"Just make the product."
I'm sure everyone has heard this saying? But things are not so simple. Designing and developing a quality product, mold, injection molding process requires a deep understanding of plastics and injection molding machine variables.
While there are many plastic molding strategies, following three methods developed by RJG can be effectively adapted to different scenarios - DECOUPLED MOLDING® I, II, III. It is very important to review and evaluate each of above methods to ensure that you are choosing right one for your product.
Here we will use a product as an example to evaluate part design, material selection, mold design, and actual process data for each molding method to determine impact of using DECOUPLED MOLDING® I, II, III on product results, and to determine , which method is most effective for this product.

For purposes of this article, we use ASTM tensile test bar D638 V, shown in Figure 1 below.

 

Figure 1: ASTM Tensile Test Bar
Tensile tests will yield a lot of data, which can be found in material technical properties table. In next section, we will f ℃us on a few of these values.

There are thousands of materials on the market, but in this experiment we will follow K.I.S. (keep it simple) principle. In addition to this, to ensure that our trials cover most of product applications in the industry, our selection of materials will cover both amorphous and semi-crystalline materials. We used Toyolac 100 (ABS) and ExxonAX03BE3 (PP) as test materials.
Table 1: ABS vs PP

Heat distortion temperature (HDT) refers to temperature at which a material deforms under a certain load. Mold designers and simulators should use this parameter when designing mold cooling systems. Target this value when injection molding to ensure product is rigid and stiff enough to avoid white stress marks or ejection marks, and (if you're very "lucky") thimble will pierce product because material is too soft . When plastic releases heat during curing, molecules want to shrink back to their relaxed state, which is often referred to as shrinkage.

Mold used for this test was an 8-cavity H-runner, using a lapped side gate, manufactured by ExtremeTool & Engineering (shown in Figure 2). Mold is equipped with RJG multi-channel button-type pressure sensors and embedded temperature sensors, so that we can see inside of product cavity and related data.

 

Figure 2: Movable side of test die
Cavity length is 63.00 mm, width is 9.37 mm, and thickness is 3.38 mm.

DECOUPLEDMOLDING® is a forming process that separates filling, feeding and packing. Each stage has its own purpose, which we'll get to in a moment.
Let's first recall that injection molding machine can control injection speed or pressure of material, which will act as a control variable. Choice of control variables depends on how we separate various molding stages, and which other variables we let to "drive proverbial car."
Decoupled I is essentially a race to get material to the end of cavity as fast as possible. It is often used for thin-wall forming, usually below 1.5 mm in thickness. This is because thin-walled products freeze very fast, and in order to fill cavity, we must fill as fast as possible to prevent under-injection, which is accompanied by high injection pressure during filling. Here, filling / packing is controlled by injection speed, as a reduced material injection speed can cause flow front to freeze, resulting in under-injection of product.
When we use standard Decoupled II, fill is controlled by speed and packing and packing are controlled by pressure.
For critical components, Decoupled III is usually chosen. In this process, a fixed and fast vel ℃ity (commonly referred to as V1) is used for filling stage, a slower and controlled vel ℃ity (commonly referred to as V2) is used in packing stage, and a fixed pressure setpoint is used in packing stage.
To ensure consistency across all process strategies in this test, we maintained end-of-fill pressure in product cavity at 200 bar.
Below (Table 2) are actual temperatures established for stable Decoupled II process.
Table 2: Molding process Temperatures

To ensure product is below HDT, we use a thermal imager to capture temperature of all cavities. Temperature control is critical to maintaining long-term production quality of products.

 

Figure 3: ABS material, product temperature during ejection

To ensure continuity between all data, we will measure a product with 8 cavities because there is a difference between cavities - this difference is due to shear imbalance. Cavity with the largest difference is located in lower left corner of Figure 4.

 

Figure 4: ABS material, Decoupled II filled sample only
The overall length (OAL) of cavity is 63.00 mm, so all length dimension data will be based on this value.

 

Data above show that product shrinks the most when Decoupled I process is used. This is because product walls freeze so quickly that shrinkage cannot be compensated for by typical feeding. According to material forming guidelines, for ABS material, actual shrinkage is outside range provided by material supplier by 0.0024 mm/mm. In contrast, actual shrinkage of PP material is around center value of range provided. Variation of OAL is 0.038 mm.
Now let's review how material performed when using Decoupled II process. For ABS material, product shrinkage is smaller, but still exceeds range provided by material supplier by 0.0004 mm/mm. Same shrinkage trend was observed in PP, but now shrinkage is 0.0017 higher than expected minimum shrinkage. For both materials, OAL measurements ranged from 0.038 mm (DI) to 0.025 mm (DII). Compared to products formed using Decoupled I process, average length of ABS increased by 0.127 mm, while average length of PP increased by 0.362 mm.
Go ahead and explore results of Decoupled III process. For these two resins, shrinkage under this process is very similar, but OAL range is reduced from 0.025 mm to 0.012-0.013 mm compared to products molded using Decoupled II process. If we examine product weight changes of PP and ABS again, we will find that their trends are consistent. With this molding process, product variation is minimal for two main reasons. First, both filling and feeding stages are speed-controlled, allowing injection molding machine to use required pressure to maintain set injection speed. This allows 2 of 3 molding stages - filling and packing - to be automatically compensated for material changes by injection molding machine. Second reason is that control of V-P switching is controlled by cavity pressure read by pressure sensor in cavity rather than screw position on injection molding machine.

 

For DecoupledI process, this trend is essentially a fluctuation of amorphous or semi-crystalline material itself with a weight change of 0.002 grams. One thing we have to consider here is that this product may never use this molding process because it is not a thin wall product.
In Decoupled II, product weight has increased for both materials. ABS increased by 0.07 g and PP increased by 0.13 g. Amount of material added is shown in Figure 5. These changes occur because a constant pressure is used in packing/packing phase to pack more polymer chains into cavity to compensate for natural shrinkage that occurs during cooling.

 

Figure 5: Left ABS, Right PP - plastic pellets of varying weight
Switching from Decoupled II to Decoupled III process, weight change is 0.0008 g for ABS and 0.0002 g for PP. Product weight variation range was reduced for both materials.

In short, DecoupledI process is best for thin-wall molded products, DecoupledIII is best for high-precision molded products, and Decoupled II is industry standard for most injection molded products.
Trying to verify a product with a tolerance of ±0.05 mm and a Cpk of 1.67 is almost impossible because OAL changes are pretty much using the total tolerance band. Proven, true Decoupled II process will work for most projects because it produces parts with an OAL range of 0.025 mm and a tolerance of ±0.05 mm, which will allow us to verify that Cpk is 1.33 instead of 1.67. OAL variation of DecoupledIII is 0.013 mm, allowing a tolerance as tight as ±0.025 and a Cpk of 1.33.
Remember, this is a very simple product and has only one gate. When product design is more complex, it is even more critical to choose correct molding process, so that mold can be designed according to appropriate shrinkage rate. Of course it's not easy, and if it were easy, everyone could do it. A team of product designers, mold designers, and process engineers should first sit down for a thorough and detailed discussion so the project can be on time, within budget, and a first-time success.
Nothing is impossible, we just need to do our due diligence as engineers.