Bending Steering Arms to Reduce Bump Steer.

Started by Duk, April 07, 2010, 03:36:11 PM

Previous topic - Next topic

Sheldon McIntosh


Duk

I found some 5/8" Heim/Rose joints that have +/- 26* of articulation (hopefully that's enough), and have plenty of load capacity so I've actually decided to use the approach as suggested by MD.
I've made 1 14x1.5 to 5/8" UNF adapter and I got a fairly long 12 mm bolt and extended the thread on it, I made 1 small bush that locates the spherical bearing of the joint on the bolt and will use 2 nuts to provide accurate location as I try different height outer joints. At least I will, once I actually get them (waiting for deliveries can be a chore  :( ) Once I get a decent height figured out, I'll make some nice solid drop legs and hold it all together with 12.9 tensile bolts.
I might still twist the steering arms because of the angle they live at, after adding the drop leg, the outer joint will be further out than normal and this will reduce the Ackerman angle of the steering. It also uses up some of the articulation that could be used to accommodate suspension droop.
When I get to doing measurements, I'll post some photos  ;D

Sheldon McIntosh

Cool , thanks for that.  Keen to see pics if you get a chance.

Duk

After some initial tests, I figure that bending and slightly twisting the steering arms is going to be required.
I've used a 100mm long bolt (note that I haven't lowered the outer rod pivot point by 100mm) and even with the rod end at the very bottom, I still can't get rid of the toe in on bump. This is just a visual observation, with no actual measuring being done yet, but the toe change is very noticeable.
And even with +/- 26* of articulation, I'm very droop limited. Tho I will admit that the tapered locating bush I made is limiting the articulation.

On another note, Vin's long shank upper ball joints add damn near 0.5* of negative camber per 10mm of bump travel  8). That's seems pretty aggressive, but I recon it's what is required to help give these cars front end grip. Note the little digital protractor gauge on the rotor.

Duk

Blatantly stolen from: http://www.vacmotorsports.com/articlesDetail.php?recordID=3


                                   Symptom:                                                                                          Cure:

Symptom 1. Toes out in compression and in on rebound all in one direction.     Cure 1.  Decrease shim on outer tie rod or lower the inner tie rod.

Symptom 2.  Toes in on compression and out in rebound all in one direction.     Cure 2.  More shim at outer tie rod or raise the inner tie rod.

Symptom 3.  Always toes in both compression and rebound.                     Cure 3.  Lengthen the tie rod as it is too short.

Symptom 4.  Always toes out on compression and rebound.                     Cure 4.  Shorten tie rod as it is too long.

Symptom 5.  Toes out on compression, then in on rebound and then starts     Cure 5.  Less shim at outer tie rod and shorten tie rod.
                  back towards out with more rebound travel.    

Symptom 6.  Toes in on compression, then moves out on rebound and then     Cure 6.  More shim at outer tie rod and lengthen tie rod.
                  starts back towards in with more rebound travel.

*Where "more shim at outer tie rod" is written, author is referring to lowering the tie rod as most cars run their outer tie rod under the steering arm.

jimnielsen

#20
I used the same approach to solve that problem, however, It think it is better to use a bolt that is long enough that it does not have any thread in the section between the rod-end and the upright arm - also I used a chrome-alloy sleeve over the bolt and a brace welded between the bottom of the sleeve and the upright body for extra rigidity. You need to ensure that you have hard bump stops so that the rod end never gets bound up. The strength figures for rod end are on-axis - they are only rated for about 10% of this figure in the way we are using them in this application, so you really need to ensure that they cannot run out of travel... You do not need either of the nuts then because the sleeve is simply trimmed to the right length to adjust for bumpsteep. I would only adjust for bumpsteer to the extent that it is actually a problem on the track, not to the extent that it can't be measured. Many racecars have practically no droop.

jim.

'95 Alfa Romeo 155 Q4
'90 Alfa Romeo 33 1.7 IE - my god! I can compete in Trofeo class!! -

Duk

#21
Quote from: jimnielsen on October 14, 2010, 04:01:54 PM
I used the same approach to solve that problem, however, It think it is better to use a bolt that is long enough that it does not have any thread in the section between the rod-end and the upright arm - also I used a chrome-alloy sleeve over the bolt and a brace welded between the bottom of the sleeve and the upright body for extra rigidity. You need to ensure that you have hard bump stops so that the rod end never gets bound up. The strength figures for rod end are on-axis - they are only rated for about 10% of this figure in the way we are using them in this application, so you really need to ensure that they cannot run out of travel... You do not need either of the nuts then because the sleeve is simply trimmed to the right length to adjust for bumpsteep. I would only adjust for bumpsteer to the extent that it is actually a problem on the track, not to the extent that it can't be measured. Many racecars have practically no droop.

jim.

Cheers Jim, I knew some of the stuff you have mentioned would come up.
The bolt in the picture is for set up purposes only. Once acceptable bump steer levels are reached, a Solid 25-30mm diameter sleeve and a 12.9 tensile bolt and Nylock nut will be used.
It's for a road car, so every effort will be made to make it safe and durable.
The outer ball joint bosses on the steering arms tilt downwards, towards the centre line of the car at about 11*. So straight away I've lost 11* of droop/drop articulation. That's nearly half of the articulation available :(.
So now I want to bend the steering arm so that it is down to the level of the joint in the picture and twisted to that at my selected ride height, the out joint boss on the steering arm is horizontal or even pointing upwards toward the centre of the car. Further bump steer testing will then be used and the outer joint will be further lowered as required.
Now I need to come up with a suitable means of making sure I bend the left and right arm evenly  ???.

Duk

I made this contraption to help keep the steering arm a constant distance from a solid reference point (obviously the brake rotor). Unfortunately I ran out of oxygen after getting the 1st bend done and will have to wait until Monday to complete the arm bending  ::).

Duk

#23
Unfortunately (for me), misinterpretation of what was required for what I was seeing happening, resulted in me making a change that was just to damn big  :-[.
Wrongfully thinking that long shank top ball joints would require big changes to the height of the outer tie rod end, in the same way that drop spindles and upside down ball joints do (the top ball joint is now a good 30+mm higher than standard, about the same as 1 1/4" drop spindle if you just look at the distance between ball joints). I heated and bent, heated and bent some more, and then twisted the arm so I wouldn't jam the rod end.
I tested and observed. I adjusted based on what I'd read. "Steering toes in during bump = lower the outer joint". So I did, more and more. But it never got any better, it only got worse. The best I got was with the rod end right under the now lowered about 50mm steering arm.
Bump from about ride height level was reasonable (still no accurate measuring as bump steer was easily observable by eye). But the toe out during droop  :o  :o  :o.
After many hours wasted, I then decided to place the rod end on top of the steering arm. It would be only a small amount lower than it would have been if the arm was still in its original shape (and the rod end mounted under the arm as per standard), and low and behold, bump steer was massively reduced to the point where I couldn't see it with out setting up a laser pointer and mirror (still have to buy a small mirror, tho)!!!!  :D :D :D

So the conundrum is, what to do now ???
Do I leave the 'adjusted' left side like it is and make the right side the same ("Upside down Miss Jane"  :P)?
Or do I do my darnedest to straighten the arm back to original shape but still with a slight twist so that I can do some fine tuning of the rod end height?
Or do I straighten the arm and just slightly lower the it to about where the pivot point currently is, use original tie rod ends and be done with it?

**** P-hoto's whill be added tomorrow for clarification sake, 'cause MD got confuzzed, again ;) ****

MD

Hey Dukster,

Very interested in what you are doing however, try as I might, your somewhat er ..rambling description is less that exacting so I cannot fully understand your testing sequence nor the outcome but I would like to.

You wanna have another crack at the write up perhaps or do some editing?

Call me thick if you want, (the list of other titles is longer.) :)
Transaxle Alfas Haul More Arse.

Current Fleet
Alfetta GTV6 3.0
Alfetta GTV Twin Spark supercharged racer
75 1.8L supercharged racer

Past Fleet
Alfa GT 3.2V6
Alfetta GTV 2.0
Giulia Super 2.0
Berlina 2.0

Sheldon McIntosh

Yeah, must admit I'm a bit lost here also.  Really keen to know what's going on, my knuckle risers arrived today  ;D

Duk

Hey Sheldon, great to read you're making geometry changes as well  8). These things really need it.

Long story short: If I have interpreted what I found today correctly, bump steer correction when using long shank ball joint/knuckle risers on our cars requires much less adjustment to get sorted than I first thought. Tho it will be affected by caster angle (as you wind in more positive caster, the outer tie rod end moves downwards).
Unfortunately I don't think it would be safe to use a drop leg style bump steer correction without slightly twisting the the outer tie rod end boss on the steering arm.

Photo's tomorrow  :D

Duk

#27
OK 2 photos to help show what it is I've done.
1st is shows the bent steering arm with the bolt I used to allow me to vary the height of the outer rod end.
The 2nd photo shows how I put the rod end above the steering arm and started to get the results I'm after.
Now there is no point in bending the steering arms this much to the turn around and put the outer rod end above the steering arm. But if you know the spindle, you'll see that the boss that the tapered hole is in is twisted so that it reduces the amount of angular articulation required of the rod end. I believe that this is required if you plan to use drop leg style bump steer correction even tho the amount of drop won't be that great.

Sheldon McIntosh

Great work, keep it up, I'll be happy to do the mods once you have a satisfactory fix!!

Fitted my knuckle risers today, interested to see how they feel on the track this weekend, will let you know.  I tried to take one corner on the way home reasonably quickly, and I did think for a minute there that I was having to turn the wheel more than I normally would, but it could have been my imagination.

I'm getting a wheel alignment first thing tomorrow, considering what's been said should I still stick with 0* toe?

MD

Seems to me that if you keep the original tie rod end ball joint and fit it in the usual manner, you wont need to bend the steering arm as much. This gives you the added bonus of using the original parts that are properly rated for the application.

After reviewing my eventual adjusted ride height, I cannot go as low as I wanted due to fouling of the guards and so the bump steer correction is going to be about half of what I originally expected or around 40mm. Bending the arm should be good enough for that.

     

Transaxle Alfas Haul More Arse.

Current Fleet
Alfetta GTV6 3.0
Alfetta GTV Twin Spark supercharged racer
75 1.8L supercharged racer

Past Fleet
Alfa GT 3.2V6
Alfetta GTV 2.0
Giulia Super 2.0
Berlina 2.0