Scientific Moulding Helps Ensure Repeatable Part Function

How to create consistency in injection moulding part production through detailed, scientific processes

Scientific, decoupled, and intelligent moulding are commonly used terms within the plastics industry to describe an injection moulding methodology that allows for consistent, repeatable parts to be efficiently produced in a system with many sources of variation. This guide covers the basic steps in setting up this process in order to create the most accurate, repeatable moulded parts for you.

viscosity versus shear rate
This chart shows viscosity change vs. shear rate. As shear rate is increased, the easier the material will flow (viscosity drops). At the point of full alignment, the viscosity will remain constant with further increasing shear rate. This is known as the Upper Newtonian Plateau.

Determining Optimum Fill Speed: In-mould Rheology Curve

An important first step is to determine a robust fill-speed setting that allows for inherent variations, which are seen both within polymers and injection moulding press performance, all while reducing the variation imposed on the polymer melt that fills the mould cavity. This will help ensure repeatable part and/or product performance related to the polymer’s bulk properties.

First, a little background and theory on polymers is needed. Polymers are made up of long chain molecules. These molecular chains are entangled with one another, and can form crystalline regions in semi-crystalline materials. When a polymer is heated beyond its melting temperature, the crystalline regions break down in semi-crystalline resins. For both semi-crystalline and amorphous resins, the free volume between molecular chains is increased to a point that the molecular chains are allowed to move freely past each other.

This chart shows viscosity change vs. shear rate. As shear rate is increased, the easier the material will flow (viscosity drops). At the point of full alignment, the viscosity will remain constant with further increasing shear rate. This is known as the Upper Newtonian Plateau.

  • Zero Shear Viscosity
  • Log Viscosity (η)
  • Log Shear Rate (γ)
  • Zero Shear Viscosity
  • Upper Newtonain Plateau

As the material is being driven to flow, shear stress is imparted on the molecular chains. The shear stress begins to align the molecular chains in the direction of flow, thus allowing the chains to slip more easily past one another. As a result, as shear rate is increased, the easier the material will flow (viscosity drops).

This is a behaviour seen with Non-Newtonian fluids. The viscosity will continue to drop as shear rate is increased until the molecular chains have reached full alignment. At the point of full alignment, the viscosity will remain constant with further increasing shear rate. This is known as the Upper Newtonian Plateau, as the material will now exhibit a more Newtonian fluid type of behaviour [see Viscosity Change vs. Shear Rate chart above].

Let’s move from background to application. The fill-speed determination process attempts to capture this shear-thinning behaviour of the material as it flows through the mould. The in-mould rheology curve plots apparent viscosity vs. shear rate using readouts from the press—actual injection pressure and actual fill time [see Rheology Curve chart below].

Rheology Curve
This example of a Rheology Curve Chart, which helps determine the optimum fill speed, shows that, as the shear rate is increased, the viscosity is reduced.

As seen in the Rheology Curve chart, as the shear rate is increased, the viscosity is reduced. At higher shear rates, the viscosity begins to level off. Choosing a fill speed that results in a shear rate in this plateaued region allows for material differences from lot to lot. Additionally at this fill speed, inconsistencies in the machine stroke speed/response will have a negligible effect on the viscosity of the material that is flowing into the mould cavity. This results in manufactured parts having more uniform molecular/filler orientation, heat history, and pressurisation from shot to shot.

Looking at the process for determining a robust fill-speed setting also allows for the moulding process to use the shear-thinning behavior of polymers to fill out smaller features, more accurately replicate mould geometry, improve weld line strength, and minimise potential for flow front hesitation. Be cautious, though: Don’t choose a fill speed that is too fast, resulting in very high shear rates. Shear rates that are too high can begin to degrade the polymer. The molecular chains can begin to break down, resulting in lower molecular weight. Lower molecular weight chains result in mechanical properties of the material being altered. This can lead to premature part failure. High shear rates are especially damaging to filled materials. Glass-fibre is a very common reinforcement for plastic materials, and often times has a shear rate limit well below that of the carrier material. When shear rates are too high, reinforcing fibers can be broken down, thus reducing their effectiveness.

Rheology Curve:

  • Lower viscosity
  • Apparenty Viscosity (psi*s)
  • Shear Rate (1/s)
  • Determining Optimum Hold Pressure

 

 

scientific moulding

This illustration highlights the changes that occur with amorphous, semi-crystalline, and liquid-crystal polymers.

open gate illustration
In this illustration, the gate is still open along with two sections: the region that is still pressurised and the underpacked region.

In order to counteract this shrinkage as the plastic cools and solidifies, pressure must be applied to compensate.

Hold pressure is used to force more plastic material into the mould cavity to increase the density of the moulded part. This helps prevent volumetric shrinkage-driven issues such as sinks and voids.

The application of hold pressure also helps to keep the material at the outer surface of the moulded part in intimate contact with the mould wall, thereby improving conductive heat transfer, facilitating cooling. Once the material has solidified, it can no longer effectively transmit pressure. This is why it is important to maintain uniform wall thicknesses in plastic part design, as well as use a thick-to-thin gating scheme.

Determining optimum hold pressure involves identifying the lowest amount of packing pressure to adequately fill the remainder of the mould cavity that was intentionally left short during the filling phase, and reduce any shrinkage related issues to an acceptable level.

Then, the highest acceptable packing pressure is also identified by finding the pressure in which the mould begins to flash, or other issues arise with the part, such as the part sticking upon ejection. These points—the lowest and highest—establish the acceptable packing pressure window for a specific melt temperature.

well packed region
Part geometry here shows a well-packed region, a region with high potential for being underpacked, and a thin section that freezes early, isolating thicker regions from hold pressure.

Selecting a point two thirds of the way between these points provides for a setting that will effectively pack the part, while minimising overpacking issues [see illustrations].

Determining Optimum Hold time (Gate Freeze)

Next up is determining a hold time setting that will allow for pressure to be transferred from the machine nozzle to the cavity for as long as the gate remains open.

Applying positive pressure to the cavity for as long as the gate remains open yields the most packed part at a given hold pressure. Applying pressure until gate freeze also prevents backflow of material from the cavity into the feed system. The more packing, the lower the volumetric shrinkage of the moulded component. Typically this results in an optimally packed part, although there are special cases where using a hold time shorter than gate freeze may be preferential. Minimising the volumetric shrinkage of a moulded component requires that the optimal packing pressure is applied for an effective amount of time.

As plastic cannot effectively transfer pressure once it is solidified, the amount of time for effective packing becomes geometry-dependent. As previously discussed, thin sections within the part can freeze off quicker than thick sections and isolate regions from packing pressure.

gate freeze
In this illustration, the part weight continues to increase as more hold time is used, until gate freeze. Once the gate is frozen, the part weight doesn’t increase because plastic is no longer flowing through the gate into the part. This is what leads to consistency and process robustness.

Following standard practices for injection moulding part and mould design, it is common that gates will be placed in an area that allows for thick to thin packing, and the gates are typically sized as a percentage of the wall thickness being gated into.

Common gate sizes typically range from 60% to 80% of the wall thickness, but certain gate designs require more restrictive sizes to facilitate separation from the part wall.

Given that the gate is typically thinner than the part wall, the gate will solidify before the part geometry. Once the gate is solidified, pressurisation of the part cavity from the feed system is complete.

There is no need to continue applying packing pressure once the gate is frozen, as this will only pack out the feed system.

Gate freeze determination involves tracking the weight of the part when different hold times are used. As long as the gate is open, and the cavities continue to be packed, the part weights will continue to increase.

Gate freeze scientific molding
This chart shows that the part weight continues to increase as more hold time is used.

Once part weights reach a stable level, it can be inferred that the gate has frozen, and the parts are no longer being packed. The hold time where the weights have leveled off can be considered the time for gate freeze.

Typically a buffer of one to two seconds is added to the hold time setting to ensure that the gate is frozen regardless of any inherent variation within the material or machine.

In the Gate Freeze chart, it can be seen that the part weight continued to increase as more hold time was used. As the gate began to solidify to a larger extent, the rate of weight gain begins to slow.

In this example, five seconds is the first point where the weight has stabilised. This indicates that the gate has frozen somewhere between four and five seconds, and a one-second buffer has been added to ensure process robustness.

If you want to find out more about how injection moulding works, our applications engineers are available to talk to you about moulding and your designs and products. Contact one of them at +44 (0) 1952 683047 or at [email protected].

cpd certification  Email [email protected] with your name, title of activity, and completion date to receive your certificate

expand_less