DIY budget 4-wheel alignment project. Update, done for under 20. Full write up!

You are 100% right but when it comes down to it if I set my tool up on the bodywork (and it is adjustable here) and get the string to say 2degrees which is pretty easy would mean a difference between front to back string hub measurement of 0.6mm about what I can measure with a steel rule anyway.

If I can get it within 1degree which I reckon I can then we are talking 0.15mm difference between front to back string to hub measurement.

Edit

So if my front bar was off by 2degrees it would work out as 0.6mm over the length of the car, break this down to the measurement over the size of the wheel, you need to work out the angle. Then use trig over the 356mm wheel area You are looking at 0.2mm. This is a small tolerance that can be lived with for me.

 

If I'm understanding the question right, the answer is 2 parallel strings down each side of the car.

The box is alligned off the hub centres - beware damaged rear axle centre grease covers - and then measurements to string on the wheel rim as usual.

Does this answer?

 

Even when the string is even to each hub centre the end bars could be twisted, this twist or angle to the bar and the string will result in a trapezium. This shape would then mean the measurements from the front of the wheel to the back will be ever so different. See post above for actual mm measurement.

 

You can use 2 axel stands front and rear then measure off the centre hubs to get it parallel and measure off the front of the alloy to the bar to square everything up

As long as the string is the same distance apart on both bars then its simple.

 

when setting up you need to make the axel stands equal distances as stated, but if you measure this at a slight angle (anything other than 90) then you start to create the trapezium. The bigger the miss measured angle the bigger the error. And it a lot easier to get this wrong when doing the axel stand method or not know about it and go at it blind which if bad enough would cause poor alignment.

 

Its a good point about using body points as references, thats never good, for the string box method you should be aligning the string using the hub centres. It doesn't matter if the bars are at an angle however as long as theyre parallel to the hub centres and equidistant side to side. a feature of a trapezoid is a pair of parallel sides so thats fine.

of course if the beam is offset or the chassis is really bent then the hub centres wont be where they should be then it wont be quite right, but at least you can still align all 4 wheels relative to each other which gets you most of the way there.

 

It does matter if the bars are off as it will mean your sides will not be perfectly parallel, see post above.

 

I can't see that. Once the car is parallel between the two strings, measured to the hub centres (and allowing for different track widths front and rear), then tracking work is ready to commence.

 

It is but only if the string at the front and back is the same distance apart measured at perfect right angles. If it is the same distance apart but right angles are not present then the strings will be off parallel, as I said if 2deg out means around 0.2mm at the wheel ie not a lot.

If someone was to not realise this and measure using axel stands, it would be very easy to measure the string distance at a greater angle without knowing and that would mess up the alignment - dependent on the measured error.

The string method is a good method and if setup correctly (to with 2deg) it would offer an accuracy to within 0.2mm

Needs a diagram to help explain my point I think as I have said a couple of times.

 

There are a lot of assumptions in this righting and wronging, because some of the discussion is on the previous thread.

If a car has an unknown wheel spacer on one corner only, the measurement to hub centres will be out on that one corner, and the said trapezium will result.

The parallel strings are gained via measurement of strings to hub. What hasn't been flagged in this thread is a separate validation of the string positions via the bars at each end. Both on this thread and the previous tracking thread, it can be seen that the bars have grooves in them to guide this process:

I suspect this isn't the axle stand version we're referring to - and that that one is one without the bars at each end. Aka bodging!!

 

Guys, didn't mean to complicate things, was just thinking as I sat here as I spent a lot of time working all this out setting up my chassis this year. Each adjustment affects the next so you need to get them in the right order. As you adjust the camber it changes the toe slightly so you need to do the camber first. This is compounded if the car is lowered and the steering arms haven't been repositioned to compensate as the toe changes far quicker etc.

My particular problems on my car are compounded by the fact that the front wishbones are adjustable in and out, back a fore plus the rear beam has some adjustment in it at the mounts and stub axle plates as well. It was therefore important to get a fixed point at the pivot points on the chassis to work from (they seemed the strongest points and least likely to be distorted, well the back ones not the front ones at the front end!) and get ride height, corner weight, wheel position and track width set first.

For standard cars where there is less chance to get it all horribly wrong it should be far simpler. As far as I can see the trapezoidal issue isn't really important as you can't move the wheels back and fore really so as long as the lines are parallel either side and square to the cars flanks then the front and rear angle isn't an issue. If however to get it perfectly square but not square with the square formed but the four wheels (i.e. not parallel to the cars flanks) then that is an issue. That said if we are talking about tenths of a mm then that isn't going to make any difference.

I reckon, having had another think (I do that a bit too much at times!!), that if you have std suspension the methods outlined above (and nicely made I must add) are just the ticket but custom suspension setups compounded by a used chassis you need a little bit more thought. Best to work from suspension mounts (or some known point) and work out from there to ensure everything is in the correct relative position before you start.

 

notches are a good shout, I've got holes drilled in mine that the string goes through.

Smudge I see what you mean about the angle of the bars, say if the front bar is straight and the rear is at an angle then the spacing of the strings on the rear is reduced, causing a slight angle in the strings? As you say though with the distance and angles involved its a small error.

So to try and clarify, I think to get it right there's 1 of 2 things that need to be done;
either
1. Ensure string lengths are equal side to side and ensure that the spacing of the strings front to rear is equal. This effectively gives a parallelogram and means any angle on the bar won't produce an error
or
2. If the string length is adjustable (i.e. with a plum weight) its important that the bars are square, or at least parallel front to rear.

Does that about right?

Great thread, thats why I love this forum!

 

OK, my gauge is the same and has adjustable string holders on the bars, the bars are the same length on my setup so setting the adjusters equal distance from the end works just like above.

The problem with the bar is knowing it is perfectly at right angles to the string.

This pic shows a perfectly setup string method.



C = distance from end of bar to string (same both sides and ends)
D = distance between string (same both ends)
B = distance from string to hub centre (same both sides)
A = distance from string to hub centre (same both sides)
E = String to rear edge of wheel
F = String to front edge of wheel


On this perfect set up B, E and F are perfectly the same distance.

If the bar was slightly wonky (made bigger in this diagram for illistarion purposes only).



A is the same on both sides, but E is now smaller and F is now larger than B. This would make for an error when aligning the wheels.

How can you make the bar perfectly at right angles to the string? some people may not even now it needs to be, thats all im saying. If it is within 2 degress the error will be minimal.

Correct me if im wrong (and I think the angles in the last one are wrong as they dont add to 360 - doh! will edit when I have time) but the images are illistrations only to help explain the importance of setting this up correctly.

 

so another question, what's the best way to ensure the rear beam is centered? As tonyB says the chassis points on a 20 year vw aren't the best references and theres a good chance the rear beams bent as a nine bob note.

 

Notches or fixed length bars, eg the BMW on p1. It all depends if this is for 1 car or several with differing widths.

Technically a parallelogram is slighly off because the widths then alter (trigonometry), but as close to 90 degrees on the bar/string should get the required accuracy for our purposes.

The main essence of this thread was/remains to 4-wheel allign a car, when most would settle on 2-wheel, thus getting a substantially better outcome.

 

Bugger you have got it! and its taken me ages to make those pics.... measure the string length side to side as long as its+-1mm all should be well

 

lol, looks like we're all over this! great minds eh

 

Aston Martin here has grooves in the string holders, and this will be identical front & rear, eliminating set up time:



BMW, fixed length bar:



Same with the Focus, fixed groove positions in end:




Sorry if this hadn't been clear from the outset. My original intention had been to buy some identical length wardrobe rails, put DIY store end caps in them, and then screw a countersunk self tapper home into it to replicate the BMW set up.

 

This method with the bars at the end on the axel stands is exactly what we use.

Sorry i i didnt make it clear in my earlier post chris.

 

Stumbled over this one!

'91 GTCC at Donny!



Appears to be bolted to a tubular frame at the front of the engine, aka radiator defender!

Schmidt Motorsport on the lorry (OT).