Debunking the Backflow Myth

Editor's note: Most molders are familiar with the concept of backflow in the mold. Conventional wisdom says backflow occurs after the mold has filled and the high pressure of injection is terminated. The assumption is that unless pack and hold pressure is high enough and held long enough to achieve gate seal, some molten material will backflow through the gate. This can lead to abnormal and troublesome variations in part weight. Roger Mansur is a technical support specialist at Polymerland with almost 30 years of plastics experience. Using simple physics he contends that backflow is a myth, and proposes an alternate explanation.

There is, in injection molding, a myth. It says that backflow can occur in a reasonably normal and well-controlled injection molding cycle. I will attempt here to dispel that myth with some established laws of physics and fluid flow. It is my contention that what is perceived as resin backflowing out of the mold is actually the system not adequately packing enough resin into the cavity. It is a failure to fully pack out the part.

The Principles

The first principle deals with fluid flow. In a system such as an injection molding machine with fluid flowing through valves, sprues, runners, and gates, there is a pressure drop with every transition. When the system goes static (no flow), the pressure becomes constant throughout the system. The greater the restriction or transition, the greater the pressure drop across that restriction or transition. Typically, the gate or gates in the mold offer the greatest restriction to melt flow.

Molten resin in this system plays a big role in how easily the static condition can be reached. A low-viscosity material, by definition, flows more readily and requires less pressure to be moved through the system. About 10 to 20 psi is enough to move water through 100 feet of garden hose. In injection molding, we think in terms of 7000 to 20,000 psi to move melt through anywhere from 6 inches to 4 feet of "plumbing." Unlike water, the melt has a tendency to lose heat and increase viscosity to the point of becoming a solid. Also, unlike water, molten resin has a measurable compressibility. Keep this in mind.

The second principle is that a mass of a liquid takes up more space than that same mass does in solid form. For instance, 1 lb of a liquid has a greater volume than that same pound of material does as a solid. Water, again, is a notable exception to this principle. For that reason, this discussion is being confined to thermoplastic materials changing state - going from liquid to solid.

Now for the tough part: in a fixed volume - like a mold cavity - when a portion of the plastic changes state from a liquid to a solid that takes up less space, the pressure of the remaining portion of the liquid drops. If the liquid were incompressible, the pressure drop would be instantaneous. With a compressible liquid like molten plastic, the pressure decay is slower.

At this point most people say, "Aha! That is what pushes the material back out of the cavity - backflow." But remember: the melt loses pressure at each transition point. If flow is still occurring, the pressure in the cavity is less than that in the runner, which is less than that in the sprue, which is less than that in the nozzle, which is less than the pressure in front of the check valve in the barrel. This is an important point and it is essential to both sides of this argument. A lot of pressure is required to move the resin, especially across the major restriction of the gate. And up to this point, this pressure has been supplied only by the molding machine.

Let's list a few other principles:

1. All things being equal, a thinner cross section of material gives up its heat and solidifies faster than a thick section. For this reason, gates are usually smaller to allow quick gate seal, and for reasons of cosmetics and ease of removal.

2. Mold surface temperatures are normally much cooler than the melt temperatures.

3. Liquids in motion are less likely to freeze than liquids that are static.

Filling in Action

Examine now the filling process by visually watching the travel of the screw. The mold has closeand the press has developed the appropriate tonnage. The screw is pushed forward by thmachine's hydraulic system. Assume the following:

If the screw is not too large, you may see it stop or even bounce back when transfer occurs. This is because of the compressive nature of the melt in front of the screw. But consider this: the amount of melt that has passed through the gate is barely compressed, and it's easier to expand into the space available in the cavity than to move back through the gate.

Back to the observed travel of the screw. The resin in front of the screw is still moving into the cavity and the machine hold pressure quickly catches up to the melt pressure and the screw moves forward again until the cavity is initially filled and its internal pressure rapidly builds to equal the system pressure. This can be observed as the screw stops forward motion. Depending upon the relationship of the screw, barrel, and cavity, sometimes you can see the screw creep forward during the balance of the packing cycle.

What's Really Happening

What is occurring at this point? If the system at this instant were filled with a fluid that was only slightly compressible and was not changing state from liquid to solid, the pressures would be equal throughout the complete system. But that is not the case. Resin solidifying throughout the system is changing the restrictions from the cavity back to the nozzle tip.

One of the largest volumes of liquid resin in the average mold/machine is that in the nozzle and in the barrel in front of the check valve. The volume of material in the system beyond the nozzle (in a conventional runner system) is trying to become solid, trying to take up less space, trying to drop in pressure. So the packing process continues, with the resin flowing through the system along the increasing resistances and associated pressure drops.

If the machine's hold pressure were dropped to zero gauge pressure, the volume of liquid resin between the check ring and mold would not necessarily go to zero. Since the resin is compressible, the screw would have to be mechanically pushed back to decompress this volume of material. This is not what one considers a "normal" molding process. If the resin had the compressed power to physically move the screw backward, it would still have the energy to move material into the cavity.

Remember, the material with the greatest pressure when hydraulic pressure is dropped is that volume right in front of the screw. Also if that volume could move the screw, it would only have the power to move the screw and the hydraulic oil behind it until the resistance of that mechanical/hydraulic system equaled the compressed force of the melt. Most molders admit that it is very unusual to see a screw bounce back when the hold pressure is dropped.

Now, without a measurable motion rearward on the screw and with a reasonable seal at the check valve, one can assume that the pressure on the liquid resin was the same or very close to the pressure prior to machine cut-off to zero gauge. This pressure would decay very rapidly with the additional solidification of the plastic in the cavity. But the available pressure to "stuff" additional resin into the cavity has been removed.

The question is, "Is there enough residual liquid volume and, thus, pressure in the mold cavity to move resin back out against the residual pressure in the melt system?" This author's contention is "No." The concept of gate seal is a good one. But my belief is that the reduction of part weight when hold pressure is cut off before gate seal is achieved is really a missed opportunity to "stuff" more resin into the cavity - not resin backing out of the cavity. Pack and hold provides the opportunity to inject more resin into the cavity until the gate seals. Variations in part weight are due to variations in melt viscosity, available pressure, tool temperatures, and melt temperatures, which can be minimized when adequate pressure is maintained on the melt until the gate seals the cavity off from further processing. Perceived backflow variations are really missed opportunities for fully packed out parts.