solo-x

i finally finished measuring the difference in bump steer for an EG with both swapped control arms and stock configuration control arms. those with a DC chassis can probably assume a very similar curve. measurements were taken a minimum of 4 times using longacre toe plates and the toe plates have been verified to be true. i didn't plot the entire curve, just absolute change from static to 1" and 2" bump and 1" droop. the resolution isn't very high either since i'm using toe plates and tape measures, but when you see the size of the delta you'll see that it doesn't matter. positive toe measurement means toe-in, negative toe measurement means toe-out and toe measurements are total toe. ride height measurements were taken from center of hub to lip of fender. static ride height is 12", maximum bump travel is 2". static toe set to zero for both configurations.

with upper control arms swapped left to right:

1" bump showed a toe change of ~.13"
2" bump showed a toe change of ~.25"
1" droop showwed a toe change of ~-.20"

with upper control arms in stock configuration:

1" bump showed negligible toe change
2" bump showed negligible toe change
1" droop showed a toe change of ~-.1"

some other interesting things i noticed were that track gets nearly .25" narrower from static to full bump and swapping the control arms affects static toe by roughly .75". that last part is important. if you remember that you can do back to back testing of both configurations and be able to quickly re-zero toe by remembering to just turn the rod ends 6 flats on each side. from stock to swapped the toe change is positive, from swapped to stock the toe change is negative.

the final thing i noticed is that having the arms in stock configuration gives you another 1/4" of bump travel before the arm hits the tower. needless to say, you won't find swapped control arms on my car.

note: depending on your ride height you might find you're in a different portion of the curve. you don't need to worry about anything lower then were i'm at, but higher ride heights definitely warrant a look at the bump steer curve.

nate - perfectly happy with 0 caster if it also means zero bumpsteer

ryan12321

Thanks for the info nate. I was considering doing this on my daily driver 00 civic but guess not.

Lyonel 13H4

Yep, thanks, Nate...I was always curious about what would happen, but never actually went to the trouble to find out for myself.

Greyout

I don't understand the significance of bumpsteer in the front of a race car.

TeamSlowdotOrg

Buh?

Tyson

so does this also show that when you lower a car, theres negligible toe change vs. ride height with stock a arms?

94eg!

iTrader: (2)

One thing that always bugged me about camber kits & shorter rear LCA's was that it would "F" up your bump steer geometry. It seems these cars can be pretty sensitive to small changes like that.

honda93

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by solo-x &raquo;</TD></TR><TR><TD CLASS="quote">i finally finished measuring the difference in bump steer for an EG with both swapped control arms and stock configuration control arms. those with a DC chassis can probably assume a very similar curve. measurements were taken a minimum of 4 times using longacre toe plates and the toe plates have been verified to be true. i didn't plot the entire curve, just absolute change from static to 1" and 2" bump and 1" droop. the resolution isn't very high either since i'm using toe plates and tape measures, but when you see the size of the delta you'll see that it doesn't matter. positive toe measurement means toe-in, negative toe measurement means toe-out and toe measurements are total toe. ride height measurements were taken from center of hub to lip of fender. static ride height is 12", maximum bump travel is 2". static toe set to zero for both configurations.

with upper control arms swapped left to right:

1" bump showed a toe change of ~.13"
2" bump showed a toe change of ~.25"
1" droop showwed a toe change of ~-.20"

with upper control arms in stock configuration:

1" bump showed negligible toe change
2" bump showed negligible toe change
1" droop showed a toe change of ~-.1" </TD></TR></TABLE>


Nate, so under compression in left handers, the LF gains toe in as well as the RF? You are listing total toe changes, so I am assuming one is to split the number in half.

solo-x

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Tyson &raquo;</TD></TR><TR><TD CLASS="quote">so does this also show that when you lower a car, theres negligible toe change vs. ride height with stock a arms?</TD></TR></TABLE>

at my ride height it would appear that way. the toe change was small enough that i was unable to detect it with the method of measurement i was using.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by honda93 &raquo;</TD></TR><TR><TD CLASS="quote">Nate, so under compression in left handers, the LF gains toe in as well as the RF? You are listing total toe changes, so I am assuming one is to split the number in half.
</TD></TR></TABLE>

well, it's really not that simple. take for example if you were trail braking into a corner, both the inside front and outside front would be in bump. if you were accelerating off the corner one or both may be in droop. roll center location and a myriad of other factors contribute. suffice it to say, with swapped arms you have toe-in in bump, toe-out in droop. most articles i've read on bumpsteer indicate that if you are to have any kind of bumpsteer it better be in the toe-out in bump direction. toe-out in bump is more stable then toe-in, at least on the front of the car.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Greyout &raquo;</TD></TR><TR><TD CLASS="quote">I don't understand the significance of bumpsteer in the front of a race car.</TD></TR></TABLE>

there seems to be two camps on this. i'm in both camps. if i were driving a formula ford i'd be more concerned with rear toe. on a fwd car, i'm more concerned with front toe. maybe i'm right, maybe i'm wrong. i do know that i'm not willing to pitch my bumpsteer out the window for 3 measely degrees of caster.

nate

Tyson

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by solo-x &raquo;</TD></TR><TR><TD CLASS="quote">

at my ride height it would appear that way. the toe change was small enough that i was unable to detect it with the method of measurement i was using.

</TD></TR></TABLE>

oh i get it, it was already lowered, and you measured up to the 2" you had left. looks like it started to be toe sensative as you moved toward stock ride height. thats good to know.

man, id really like to see a full range profile of bumpsteer.

honda93

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by solo-x &raquo;</TD></TR><TR><TD CLASS="quote"> well, it's really not that simple. take for example if you were trail braking into a corner, both the inside front and outside front would be in bump. if you were accelerating off the corner one or both may be in droop. roll center location and a myriad of other factors contribute. suffice it to say, with swapped arms you have toe-in in bump, toe-out in droop. most articles i've read on bumpsteer indicate that if you are to have any kind of bumpsteer it better be in the toe-out in bump direction. toe-out in bump is more stable then toe-in, at least on the front of the car.

nate</TD></TR></TABLE>

I could see where this would make the car more apt to swap ends because the front will be more inclined to rotate about a tighter axis. If toe-out exists in the rear (as it does for my setup) it could make the car nervous in certain situations.

solo-x

yeah, i only measured to the limit of what i felt i could expect to see in actually race conditions. i wasn't very concerned with what happened at closer to stock ride heights as i don't plan on running the car any heigher then it is right now. lower maybe, never higher.

nate

stmotorsports

i swapped the arms on my car (5th gen ex coupe), and it actually feels like it corners worse, if anything- real sloppy and tends to push worse in hard cornering. After finding this thread now I can see why. I was planning on swapping the arms back to stock, but now I'm wondering if there's a somewhat reasonable way to keep the swapped arms but correct the bumpsteer? I don't have a svc manual or anything handy at the moment but if I'm picturing the civic suspension correctly in my head, I'm assuming the length of the tie rods are the problem. If you could shorten the outer tie rods accordingly you would effectively shorten their curve to coincide with the new rearward-biased camber curve. The only problem is to do this and be able to keep toe settings in the correct range, you would have to move the tie rod pivot point outward ie- lengthening inner tie rods + shortening outers...

anybody have any thoughts?

DB2-R81

Tie rod length has minimal effect on bump steer. Of greater importance is the difference in steering rack height referenced against the height of tie rod attachment at the steering knuckle.

stmotorsports

hmm I hadn't thought of the height of the rack...
NAte noted that after swapping arms the toe change was positive (inwards), and if I recall correctly I found the same with mine. Which makes sense because the steering arm on the knuckle would be leaned back & down, putting it back into the lower part of the tie rod's curve(?)...
so I'm thinking however much the tierod balljoint is moved from the arm swap, the rack would have to be moved proportionately in the same direction? It probably wouldn't have to be much, around 1/2" or so I'm guessing ??

honda93

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by stmotorsports &raquo;</TD></TR><TR><TD CLASS="quote">NAte noted that after swapping arms the toe change was positive (inwards), and if I recall correctly I found the same with mine. Which makes sense because the steering arm on the knuckle would be leaned back & down, putting it back into the lower part of the tie rod's curve(?)...
so I'm thinking however much the tierod balljoint is moved from the arm swap, the rack would have to be moved proportionately in the same direction? It probably wouldn't have to be much, around 1/2" or so I'm guessing ??</TD></TR></TABLE>

OK, so the gain in negative camber on the outside wheel in a turn is actually disadvantageous? AND if positive toe was gained by doing the swap, realign it back to 0" or toe out... You'd have to anyway because the entire front suspension is practically disassembled to do this swap.

Or am I looking at this to simplistically again?

stmotorsports

no- you DO want camber gain when cornering, thats one reason why you want positive caster which is the whole point of doing this mod. What you don't want is toe gain which is what nate figured out results when you do the arm swap.

Weston

honda93: I think what he is saying is that this creates dynamic toe-in as the suspension is compressed. It's not something that you can zero out by setting the alignment.

If I'm interpreting this correctly, in a right turn the left suspension is more compressed and would have toe-in, while the right suspension is less compressed and has toe-out. In other words, the left wheel is pointing more right than normal and so is the right wheel, but we are turning right anyway, so I don't think that's a problem.

It sounds like the real problem is with front&lt;-&gt;rear weight transfer and bumps, as that would cause the front wheels to have toe-in when it has weight on it or hits a bump, and toe-out when it doesn't. Does anyone have any thoughts on how much of an issue this really is?

Thanks, solo-x, for taking the time to do this and posting up the numbers. I had suspected that this extra caster came at a price.

stmotorsports

I found this while doing some research:

"if toe links/tie rods are not perfectly parallel to the control arms at all extremes of suspension travel, there will be a change in toe angle, thus bump steer."

so the reason why you don't run into this when lowering is because the suspension components are all still within their correct travel arcs- all the arms/rods remain parallel to each other. When you move the upper ball joint rearward, the control arms are still parallel to each other, but the tie rod gets swung down and to the rear thus putting the tie rod out of alignment. So DB2-R81 was right in that relocating the rack would probably fix this.

Now the next question is, who's willing to hack their car to pieces to test out a shaky theory posted on the internet?

honda93

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Weston &raquo;</TD></TR><TR><TD CLASS="quote">honda93: I think what he is saying is that this creates dynamic toe-in as the suspension is compressed. It's not something that you can zero out by setting the alignment.

If I'm interpreting this correctly, in a right turn the left suspension is more compressed and would have toe-in, while the right suspension is less compressed and has toe-out. In other words, the left wheel is pointing more right than normal and so is the right wheel, but we are turning right anyway, so I don't think that's a problem.

It sounds like the real problem is with front&lt;-&gt;rear weight transfer and bumps, as that would cause the front wheels to have toe-in when it has weight on it or hits a bump, and toe-out when it doesn't. Does anyone have any thoughts on how much of an issue this really is?

Thanks, solo-x, for taking the time to do this and posting up the numbers. I had suspected that this extra caster came at a price.</TD></TR></TABLE>

I understand this, but I am one of the few people who like to run a lot of static toe out (-3/8" total) up front. I am assuming that any dynamic toe-in would cancel out. Also, I run my car a lot higher than most (1.5* static) due to the fact that I don't have flared fenders and the 13x8.5 setup I have would rub otherwise. I believe this swap is affecting me less than it would to someone like Nate being we have the same chassis but different philosophies in car setup in all other areas.

GSpeedR

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by stmotorsports &raquo;</TD></TR><TR><TD CLASS="quote">Now the next question is, who's willing to hack their car to pieces to test out a shaky theory posted on the internet? </TD></TR></TABLE>

The theory isn't shaky, it's just mechanics. Think in terms of 4-bar linkages and it gets simpler to visualize. You need to change the rack position and the tie-rod mounting point on the hub.

stmotorsports

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by GSpeedR &raquo;</TD></TR><TR><TD CLASS="quote">
The theory isn't shaky, it's just mechanics. Think in terms of 4-bar linkages and it gets simpler to visualize. You need to change the rack position and the tie-rod mounting point on the hub. </TD></TR></TABLE>

sounds to me like we have a volunteer :D

GSpeedR

Who knows what autocross class rack relocation bumps you into...

I have my car apart right, so I'll "visualize" the setup for you and let you know the results

chad

i really don't understand this droop....

solo-x

i can't move the rack or change the knuckle so i'm not even going to spend time thinking about it. if you REALLY want the caster, try spacing the lca further foward. a smaller movement on that end will get a larger increase in caster (relative to uca relocation) and would likely have a smaller bump steer change. just a guess here, but it might be worth checking out.

dynamic camber change from caster.... um, don't worry about it. from what i've seen of the average honda chassis setup most people are about 2-3 degrees away from ideal static negative camber for cornering. the dynamic camber gain from increased caster at the steering angles i expect to see in the tightest corner is only about .1 degree. even assuming the camber gain was significant it would be so closely tied to steering angle that the car wouldn't have enough camber in high speed corners and enough or too much in slow corners. i was using caster for a stability and steering feedback improvement only. if i need more camber i'll either increase my roll stiffness or lower the car more to increase my static camber setting, not increase caster.

tony, bump steer is not something you can "dial out" with a static toe setting. i can explain it much better in person, but suffice it to say that zero bump steer is typically what chassis builders aim for. with your ride height being that much higher then mine you might be in a different portion of the curve. i don't think you are, but the only way to know for sure is to take some measurements. imo, its a waste of time and effort to even swap the arms. the caster increase is minimal, and the change in steering feel is placebic at best. i'd invest the time else where personally, but that's only because i've already wasted it.

happy new year!

nate
*shakes fist at chad*