bump steer adjustment?

[55EXX]

I have done the upper control arm swap and whiteline caster kit install. many sites say the upper control arm sway introduces more bump steer. Now i have looked into bump steer and what it is exactly and some good info was found on good ol wikipedia.

Bump Steer is the term for the tendency of a wheel to steer as it moves upwards into jounce. On modern cars the front wheels toe out as the suspension is raised. This gives roll understeer. The rear suspension is usually set up to minimise bump steer, where possible.

A typical value is two degrees per metre, or perhaps more, for the front wheels.

Excessive bump steer increases tire wear and makes the vehicle more difficult to handle on rough roads.

The linearity of the bump steer curve is important. If it is not straight then the length of the tie rod needs to be adjusted.

Bump steer can be made more toe out in jounce by lifting the rack or dropping the outer tie rod, if the rack is in front of the axle. The reverse applies if the rack is behind the axle. Usually only small adjustments (say 3mm) are required.

so i am aware that this increased rear tilt on the knuckle lowers the tie rod end and is the cause to the increased bump steer. is there anyway to lift it back up and reduce bump steer? washers could maybe be used to lift the tie rod end up but then would make it impossible to get the locking pin in with the castle bolt.

i have yet to experience bad times of bump steer but would like to reduce it if possible.

thanks

[JohnL]

OK, the suspension designer will typically try very hard to introduce substantial amounts of understeer by all sorts of means. With the stock caster angles you most probably have some intentional bump steer that causes the wheels to gain toe-out in bump / jounce / compression, as a means to induce a tendency toward some degree of roll understeer. The designer doesn't really want the bump steer per se (and will try to minimise this), but does want the roll understeer tendency that comes with the bump steer.

This will be achieved by mounting the steering rack 'too low', and / or positioning the tie rod end 'too high', and / or by making the tie rods 'too short'.

The bump steer is caused by the disparity between the arc in which the tie rod end moves, and the arc in which the upright / knuckle moves (or more accurately, the arc of a point having a fixed relationship to the knuckle that occupies the same location as the centre of the tie rod end ball).

If the caster angle is increased then the end of the steering arm / tie rod end is lowered and the tie rod will become angled more downward toward it's outer end. Now, as the suspension moves upward in bump the arc of tie rod end motion will be more lateral (i.e. the lateral component of tie rod end arc motion will be increased to some degree relative to with the lesser caster angle), and this will create some tendency to toe-in with bump motion (i.e. the opposite effect to what the suspension designer intended, which may or may not be a good thing...).

Assumimng that the OE bump steer is to toe-out (in bump motion, which also may or may not mean to toe-in with droop motion); it's likely that increasing caster as you've done will decrease the factory toe change and thus reduce any existing intentional bump steer. It's also possible that lowering the tie rod end far enough might move things past zero bump steer into actual bump toe-in and thus toward roll oversteer (the opposite of the designer's intention).

I hope that makes sense, it's not easy to describe without diagrams...

You can make a simple bump steer guage with two pieces of plywood, a couple of hinges and some pointers (screws will do). Cut some plywood so that it's a bit wider than the wheel rim diameter, and cut again so one piece is a bit taller than the hub centre (e.g. if the rim is say 15" diameter and 300mm radius, then make one piece of plywood say 16" X 320mm, and the other piece 16" X about 300mm or whatever). Now hinge the two pieces together along the 16" sides so you have two 'pages' that can open and shut.

Shut the 'pages' and place the 'book' flat on the ground near a front wheel with the hinges near the tyre. Now open the 'book' so that one page leans gainst the rim. You now need to make some pointers that can rest against the leading and trailing edges of the rim at more or less hub centre height, screwing screws right through the ply works well (screws will be near the corners of the vertical ply sheet / 'page').

Position the bump steer guage (pages open at roughly 90°) with the front and rear pointers touching the rim, now place some substantial weight on the other 'page' to prevent it moving. You can now jack the chassis up and down while noting the change in toe as measured by the pointers touching / losing contact with the rim (one pointer will always touch the rim, but it may need to be 'helped' a bit, especially if the hinges are not frictionless).

The vertical 'page' of the guage will move in / out pivotting on the hinges while maintaining the original toe measurement, so any toe change change with suspension motion (i.e. bump steer) can be seen as a gap appearing between one of the pointers and the rim. Removing the springs will be required to make this measurement possible in bump motion.

[nd55]

I'm still digesting it all John, but ...

> You can make a simple bump steer guage with two pieces of plywood

Ozhonda's resident McGuiver.

Nick.

[nd55]

> i have yet to experience bad times of bump steer but would like to reduce it if possible.

I doubt you'll need these but 'J's Racing' & 'M Factory' make some 'roll centre adjusters' to suit the Eg/Ek knuckle.

How much of a drop are you considering?

Given the popularity of "slamming" Honda's, I think it would be a common issue if there was a problem.

> washers could maybe be used to lift the tie rod end up but then
> would make it impossible to get the locking pin in with the castle bolt.

By moving the tie rod further up from where it's supposed to be with washers, the forces (torsional moment) increase dramatically.

Tread carefully grasshopper.


I believe there are inverted tie rods for the newer strut front suspensions, but this doesn't help you.

Nick.

[JohnL]

I'm still digesting it all John, but ...

> You can make a simple bump steer guage with two pieces of plywood

Ozhonda's resident McGuiver.

Nick.

If you can find a copy of 'How To make Your Car Handle' by Fred Puhn, you'll find a description and pics of the plywood bump steer guage (I think it's that book, I 'lent' my copy to someone so I can't double check).

[JohnL]

I doubt you'll need these but 'J's Racing' & 'M Factory' make some 'roll centre adjusters' to suit the Eg/Ek knuckle.

'Roll centre adjusters' aren't intended to fix bump steer problems, but rather to restore (in some degree) the OE suspension geometry that is messed up by substantial lowering (i.e. by changing of the wishbone angles). They may well have some affect on bump steer, but it would be luck if it were the affect you wanted.


Given the popularity of "slamming" Honda's, I think it would be a common issue if there was a problem.

"Slamming" doesn't alter the caster angle, which is the mod responsible for the bump steer. This is because as the upright / knuckle is leant backward the end of the steering arm lowers, thus changing the 'horizontal' angle of the tie rod (which is what can create additional bump steer).


> washers could maybe be used to lift the tie rod end up but then
> would make it impossible to get the locking pin in with the castle bolt.

By moving the tie rod further up from where it's supposed to be with washers, the forces (torsional moment) increase dramatically.

Tread carefully grasshopper.

Don't walk there at all. The tie rod end is attached to the steering arm with a tapered fitting. It's not possible to shim it higher with spacers, the male and female tapers have to be fully seated / inserted or else you will have a dangerously weak connection.

Taper fittings are almost universal in ball joint and tie rod end attachment, but there are a few that use a straight shanked fittings (and this works perfectly well). For the life of me I can't see why taper fittings are so common with ball joints etc, they are a pain to seperate if they have been overtightened, even by only a small amount (and most mechanics overtighten them).

[55EXX]

i have no bump steer problems at all after gaining 5 degrees caster with both the whiteline shim kit and upper wish bone swap in my eg.

i've taken my car for a few hot laps of mt cootha both clockwise and anti clockwise and it was fine. this is a smoother road tho so i took it up to the waterfall at Springbrook down the coast which is not smooth at all, and in a few sections painful (missus very sore ribs can vouge for that) and it had no problems with keeping the line i choose, understeering, wandering, or front end sudden loss of traction.

the car just felt more direct with road with more push back to centre in the wheel for an over all more go kart like ride, which i enjoyed.

I'd say JohnL that your on the ball with the reversed some of the engineered bump steer. i cannot feel any bump steer, tho i am not fully aware of how to diagnose it as it has never been an issue that has popped up in the past.

my car is only low not super low or even dumped.

[JohnL]

I have 6° caster on my CB7 Accord (2° - 3° stock), and like it very much for the reasons you stated. I've also deleted the PS belt, and like that very much as well, though the steering can load up quite a lot in some corners and parking is a pain.

With bump steer, most cars will have at least some, and some changes to geometry will make it either better or worse, or make a change not great enough to feel.

In my last post I said something that could be read as meaning that a substantial lowering won't cause bump steer, which wasn't my meaning. Substantially lowering the chassis can often (but not always) result in significantly increased bump steer because it changes the orientation of the arc in which the tie rod end moves. This is another reason to avoid substantial lowering.

If you are also making other changes that could affect bump steer (e.g. increasing caster), then you could possibly end up with positive and negative affects partially cancelling each other out, but equally could have cumulative affects resulting in substantial bump steer.

[55EXX]

post from Soichiro's way@MF on honda tech

"rollcenter adjusters will not have much of an impact on bumpsteer. This is because the knuckle and upper control arms will stay in their relative positions- providing that the ride height is set to the same level as before installing the "rca" kit. "

i've read nothing bad about these only great comments. definately something i am going to look into down the track when i get height adjustable coilovers. my bump steer is not a problem but i am very keen on these roll centre adjusters. my car isn't really low but i understand the RC and CG concept in relation to body roll and agree with what they do. i don't know the geometry tho which would be interesting to see.

[JohnL]

...but i understand the RC and CG concept in relation to body roll...

Very few people really do understand this stuff. Roll theory (or rather, weight transfer theory, because roll is only a symptom of weight transfer, not a cause) is very complex and hard to get your head around. What you read in the popular suspension books etc is only the tip of the iceburg, is sometimes wrong, and even when correct tends to be so simplistic as to be misleading.

I've been investigating roll theory for years, and I only understand it to a limited degree (the more I know the more confused I get...). It's the source of much debate and disagreement between proffessional suspension engineers, but is really at the core of chassis dynamics...

[55EXX]

i only have a basic understanding from what i have read in mags and on the net. but the thing is does moving the rc higher to meet with the cg increase roll stiffness? or change it some other way? increasing roll stiffness can easily be done with sways or spring rates what is the advantage of the roll center adjusters then beside smoother ride. sharper turn in higher corner speed etc can be had with bigger front sways too.

[markoJEK1]

post from Soichiro's way@MF on honda tech

"rollcenter adjusters will not have much of an impact on bumpsteer. This is because the knuckle and upper control arms will stay in their relative positions- providing that the ride height is set to the same level as before installing the "rca" kit. "

i've read nothing bad about these only great comments. definately something i am going to look into down the track when i get height adjustable coilovers. my bump steer is not a problem but i am very keen on these roll centre adjusters. my car isn't really low but i understand the RC and CG concept in relation to body roll and agree with what they do. i don't know the geometry tho which would be interesting to see.

Interesting I didnt find that post =/ I found something different, oh well sorry for the misleading, RC are great and Im definatly looking into them myself

[nd55]

> but the thing is does moving the rc higher to meet with the cg increase roll stiffness?

You can eliminate body roll by placing the RC at the CG, whereby roll stiffness goes infinite.

The problem is that RC's move as the suspension goes through the range of motion.

A downside problem with co-incident RC's and CG's is that jacking forces (like anti lift/squat in the fore and aft plane) become significant.

I read in one suspension book, the best approach is to have the RC's about 10cm below the CG points, and a line joining the front and rear RC to be sloping down slightly toward the front.

In another, more directed to open wheelers with very low CG's, there was a comment to ignore where abouts the RC is, just make sure it doesn't move and make doubly sure it doesn't move up and down through the road surface because then the forces on the inside wheel go opposite (ie go from pushing down to pushing up) which makes for a very interesting handling.


All very academic, and getting off the topic of bump steer.

Nick.

PS> Standard disclaimer: I usually don't know what I'm on about.

[55EXX]

i see my understanding as being the rc is the point at which the car rotates around and the cg is the mass centre and determines the length of the moment arm around the rc. the further the cg is away from the rc the longer the moment arm and therefore the greater roll force produce from the same cornering g forces.

so by creating a environment that the rc cmes closer to the cg as beneficial for increasing roll stiffness geometrically by shortening that moment arm other than mechanically by the use of sways and springs rates etc. this would then mean with softer springs and sways a softer ride can be enjoyed and i guess the wheels would follow the undulations in the road better but what other advantage is there? like i said before increased front roll resistance can be increased by bigger sways, and which with soft springs can increase front roll rsistance and follow undulations in the road.

Would like to hear of anyones experiences with them.

maybe this needs a new thread.

[nd55]

Post Deleted.

[chargeR]

> but the thing is does moving the rc higher to meet with the cg increase roll stiffness?

You can eliminate body roll by placing the RC at the CG, whereby roll stiffness goes infinite.

The problem is that RC's move as the suspension goes through the range of motion.

A downside problem with co-incident RC's and CG's is that jacking forces (like anti lift/squat in the fore and aft plane) become significant.

I read in one suspension book, the best approach is to have the RC's about 10cm below the CG points, and a line joining the front and rear RC to be sloping down slightly toward the front.

In another, more directed to open wheelers with very low CG's, there was a comment to ignore where abouts the RC is, just make sure it doesn't move and make doubly sure it doesn't move up and down through the road surface because then the forces on the inside wheel go opposite (ie go from pushing down to pushing up) which makes for a very interesting handling.


All very academic, and getting off the topic of bump steer.

Nick.

PS> Standard disclaimer: I usually don't know what I'm on about.

What I have read agrees with this too. A roll centre coincident with the CG is generally not desirable as it gives significant jacking and unpredictable behaviour. It is better to have the roll centre below the CG by a certain amount and to ensure that the roll centre doesn't migrate below the ground.

[chargeR]

post from Soichiro's way@MF on honda tech

"rollcenter adjusters will not have much of an impact on bumpsteer. This is because the knuckle and upper control arms will stay in their relative positions- providing that the ride height is set to the same level as before installing the "rca" kit. "

i've read nothing bad about these only great comments. definately something i am going to look into down the track when i get height adjustable coilovers. my bump steer is not a problem but i am very keen on these roll centre adjusters. my car isn't really low but i understand the RC and CG concept in relation to body roll and agree with what they do. i don't know the geometry tho which would be interesting to see.

This stuff you have quoted here seems to be a bit dubious to me. It states that the knuckle and the upper control arm are going to stay in their same positions if the ride height is the same which is true, however if one has spaced the lower control arm down to change the location of the roll centre then the arc through which the knuckle and hence the steering arm travels will have changed also hence affecting bump steer behaviour.

[55EXX]

which is what i thought as the knuckle would be sitting higher along with the upper control arm. it all seems bit weird. i would have thought it would have the same effect on bump steer as lowering your car even further.

i've just got back from a awesome holiday up in the sunshine coast hinterland and had the absolute pleasure of driving some of the most beautiful scenic roads in queensland. many tho were bumpy and really really windy and with all sorts of mixed up cambers. anyone who has drivin the small road to montville would know. and through all those roads i had no trouble with bump steer. i even had a trip out to rainbow beach and got to tear down a awesome road at unmentionable speeds though some smoother corners with out any trouble.

on the way home and after a good kidney bash my missus wanted a softer ride so i backed my dampers to thier softest setting and took a endlink off my 22mm rear sway to smooth the whole thing out. the ride was nice a soft and comfy but i did feel my car wander more and i found myself having to adjust the wheel more to just drive straight? Does this sound right or do you think that by softening it all up and making the wheel actually use some of its travel it may have awaken the bump steer that i previously never felt? either way once its all hard again its fine so i am still just going to leave it.

[Tegzieboi_BAR]

i am a dc4 owner and after puttin 17s on my car noticed a huge increase in bump steer... after a while ive gone down 2 lightweight 15's n after some minor suspension mods have noticed more n more bump steer.

Currently im sitting on tein coilovers with front n rear strut braces n stock sways. Th stiffer the suspension the more bump steer.

Goin over bumps dramatically jerks my steering wheel, no matter the speed, after reading through this forum quite thoroughly i hav an understanding of the causes but cant get a grasp of what exactly is required 2 reduce this n if there are aftermarket kits available 2 account 4 this.

Sorry 2 interrupt!

[chargeR]

i am a dc4 owner and after puttin 17s on my car noticed a huge increase in bump steer... after a while ive gone down 2 lightweight 15's n after some minor suspension mods have noticed more n more bump steer.


Changing tyres and wheels doesn't affect bump steer only changes to suspension geometry like ride height do.



Currently im sitting on tein coilovers with front n rear strut braces n stock sways. Th stiffer the suspension the more bump steer.


The stiffer the suspension the less bump steer as the suspension travels through a smaller range of motion and hence less toe change, if all else is equal. The phenomena you are describing are probably not due to bump steer ;).

[Tegzieboi_BAR]

hmmm i dno man... whatever it is its fn annoying sometimes, cant let go of the steering wheel sometimes coz if i go over a bump my car will instantly change direction sometimes.

[55EXX]

what you are experiencing is from the non compliance of your low profile tyres. the ability for your tyre to deform over bumps in the road is gone and is what is making your ride jumpy like that. bet its terrible over heavily worn raod that have got the 2 dips where trucks have squished the road in and your car just has to tramline along these and getting out makes the wheel jump around. this isn't really bump steer. higher profile tyres will help out. i had 18's on my last civic and i hated them for that... and the loss in power and handling. 15's for the win

[string]

hmmm i dno man... whatever it is its fn annoying sometimes, cant let go of the steering wheel sometimes coz if i go over a bump my car will instantly change direction sometimes.

A car with zero bump-steer will still track the road. You can't let go of the wheel and expect it to track straight over a non-flat surface.

[JohnL]

Tegzieboi_BAR ,
When you increased your wheel size you must have decreased the aspect ratio (which will increase the casing stiffness, all else being equal), but did you also increase your tread width and / or tyre pressure?

I suspect what you're getting isn't bump-steer (though I can't definitively say it isn't), but more likely to be 'kickback' which is related to tramlining (a more severe manifestation of tramlining). This is likely to be worse the wider the tread, the stiffer the tyre case, the higher the psi. It's also likely to be excacerbated by stiffer springs, dampers etc.

[55EXX]

after driving with softer sussy i would now definately say there is more bump steer. when i go over a bump with one wheel the wheel jumps away from that side but car remain straight. it is more wanderery. i feel this mod would be for someone with already very firm sussy as i never felt these issues before, tho its not too terribly strong with soft sussy, and i probably wouldn't notice too much if i weren't looking out for it.

new oem bushes put in today final alignment setting on monday. rear camber kit arrived will be setting the rear to 1.2-1.3 degrees -ve and 0 toe all round with ~4.6 degrees caster both fronts and ~ 2.1 degrees camber front and complete my assesment in the diy section after a cruise up nebo tuesday

[JohnL]

Almost by definition bump steer will be worse the greater the suspension travel is, so if you have an existing bone fide bump steer then softer spring / damper rates are likely to make it worse.

But, you can't make any judgements if any of the suspension articulations are suspect...

[55EXX]

got my my car back and the bushes were ****ed as. the centres weren't really attached to the outer as some of the rubber inside has disintegrated. this leaves me to think to not change any washers in the front castor this time round as both were pretty bad and the setting would not be true as the control arm could move around before giving incorrect readings. what are the downsides to uneven camber? i'm guess uneven trail leading to pull to one side? pure guess

with the new bushes the car feel more responsive and accurate. man didn't know the difference they made. bump steer is less evident but still slightly there maybe what i felt before was the control arm moving with the bumps. in the wet tho it can barely if not all all be felt. rear camber kit in now to what my estimate is ~1.2 degrees. put my more fresh tyres onto the front so i'll see how she goes.