On Thu, 8 Dec 2016 20:12:12 -0000 (UTC), John Harmon
wrote:

actually said:

0.01deg ?? I don't think so...

We really must know to what accuracy we need the measurements to be becuase
every measurement tool ever made has this as its basic issue.

Do you think it's less, or more accurate that we need for camber
measurement?

As just one reference, page 8 of this document says that camber (and toe)
measurements must be accurate to "2 angular minutes".
http://www.bimmerboard.com/members/s...tem%5B1%5D.pdf

The question then becomes how to translate 2 angular minutes into inch
measurements.

Well, 2 angular minutes is 1/30 degree, 03 0.03 degrees.

On page 10 of that document it says the camber tolerance of another vehicle
model is ? 10' (plus or minus 10 minutes).

So what is 10 minutes in inches?
It is .01 degrees

If your car doesn't pull to one side and the tire is not wearing un-evenly,
the camber is fine.

I realize there are many ways to measure things, and I understand that
you're using the tire wear and handling to measure camber, but I would like
to try to get a bit finer in granularity (especially since lots of other
things can cause both those issues).

I have used an ordinary carpenters bubble level to check it.

I have plenty of carpenters bubble levels, one with digital output, so
that's also another option.

If it is within 1/4 bubble it should be OK.
I understand what you're saying which is that the negative camber on my
rear tires can be anywhere between 0 and minus 2 degrees.

But I would like to get a bit more accurate than 1/4 bubble!

One of my cars specifies the following static camber range, for example:
Front (non-adjustable) camber = -0.7? minimum, 0.3? maximum
Rear (adjustable) camber = -2.2 ?mimimum, -2.0? maximum
( http://www.bmwdiy.info/alignment/index.html )

Most roads have crown so the camber is not as critical as you might think.

Some cars compensate for that by specificying cross camber specs, but mine
are symmetric.

And the caster has an offset instead.

The static negative camber is "supposed" to increase lateral grip. At the
same time, it certainly increases inner tire edge wear and decreases
straight-line braking traction. On uneven road surfaces, you can get camber
thrust (where the tire moves toward the camber).

Problems with this method a
1 ground where the car is parked needs to be both flat and level

Yup. That's a measurement and calibration issue for sure, but luckily, my
garage is extremely flat (I measured it once long ago).

2 ordinary tire bulges out on the bottom, need to set the level
against the tire away from the buldge

That's excellent advice. Since the tire bulges, I wonder if it's best
to use the wheel lugs to mount a jig which is what we measure to?

Like I said in an earlier post - make a "jig" - a kind of trammel
device - that contacts the lip of the rim and transfers the
measurement out to a straightedge that spans the rim fiving a flat
surface parallel with the wheel centerline. Best to use this for toe
adjustments as well.

Sometimes you can simply compare the reading on the front wheels
to the back wheels.

This is a good hint, which is that we can just note what the *delta* is
between the front and back, and measure that delta, over time, with a handy
instrument.

Also note many cars are designed to have the front wheels tilted
inward at the top slightly for stability

Mine has negative camber on both front and rear, but front isn't adjustable
without adding camber plates.

Unless you like this as a hobby, it probably doesn't pay to DIY.

I disagree but I understand your point.
On sheer economy, there are only 3 measurements I need for my sedan:
1. toe front
2. toe rear
3. camber rear

So all I need, to do a "pragmatic" alignment check, is to check those
three.
A. If they're off, then I can get the car aligned for $100 or more.
B. If they're on target, then I save $100 each time I measure them.

Measuring toe in is much more fun.

On page 14 of the document above, it tells me that the static toe and
camber accuracy needs to be:
Toe measuring accuracy ?2' in measuring range ?2? in total range ?18?
Camber measuring accuracy ?1' in measuring range ?3? in total range ?10?
http://i.cubeupload.com/cfaDWp.jpg

Does anyone here know how to convert the 1 and 2 minutes to inches?

minutes are 1/60th of a degree.. Converting that to inches requires
offset and trig. The farther you can extend the "offset" the more
accurate you can be. I use a laser level on the trammel jig to extend
the line out about 10 feet from the spindle. Then do the calcs to find
out how much toe-in you want at 10 feet for the angle specified. If
the toe is specified in inches it is the difference between the front
and back of the tire circumference. Again, some calculations will
allow you to measure farther out for more accuracy.