Feedback to this page: click here

The Polymer Injection Molding machine is no longer available as an open access tool at DTU Nanolab. The machine has been moved to DTU Construct.

Polymer injection molding

Polymer injection molding is a high-volume polymer repliction method where a viscous polymer is injected into a custom-made closed mold. After solidification of the polymer, the cavity is opened and the polymer part is ejected.

Description of Machine

Injection molding machine in the basement under the cleanroom

The injection molder at DTU Nanolab was an Engel Victory Tech 80/45 which is a hydraulic machine with single motor. The machine was also equipped with a robot which could pick up finished samples and place them on a conveyor belt. See the table below for key capabilities of this machine.

Equipment Performance

The polymer injection molding machine previously installed at DTU Nanolab was a hydraulic system with the following capabilities:

Parameter		Value
General	Tools	Flat disc, Disc with Luer connectors, Microscope slide
General	Clamping force	Up to 450 kN
Processing temperatures	Mold temperature	~15°C to 140°C
Processing temperatures	Heating cylinder	Absolute max: 350°C
Molding parameters	Injection speed	Up to 53 cm³/s
	Injection pressure	Up to 2200 bar
	Process types	Constant or variable mold temperature

Polymers

At DTU Nanolab the following polymers were used routinely:

Topas 5013L-10
Topas 8007S-04
PP
PMMA

Process Parameters and Optimization

There are no standard injection molding processes because it varies significantly how users prefer to optimize their process and how they prioritize parameters such as polymer type, replication fidelity, dimensional accuracy, optical properties, residual stress and cycle time. However, browsing through the Process Log it's usually not a problem to find a program made for the polymer and tool at hand which can be used as a starting point.

The injection moulding cycle consists of the steps outlined below:

Mold heating setup
Closing mold
Injection
Switch-over type
After (holding) pressure
Cooling
Dosing (plastizising)
Demolding
Ejection
Robot sample pickup
Nozzle settings

Process overview

Once a process is running satisfactory, the machine also has tools that can be used for monitoring a running process. These tools can help to ensure desired sample quality and reproducibility. Press the second button from the top on the right column of buttons to get to the 'Cycle time analysis' screen:

The 'Cycle time analysis' screen provides an overview of the entire injection molding cycle.

'Mold Close' to 'Part removal': In the left part of the screen the individual phases of the injection molding cycle is listed. Just to the right of these names there's a small box. If it's gray it means this step i currently not active. If it's red it means this step is currently active (so in this example the sample is currently being ejected by the ejector pins and the part (sample) is removed by the robot).

Progress bars: A little further to the right (in the center of the screen), progress bars will indicate position and timing of each individual step in the cycle (both last cycle and current cycle). Note that the 'Part removal' step of Variotherm processes often takes quite a long time. This is because the 'Mold pause time' delay (mentioned in section '9. Ejection') used for reheating the mold is included in this step.

Last and Current columns: In these columns it's possible to see how long time each phase takes in the currently active cycle. For comparison the corresponding time during the last cycle is also shown.

Total time: In the bottom of the screen a progress bar will indicate progress for the currently running (yellow) and previous cycles (green). Please note that the time axis of the progress bars do not autoscale. Min and max time scale values must be entered manually (but since the total cycle time for both the current and last cycle is shown right above, it's very easy to enter a sensible value).

The injection molding machine also has tools for monitoring process parameters during the injection molding cycle. Usually the most interesting events happen around injection, so data logging is often centered around this event. But there are many and almost endless possibilities. To ensure reproducible results it can be beneficial to monitor parameters that are tightly linked to sample quality/properties which include injection speed and injection pressure. Pressing the second to last button on the left side of the screen will take you to the 'Micrograph' tool:

The Micrograph tool can be used for monitoring process parameters.

Setup: To get started you can press 'Setup' and then 'Manual setup' and select relevant parameters. On the tab 'Measurement duration' a start condition (in this example 'Start injection' was chosen) and a measurement duration (in this example 1,5 sec) can be selected. On the tab 'Curve selection' parameters to be plotted are selected (in this example specific injection pressure, injection speed and shot volume were selected, but there are many other possibilities). Since all selected process parameters are plotted on the same y-axis, this axis will always show the unit per cent and always be scaled 0-100%. The 100% value is then set independently for each plot (in this case 800 bar, 40,0 cm³ and 15,0 cm³ was selected for the 100% value of specific injection pressure, injection speed and shot volume respectively.

Other settings: There are many other possibilities for setting up monitoring features, logging process parameters to file, monitor sample quality and e.g. discard samples that do not meet set requirements. Consult the original machine manual or ask for details.

Mold temperature setup

Injection molding processes can be divided into two different types: Constant mold temperature processes and Variable mold temperature (Variotherm) processes.

1. Constant mold temperature processes

In this type of process the mold remains at the same temperature during the entire injection molding cycle. The temperature should not exceed the glass transition (or melting point) temperature of the polymer being used. Otherwise the polymer will never get cold enough to solidify to a degree where it can be demolded without being deformed/damaged. How close one can get to the glass transition temperature depends on several factors like e.g. polymer type, cooling time and whether the polymer has a tendency to stick to the shim (especially an issue when replicating high aspect ratio structures). With Topas 5013L-10 (which has a Tg around 135°C) one can often go up to 100°C and for samples that are easy to demold even higher (110-120°C). For polymers with very low Tg (such as e.g. flexible polyurethanes) one often has to run as cold as possible. With the mold heating switched completely off and mold cooling water fully open, the lowest achievable mold temperature is around 15°C.

Parameter	Benefits	Drawbacks
Low mold temperature	Very fast cycle times due to short cooling times	Problems with replication of small features (incomplete filling). Stress in samples.
High mold temperature	Improved replication	Problems with demolding. Problems with diesel effect. Slightly longer cycle times.

For constant mold temperature processes it is a good idea to manually close the mold cooling water valves e and g. For reproducible results it is also a good idea to make sure that you always run at the same temperature setpoint for the mold casing cooling water controllers (see Chapter 4.11 in the manual). A setpoint around 50-70°C for both sides works well in most cases. If the tool casing is allowed to heat up to 90°C or higher it will often result in problems because the internal parts of the mold will get stuck (especially the ejector pin system).

2. Variable mold temperature (Variotherm) processes

In this type of process, the mold changes temperature during the cycle. There are two principal ways to achieve this:

A: Switch off mold heating during the molding cycle. This case is best suited for situations where a slow and well-controlled cooling is desirable. Since cooling happens quite slowly (heat dissipates out in the tool casing) this method is usually only of practical interest when the mold temperature only has to drop slightly (maybe up to ~10-20°C drop in temperature).

B: Switch off mold heating and enable mold cooling water during the molding cycle. This is basically the same procedure as above except we also send cooling water through the mold. This allows for rapid cooling of the mold making it suitable for situations where large or quick drops in temperature are desired.

A: Variable mold temperature: Switch off heating without insert cooling

To set up the machine for this mode do the following:

Set up the points in your cycle where you want the mold heaters to switch on and switch off (See Chapter 4.12 in the manual)
Close valves e and g as described in Chapter 4.11
Heat up the mold to the setpoint (you have to fulfill the condition set in your program to switch on the mold heating)

There are many possibilities for defining the point where heating should activate (switch on) and deactivate (switch off). Selecting 'Ejector advance' as Activate point and 'Inject' or 'Holding pressure' as Deactivate point usually works well. But there are many other options, including delays and defining the activate/deactivate points directly in the process sequence.

B: Variable mold temperature: Switch off heating with insert cooling

To set up the machine for this mode do the following:

Set up the points in your cycle where you want mold heaters to switch on and switch off (See Chapter 4.12 in the manual)
Set up the points in your cycle where you want insert cooling valves to open and to close. Valves are set up for each half of the mold separately.
Close the magnetic valves f and h by fulfilling the 'Activate' condition set above (you have to fulfill the condition set above to close the valves)
Open valves e and g. Since magnetic valves f and h should be closed the propellers should NOT be spinning. The cooling water flow can be adjusted with valves b and c. For fastest possible cooling rate the valves should be fully open. If better control is preferred the flow can be reduced.
Heat up the mold to the temperature setpoint (you have to fulfill the condition set above to switch on the mold heating)

The machine should now be ready for your process. Please remember that:

Every time you open the gate (door) at the mold (either the front gate or the back gate) mold heating will switch off and the insert cooling water valves (f + h) will open due to a safety interlock. To switch heating back on and close the cooling water vales, the 'Activate' conditions have to be fulfilled again.
Avoid circulation of mold cooling water while the mold heating is switched on. The machine will give an error stating that the mold heating is defective. Mold heating should always be switched off before opening mold cooling water valves.

Injection

Injection parameters are very important in injection molding optimization. The injection speed will of course determine how long time it takes to fill the cavity. It also has a large influence on the rheology of the polymer melt. Generally, viscosity of polymer melt will decrease with increasing injection speed (an effect known as shear thinning). The injection speed also influences how structures are filled, and to what extent the air inside the cavity has time to escape as polymer is injected. It therefore often takes some optimization to achieve all desired sample properties. Injection speed settings are adjusted in the 'Inject' screen:

The injection velocity can be adjusted in several ways. One can tap the '=' sign to enter a constant velocity or alternatively use the up/down arrows to move the curve up and down. For more complicated injection speed profiles it is also possible to tap the curve to enter the graphical curve editor where the number and position of individual data points can be edited. This makes it possible to make custom injection speed profiles.
The 'Specific injection pressure limit' is an upper safety limit that can be set to protect the mold and/or shim (e.g. if using shims that are soft and thus prone to deformation). Note that this setting does not in any way set the injection pressure. It is only an upper safety limit that will not be exceeded.
The 'Minimum injection time' is usually set around 0-0,05 sec and 'Max. injection time' around 5-8 times the actual injection time. The actual injection time can be seen in the blueish/gray filed just over the max injection time setpoint (in this case it says 0.00 sec because no sample has been produced yet). Actual injection times are usually around 0,2 to 2 seconds so the 'Max. injection time' can typically be set to 1,0 - 15 seconds.

Switch-over Type

The switch-over type is a very important setting! It has a significant impact on the properties of the sample. This setting defines when the injection-phase is done and thus when the machine will continue to the next phase of the injection molding cycle (after pressure / holding pressure). Since this setting defines when we consider injection is done it significantly affects the properties of the produced sample, in particular filling.

Volume-dependent switchover: A certain volume of polymer is injected. Note that the volume entered (3,2 cm³ in this example) is not the volume injected, but the amount of polymer left after injection! Also note that changes in shot volume (the amount of polymer in the heating cylinder before injection) will also change injected volume unless the switch-over point is adjusted accordingly. Volume dependent switchover is particularly well suited for developing new injection molding processes (so you can gradually find the point where the cavity is almost filled and then work from there) or if you already know the exact volume of polymer required.
Time-dependent switchover: After a certain amount of time, injection is considered complete. Not used very often because it offers very little control over both injection volume and pressure.
Injection pressure-dependent switchover: Polymer is injected until a certain pressure is reached. Please note that the pressure in question is measured in the injection unit (heating cylinder). The pressure inside the cavity is not known. Special molds with an integrated pressure transducer is required to determine cavity pressures (very interesting, but unfortunately not available at DTU Nanolab). This switchover type is very easy to use and is not influenced by e.g. changes in shot volume. The selected switchover pressure (in this example 800 bar) is critical for replication. Very low switchover pressures can result in bad replication or even incomplete filling while too high pressure can cause the sample to stick so well to the shim, that it gets very difficult to demold.
External switchover: Injection continues until a signal is received from an external device (e.g. an external pressure transducer). Can not be used, since no external devices are currently available.
Parallel switchover: Injection is defined by a combination of parameters. Has never been used until now.

After (holding) pressure

After injection it is generally necessary to maintain the injection pressure for a certain time. The time required will depend on processing parameters (especially mold and melt temperatures), but generally it is desired to maintain a pressure for long enough time for the gate to have solidified. Too short duration or too low after pressure can cause the polymer to flow backwards (out of the cavity) or that the sample gets an uneven surface (sink marks) because the sample shrinks as the polymer cools down. On the other hand too high after pressure or time can cause high levels of stress in the sample (due to excessive packing of polymer) and at some point increasing after pressure and time will cause the sample to get stuck in the mold, masking it very hard to remove.

Holding pressure profile: Holding pressures can be set as a constant pressure (by pressing the '=' next to the curve and entering a value) or as a pressure that varies with holding time (by pressing the graph areas and editing the points). It is preferable to let the holding pressure gradually taper off as shown here to avoid very abrupt changes in pressure. Just letting go from e.g. 1000 bar to 0 bar causes the screw to fly back violently and should be avoided.
Holding pressure time: The holding time is entered here. The time axis in the graph will update automatically when the value is changed.
Cushion monitoring: The cushion is the amount of polymer left in the barrel (heating cylinder) after injection. A small amount must be left in order to be able to maintain a holding pressure. It is recommended to keep the cushion around the center of the interval (i.e. ~1,5 cm³). If the cushion is too small the shot volume should be increased and if the cushion is too large the shot volume should be decreased accordingly. Please note that the cushion is pressure dependent. Increasing the holding pressure will decrease the cushion (because the higher the holding pressure the higher compression of the polymer).
Shot volume: If the cushion is too small or too large, adjustments of the shot volume can be quickly made here, as described above. All other settings relevant for dosing (plasticizing) are found on the Plasticizing screen.

Cooling time

When the holding time has elapsed the machine will switch to cooling time. The optimal cooling time is very process dependent, but the overall purpose is of course to give the sample enough time to cool sufficiently below the polymer's melting point (or glass transition temperature) so that the sample can be removed from the mold without deforming.

The optimal demolding temperature (temperature of the mold when the mold opens) depends on both the polymer and mold used and the structures being replicated. Generally the lower the demolding temperature the better because this makes the polymer more rigid and in some cases the shrinkage of the polymer as it cools down can also help releasing the sample from the shim. But lower demolding temperatures come with a time penalty. Generally the Luer tool is most tolerant to varying demolding condition because the 12 Luer connectors help in pulling the sample off the shim. The flat disc tool can require a little optimization for reliable demolding while the microscope slide tool can be quite challenging (especially with deep or high aspect ratio structures) because of its geometry. If you are running a variotherm process remember that mold heating and cooling water timing will greatly influence the cooling phase.

Dosing (Plasticizing)

Plasticizing is the process of dosing polymer pellets into the heating cylinder and press it forwards to melt and circulate (knead) the polymer melt inside the heating cylinder. A major part of the heat required to melt the polymer originates from friction energy during this kneading process. Some polymers are very sensitive to heat and thus must be dosed/plasticized gently to avoid decomposition. If samples turn out yellowish or brown it is likely to be caused by too harsh plasticizing settings (or too high heating cylinder temperatures). This problem can thus often be solved by decreasing dosing speed and/or back pressure.

Dosing speed: Dosing speed is adjusted by pressing the left (green) '=' sign and entering a value. Alternatively the up/down arrow can be used or the points of the green graph may be edited directly. It is recommended not to increase the speed above ~0,34 m/s. For sensitive polymers it may be necessary to lower the speed.
Back pressure: The back pressure is adjusted by pressing the right (red) '=' sign and entering a value, using the arrows or editing the red curve points directly. During dosing the only thing that moves the screw backwards is polymer entering the cylinder and thus pushing the screw backwards. A back pressure is used to "press against" this motion. The higher the back pressure the harder 'kneading' of the polymer. Again, some sensitive polymers might require reduced back pressures to avoid thermal damage (decomposition) of the polymer. It is usually sufficient to keep the back pressure at 120 bars or below, but in some cases (e.g. when using pigments) a higher back pressure can give better mixing of pigment and polymer. But be aware of the risks involved regarding decomposition of the polymer.
Shot volume: This is the setpoint for the amount of polymer that is loaded into the heating cylinder. Note that this number in no way represents the amount of polymer injected into the mold cavity (besides the fact that the amount of injected polymer can of course never exceed the shot volume). The volume of polymer actually injected is defined via parameters on the "Switchover" screen mentioned earlier. As mentioned earlier a shot volume resulting in a cushion of about 1,5 cm³ is recommended. Loading much more polymer than required for the shot increases the residence time of the polymer in the heating cylinder. This is undesired since it increases the risk of thermal degradation of the polymer.
Plasticizing delay time: Adjusting this parameter it is possible to insert a delay between end of holding pressure and plasticizing. This can be beneficial if running processes with long cycle times (usually Variotherm-processes) in which case the polymer is in risk of thermal degradation inside the heating cylinder because it will stay there for a long time. In this case it makes sense to enter a delay so that plasticizing finishes briefly before cooling time has elapsed (a delay which is ~5 seconds less than the cooling time would be a good starting point). For processes with constant mold temperature (where cycle times are usually comparatively low) this value can be left at zero to minimize wasted time.
Plasticizing time monitoring: It is recommended to always keep this enabled (it must remain enabled if leaving the machine running unattended) to make sure the machine stops in case of errors. Usually the maximum plasticizing time is set around 4-8 times the actual plasticizing time (which can be seen in the light blue field just above the setpoint). In this particular screenshot it shows 0,00 sec because no samples has been produced yet after the machine was started. A good starting point for 'Maximum plasticizing time' is around 2-5 times the normal plasticizing time. Then there's amble time for normal fluctuations.
Decompression: To avoid problems with stress in the polymer melt it can be a good idea to enable decompression. A setpoint of 1 cm³ works well for most cases. This means that after plasticizing the screw moves a little further backwards to provide extra volume for the polymer melt to relax. This is particularly beneficial if running at high back pressure.

Demolding

Demolding is the process of opening the mold and thereby pulling the sample off the shim. Many of the previously mentioned parameters will influence the demolding process. Generally, the higher the mold temperature, the faster injection speed, the higher switchover/after pressure and the larger the structured area, the more difficult it can be to demold the sample. If the shim thickness is not optimal, polymer may also get in between the shim and the holding plate, making demolding almost impossible. A very critical parameter in easy demolding is designing the structures in a way, that makes them easier to demold (side walls of deep structures should not be completely vertical but have appropriate slip angles to aid in release and roughness of side walls can also severely impact the release properties).

Opening profile: A constant speed can be set using the '=' button or the arrows. Alternatively individual points of the graph can be edited. If samples tend to break during demolding it might help lowering the opening speed. In general the speed should be low towards the end position (fully open position) to avoid violently slamming the mold against the end stop.

Mold stroke position: This value defines how far the mold opens. It should generally be avoided to open the mold all the way to the end position, since this causes the mold to bang violently into the end stop. If this happens (you hear a bang every time the mold reaches the open position) decrease the mold stroke position by a few millimeter (e.g. change the stroke from 235 mm to 232 mm). Remember that changing the opening stroke will also change sample take-off position, so remember to adjust the take off position (under robot settings) accordingly.

Sequence settings: These settings change how the robot is allowed to move around the mold. These settings should generally not be changed (be very careful if you do) to avoid crashing the robot into the mold.

Ejection

Once the mold has opened, the sample must be ejected from the mold to enable the robot to pick up the sample. This is done by pushing out ejector pins from within the mold, which in turn will push out the sample. Once samples are picked up reliably by the robot it is usually not necessary to adjust these settings. Parameters for the ejector pins can be found on the 'Ejector' screen:

Advance profile: These settings control how the ejector pins advance. The advance speed (green curve) and advance force (red curve) can be edited using the '=' or arrow buttons or by editing individual points on the graph. Generally it is not advisable to increase the advance speed above a few mm/s. And importantly: Under no circumstances should the advance force be increased above 8 kN! Otherwise the ejector pins may be damaged (bend or even break)! It is recommended to keep the advance force at 6 kN or below and only increase it to 8 kN if samples cannot be demolded reliably.

Retract profile: These parameters control the ejector pin retraction and is edited in the same way as described for the advance force.

Ejector stroke position: These settings control the positions of the ejector pins in the retracted and advanced position. It is recommended to leave the 'Ejector retracted position' around 25,6 mm. The 'Ejector pin advanced position' can be fine tuned if you have problems with unreliable robot pickup, but 56,0 mm usually works well. If you make changes to the 'Ejector advanced position' make sure to adjust the robot take-off position as well. Otherwise you risk missing the sample or that the sample is pushed hard into the robot arm and thus risk damaging both the robot and your samples. As mentioned in the 'Closing mold' chapter, remember that the the 'Opening stroke' parameter also influences the position at which the robot must pick up the sample.

Ejector shake counter: If samples are difficult to release from the moving part of the mold, this setting makes it possible to move ejector pins in and out several times to help releasing the sample in order for the robot to be able to grab the sample. Usually leaving it at 1 works fine (meaning the ejector pins will move out once and the robot will immediately attempt to pick up the sample), but in some cases it may help to increase the shake counter to 2 or 3.

Ejector shake position: This position defines the position the ejector pins move to during shake procedures.

Mold pause time: This setting is often used when running variable mold temperature ("Variotherm") processes where a delay is required in order to give the mold time to heat back up to the desired mold temperature before the next cycle can be started. The required time will depend on how cold the mold is at this point in the cycle and the mold temperature setpoint. If the mold has cooled down to 40°C and needs to heat up to 140°C it will usually take several minutes to do so (usually 3-4 minutes). The 'Mold pause time' is an easy way of achieving this, since this a delay at the end of the cycle. When running constant mold temperature processes, the mold pause time can be set to zero, since no delay is required (the mold temperature is always at the correct temperature).