[Mykl]

Originally Posted by drees

Ideally, whenever you lower the car, you'd increase the bump travel and reduce the droop travel by the same amount you lowered the car.

Another reason to increase bump travel is if you're running significantly stickier tires. Such as, if your car is adjusted to run RT615's and you make a switch to running V710's. The additional grip of the much stickier tires may cause the suspension to bottom out.

[Splash]

Originally Posted by stretch

Right, but a matched spring/strut set should not have droop travel that extends passed the length of the spring. Sorry if I was unclear of the point I was trying to make.

No set, matched or not, should allow this. Bad things happen if the springs unseat.

Originally Posted by stretch

I agree that we must use progressive or tender springs on stock or stock-length struts, but not at all for the reason you state.

Extension IS the only reason they are used, and the reason the OEM springs are progressive. The total amount of travel an STi comes with PRECLUDES the use of any linear spring.

If one were to make a linear spring for the stock amount of travel, the spring would either be too short and would unseat on extension, or it would be too long and coilbind at full compression.

This is why coilovers with linear springs alone will also have significantly reduced total travel.

Originally Posted by stretch

The reason lowering springs must be progressive is because firmer-than-stock springs would run out of extension before the strut does and would become unseated. The problem is worst in the rear, where each spring supports roughly 685lbs. A 250lb/in linear spring will support its 685lbs with 2.75 inches of compression. However, the car has more droop travel than that even at stock ride height, and more if lowered. Thus, the spring would not stay in place without a helper spring.

My RCE's are firmer than both OEMs and Pinks. The S-techs almost mirror the Pink rates. The OEM springs are a combination of taller than the aftermaket ones, AND LESS progressive. You need the progressive part of the spring to hold it in with the full droop travel available on the OEM struts, especially when lowering on them because you are shifting bump travel TO droop travel to obtain the lowering amount... My S-techs had a crapload of droop available, but only about 10mm of bump.

Originally Posted by stretch

The solution is to either run a shortened strut (of which none are available off the shelf), run a helper spring, or run a progressive spring rate and avoid the problem. I, however, do not like progressive springs- if I did, I'd just let the car ride on its bump stops! I don't even like having to use helper springs because they make a linear spring slightly progressive, even if their impact is minimal. A shortened strut is the preferred solution here, but currently one must buy coilovers to get that.

Here is why coilovers don't buy you this... They will only add bump travel if the lowest point of the shaft is lower (in relation to the ride height of the car) than stock. Since the bottoms of the stock struts are already right above the driveshafts, lowering the lowest shaft point is not feasible. So ANY lowering of the ride height results in reduced bump travel, no matter what you do to the strut's overall length.

The only REAL way to increase bump travel is to raise the ride height. Rally cars do this via the DMS 50 coilover which is really long, raises the car, and has a 12" progressive spring.

You could redesign the Subaru front suspension to allow the strut body to be lowered in relation the chassis, but it would have to be on either side of the drive shafts to do so.

Originally Posted by stretch

Bilstein often makes two versions of their struts for cars, one OE size and one shortened for use with lower springs. They offer the shortened strut for the above reason. Too much droop travel is not good if you're springs can't use it.

These won't help bump travel for the reason stated above. You will get your lowering, and you will lose droop travel, but you will get no more bump travel. By the time you used the collars to raise the car back up on the shorter bodies to get more bump, you will run into the situation like that WRX in the link I posted where his droop travel has vanished.

Originally Posted by stretch

Coil bind shouldn't be an issue unless running tightly wound, short springs. The stock front springs, for example, only make four winds. Fully compressed (in a vice), the entire spring would be about two inches! In fact, coil bind is more of an issue with progressive springs because they typically have much more winds in them.

Keep in mind that, in some progressive springs, the winds that are closer together are SUPPOSED to coilbind - just like a tender spring. They aren't there to help the spring to be more comfortable, they're there to keep the spring from falling out at full droop.

The S-Techs are a prime example of this because it's how they get a 2" drop and still not fall out of OEM struts. They come with plastic coil wraps to keep the progressive part from making noise when they collapse fully, which happens at static loads.

A linear spring will coilbind easily, and a coilover spring (being only 2.25 or 2.5" diameter) will do it REAL easily. My old H&R coilover springs were 6" long at rest and were in full coilbind at 3.5"... 2.5" is NOT much range...

Coilover spring companies have the ranges of free length and max compressed heights listed... See if you can find a spring that will accomodate an '04 STi's travel range (65mm bump, 95mm droop - 160mm total, which is almost 7 inches)... When you find it, see how long it is, and how high the car has to sit to use it.

THEN, keep in mind WHERE the strut body sits in relation to the chassis, recognizing that any increase in the body length happens ABOVE the static ride height.

I maintain that, as long as you can't lower the strut body in the chassis, ANY amount of lowering of the chassis will reduce bump travel, no matter how long or short the struts/springs are.

The only thing you can do, is make a coilover not kill off all the droop travel, like the ZZYZX units do.

Mike

[drees]

Mike, I think you're mistaken about the shorter/strut body and increased bump travel.

If you are shortening the strut body to increase bump travel, keep the bottom position of the body housing the same which lowers the top position of the body.

That will increase bump travel the same amount you have shortened the strut body as long as you don't run into coil bind.

Just because you shorten the strut body, does not mean you have to move the spring perch.

If the highest you can position the spring perch on the shortened body ends up lowering the spring perch, you can use a longer spring. Of course, if a longer spring binds, then you've gone about as far as you can go without moving the strut mounts higher.

When that happens, RCE lowering plates will get your tophat 3/8" higher.

[Splash]

You CAN use a shorter body (which also means a shorter shaft) to get more bump, as long as the ride height doesn't change. But you are exchanging droop to get it. You can do the same thing by rasing the car with stock strut body and shaft. The same exchange is made. If you then lower the car on the shorter body/shaft, you are right back where you started, plus you now have less total travel.

Though the RCE plates DO get you 3/8" lowering of the chassis with no travel penalty. You could get the same effect by moving the spindle bracket up the strut body, but you can't do that without cramming the strut body into the CV joint.

The thought had occured to me that this is kinda off topic now... In so far as the bumpstops go, I see the bumpstop as being about 63mm long.

If that table from SOA includes the bumpstop as part of the 65mm bump travel number, then stock cars only have 2mm of free bump travel, which tells me that a 2" lowering spring must overcome almost 2" of a 2.5" bumpstop to get that drop, and even then either have no bump travel left, or attempt to compress that last 0.5" of bumpstop before it crashes. That does not sound right to me.

What sounds more right to me, is that the bumpstop is NOT included in the SOA number, meaning OEM springs make the strut have 65mm before they hit the stops. In this guise, a 2" lowering spring would only leave you 15mm of free bump before hitting the stops, which not only makes more sense, but mimics what I felt with a car in this configuration.

Mike

[mxpop]

Bumpstop (rubbers) in motocross are an essential and tunable part of the travel for the rear shock. We have long tapered ones and some with steps and some with diff densities. We use full travel (almost 13 inches) very often each lap while most cars dont, making the bike's bump stop needs more critical.

Total spring rate must be calculated with all aspects of support including the effect of nitrogen pressure, spring, bumpstop, etc.

[stretch]

Originally Posted by Splash

You CAN use a shorter body (which also means a shorter shaft) to get more bump, as long as the ride height doesn't change. But you are exchanging droop to get it. You can do the same thing by rasing the car with stock strut body and shaft. The same exchange is made. If you then lower the car on the shorter body/shaft, you are right back where you started, plus you now have less total travel.

Mike, that's not quite true. The stock struts are monotubes. The use of a twin-tube strut would allow a shorter body and longer travel because there would be no floating piston wasting space inside the strut body.

Zzyzx coilovers are twin tubes, which is a reason they can increase travel. The Koni's I just installed increased travel in the rear as well, mostly in the bump area. I'd estimate I gained over an inch, but I didn't actually measure.

With monotubes, you're right that total travel can't be increased (without sacrificing bump travel) because there's not enough room for a bigger strut. However, I'm not sure I agree of the importance of this- in the case of a lowered, firmer car, there's no need for stock-like droop travel. In this case, a shorter shaft and shorter body could be used to trade droop travel for bump travel. Droop travel would be lost at twice the rate that bump travel is gained, so overall suspension travel decreases.

However, one can still have adequate bump travel. Consider a car lowered 0.5 inch with 500lb/in, linear springs, a choice I think is reasonable for a track car. On a stock strut, this would mean 4" of droop travel (would need a helper spring) and 2" of bump travel. However, one could design a matched strut to have a one inch shorter body and one inch shorter shaft for a total loss of two inches of droop travel. That would bring the total amount of droop to 2 inches (down from 4"), which is all a 500# spring could use anyway (1000lbs per front corner = 2 inches). The 1" shorter strut body permits 1" more bump travel, bringing total bump travel to 3 inches. Half of that can be bump stop, so we still have 1.5 inches of bump-stop free travel and more total bump travel. Total suspension travel has decreased from 6 inches to 5 inches, but that's enough for spring rates double the stock rates. The car has been lowered and bump travel gained at the expense of otherwise useless droop travel.

I meant to take apart a rear strut last night and measure travel. Perhaps I'll do that tonight.

[psuLemon]

Just wanted to say this is an awesome thread!

[Guest_1u53r]

Originally Posted by stretch

OK, I just measured 3.5" of wheel travel when lifting the car from rest (no driver). That should correspond to 3.2 inches of strut travel, which is close enough to my original statement.

Strut based suspensions have a 1:1 motion ratio so the strut is compressing 3.5" as well.

Good work on the measurements. More ammo for the anti-big-drop argument!

[stretch]

Originally Posted by nhluhr

Strut based suspensions have a 1:1 motion ratio so the strut is compressing 3.5" as well.

Yeah, I was thinking about that today and came to the same conclusion. I had read earlier that the motion ratio was 0.95, but that doesn't make much sense when the strut is connected directly to the hub, now does it?

That, by the way, means the STI is on its bump stops even without driver and passenger weight on stock springs.

Good work on the measurements. More ammo for the anti-big-drop argument!

Thanks. And for that, here's the rear strut. There's 4.5 inches of strut travel before the slightly over 3-inch-long stops. Anyone want to measure the distance between full droop and their ride height? I can't measure this myself anymore because I now have Koni's installed, and because they are hydrolic struts, they lower the car a smidge.

The rears have an extra progression in their longer bump stops. Overall travel is greater. However, I wouldn't yet rule out the rear's riding on those stops, too. It looks pretty close.

Incidentally, those Koni's I installed also had at least 12 inches of total strut travel while only being slightly longer overall than the stock STI struts. I didn't actually measure travel, but there was a huge amount. I'd bet at least two inches of bump and about an inch of droop was added.

[Guest_1u53r]

Originally Posted by stretch

Yeah, I was thinking about that today and came to the same conclusion. I had read earlier that the motion ratio was 0.95, but that doesn't make much sense when the strut is connected directly to the hub, now does it?

Forgive my oversimplified statement. I believe the motion ratio of the suspension in vertical compression will be = Cos(SAI). This is because the strut is angled and therefore compresses slightly more along its axis for a given amount of vertical travel of the wheel. The SAI on the STI is around 15* so the motion ratio becomes 0.966. But still almost 1:1 though.

That, by the way, means the STI is on its bump stops even without driver and passenger weight on stock springs.

[drees]

Originally Posted by stretch

Yeah, I was thinking about that today and came to the same conclusion. I had read earlier that the motion ratio was 0.95, but that doesn't make much sense when the strut is connected directly to the hub, now does it?

The motion ratio must be less than 1 as long as one end of the suspension arm remains fixed and the spring is not the same distance from the pivot point as the wheel. The longer your suspension arm, the closer the motion ratio gets to 1. Not sure what the motion ratio is on the Impreza...

Edit: I see that nhluhr supplied a real motion ratio and also notes that the SAI will affect the MR as well.

[Guest_1u53r]

Originally Posted by drees

The motion ratio must be less than 1 as long as one end of the suspension arm remains fixed. The longer your suspension arm, the closer the motion ratio gets to 1. Not sure what the motion ratio is on the Impreza...

Edit: I see that nhluhr supplied a real motion ratio.

The fact that the wheel, hub, and strut move together as a non-pivoting unit mean the MR is 1 except for any inclination of the strut. If the 'damper' was attached inboard of the ball joint, it would be a straight up ratio (+ inclination effect) of the lengths involved.

Nonetheless, don't want to stray too far from the point of this thread.

I say this explains a LOT of the bobblehead and rough riding of the impreza, but I have to question the motives of the Japanese engineers who put this package together...

[stretch]

The fact that the wheel, hub, and strut move together as a non-pivoting unit mean the MR is 1 except for any inclination of the strut. If the 'damper' was attached inboard of the ball joint, it would be a straight up ratio (+ inclination effect) of the lengths involved.

Ah, I hadn't thought of that since the angle is so slim. Thanks again (and to Drees, too). So, camber plates can actually decrease your wheel rate by a percent or so.

I say this explains a LOT of the bobblehead and rough riding of the impreza, but I have to question the motives of the Japanese engineers who put this package together...

The bobblehead is undamped motion, which means I doubt the bump stops are to blame for this. The bump stops cause a spike in spring rate, but the motion should still be damped enough not to cause this bobblehead. The stock struts are widely considered overdamped and can probably handle the increase in spring rate from the bump stops.

I no longer have any bobblehead, but I trimmed my bump stops and installed Group-N tops at the same time. I can't say for sure which did the trick, but I'd bet it was the tops.

[Guest_1u53r]

Originally Posted by stretch

Ah, I hadn't thought of that since the angle is so slim. Thanks again (and to Drees, too). So, camber plates can actually decrease your wheel rate by a percent or so.

Adjusting the camber plate from its '0' setting to a max of around '-2' would only increase the total SAI to about 17*...

cos15 = 0.965
cos17 = 0.956

So yep, about a 1% change in MR. That's almost nothing though... a 250lb/in spring becomes effectively a 247.5lb/in spring.

There are other drawbacks to using a camber plate as well, such as altering the kingpin offset (and thus, the scrub radius). With a camber-plate car, it makes sense to use wheels with less offset (say, 48mm instead of stock 53) to maintain a 0 scrub radius.

Also, camber/caster plates should ALWAYS be adjusted to exactly the same amount on both sides, since even a small imbalance in SAI will definitely result in a persistant pull. If you need to 'equalize' the camber left to right, do it at the camber bolts on the hub, not the camber plate.

[stretch]

Ah, good discussion.

There are other drawbacks to using a camber plate as well, such as altering the kingpin offset (and thus, the scrub radius). With a camber-plate car, it makes sense to use wheels with less offset (say, 48mm instead of stock 53) to maintain a 0 scrub radius.

Is geometry really to blame for a drifting car, or the alignment? On a heavily cambered car, the insides of the tires have more weight on them than the outside (when going straight). Thus the center of drag on that tire is no longer in the center of the tire. The scrub radius is measured from the center of the tire, but that no longer is the center point of the forces on the tire. This shift is probably greater than the shift due to a change in scrub radius and I think the more likely reason that lower-offset wheels should be used. It helps to center tire forces on the scrub radius.

Increasing kingpin inclination (SAI) increases dynamic camber just the same way caster does. Most people would find this desirable, although I suspect it would have the same effect on bump steer as caster does. It also improves your camber curve just as the extended ball joints do, only to a lesser degree. This is done at the expense of roll center height.

Overall, I think it's a tough argument to say that the suspension geometry cons outweigh the pros with camber plates (though I don't think you were saying that either), even ignoring the camber gains and ability to fit wider tires.

Also, camber/caster plates should ALWAYS be adjusted to exactly the same amount on both sides, since even a small imbalance in SAI will definitely result in a persistant pull. If you need to 'equalize' the camber left to right, do it at the camber bolts on the hub, not the camber plate.

I agree with this if your ride height differs left-to-right due to corner balancing. However, if ride height (thus strut length) is equal left-and-right and camber still differs with your camber bolts in the same position, wouldn't that imply that the frame is slightly off-center and your SAI is already not identical left-and-right? In the case of an off-center frame, adjusting with plates would plates instead of bolts would ensure equal SAI left-and-right.