Re: digitizing

From: Kurt Skifstad (
Date: Wed Jun 05 1996 - 16:35:34 EEST

We deal with both laser and CMM data on a daily basis, so, just to muddy the
waters a little bit, I though I'd put in my two cents worth on this "laser
vs. CMM" discussion.

First of all, we need to look at our terminology. Its a technicality, but
the discussion should be "non-contact vs. contact", not "laser vs. CMM".
Although laser technology is the basis for most existing non-contact
scanning technologies, there are a number of very interesting digitizers on
the market which are not laser based. In particular, the Moire, structured
light, and Digital Photogrammetry technologies should not be excluded from
the discussion.

Steve Farentinos brought up a number of very good points about the different
technologies. I'd like to add a little bit to a few of them:

>The CMM is better because:
>1) the CMM is very good at setting up part coordinate systems according to
a datum
>structure. This allows you to easily digitize in stages, i.e., move or
turn the part >and re-establish the previous coordinate system.

This is true for prismatic parts and parts in a fixture, but not necessarily
so for free-form geometry. In fact, the "higher density" data you typically
get from a non-contact scanning device may allow you to much more
effectively "digitize in stages" as you describe above. The coordinate
systems for the multiple scans can in many cases be more accurately aligned
(registered) in software, with or without datums. We have a number of
customers who are using our software-based alignment tools for complex
free-form sheet metal parts that have no slots, holes, planes or prismatic
sections to align off of.

>2) The CMM is more accurate than the laser

This may be a pretty good rule of thumb to go by, but it is not necessarily
true in an absolute sense. Accuracy is clearly product dependent, and more
importantly, the accuracy you need for your application may differ
significantly that that which someone else requires. Certainly, either
technology is adequate for the range of manufacturing application we deal
with daily in the automotive, aero, and consumer products industries.

>3) The CMM can acquire point data and measure features on top, on the side
and even
>underneath a part.

Perhaps more importantly, a CMM can digitize down in holes or deep pockets.
Laser and other optical triangulation-based technologies typically cannot
because the receptor is occluded by the wall of the hole or pocket.

>4) The CMM can measure features of position and form in the same set-up as
the digitizing
>is done. For example, you might want to digitize a contoured surface then
measure the
>position os several mounting bosses on that surface. It is far more useful
to get exact
>numbers for the size and location of these bosses than it is to have a
number of points on
>5) The CMM allows you to control the point density during digitizing: more
points in
>convoluted areas and fewer in flatter areas. An obviously planar area can
be digitized
>with exactly three points.

With the non-contact technologies, you typically have the option of keeping
all, or just a filtered portion of the data. Sure, you ideally only need
three points to define a plane or a circle, but this approach may be a bit
minimalistic. Given the option, I'd rather have the option of ignoring extra
data, than not having the option at all. One of our customers just
completed a major study of the tradeoff of using high-density data (in this
case, industrial CT) vs. traditional CMM for the inspection of cast parts.
The parts were completely "prismatic", so the CMM inspection took the
traditional 3 points for a plane, 4 for a cylinder, etc. approach. The
dense data approach actually took one third of the time (8 days vs. five
weeks), but was more successful in finding manufacturing defects, due to the
more complete description of the shape provided by the dense data.

>6) Since the CMM can measure size and location of primitives, it is better
suited to
>working with feature-based solid modelers.

This seems to me to be more of a process issue, rather than a sensor
technology issue (see above comments)

I would add a seventh advantage for CMM's: you can measure any surface that
is rigid enough to deflect the probe tip. The quality of your results is
independent of the surface finish or material which the part is made out of.
This is not necessarily the case for non-contact scanning technologies,
where you often have to prepare shiny or metallic parts by spraying them
with matte paint or a diffusing powder.

>The Laser is better because:
>1) the laser can digitize hundreds of times faster. This is very useful
for creating
>surface models of parts that must be built as "free form" surfaces (for
example, a human
>hand). If this is the type of work you're doing, I highly recommend a good
>tool for dealing with the thousands of points the laser will provide.
>2) The laser provides point data that are ON the surface of the part being
>This is not the case with the CMM, where the points represent the center of
the ball
>stylus and a later offset operation has to be performed to compensate.
This is like
>creating the cutter path for a ball-end mill, but in reverse.

Keep in mind that, since the laser point or line has a finite thickness,
there is potential for errors due to "ghosting" or point localization
problems, depending on the shape you are measuring and the angle at which
you are measuring that shape from.

Here's a question for the CMM guru's out there: all the accuracy numbers
you see from CMM's are based on benchmarks against (NIST ?) standard
geometry like blocks and ball bars, etc, right? Since, like Steve says, you
have to offset the data to get true surface location, can you have the same
level of confidence in feature measurements, like hole locations or edges?
Don't the offset calculations add an uncertainty to the data? and, doesn't
the process then become highly dependent on the skill of the operator (to
align the probe as optimally as possible)?

>3) Some laser systems can go directly to an STL file after digitizing a part.

Actually, in addition to what's available from some of the digitizer
vendors, there is software on the market that will take any (relatively)
dense scan and create an STL file.

>I hope this helps.

Me too.


Kurt D. Skifstad Imageware Inc, 313 N. First St, Ann Arbor, MI 48103
Email: Voice: (313) 994-7300 Fax: (313) 994-7303

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:37:23 EEST