realize it would be this easy! If you were planning on using torsen bars and stock A-arms, the only real hurdle would be moving the upper mount (we're not yet sure how far it has to be moved).

If you were planning on running coilovers with non-stock A-arms like Kevin's, would it be better to use aftermarket boxter struts like the ones on Bill's Spec Boxter? Those threaded aluminum bodies looked sweet!

Posted by: Brad Roberts Oct 29 2006, 01:54 AM

Here is a really good shot of the Boxster front hub and Bilstein PSS9



B


Attached image(s)

Posted by: J P Stein Oct 29 2006, 07:54 AM

Thanks, Brad, the pic tells the tale.

It appears the strut leads the ball joint by an inch or 2 at the bottom.
also.....running a 914 strut thru the carrier would hose up the angles (2)
axle to strut. Using the Boxster strut would take care of that, but then there's the ball joint problem. Those in the pic appear to be pressed into the A arm....true?....and the angle of incidence is different than a 914 A arm.....not to say that that is insurmountable, but some serious fab work would be necessary.

Would it be worth it to gain 1-1.5 inches of backspace for a wheel without getting into
scrub radius problems? I suppose it depends on what a guy could come up with for a ball joint, but it's beyond my fab skills.

Posted by: ottox914 Oct 29 2006, 08:05 AM

Don't forget that for our purposes, we are considering use of the complete boxster coil over/spindle/disc/caliper, with custom lower control arms. If the 914 strut and T bar set up looks better on paper, we can go that way with the boxster spindle/disc/caliper. Maybe an inspired design for the lower control arms will take care of some of the points you make, and allow benefits to be gained in lighter weight, bigger brakes, camber curve, coil over vs T bar, and wheel spacing, so much the better. We've got time on our side for now. Lets keep talking this thru and see what comes of it.

Posted by: J P Stein Oct 29 2006, 09:07 AM

Heh....you must be younger than me.

Talk is easy tho, so.......
The gain (to me at least) would also be in getting the coil overs for ease
of adjustability. I doubt there would much (if any) weight loss. One would have to use the Boxster brakes....kinda comes with the package, eh? No gain for me there unless ABS came with it....not likely.

The trick is gonna be getting the axle in the correct plane.....x,y, & z
Since there is no "axle" (instead, a hub carrier) on the Boxster, this gets tougher.
One would have to precision fabricate a substitute for this.....not insurmountable again. As I see it, you'd have to fabricate a jig that would be set in place while the 914 axle is in place, then remove the 914 strut. A thousand pound jig base would work on a garage floor ..."Simply" then install the Boxster "Axle" to suit. An inspired design on a 914 A arm (cut off the old balljoint mount & start over) would be gud.
The car & its steering can't move till your done with both sides. All this assumes that one can get the strut top mounted in about the same place as the 914 strut.

My personal time factor doesn't have room for this....that is insurmountable

Posted by: Brad Roberts Oct 29 2006, 10:21 AM

Ira from Tarret engineering has a live simulation complete 911 front suspension that he takes to trade shows. I'm going to borrow it for this experiment.

I agree.. the benefit is using the lower control arms (or something equivalant) The ball joint is pressed in.

Your looking at GT3 street control arms. They shim out for more camber at the pivot point on the chassis. $385 per side with X number of shims.



B

Posted by: Brett W Oct 31 2006, 01:02 AM

How does the wheel bearing mount? Integrated hub/bearing or spindle and tapered rollers? Trying to figure out some things.