RE: Stratasys Press Release

From: Atif Mehmet Yardimci (ayardi1@uic.edu)
Date: Tue Jan 27 1998 - 07:42:18 EET


On Tue, 27 Jan 1998, DAN DAVIS, RAPID PROTOTYPE wrote:

> > reliability. This system builds parts up to 2.5 to 5 times faster
> > than
> > the company's current FDM systems. Additionally, a
> >
> Didn't they say that about the 8000, too? Is this a multi-head unit?
> If not, it seems to me that physics says that it cannot be much faster
> -- you can extrude plastic only so fast. Show me!
>
>

My 2 cents:

Assuming that 'standard' road cross-sections remain the same, i.e. 20 mil
wide x 10 mil thick:

Area = 1.29 10^(-7) m^2,
Projected translation speed = 5 "/s = 0.127 m/s,
Flow rate = 1.64 10^(-8) m^3/s,
Nozzle Area for T12 = 7.3 10^(-8) m^2,
Average velocity for 12 mil diameter nozzle = 0.225 m/s,
Estimated shear rate ~ 4 * V / r ~ 6000 1/s (under-estimation due to
 shear thinning nature of polymer melts)

For ABS : density * specific heat = 2 10^6 [J/((m^3) K)],
T_tip = 270 C,
T_env = 70 C,
power required for phase change =
  rho * cp * flow rate * (T_tip - T_env) ~ 7 W

Accordingly, in current configuration (1 "/s) shear rate and heater power
can be estimated as: 1200 1/s, and 1.4 W. Clearly heater requirements
are not critical. If the current filament delivery system
can generate five-fold increase in pressure without filament buckling
or slippage in rollers, FDM hardware appears to be scalable. Then again,
high pressure delivery systems would increase the scalability window
further.

IMHO, extrusion does not appear to be the critical step in FD
technology.

On the other hand; as fabrication platforms are being
scaled up (geometrically and kinematically), material properties
of modeling materials remain the same. Feature resolution
and shape stability problems may appear at high deposition
speeds, due to longer dwelling times at high temperatures.
In other words, relevant heat transfer processes (convective
cooling, conduction and advection) do not scale-up accordingly.

Best Regards,

Atif Yardimci



This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:44:46 EEST