Suspension alignment

As I work within the car industry, and specifically within the suspension design and development area, I am all too aware of the importance of getting the suspension set up accurately.  Suspension alignment relates to the angles of the 4 wheels relative to the road, i.e. the toe angle (the angle of the wheel relative to the straight ahead) and the camber angle (the angle relative to the vertical).  To get the toe roughly setup, I had originally used some long lengths of metal tubing strapped to the wheels protruding beyond the bodywork, and then measured the relative distance between the poles at 2 points.  Some trigonometry gives the total toe angle (which is the angle of one wheel relative to the other).  Hopefully this diagram (with rather extreme toe in!) expains what I mean:

For the camber I performed a rough check using a spirit level to give a vertical reference, and then used some more trigonometry.  Camber settings are generally less critical than toe settings- a general rule of thumb is that the vehicle is ten times more sensitive to toe than camber.  Hence, a toe variation of 1 tenth of a degree (or 6 minutes) would be seen as equivalent to varying the camber by a whole degree.

This method was OK for getting me through SVA (in fact, I had to mess around with front toe settings to get some degree of steering self-centring, but that's another story...), but given the tolerances we set vehicles up to at work (ideally to within a few minutes tolerance), I was keen to get the Mojo setup properly.  Once again, I was lucky in that I had access to a full optical alignment 'Hunter' rig after normal working hours at work!

One thing I was keen to investigate was how front toe settings changed the steering feel.  Since backing off from the extreme toe settings I used for the SVA test, I had noticed a slightly 'dead' feeling around the straight ahead, followed by a sudden eagerness to turn.  This 'two stage' feeling made the car feel quite nervous.  I had also noticed a tendency for the car to follow ruts in the road quite violently- the kind you get when HGVs regularly use a route- and this also made the car a little unnerving to drive.

So, with the Mojo on the Hunter rig, we pulled off the current setup:

So, the car is setup with a small amount of toe out at the front, and is very assymmetrical in terms of toe at the rear.  The camber values are reasonably consistent, so I chose not to adjust them at this stage.  As a baseline setup for toe, Jeremy Phillips at Sylva had suggested using Lotus Elise settings.  However, this runs with toe out at the front and I wanted to try running with a little toe in to see if this improved the 'dead' feeling.  Caterham suggest around 0.3degree toe in for their cars, so I decided to aim for something similar on the front, and aimed for the Elise setting of 22 minutes total toe in at the rear.  This is what we ended up with:

So, the rear toe assymmetry is now pretty much fixed and set up as an Elise, and I'm running with 0.24 degrees of total toe in at the front.

Driving impressions with these settings are positive: The 'dead' feel to the steering is now much better, the car responds sooner to steer inputs but is more progressive, and the steering feel once in a bend also seems better.  This demonstrates how a change of less than a fifth of a degree on each front wheel can make a difference!  Ideally I should have test driven the car after making the first adjustment rather than changing both front and rear toe settings together, but in practice this was going to be a pain- it takes quite a while to get the car setup on the rig.

April 2008 Update:  One of the few aspects of the Mojo that I felt could be improved upon was the rear toe link arrangement. The toe link runs from a bracket on the lower wishbone to the rear end of the lower upright through-bolt. 

On my car, the outer end has a bush as used elsewhere on the suspension, and the inner end is attached to the wishbone with 2 M8 bolts. Adjustment is therefore carried out by releasing the 2 bolts and slotting the holes if necessary. However this is a very imprecise adjustment, normally requires 2 people (one to hold the wheel, one to tighten the nuts) and is prone to being knocked out of position.

Later Mojos switched to a better setup (inherited from the Riot) with a rose joint at the outer end and a single inner bolt. Adjustment could therefore be performed by undoing the inner bolt, rotating the link (in half turn increments) and re-attaching the inner bolt. However, this still required the inner bolt to be removed, and so each half turn on the rose joint won't necessarily give equal adjustments due to tolerances inevitable when removing and replacing the bolt.

I therefore devised a further improvement to the arrangement, using a similar outer rose joint setup to the later Mojos, but attaching the inner end of the link to the plate on the wishbone with a second rose joint. By making one of these rose joints a left hand threaded item, adjustment can be performed by rotating the link in much the same way as the front trackrod.

The obvious choice for the link was to use steel with a weld in threaded insert on each end. However, this meant buying the steel, 4 inserts, getting someone to weld them up for me and then priming and painting. I looked into an off the shelf solution and found some hex aluminium 'turnbuckles' from http://www.midwestcontrol.com. These looked perfect for the job, with left and right hand threads in various sizes and lengths. Rather than facing importing from the USA, I decided to replicate these parts, so bought some hex aluminium to get cut to length, drilled and tapped by a local fabrication company.

I used 3/4" aluminium hex which was bought from ebay (500mm length for £8 plus delivery) and then ordered the rose joints from McGill Motorsports after seeing many recommendations on www.locostbuilders.co.uk. I ordered enough parts to convert 2 Mojos as a fellow owner wanted to do the same to his car.

The centre to centre length of the link needed to be approx 160mm, which translated to the aluminium section needing to be approx 90mm long.

With the parts sourced and made up, it was a simple job to swap them over.