SAMPE Journal - July/August 2016 - 10

Feature
rticle
eatureAA
rticle

Figure 3. Boiling point of water as a
function of vacuum level.

parts, the super-ambient dwell range of 50°C - 60°C
which appears to work well for the CYCOM 5320-1
prepreg system clearly has the potential to approach
or exceed the boiling point of water and is a probable
reason why a dwell in this temperature range is much
more efficient in removing this volatile than one at
lower temperatures.
It is suspected that it may be possible to further
shorten the super-ambient dwell time from the 4
hours used for the panels shown in Figure 2 and a
program of work is in progress to further optimize
the cure cycle. In addition, further work will also be
carried out to determine whether the use of a superambient dwell can eliminate the need for both the 16
hour room temperature dwell and the 121°C dwell
and permit a ramp straight from the super-ambient
dwell temperature to the final 177°C cure. This latter
cure cycle has been verified in small scale trials, but
larger discriminator panels will need to be made to
confirm this option for large parts.
Part Curing Scenarios - In-Tool Cures vs. Demold and Freestanding
Postcure
As previously noted, much of the published work on
VBO structures has been concerned with prototype
and limited production parts which are often cured
initially on low cost, low temperature capable
tooling, demolded and postcured freestanding to
the final cure temperature. For CYCOM 5320-1,
cure cycles used include a) a 93°C/200°F 12 hour
initial cure followed by a slow ramp at 0.3C/minute
to 177°C/350°F and b) a 121°C/250°F 3 hour
initial cure followed by a slow ramped postcure to
177°C/350°F. These cure cycles are illustrated in
Figures 4 and 5 including predictions of viscosity, Tg
and degree of cure from the Raven® cure modeling
software developed by Convergent Manufacturing
10

Technologies. Cytec funded the Raven kinetic and
viscosity models for CYCOM 5320-1 and they are
publicly available.
For the scenario of initial cure - demold - postcure,
three general rules of thumb have traditionally been
used:
* Tg > initial cure temperature after the initial cure
dwell
* Degree of cure > 50% after the completion of the
initial cure dwell
* Postcure ramp rate to be slow enough such that
Tg > oven temperature during the ramp
It can be seen from Figures 4 and 5 that both cure
cycles actually break some of these 'rules', the 93°C
initial cure giving less than a 50% degree of cure,
the 121°C initial cure giving a Tg slightly less than
121°C and for both cure cycles the Tg is marginally
exceeded during the initial stages of the postcure
ramp. Both cycles have however been successfully
used for the manufacture of large parts which
illustrates the fact that applying the rules strictly to
the results from a kinetic model may be conservative
in some cases.
The cure cycles illustrated in Figures 4 and 5
assume the use of low cost tooling in a prototype
environment and in such a scenario the length of
these cures with or without the addition of a 16 hour
room temperature dwell or a shorter super-ambient
dwell is not an issue as production rate is not a
concern. For rate production, however the duration
of the cure is very much an issue and a different,
shorter cure cycle is required. Figure 6 shows a cure
cycle for CYCOM 5320-1 typical of those used for
production parts. The cycle consists of a 2 hour
dwell at 121°C/250°F followed by a 2 hour dwell at
177°C/350°F. The ramp rate used was 2°C/minute
throughout.
SAMPE Journal, Volume 52, No. 4, July/August 2016

Table of Contents for the Digital Edition of SAMPE Journal - July/August 2016