I used a smart strings setup yesterday and strung up my car just to get an exact box around the car and see what the measurement difference is front to rear on my ap2 with stock ap2 wheels. Using the extremely accurate smart strings setup, I found each side to be 1.125" or 1 1/8" exactly front to rear EACH SIDE. For a total track width difference of 2.25" When using this string up measurement, my alignment settings were exactly what we had set it for on an real laser alignment rack at a buddies shop.

I know it goes against all things I've read as far as track widths, but I have no reason to believe my 2.25" total track width is wrong. Or 1.125" on each side.

 

Andrew, I measured 2.375" less distance between the front and rear wheel mounting surfaces using a tape measure and a buddy so I'm going to go with your 2.25" measurement and update my website. It does appear that my AP1 is within about 1/10th of an inch of Andrew's AP2--they could be exactly the same figuring in my tape measure measuring error.

Andrew, what you're measuring isn't track width though--that's measured at the outside wheel (or tire). The wheel's offset and backspacing affect track.

 

The plumb bob says 1 3/16" difference from the hub faces. And by plumb bob I mean nuts on the end of fishing line. Anyway, using the earlier numbers that would make a 1.5" track difference measured from the rim of the tire, and presumably also the center line of the tire, although I didn't verify that.

That would make my new number for the center cap setup difference 3/4 (half the track difference) + 1/8 (difference between the center cap to rim edge) = 7/8. Which is fairly close to the original 13/16.

Conclusion- time to build a smart string setup, because it gives you a known square, or at worst a slight parallelogram. Also, there's a lot of potential inaccuracy built into my measuring, so take all these numbers with a grain of salt.

 

So far we have 3 wheel mating difference measurements: My 2.375", Andrew's 2.25", and George's 1 3/16". We need more AP1 & AP2 measurements to confirm what's the real number(s).

Anyone have these AP1 wheel measurements?:

 

I think I may be misunderstanding what the question is but I believe the differences being noted in track width are due to different ride heights of the comparing cars. The more horizontal the arms are the wider the track will be.
The important thing to start with is to verify square of pickup points (subframe). After that is verified and corrected if needed the relationship of each corners toe to the centerline of the chassis is what we are trying to adjust that is why the strings need to be parallel to center of the chassis not the same distance from front and rear hubs. Does that make sense? So the string should be the same distance from the hubs (left and right) in the rear first since it will be wider than the front and then this distance from the chassis centerline noted and duplicated at the front of the car so that the strings are parallel to the centerline and to each other. I don’t think it’s absolutely necessary to start at the rear but when setting up it might save a step.

 

My car was on jackstands and all 4 hubs were at full resting extension when I measured the width of the wheel mating surfaces.

 

measuring front & rear at ride height is going to be the only way to get it precisely accurate.

if your front & rear droop travel is different you'll end up with skewed numbers and be back to a parallelogram - albeit a very small one.

the easiest way to get that would probably be measure a known distance at ride height - pop the springs out and use jack stands and set a front and rear suspension to the same height from the body and then get the plumb bob out and find the difference between the front and rear - it should be wider at ride height than at full droop.

And as a few others have said - trackwidth is another one of those fun measurements that different people measure different ways - it's 'supposed' to be the middle of the tread width of the tires at either end of a car. it's very frequently measured as the outside of the tread width of the tire and just as commonly measured as the outside of the section width of the tires.

 

I just came up* with a $100 DIY alternative to smart strings that should be just as accurate. The design basis as I understand it is that as long as the front and rear distance between the strings is identical, the worst you can do is make a slight parallelogram. You can pick any point on the front and rear that is the same side to side but not necessarily front to rear (front hub face, axle stubs) to square it up within a 32nd, and the parallelogram aspect will be very small indeed. Then all of these numbers we're wondering about will be irrelevant. I'll post it this weekend in a new thread when my girlfriend stops yelling at me for working on the car.

*By I came up with, I mean I ripped off my brother who runs spec miata and won't see this. I just added a couple of user friendly options.

 

Quote:

if your front & rear droop travel is different you'll end up with skewed numbers and be back to a parallelogram - albeit a very small one.

^This. A small difference in droop travel leads to a small error in measurement that leads to a very small error in toe-in--the error is the angle of a triangle made up of two 94.5 inch sides (wheelbase) and the third side is the error (1/8" = 0.038 degrees or 16 thousandths of an inch of toe).

Quote:

measuring front & rear at ride height is going to be the only way to get it precisely accurate.

No need to "pop the springs off," just measure one axle at a time. Put the car on jack stands, remove the front wheels, jack the two front hubs up to equal height just enough to get the chassis off the jack stands, measure the wheel mating surfaces, then repeat on the rear.

 

Thanks for the earlier posts. Good info.


Let's talk about the tolerance stackup. Lots of things to consider here but let's start with the biggest contributor for most. Being able to measure to 1/32nds sounds impressive but it's not that great. Let's say everything is boxed up perfectly and you are measuring the leading and trailing edge of the wheel to within a range of 1/32" or +- 1/64". You are making 4 measurements per axle for toe and worst case scenario you could make all of them wrong by 1/64". That is 1/16" of total measurement error per axle. For a 17" wheel that results in about 0.20 degrees of total toe error using this trig calculator. I believe the factory says that total toe range is about 0.40 degrees so you've eaten up almost half of the entire range.

Obviously, the key to a good alignment is to reduce that per measurement error. Some people are good doing it consistently and looking for the diffused light through the fishing line as the lines on the scale underneath show up. I personally chose to use a Vernier caliper with a dial in inches. I took the same measurement multiple times and found I could generally get the values to within a range of 0.015". That is +- 0.075" times 4 = 0.03" of total measurement error per axle. That is total toe error of 0.10 degrees which is almost twice as good as using a ruler that reads to 1/32nds.

So the moral of this story, ditch the little ruler and use some Vernier calipers. Also, do a 'Gage R&R' on yourself and for your first time, take those same 8 measurements 10 times over. See what kind of measurement range you get. I was within 0.015".


The error in trackwidth between Rob's and Andrew's measurements by comparison is about 1/32" over 94.5" which only results in an error of 0.02 degrees.


Also there are other variables like wheel runout which can be quantified by a dial gauge which wouldn't be a bad idea to do.