Re: SLA scan speed vs. laser power

From: Lightman, Allan J. (
Date: Thu Jun 19 1997 - 17:17:00 EEST

The question was raised about the relationship between laser power and
scanning speed. The answer given, that the two should be directly
proportional, should have been prefaced with the phrase "all other
things equal." The answer was based upon the proposition that the cure
depth directly depends upon the energy deposition per unit volume and
doubling the laser power should double the energy flux. There are a
couple of implicit assumptions made here that the users should be aware
of. These factors are typically quite small and, in the past,, under
usual operating conditions they would not have been a noticeable factor.
But, with the accuracies currently being achieved by best operating
procedures, the operator must check carefully. As Paul Jacobs was fond
of saying, when you improve the process you first take care of the
boulders blocking progress, then the rocks, and then the small stones.
SLA has reached the capability of quite small errors, so the assumption
of direct proportionality over an extended range (2:1) should be
carefully checked.

Here are a couple of issues which need to be addressed. If the increase
in laser power is achieved by changing the laser operating condition
(ex., turn up the current on an argon laser), the laser footprint at the
vat needs to be measured. Often, raising the current will change the
beam diameter in the tube. Depending upon the focusing conditions, this
may result in a smaller footprint resulting in the deposited energy
density going up faster than the laser power increases. Also, the laser
mode structure may change. At higher gain the resonator may support
higher order modes. The resulting beam divergence will change. This
will also impact the spot size (make it larger) and the collimation
distance (i.e., the footprint change from centerline to best focus to
farthest edge of the vat may go beyond the accuracy control needed).

Other issues of concern are the scanners and the material response.
There is a range over which the scanner response will be in direct
proportion to the control. It is possible to move so fast that the
spindle in the galvanometer shifts in the bearings (due to torque)
resulting in the mirrors cocking just a bit. Also, there may be
response lag and other adverse effects. The resin polymerization is a
chemical reaction, often exothermic. If the process is calibrated for a
particular rate of reaction, dramatically changing the reaction can play
havoc with very fine control.

There are other factors which also need to be checked. In summary, if
you change the operating parameters, you should verify your process.
The simple approximation is a good starting point. But if your goal is
to achieve the ultimate performance, every detail must be taken into

Allan Lightman
University of Dayton Research Institute

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:39:44 EEST