[alfa] more on paint, does it matter?

[Tessie McMillan <tessmc@xxxxxxxxxxx>]

Hey, this is an interesting topic!

You know, I used to work in a clinical chemistry laboratory. It was a U.S. 
reference laboratory for measuring cholesterol -- one of only seven (at 
the time) labs worldwide, standardized to what we called the gold 
standard at the NIST. I actually used to present a discussion of our 
procedures at scientific meetings. 

You'd think an analyzer is a simple procedure, but it is not. And 
depending on the type of analyzer, there are errors from a huge variety of 
sources.... These errors are not limited to the analyzer, but include the 
way the samples were taken, any differences among the samples (like 
additives, viscosity, temperature, etc.), the optics of the system used, 
the way the operator performed or interpreted the analyzer, the way the 
system was calibrated and the type of material used to calibrate the 
system, the type of measurement system used (enzymic, chemical, 
colorimetric, etc.), and on and on.

So... If coming up with a paint color is anything like performing a 
cholesterol determination, I see a good potential for different results 
here. &:-)

To prove it, I did an experiment myself with house paint at the local Home 
Despot when I was painting my house, which is an.... unusual ... color, 
so color subtleties are fairly noticeable. Having painted it by hand and 
brush it took me quite a while and several trips to the hardware store. 
You'd be surprised at the variability in paint color I got, even using 
the same code and checking the code on the label. The system they use 
there for mixing house paint seems to be poor in both precision and 
accuracy; it can't get the _same_color each time, nor can it get the 
_right_ color each time. 

Not knowing anything about the machines or how they worked, I thought at 
first it was a question of calibration; that all you need to do is 
calibrate to the correct color standard and once you find a color 
standard you can trust, you can repeatedly get the same good paint with 
that machine. So I tried different Home Despots, finding a little bit 
better results at the new store in Redmond. But then I realized -- having 
purchased what turned out to be a funny looking color in a can bearing the 
correct label at the trusted store -- the machines must be dumping in, 
with variability, slightly higher or lower amounts of color additive than 
what is required by the measurement. It could be that the machines are 
variable in this capacity, or that there's a large variety in the paint 
quality. I then sampled paint lots and found that some lots of paint 
tended to be much thinner than others, so they probably injected 
differently into the mixture when it was calculated. I have no idea what 
kind of optics are used -- whether it's a "calculate, pump color, and 
you're done", or whether there are before and after measurements, but 
judging on the color I received, I'd guess it's more like the former.

Getting back to auto paint, aside from the measurement aspect, I believe 
there are different paint techniques for the different paint brands, is 
that not so? Time for -- what is the term -- enzyme accelerating or 
something like that -- could be too long for a particular procedure given 
the correct pigmentation, and this leads to a brittleness that then causes 
a different appearance in the paint because the paint then reflects light 
differently than it would with less enzyme acceleration (I'm making the 
word up because I can't remember what this is called). So that in and of 
itself could cause an additional difference between paints.

But since we're talking about ALFAs, does that really matter? I thought 
reds from different years were different. I would expect some colors to be 
hard to find during war years because the components might be more 
expensive, and so perhaps economies might be made in those years. 
Further, if you really examine any one car from a period and compare it 
to another, it might have different mats, door knobs, turn signals, or 
trim, for example, because that's what was handy when the car was built. 
So if you find a PPG paint that says it's Alfa yellow and computes to 
Alfa yellow, would the difference be SO significant from a Glasurit Alfa 
yellow that it really matters? I mean, who is around to decide whether it 
really matters any more? I look at my Spider and usually conclude it was 
CERTAINLY put together by someone who hadn't read "Alfa Romeo Spider: the 
Complete Story" &:-).

I'd think getting the most visually pleasing (to me) color would be the 
key. (And then purchasing enough for the entire car so it's all the same 
lot... &:-)) What do you think?

Tess
in Bellevue, WA USA
--
to be removed from alfa, see /bin/digest-subs.cgi
or email "unsubscribe alfa" to majordomo@domain.elided