Front or rear, and what are the symptoms, and how did you diagnose that it has bump steer?

I would not expect a 20mm difference in ride height to make much difference to how its bump steer behaves. (I'm not saying it has "none". I'm saying if it had bump steer at stock ride height - it may have some designed in, and it's near impossible to have precisely none - then it'll have comparable bump steer 20mm lowered.) MacPherson strut front, from what I can tell online the lateral arm is pretty similar length to the steering tie-rod and pretty close to the same angle in both top view and side view. Its bump steer should be pretty insensitive to ride height.

If someone put something together wrongly then of course, all bets are off.

Let's hear what your symptoms are.

Hello Brian


The way I experience bumpsteer is when I go over the asphalt joints on highway bridges. When I pass over those types of joints at speeds over 140 km/h, the steering moves on its own, it turns by itself. I believe this is bumpsteer in the front suspension.

When my car arrived from the factory with a 10mm lowering, I didn't experience any of this steering movement when passing over the same spot on the highway. However, after lowering it an additional 10mm following directions and using Porsche parts, my car now undergoes a slight steering rotation when passing through the same point.

The Porsche service manager who performed the retrofit says everything is in order but that bumpsteer is normal.

Thanks !

M.

"Slight"?

Can this be quantified ...

Does it cause a real problem, or is it just a something that was noticed?

This mid-engine car has a slight rear weight bias. I would expect that part of the understeer budget (and some understeer is necessary for stability) involves designed-in roll understeer, and "roll understeer" is another way of saying "bump steer" because the things in the steering geometry that cause roll understeer are the same things that cause bump steer.

If you have someone at an alignment rack who knows what they're doing, they can quantify the bump steer. Measure toe at nominal ride height (with the car's weight on the wheels as normal), very carefully measure the height of a reference point above the ground at the front wheels (piece of tape on the fender with a marking on it works), jack the car up precisely (let's say) 25mm or 50mm if you have it, measure toe again.

Hang on, are you hitting the joints with both wheels?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

"Interestingly, my friend has the same specifications but their car came directly from the factory, and they don't experience any bumpsteer.

I'm at a loss about what to do next"

I would ask the friend if you can make measurements of every adjustable link and angles for toe and caster, et al, of his car. This difference is going to be several degrees or something is off several millimeters.

Also, see if he'll let you drive his car over the same joints.

That identical cars behave much differently is unlikely.

Is this on joints that are hit perpendicularly or are they diagonal joints.

Your diagnosis is all wrong. Don't call it bump steer, professionally it's ride-steer because ride steer and roll steer are two completely different phenomenon because it's symmetric tie-rod loads vs. asymmetric tie-rod loads. Ride-steer and roll-steer are closely related, involving the signs of loads on the steering gear.

Going over tar strips at any speed would cause symmetric tie-rod loads, hence NO change in steer gear action and no change in wheel position.

You have an asymmetry in net tierod load that could be due to many things:
1) You have two different shocks, maybe 1 is snafu. so wheel dynamics is causing the asymmetry, even if you had zero ride-steer.
2) You have different caster or caster offset in the front spindles. Slipped part, bad bushing, bent part. wrong strut mount, different LCA, wrong 'upright', etc.
3) You have different tie-rod lengths due to carelessness in positioning the steering wheel & or rack during reassembly. Count the threads on each side. AND then turn the steering wheel to full lock in each direction The thread count should be identical. The steering wheel position should be exactly the opposite number of degrees off of zero (straight ahead) reference.
4) The car is off kilter from wrong springs or content loads. Ride steer would be determined by the front view angle of the tie-rods when the vehicle is level. If yours is not level, there's a reaction from it. SKip the donuts and fried chicken for a week or two.
5). My bet is that when they changed the 'package', they might have had to use a different steering steering intermediate shaft that's different because of who knows what. If the column or gear is a bit out of position, then the I-Shaft phasing is off, which often can cause the effect you describe.
6) you have differences between left & right tire properties. Brand & size & pressure is correct, tire Mx is different. If tires are directional for rotation, 1 is mounted wrong. One could be a real Porsche factory tire, one could be a dealer supplied imitation. I've seen this frequently and dealers are either unaware or will to take a chance. I forget the special marker on a Porsche factory tire. On a BMW its a star symbol.
7) You got a different steering gear and the valve and/or pinion is staked in the wrong position. That's a ZF gear, correct ?
8) you have a pinched steering hydraulic line, or if you were unlucky and have electric power assisted steering, the cal is messed up.

It may be considered normal and to spec. Porsche's have quite a bit of front & ride steer because the architecture doesn't provide enough total vehicle understeer to obtain decent response times/bandwidth as well as directional stability at Ay levels below let's say 0.5 or 0.6 g's. The flaw in this design philosophy is that it's tires don't give a crap about any such small changes in front & rear tire slip angle due to any 'roll steer' changes above that Ay level.. They only listen to Fz loads on them and a bit of camber (depending on the brand, construction, pressure, rim width, etc. stuck on the car)

Measurements of Cayman GTS cars in my K&C database indicate front roll-steer is about 14 percent while rear is about 2 percent. That's 0.14 degrees of FRONT understeer per degree of roll (Which is A LOT). The outer tie-rod ball is above the rack ball height by a noticeable angle. Lowering the body/chassis makes this worse. I would think that tire wear is also apparent by now in your car. That's a lot of Mz scrubbing.

So come back with some preliminary measurements of the suspension & chassis parts & locations. And clarify whether the tar-strip is going under both front tires or just 1.

Hey everyone,

Thank you so much for taking the time to respond in such detail. I really appreciate it.

I'm going to add more information that might help with the diagnosis.

Regarding the question "Hang on, are you hitting the joints with both wheels?", the problem is much more noticeable when I hit the joint at an angle. When I hit it with both wheels perpendicular, the steering movement is almost imperceptible. However, it becomes much more pronounced when I go over a joint that is slightly offset, meaning one wheel hits it a fraction of a second before the other.

@cibachrome, thank you for your comments. I will ask them to check the length of each tie rod and follow your suggestions. I will also share your comments with the Porsche workshop manager.

I'm thinking of taking my car and my friend's car to a sophisticated Hunter alignment shop, as you suggested. What specific things should I ask the operator to check? What measurements should I request them to take? What should they pay close attention to?

Thank you so much.

M.

Do what cibachrome says. Pay particular attention to effective tie-rod lengths (his point 3).

If you have access to a good alignment machine, you can quantify the bump steer (which implies the roll understeer). Toe-out with suspension compression = roll understeer. Find a way to apply upward force on the front of the vehicle from underneath - you're not going to lift the whole front off the ground, just unload the suspension. I don't know what it looks like underneath those but if there's a front subframe (highly likely) then a piece of lumber underneath a crossmember with a mechanical jack dead-center underneath that ought to be able to do the job. Use masking tape to make a reference mark on the front fenders directly above the front wheel arches - someplace you can accurately measure straight down from. Measure ride height at your reference points with the car sitting normally and the steering centered. Measure toe with the alignment machine. Jack it up 10mm evenly both sides ("PASM" height i.e. -10mm from standard, relative to the -20 where you are now), being absolutely sure to not touch the steering wheel or let the steering wheel move at all, measure toe again. Jack it up 20mm evenly both sides ("stock ride height"), measure again. Take your jack out from underneath, feel free to double-check your "as-it-sits" toe measurement. Put something in the "frunk" heavy enough to lower the front 10mm below your current nominal ride height, measure toe again.

You now have enough information to produce a bump-steer chart, and if you were really careful to not let the steering wheel move during any of this, you have enough information to produce a separate chart for both sides.

Taking cibachrome's numbers, expect to find significant bump steer. It's designed in from the factory. It's there to help you keep the car between the hedges at high speed, and it's there because unlike most cars, yours is a smidge heavy in the tail. What would be interesting, is how the toe change from +0mm to -10mm ride height compares to that from -10mm to -20mm.

I still think you're going to find something wrongly assembled, or maybe a bad tire.

Check tire pressures as well.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

A modern Hunter aligner has the capability of a pull-down & push up test with digital data output. You don't want just 2 readings at a couple of different ride heights. The goal is to see if the wheel functions have curvature & hysteresis as well as slope. If your car has the active ride control feature, the Shop ought to have an analyzer on-hand which tricks the suspension into thinking its at a non zero speed via the transmission buss. Ramp it up. It's interesting to watch, especially if the amount is close to what I would expect. If you set some inclinometers on the tierods, keep notes on what the angles are.

Have them show you a ride caster plot for both wheels, too. Since caster amounts to the change in camber per steer, the convolution from wheel vertical motion adds another player to this investigation.

BTW: a 20 mm ride height change ought to make a big ride steer change but no net tierod load change, unless the inputs you described are actually not parallel.

I hope this message finds you all well.

It's been some time since our last correspondence, but just recently, we had the opportunity to use a Hunter Alignment Machine with a bumpsteer reader kit, and I believe the results warrant a discussion.

To give you a brief background, my car was initially ordered with a 20mm lowered suspension. However, due to a dealership error, it arrived with only a 10mm drop. To rectify this, I requested a retrofit to the original 20mm drop using genuine Porsche parts. This retrofit included the installation of new shocks, springs, and bars.

Post-retrofit, I've noticed a concerning issue. When driving over bumps, the steering wheel tends to move on its own. This is hardly noticeable at low speeds, but above 60MPH, it becomes quite apparent. This problem is particularly frustrating as it significantly detracts from the driving experience.

Yesterday, we took the car to analyze this issue more thoroughly. The car was tested on a Hunter Alignment Machine equipped with a bumpsteer reader kit. I have attached the graphs from this analysis for your review. The Porsche service manager is ready to take necessary actions, but it seems we are in need of more technical feedback to properly address this issue.

I wonder if readings show off-range values ? normal values ?

I would greatly appreciate any feedback or guidance you can provide on this matter. Your expertise is invaluable, and I am eager to resolve this issue as soon as possible.

You may see the graphs here : https://postimg.cc/gallery/dLYvfZ0

Thank you for your attention to this matter, and I look forward to your response.

Thanks again !!

Marcelo

• https://files.engineering.com/getfile.aspx?folder=3f1f59d8-d5f5-4b7b-8d08-2

Front right in the first photo seems to be showing about 27 deg/m of bump steer. That is an awful lot, roughly twice as much as I've ever used.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Well if nothing else the car is waaaaaayyyyy out of alignment. If the dealer told you they aligned the car after the suspension change, they either didn't actually do it or their alignment tech is not very good.

Start with a comprehensive alignment.

I would expect toe-out with suspension compression in the front as that's the direction of kinematic roll understeer. I gather the graph in the left image is "height versus convergence front" in which case I interpret "convergence" as "toe-in", and it's negative numbers so that's toe-out. Now ... In the units along the left side of the graph, do positive numbers imply suspension above nominal ride height (extension) and negative numbers imply suspension below nominal ride height (compression)? And is 0mm the "new" nominal ride height (lowered 20mm from stock) with thus 20mm on the graph representing where the OEM original no-lowering-kit-installed nominal ride height was?

If that's the case then you had a little static toe-out from standard, you now have a lot of static toe-out at the new (lowered) nominal ride height, and you've got an lot of toe change with suspension movement (bump steer).

I agree with the above post ... sort out the static toe first and see what happens. If it were my car, I'd aim for a wee tiny bit of static toe IN, not toe out, so that passengers, luggage, etc would put it somewhere close to zero toe.

Static toe doesn't affect bump steer to any significant degree.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Bumpsteer appears to be symmetrical,
I would lean towards Cibachrome for some asymmetry in the system
I assume that the car was not crashed and that there are no idiots in the workshop to use different parts for the two opposite sides
the only thing that is not visually apparent could be little difference in the characteristics of the shock absorbers, (I don't know if this is possible) this would cause a different ride height when crossing a bump and with such a large bumpsteer a significant difference in the steering angle of the wheels.

Radek

Thank You for your feedback.

Do you suggest that the technician should review the installation of the parts, considering the unexpected symmetrical toe movement when the suspension compresses?

Seems pretty obvious, they lowered the sprung mass, but failed to change out the knuckles/uprights to keep the ride/roll steer to the original spec (which is kinda big on that car anyways. The front view tierod angles should give it away, so measure them. They might have tried to compensate for this by changing the caster to drop the outer tierod ball height, so check this out, too.
As a last resort, they might have shimmed upwards the steering rack, but this would most likely add a lot of compliance understeer, But who really cares. When was the last time the car saw +.8 gs Ay ? I'll bet never.

Quote (FreshDriver)

Do you suggest that the technician should review the installation of the parts, considering the unexpected symmetrical toe movement when the suspension compresses?

You really need to get this car aligned. It's out of alignment by a large margin, and there are asymmetries in the parts of the alignment we can review, which would lead me to believe there may be other issues.

This behavior could be easily explained by a caster imbalance, which could be a consequence of a bad alignment.

Proper alignment may correct this issue without any further drama.

Hello All

I had my car aligned using the Hunter alignment values specified for the -20mm Porsche Cayman 718, which are included in the Hunter machine's database. I've attached the graph showing new reading, as well as the charts comparing the alignment before and after the procedure. Any comments or insights would be greatly appreciated.

Thanks for your comments

FD

Take video of the car being driven over a moderate symmetric bump. There's a good chance one of your electronic controlled dampers is out of action. (But the ride steer is still horrid)...

Hello Cibachrome

Will take a video this weekend..

When you say "But the ride steer is still horrid)..."
You refer to the reading shown in the graphic after alignment ?

So your camber was out of whack on one side both front and rear (bad), and is now OK, and you had static toe-out before on both front and rear (bad), and now have slight toe-in both front and rear, which should be good. There's still a lot of bump-steer by the numbers, but it's now very close to the same left to right.

Repeat your drive-over-a-bump test, and report back with findings. I have a funny feeling that the car will behave now.

It's on the high side of just about every one of the many, many, many sports cars I have K&C data for. If so, 1 or more of your front tires ought to start showing shoulder wear.

Static toe-out is a cheap Ackermann steering patch, also bad for adding understeer, reduces max lat, and makes the steering gain nonlinear thruout the range of 'sporty' driving. Especially on tires as oversized as come with this make/model.

We took two cars to the hunter machine: my car, equipped with a retrofitted 20mm drop Porsche suspension, and another identical car with the same suspension, but factory-installed. Essentially, we had two identical cars, both with the same suspension setup; one was installed by a Porsche dealer, while the other came with it as a stock setup. The car was slightly out of alignment, but you can still check the toe movement during compression of the suspension on a factory stock suspension.
How do you see it compared to my car ?
Here is the graph from the other car:

Are very similar bumpsteer values for both cars,

after alignment your problem still remains?

then it really seems that "only" damping problem
when different damping from left to right creates different instant ride height this mean different toe angle for opposite wheels
or
also different damping will create different wheel grip (during wheel vertical vibrations) over highfrequency road unequalities

what your wheels size? (unsprung weight)

Radek

Go to damping test ,
of course your problem also can comes from rear axle

Quote (sierra4000)

Are very similar bumpsteer values for both cars,

after alignment your problem still remains?

then it really seems that "only" damping problem
when different damping from left to right creates different instant ride height this mean different toe angle for opposite wheels
or
also different damping will create different wheel grip (during wheel vertical vibrations) over highfrequency road unequalities

what your wheels size? (unsprung weight)

Radek

I have 20" inch wheels
Any way to measure the damping ?
After alignment have not checked carefully, will do it today.

At this point I'd say the bump-steer that you have in the car, is what Porsche designed it to have.

Damping is not easy to measure on the car. Still ... Do your drive-over-an-uneven-bump test (ideally with both cars), and get back to us.

You might want to compare your car's data with values from a similar model. Even their best is not exactly symmetrical, eh ?

https://i.imgur.com/IY4I6x3.jpg

According by wikipedia, Taycan have different weight distribution, can be reason? Interesting asymmetric values,but is really under 2Dg/m difference too much?
I love real cars K&C data !!!

it would also be interesting to know the compliance data of both cars

Some workshops measure damping on the car,i think in this case can be sufficient for dampers comparison.

https://www.youtube.com/watch?v=mGxNRF7W9jA

Radek

Your damping is speed dependent and most owners have the wheel rolling when they drive over bumps. Somebody would have to trick the car into setting all the speed dependent settings. Usually done with a transmission hack.
As for the compliances, I see no reason that they would be any different essentially for the same basic architecture. Manufacturing issues and net build processes are the reasons for most differences.

BTW: you can easily figure out what the ride steer would be with a 20 mm change in static ride height. It's probably quite a lot. BTW2: the car I showed you data for has ROLL STEER values of 8% exactly on both front wheels. So, same symmetric roll steer but asymmetric ride steer tells me exactly how carefully your OEM build was done. Guess the cause !!

Yes, Cayman (original post) is not Taycan (that link) but same manufacturer so maybe similar design philosophy? Internet says Cayman has 55% rear weight distribution, Taycan is very close to even split. That kinda suggests the Cayman needing more understeer from its geometry. The way I'm reading it is that Cayman (original post) has something like double the bump-steer of Taycan, but what's also striking is the substantial difference left-to-right in the Taycan.

A 20mm one-wheel compression bump in the original-post Cayman makes a 0.18 degree toe-out steer angle change on that wheel. If it's a true one-wheel bump then the other wheel stays put and the steering mechanism splits the difference roughly speaking, so 0.09 degrees of steer-angle change. If the steering ratio is 15:1 then the steering wheel would, ideal-scenario, jiggle by 1.4 degrees. That's not much. Reality is probably less than that because presumably the front instant-center is probably above ground level which means that wheel would simultaneously have its contact patch move outward from the vehicle a smidge which kinda-sorta agrees with that wheel toeing out a smidge ... keep in mind also that the rear is probably toeing in a smidge when it goes over that same bump one wheelbase-length later ... still, all these effects are looking pretty small.

We still haven't heard from the original poster whether getting a proper alignment sorted this out.

I tend to think hes getting a wheel kick from the power assisted steering. Either electric or hydraulic reaction. Can't clamp it down because of evasive maneuver requirements. Flunko de Mooso Testo.

I usually find an asymmetric bumpsteer on steering rack inclination, different KPI or with different caster.

Radek

Here are two videos.

First one is using normal suspension setting, second one is using sport suspension(stiffer) setting.

First Video
https://www.youtube.com/watch?v=37jCYJl2WpY

Here, the vehicle is passing over a diagonal highway joint, and you may notice steering shake, indicating that the joint is not parallel to the wheel links. When passing over a joint that is parallel to the wheel links, there is no steering shake. https://youtube.com/clip/Ugkx9fSjZ4eFsg5CYtQAgAu9h...

Here is speeding up to 110MPH https://youtube.com/clip/UgkxZD4Xa75Fqh8bTn2wvvC1u...

Second Video using stiffer suspension setting.
https://www.youtube.com/watch?v=cKVlCk-lgqE

Here is a jumpy road surface https://youtube.com/clip/UgkxB_2ZHgwbrhk-1hV4L9OUq...

Your feedback is very much appreciated.

Looks pretty normal to me.

I noticed something similar with a series of offroad vehicles with oversized wheels (low ET big width,big diameter,big unsprung)

Radek

I see nothing out of the ordinary.

A video showing the TIRES taking on some severe bumps would be better for any diagnostics. The power steering system masks the most important causal events. Follow someone with a passenger videoing both front tire. Flash your lights when you detect a complaint level event. looking for asymmetric responses from symmetric roadway inputs.