Checking and Removing Bumpsteer
From time to time the question of the poor handling of the MKIII Sumo crops up on the Cobra forum, the subject quite quickly turns to bumpsteer! Many people have also stated that their car handles just fine and that their car does not have the bumpsteer problem……They usually need to sit down with a strong, hot cup of tea after the modifications have been carried on their car, as it makes a huge difference to the way that the car handles.
Believe me, there is a great handling car inside every Sumo MKIII just waiting to get out!
What is bumpsteer?
When the car is travelling in a straight line both front wheels should be pointing exactly straight ahead, this is because the toe angle (the angle that would be formed between two lines projected a head from the front wheels) should be set to zero degrees on the MKIII Sumo. In other words the wheels are parallel to each other. If this angle changes in any way as either front wheel travels up and down due to the action of the suspension then bumpsteer is being induced.
What causes bumpsteer?
As the wheels travel up and down with the suspension there is a fairly complicated set of movements occurring in the wishbones. This is due to the angle of them relative to each other along with the fact they are different lengths (by design).
All this movement and swinging about of the uprights or hubs affects their relationship with the steering rack and track rods. Some people have suggested that the track rods just need to be parallel with the bottom wishbone. Personally I think that there is a bit more to it than that which why it’s easier in the long run just to measure the problem and then move the rack up or down to correct the problem.
IMPORTANT: - You cannot just go moving the rack unless you’re prepared to re-set the toe angle to zero degrees. If you do not re-set the tracking, the result will be a poor steering set up that may not self-centre very well at all. You could take the car to a professional tracking centre but you must make them aware of the fact that the rear wheels can not really be used to set the fronts up as they have a toe out issue. ANY adjustments that you make to ANY of the various settings of the front wheels will always mess up the tracking so you may as well get them all done and dusted before you start!
Bad tracking will always result in a car that handles badly!
Elsewhere in this Wiki there is a method of increasing the castor angle; it is advised that you should carry out this modification. The other modification that you should do is to reduce the excessive camber that results from using the Pilgrim ‘vertical at full droop’ set up method. As a rough guide you can set the camber to vertical at full droop but then wind out each top ball joint two full turns to reduce the camber. The figure for my car is 0.8 degrees at normal ride height. (Measured using another method!).
Some combinations of donor and Pilgrim parts make setting up the castor quite tricky as the ball joint will not have the 2 turns of thread poking out the back of the wishbone, this is required in order to keep the SVA man happy. Tall mushrooms may help in this respect if you are using sierra uprights. Whatever method you use to sort this out, sort it out you must.
I believe that the excessive chamber also has an effect on the self-centring of the steering (reducing it). You should do all the other mods before you attempt to correct the bumpsteer problem as the other modifications will have an effect on the final rack height setting.
Note: The wishbone bolts, if disturbed must only be tightened up when the suspension is at ride height, if you tighten them when the suspension is at full droop the bushes will be under a lot of unnecessary stress when the suspension is under load.
What does bumpsteer feel like on the road?
This is quite hard to describe, but if the steering wheel needs small but constant corrections in both directions as you travel down the road then bumpsteer is the issue.
If you never feel happy about letting go of the steering wheel with both hands then it’s because subconsciously you know that the car will ‘bumpsteer’ you into the nearest hedge! To give you some idea what a modified car is like I would be totally happy driving with no hands even at 130 mph (on a private test track).
If you have to concentrate much harder when driving your Cobra than you do when you are driving your tin-top even at the same speed, that’s bumpsteer.
Basically bumpsteer spoils all the fun!
What part of the chassis needs to be modified?
The steering rack on the MKIII chassis is bolted down using two long bolts that go through two welded tubes or posts. Changing the height of these posts is the modification that will get rid of the bumpsteer.
The early MKIII chassis had posts that were 120mm high when measured in total length from top to bottom. On the later MKIII chassis the posts are 110mm high. At the time of writing Pilgrim is looking into reducing the posts to around 101mm high to finally solve the issue.
Pilgrim will also specify the components that this post height will work with so DON’T just cut your posts down to this height unless you are 100% sure that you are using the specified components.
There is also some talk that suggests the various racks that Ford used over the years have had different heights, again, if true this will affect your final post heights and should be checked out.
How do you cut the posts down?
This depends on many things such as whether your car is fully built, whether you have inner wings fitted, whether you have the radiator fitted. All I can say it that this can be the worst part of the whole modification. You need to mark the post up accurately and cut them nice and square. I used a jubilee clip round each post to help keep the cut square. I did the cutting with a hacksaw blade held in my hand and had radiator and inner wings fitted, it took two days!
If you have an air hacksaw or an angle grinder with a thin cutting disk it might take 10 minutes to cut the posts if you can get to them ok!
How much should you cut off?
This is tricky as you don’t really know if the height is right until you bolt the rack back into the car and carry out the bumpsteer test. (The method is described below)
Personally I would cut the posts down to 95-98mm as it is easy to lift the rack back up with shims (washers). Cutting yet more off the posts if they are still too high would be a real chore!
Make sure that you have plenty of shim washers that are 1mm thick or thinner.
It is quite likely that the pinch bolt for the lower UJ at the rack end of the steering rod will foul the chassis upright, it is very important that this does not happen, a small dent in the chassis soon sorts out the problem! (make sure that the bolt is no longer than it needs to be).
You must at least slacken off the UJ pinch bolts for both UJ’s and the bolts that form the clamp that grips the triangular steering rod where it pokes through the bulkhead to aid moving the rack. Don’t forget to tighten all of the bolts back up AFTER the rack is bolted down at its final height.
The long bolts that hold the steering rack in place will be slightly too long after the modification has been carried. The nuts will seem to tighten up but in reality the rack will not be held tightly against the posts as you will be tightening up against the shank of the bolt. You could (if you're clever) use the bits that you cut off the posts as spacers above the rack to ‘take up the slack’ (make sure that the ends are squared off). You could also drill out the threads of a couple of 19mm nuts and use them.
Making adjustments to the camber or castor could effect the final height of the rack so all the necessary with respect to these settings should be carried out first.
How do you measure the Bumpsteer?
There are many methods that can be deployed to measure the bumpsteer, all have advantages and disadvantages. The method that I used requires a cheap laser; even a laser pointer pen could be used at a pinch. The only other items are a few scraps of wood or steel and a piece of paper.
The first thing that you should do is measure from a reference point on the wheel to the top of the wheel arch whilst the car is sat on its wheels. You could use the hole in the wheel where the false spinner fits as the reference point. (Assuming that you are using image Halibrand replica wheels). Make a note of this measurement, as you will need it later.
You will need at least 3 metres of space in front of the car.
You then need to jack the front of the car up so that both front wheels are at full droop. Place a piece of wood under the passenger side wheel and place two axle stands under the chassis near the engine (either side of it). Lower the car onto the stands. The driver side wheel needs to be just off the ground but not miles up in the air, as this may slightly alter the readings. The passenger wheel must be resting on the wood but the wheel must NOT be supporting the weight of the car. This is required as the rack must have something to push against.
The ground needs to be reasonably level and the car also need to be level from left to right. The axle stands should be set to the same height.
Set the steering wheel in the straight-ahead position.
The laser needs to be attached to the driver side wheel so that it is horizontal and lined up with the centre line of the wheel. It could just be taped to a piece of 1x1” wood that it jammed into the rim ‘returns’ of the wheel. The image below shows slightly more complicated fixing method as I use a pair of lasers to set the tracking up. The wedges that tension the wire are very high tech! (Dismantled clothes pegs!)
As we want to measure the suspension from full droop to full compression the pre-load collars for the spring need to be backed right off. (Gaz shocks have a small gub screw in the collar, dont forget to loosen it off!). Count the number of turns so that you can reset the collar to its original position. The bumpsteer after the shock has compressed past its normal ride height distance is in fact more important than the ‘droop’ settings in many respects. A neat trick is to place a large amount of weight in the inner rim of the wheel so that it will self level the laser as you jack the suspension up through its travel. (I used two half bricks!)
Make sure that the laser is truly horizontal with a spirit level then turn the laser on.
Arrange a piece of paper so that it is held vertically 3 metres away from the car. The laser beam needs to be at the bottom of the paper. The paper should be taped or pinned to a piece of ply wood board or something similar. The edge of the paper should be vertical as this is a reference with regards to the marks that you are about to make on the paper. You can use a plumb weight to make sure that the paper is vertical.
Now you have a choice of methods, both will work very well.
Method 1, (slightly less accurate than method 2 but is very easy to do)
Mark the paper where the beam hits the paper.
Now jack the wheel up 10mm with a trolley jack under the lower wishbone, rotate the wheel slightly if required to get the laser level again. The beam should be hitting the paper 10mm higher than the previous, mark where it hits, the horizontal movement it what we are interested in.
Repeat the jacking up and marking procedure (adding 10mm each time) making sure that the laser is horizontal throughout.
At some point the wheel will be jacked up so that the distance between it and the top of the wheel arch is the same as the distance that you measured earlier. Make a special mark on the paper at this point as this can be classed as the zero, or ride height point. All the marks on the paper below this point are suspension ‘droop’ points. All marks above are suspension compression points. (The suspension compression points are slightly more important than the droop marks).
Keep jacking the wheel up 10mm at a time and marking the paper until the shock absorber runs out of travel.
Examining the results
Join all the marks up and check if the line goes straight up and down the paper (i.e. is it a vertical line when the paper is held up?). If the line is not vertical then you have bumpsteer!
Method 2, (slightly more accurate than method 1 but a bit more work is involved)
This method will allow for the fact that the wheel does not travel up and down in a straight line, it travels outwards in an arc from full droop as it goes up but at some point it will travel back in towards the engine.
The set up of the laser and the paper is exactly the same as before but you will also need a vertical reference point that is 100mm away from the centre of the wheel. You will need to use a plumb weight to make sure that the reference bar is vertical. In the image below I used my pillar drill vice with a piece of aluminium clamped in it. (Use a plumb line to make sure that the bar is vertical)
With the wheel at full droop, position the bar 100mm away from the centre of the wheel. Now jack up the wheel 10mm. Measure the gap between the wheel and the bar. Lets assume that it has reduced to 98mm, this means that the wheel has moved outward 2mm. Therefore the mark on the paper will always be 2mm out even if there was no bumpsteer at all as the wheel has moved out 2mm. The actual pen mark placed on the paper should therefore be offset back towards the projected centreline of the car by 2mm.
Jack the wheel up another 10mm, lets assume that the wheel to bar measurement is now 96mm, again the pen mark needs to be compensated for by 4mm.
Repeat the process until the shock absorber is fully compressed.
In all honesty method 1 is more than good enough to work out the rack height and it is a lot quicker and easier to do!
Below is an image of the vertical reference bar required for method 2
Method 3 - the Rowly Method.
this approach is similar to the above but uses a mirror to reflect the laser back to the wheel, this removes the arch of the wheel so you don't have to. Rowly has done a superb video which helps explain a lot.
The results on my car
The image below shows 3 sets of results with my rack height being set to 101mm, 110mm and 120mm.
With a 120mm rack height the horizontal movement was terrible, this shows that as the suspension goes up and down, the wheel is being turned, meaning if you his a lump in the road with one wheel, the car will pull noticeably to the side. With the rack lowered to 110mm, the bumpsteer is reduce but still not good. The final mark shows my modified rack height of 101mm, it can be seen that my rack could benefit from being lifted up 1mm to get the line above the ride height to be trued up to vertical. Remember that the more vertical the line the less bumpsteer there is. (I will do this very fine tuning and then re-set the tracking in the summer!)