Info

^ Closed Valve (No Bleeding) | | Open Valve (Bleeding)

Fig. 7.34. Resin Flow Analogy1'

^ Closed Valve (No Bleeding) | | Open Valve (Bleeding)

Step 7 No further bleeding occurs because the pressure on the resin has now dropped to zero, and the entire load (100 lb) is being carried by the spring. If this condition occurs during actual autoclave processing before the resin gels (solidifies), it would be quite easy for dissolved volatiles to vaporize out of solution and form voids.

Although this analogy greatly simplifies the composite flow process, it does illustrate several key points. In the early stages of the cure cycle, the hydrostatic resin pressure should be equal to the applied autoclave pressure. As resin flow occurs, the resin pressure drops. If a laminate is severely overbled, the resin pressure could drop to a low enough value to allow void formation; therefore, the hydrostatic resin pressure is directly dependent on the amount of resin bleeding that occurs. As the amount of bleeding increases, the fiber volume increases, resulting in an increase in the load-carrying capability of the fiber bed. It should be noted that resin flow and bleeding can be intentional or unintentional. Intentional bleeding is, of course, the bleeder cloth used to remove the excess resin from the prepreg during cure. Examples of unintentional bleeding are excessive gaps between the dams and laminate, tears in the inner bag that allow resin to flow into the breather material, and mismatched tooling details that allow escape paths for the liquid resin. Therefore, the hydrostatic resin pressure is directly dependent on the amount of resin bleeding that occurs. As the amount of resin bleeding increases, the fiber volume increases, resulting in increase in the load-carrying capability of the fiber bed and a decrease in the hydrostatic resin pressure. Referring back to the cure cycle shown in Fig. 7.31, the second ramp portion of this cure cycle is critical from a void nucleation and growth standpoint. During this ramp, the temperature is high, the resin pressure can be near its minimum, and the volatile vapor pressure is high and rising with the temperature. These are the ideal conditions for void formation and growth.

Unfortunately, the void problem cannot be resolved simply by maintaining the hydrostatic resin pressure above the potential void pressure of the volatiles (although this is a good start). During collation or ply lay-up, air can become entrapped between the prepreg plies. The amount of air entrapped depends on many variables: the prepreg tack, the resin viscosity at room temperature, the degree of impregnation of the prepreg and its surface smoothness, the number of intermediate debulk cycles used during collation, and geometrical factors such as ply drop-offs, radii, and so forth. An obvious place where entrapped air pockets form is at the terminations of internal ply drop-offs. In addition, air can be entrained in the resin itself during the mixing and prepregging operations. This entrained air can also lead to voids, or at least serve as nucleation sites. To summarize, the void formation and growth process is complex and yet to be fully understood. However, a number of basic principles are fairly well understood and have been investigated through several research studies. Several of these studies are summarized in the next section.

7.9.3 Hydrostatic Resin Pressure

Considerable research17 has been conducted to develop a better understanding of resin pressure and the many variables that influence resin pressure. The majority of this work has been done with unidirectional carbon/epoxy; however, woven and unidirectional carbon/bismaleimide have also been studied. The experimental setup used for these studies is shown in Fig. 7.35. To measure the hydrostatic resin pressure, a transducer was recessed into the tool surface and filled with an uncatalyzed liquid resin. To assure that the laminate did not deflect under pressure and contact the transducer, a stiff wire screen was placed over the transducer recess.

Vacuum Bag Breather Inner Bag Bleeder Porous Release Carbon/Epoxy Lay-up

Pressure Transducer

Tooling Plate

Pressure Transducer

Tooling Plate

Fig. 7.35. Resin Pressure Setup1

Studies were initially conducted on unidirectional carbon/epoxy composites to evaluate the effects of (1) the type of epoxy resin (Hexcel's 3501-6 and 3502), (2) laminate bleeding (normal and overbleed), and (3) pressure application (normal applied autoclave and internally pressurized bag). In all cases, the lay-ups were cross-plied containing 0°, 90°, and ±45° plies.

These tests showed that:

(1) A high flow resin system experiences a larger pressure drop than a low flow system.

A comparison of the resin pressures for 3501-6 and 3502 (Fig. 7.36) shows that the higher flow 3502 resin system experiences a larger pressure drop than the lower flow 3501-6 system. The 3501-6 system is a lower flow system because it contains a boron trifluoride (BF3) catalyst that significantly alters the cure behavior, resulting in a lower flow system that gels at a lower temperature. Since high flow resin systems are more prone to bleeding, additional care must be taken when they are tooled or bagged. High flow laminates should be tightly bagged and sealed to eliminate leak paths. Since high flow resin systems also typically have high gel temperatures, additional care must be taken to insure that the potential void pressure does not exceed the hydrostatic resin pressure prior to resin gellation.

(2) Overbleeding a laminate causes a large drop in resin pressure.

A comparison of normal bleeding with overbleeding is shown in Fig. 7.37. Overbleeding was accomplished by using three times the normal amount of glass bleeder and by removing the inner bag to create a freebleeding condition. While this would normally not be done in composite

AS-4/3501-6 AS-4/3502

Normal Bleed Normal Bleed

AS-4/3501-6 AS-4/3502

Normal Bleed Normal Bleed

Fig. 7.36. Resin Flow Comparison17

Normal Bleed Over Bleed

Normal Bleed Over Bleed

Fig. 7.37. Overbleeding Causes Pressure Loss17

part fabrication, overbleeding can, and does, occur because of leaky damming systems or leaky matched die molds. The analysis of these cured laminates included non-destructive testing (NDT), thickness measurements, resin content determinations, and materiallographic cross-sections. The resin contents and thickness measurements showed the dramatic effect of overbleeding. The resin contents and thickness values of the overbled laminate were significantly lower than those for standard bled laminate. Since the resin pressure actually dropped to below 0psi due to the vacuum pulled underneath the bag, little resistance to void growth existed. As expected, the ultrasonic NDT results and materiallo-graphic cross-sections revealed gross voids and porosity in the overbled laminate.

(3) Internal bag pressure can be used to maintain hydrostatic resin pressure and reduce resin flow.

Internally Pressurized Bag (IPB) curing was originally developed during the Reference 18 program. In this process, two separate pressure sources are used: (1) a normal external applied autoclave pressure, which provides the ply compaction, and (2) a somewhat lesser internal bag pressure, which applies hydrostatic pressure directly to the liquid resin to keep volatiles in solution and thereby prevent void nucleation and growth. An autoclave setup for IPB curing is shown in (Fig. 7.38). In the cure cycle used for this experiment (Fig. 7.39), an external applied autoclave pressure of 100 psi was used along with an internal bag pressure of 70psi. This results in a compaction, or membrane, pressure of 30 psi (applied autoclave pressure—internal bag pressure = 100 psi — 70 psi) on the plies and a hydrostatic pressure of 70 psi minimum on the resin. There is nothing magical about these pressure selections. If desired, the compaction pressure could be raised back up to 100 psi by simply increasing the applied autoclave pressure to 170 psi. The only restriction is that the applied autoclave pressure must be greater than the internal bag pressure, to prevent blowing the bag off of the tool. Of course, the internal bag pressure needs to be high enough to keep the volatiles in solution to prevent void nucleation and growth. To test a worst-case condition, the IPB laminate was bagged in the same manner as the previous overbleed laminate in which the resin pressure had dropped to zero. Even though this

Autoclave Pressure Vessel

Autoclave Pressure Vessel

Vacuum

Port Closed

Fig. 7.38. Schematic of Internally Pressurized Bag CuringJ

Was this article helpful?

0 0

Post a comment