I know that ideally, you would have 0 bumpsteer through the suspension travel. But practically, i'm working with a 50+ year old suspension system, and i'm sure there will have to be compromises for packaging. Is there a rule of thumb for how much bumpsteer is "tolerable" for a street car? Not a track car so i'm mostly interested in being as pleasant to drive as reasonably possible. Thanks!
Great question. Hard to answer.
If you don't let the suspension move, you won't have bumpsteer.
So, likewise, if the suspension moves a lot, you will have lots of tire change of direction.
I can tell you that when I moved my rack down about 1.5 inches in my rx8 that it was not very pleasant to drive. Two hands on the wheel at all times and seemingly did what it wanted to.
Moved the rack back up 1" (so a total of .5" down) and it feels great!
I attempted to actually measure bumpsteer and it was so drastically bad that I didn't need to measure. So unfortunately I don't have a number to give you saying "bad" or "good".
I think it was Pete that did some bump steer measuring and will probably be along to help give actual number data.
Interested to see what the hive thinks. My little Dodge has some bump steer as well and minimizing it would be nice.
From what I've heard from friends with older cars, if it can't be avoid, they aim for around 1/2 inch to an 1 inch at most. What car are you working on?
It's fine until it isn't.
I nearly killed myself in my turbo Miata when I hit the bump stops and spring rate went instantly to infinity. I was going through a transition from right to left at the same time I was transitioning from downhill to uphill. 40 mph. Sent me airborne, bounced me off a tree, spun me 180* into an embankment, then added another 180* for good measure and dropped me back in the middle of the road.
Don't underestimate the significance of loosing all movement right when you want it the most.
If you are autocrossing in a momentum car and expecting the bump stops on a reasonably flat surface it's fine. For street driving not so much.
Car is a 1972 Ventura.
I guess to clarify: I understand that bump steer can be really bad, and I plan to aim for 0 bumpsteer. Perhaps cynically, i'm expecting that there will be some reason (packaging or whatever) that I cannot locate the steering parts to actually give me 0 bumpsteer, so I'm wondering what i can aim for.
I think i also read that if you have to have bumpsteer, to-in is preferable to toe-out, but surely there is a threshold where some amount of toe-out is better than 3x as much toe-in.
Yes, toe in bump steer is preferable if you can pick. I had a very lowered truck that had some toe out bump steer because of some knockoff drop spindles. It toed out both on suspension compression and on extension from ride height. Turn-in was really strange because you would lose steering angle with body roll so it felt constantly understeery with constant steering corrections, and hitting a bump in a straight line tended to try to launch you into the oncoming lane, or off the road entirely. It was exhausting to drive. This was combined with added scrub radius, so had multiple problems.
I never measured the total toe change but I would guess it was on the order of at least an inch. It was very visible when cycling the suspension through its travel on jack stands.
budget_bandit said:Car is a 1972 Ventura.
I guess to clarify: I understand that bump steer can be really bad, and I plan to aim for 0 bumpsteer. Perhaps cynically, i'm expecting that there will be some reason (packaging or whatever) that I cannot locate the steering parts to actually give me 0 bumpsteer, so I'm wondering what i can aim for.
I think i also read that if you have to have bumpsteer, to-in is preferable to toe-out, but surely there is a threshold where some amount of toe-out is better than 3x as much toe-in.
If you are looking for a target, maybe shoot for 1/8 toe change per inch of wheel travel or less. I would think that would be about the limit.
In the early stages I couldn't get any better than 1/16-3/32 of bump steer over only 3" of travel (2 compression , 1 droop). Got significantly worse...like 1/4"+ in higher droop travel.
After redesign of the front end I was able to get down to less than .015 over 5" of travel. Huge improvement in tire wear! Did i notice it in handling? Not really that I remember. I suppose it is more stable over bumps and curbing.
Now with more changes I'm at about 1/32 (.03-.04) and didn't notice a difference in handling related to bump steer.
In reply to Asphalt_Gundam :
that is a great data point! is that per wheel or total?
SV reX said:It's fine until it isn't.
I nearly killed myself in my turbo Miata when I hit the bump stops and spring rate went instantly to infinity. I was going through a transition from right to left at the same time I was transitioning from downhill to uphill. 40 mph. Sent me airborne, bounced me off a tree, spun me 180* into an embankment, then added another 180* for good measure and dropped me back in the middle of the road.
Don't underestimate the significance of loosing all movement right when you want it the most.
If you are autocrossing in a momentum car and expecting the bump stops on a reasonably flat surface it's fine. For street driving not so much.
Bump steer is the change in toe as the suspension moves through its range of travel. Not the same thing that bit you in the Miata.
It's always going to go a little nuts near the ends of suspension travel. What you want is as little change as possible around your static ride height. I'd probably prioritize minimizing the change at the limit of compression over the limit of droop, as the tire is less loaded in the latter scenario.
budget_bandit said:In reply to Asphalt_Gundam :
that is a great data point! is that per wheel or total?
Per wheel, using a bump steer gauge to get accurate measurements.
Maybe there is a better way to do it, but the guys at project Binky did a thing (the bump steer episodes 10 & 11 ): https://youtu.be/X3DTTL9xRQ0?si=8W7pjrMmH_DlR6q9 (toward the end of episode 10 at the 25 minute mark) https://youtu.be/ob-FUF0Sh_E?si=L5YxQFtBb5nijwx5 (throughout 11)
In reply to Keith Tanner :
Will going to a wider tire aggravate bump steer?
You can probably minimize it by moving the rack or steering linkage up or down as needed...
... but to truly reduce or eliminate it you may need to lengthen or shorten the steering rack (or linkage arms) as well.
Depends how far down the rabbit hole you want to go.
Recon1342 said:In reply to Keith Tanner :
Will going to a wider tire aggravate bump steer?
If you define bump steer as the geometric change, no. If you define it as the handling effects of the geometric change, I expect so.
I appreciate all the feedback. I think my next step is to measure all my suspension points so i can draw it up in CAD and figure out what rack length I need. It needs to be rear steer, which is going to limit my options.
The stock tie rods are 18.25" apart, but I suspect that is not close to 0 bumpsteer so i'm going to measure it myself. From what I can tell on the internet, the consensus is that the 68-74 X-body (which shares front suspension with 67-69 F-body) has a pretty compromised front suspension, including positive camber gain under compression
In reply to budget_bandit :
Iirc, rack length is easy enough. Draw a line between the upper and lower inner control arm mounting points. Rack width (more accurately, inner toe rod pivot width) is the distance side to side on that line at the height of the steering arm. No need for CAD, you can do this quickly with a piece of paper. I seem to recall that there are a lot of racks right around 24".
If you want rear steer, maybe a RHD front steer rack flipped over will suit.
Interesting, i may be overcomplicating things...I presumed that i would need to draw a line between front/rear pivot points of both control arms and then draw the line between there which would also change based on fore/aft position, since my LCA pivot points are angled somewhat severely. not so?
If that's the case, then the vertical(ish) line should be in the same plane as the steering arm. Interesting that the LCA points are so different front to rear, you're going to have some fore/aft movement of the wheel on suspension movement which will make this so much more fun. The goal is to put the tie rod pivot points as close as possible to the lines the other components are pivoting on.
What car is this (or what car is it based on)?
It's for my 1972 Pontiac Ventura (sister car to the Nova, the X-Body platform).
I just did some back of the napkin math and the LCAs are each tilted off the fore-aft axis about 13.5 degrees. so maybe that's not super severe? The UCAs are practically parallel. See attached crude paint drawing lol
It's hard to find pictures of stock Nova (Nova II?) steering. Looks like there are lots of options for throwing it away and installing a rack already. Maybe you can get rack dimensions from one of those? And with bolt-on arms, conversion to front steer is a possibility. Heck, some of the pictures make it look like you could swap the steering arms from left to right and end up with front steer.
If you're using the stock steering arms, it looks as if they're placed right beside the lower ball joint and parallel the control arm. So your rack length is going to be awfully close to the distance between that pivot point on each side. Is this what you're dealing with?
That does look like what I have. FWIW i think the "II" was dropped from the nova name in '68.
I appreciate the suggestions. I hadn't really considered copying a kit, most that I've seen are for drag race applications so i (incorrectly?) assumed they wouldn't have cared as much about geometry as i would.
I also hadn't considered just making new steering arms to go to front steer, that might be a nice way to do it!
I repeatedly measure (or maybe I should say, approximate) bump steer on cars I inspect for road legality, and I also test drive them to judge the practical behavior.
As has been said, there are IME no exact limits for when bump steer is safe and not. The thing that I found has the greatest impact (apart from suspension travel and stiffness) is scrub radius - a large scrub radius will make bump steer feel way worse.
When it comes to wanting toe in or toe out on bump that is somewhat debatable. If you tune for "straight bumps" ie when both sides compress equally Id' agree toe in might be "better" (increased stability). However, I have gravitated towards tuning more for roll characteristics and in that case I prefer toe out on compression. This will give the car a slight understeer feel as it rolls over the outside wheel and that is usually a safer feel than a car that will steer more as it rolls.
For practical reasons I usually approximate bump steer using a laser pointer straight ahead from the wheel or brake disc, and track the dot on a board or something about 1 meter in front of the wheel center. My experience is that if the dot moves something like 10 mm or thereabouts, it is not an issue at all. More movement (and quick changes) can be more troublesome. Usually the dots will form an S-like shape.
It may be important to stress that this is for road car use and not maximum performance on a race car.
I have also found that theoretical calculations will only go so far - measuring what the suspension really is doing, with a bump steer gauge, preferably with the steering rack possible to shim upwards and downwards is what will give the results.
Gustaf
You'll need to log in to post.
