A brief primer on how the injection molding process affects flight characteristics

[sidewinder22]

OG link and DISCussion: https://www.reddit.com/r/discgolf/comments/8jc7zf/i_was_in_the_pool_a_brief_primer_on_how_the/

I design plastic injections molds for a living and thought I would share some of my acquired knowledge about the subject with my fellow disc golfers. This post will mainly focus on how the injection molding process affects the flight of a disc and why there are inconsistencies between runs and plastic types. I see a lot of misinformation on this sub (which is understandable if you're not in the industry) and wanted give everyone some knowledge that they can drop on new users in the future. I used a simulation software (Autodesk MoldFlow) to help visualize the examples. But first, a few disclaimers:

I have no idea what the propriety resin blends are for any disc company, so I picked a generic pure polyurethane to emulate Innova's Champion Edition plastic and a generic polypropylene to emulate baseline plastic. The magnitude of the results will not be comparable to molded discs, but the trends between variable changes will hold true.
I design molds for a living, meaning I spend 95% of my time behind a desk using Solidworks. Setting up and running a tool in the press happens in a completely different building, so some of the processing inputs for the simulation software are done automatically by the software since I would be completely guessing if I tried to enter them in myself. Again, this will affect the magnitude of the results but not any trends.
Now, let's get down to business.

Warpage

I see the term "differential cooling" thrown around a lot as the main cause for why the same mold of a disc flies differently in different plastics. This is a misnomer and we'll get to why in a bit. Warpage is the term used to describe how an injection molded part deviates from the ideal model. Every molded part warps, which is why part designers allow for a certain amount of tolerance in the final part dimensions. There are three main causes of warpage: Differential cooling, differential shrinkage, and fiber orientation. Fiber orientation doesn't apply to discs since (as far as I know), fiber-y fillers such as glass strands and carbon fiber aren't used in any production discs so we'll only be discussing the first two. (In case you were interested, plastic shrinks less in the direction the fibers are oriented since the the fibers prevent the plastic from "pulling back" in that direction.)

Here's what the total warpage looks like for our premium and baseline discs:
https://i.redd.it/hobblpdsntx01.jpg
Top is premium, bottom is baseline. The translucent outline represents the ideal shape.
Woah! The parting line for the baseline disc is way lower, confirming what many of us have observed: baseline plastic discs are more understable than their premium counterparts. Let's dissect what causes this.

Differential cooling

I see this term thrown around to describe how discs in different plastic types and/or production runs warp differently as they cool outside of the press. What this term actually refers to is how the water lines that cool the tool affect the final warpage of the part. The steel cavities and the mold base need cooled liquid to run through the tool during the molding process or else the parts will always stay at their melting temperature and never solidify. Complex or large parts usually have trouble getting efficient heat transfer in all areas due to limitations in manufacturing water lines around inserts, slides, bolts, etc. These differences in the cooling of a part cause the part to warp differently than if cooling is assumed to be 100% efficient.

Let's take a look at how differential cooling affects our two molds (the top is premium, bottom is baseline):
https://i.redd.it/qi71uzgzntx01.jpg
Top is premium, bottom is baseline
And the answer is… not much. That's because both the tool design and geometry of a disc is very simple, which allows for efficient heat transfer. We're able to put the cooling lines close to the cavity impression without any obstructions. We'll touch a bit more on cooling later, but first let's look at what really drives warpage.

Differential Shrinkage

Nearly every resin will expand when heated and shrink as it cools. When a mold is designed, we "add shrink" to the part by multiplying the volume of the part in CAD by the resin's known shrink rate and using that larger part to design the tool. The shrink rate is a material property that is the measure of the percentage a material shrinks per every inch/mm of its length. Each type of resin shrinks differently and blending resins or adding additives will change the shrink rate. The shrink rate is useful guideline for telling us approximately how much the overall dimensions of a finished part should turn out, but the final volumetric shrinkage may differ in different zones of the part. Processing and part geometry play a key role in how evenly a part shrinks. For instance, a part with uneven wall thickness, like a disc, will shrink more in the thicker sections since it takes longer for the thick plastic to cool to its transition temperature.

Here's how shrinkage affects the warpage of our discs:
https://i.redd.it/01apnhq8otx01.jpg
Top is premium, bottom is baseline. The translucent outline represents the ideal shape.
Wow. I think we found the driving factor for warpage. But why is this? It all has to do with the hoop stress caused by the resin shrinking to its nominal size as it cools. When the part cools, it wants to shrink in towards the central axis. This is how a disc without a flight plate will deform:
https://i.redd.it/qmhcit6eotx01.jpg
https://i.redd.it/rdr8ggscotx01.jpg

There's no change in the parting line height. For a ring-shaped object, this deformation will be even. The flight plate of a disc, however, prevents even deformation. The bottom rim of the disc is free to shrink in towards the center, but the top is constrained by the plate which forces the middle of the plate to pop up and the parting line to drop down.
As shrink rate increases, so does the hoop stress and consequently the force that the flight plate exerts on the top rim of the disc, causing a greater shift in parting line movement.

Other factors

Even though differential shrink is the main cause for why discs of different plastic types fly differently, there are many other factors that can cause different runs of discs in the same plastic to fly differently. Note that some of these variables may have an inconsequential effect on the final flight characteristics on their own, but may have a measurable effect when combined. Here's a rapid-fire list:

Additives:

-Different colorants/ratios of colorants will sometimes have an affect on the shrink rate. Yes, I am confirming that discs of different color runs may fly differently, even though manufacturers will deny it (though it's rare). Metal flakes reduce shrink, which is why metal flake discs tend to be more overstable.

Processing:

-Raw resin pellets that are used in the injection press absorb moisture and require a trip through a dryer before being being processed. If there's deviance in the time/temperature of the dryer or relative humidity in the plant, there may be slight differences between shrink rates between runs.

-Every brand of injection press and the barrel used to melt and inject the plastic will be slightly different. If the resin is heated and sheared differently in the barrel screw, there will be deviances in the final residual stress of the disc and therefore the warpage will differ.

-The processing parameters (melt temperature, injection time, pack time, pack pressure, cooling time, etc.) should be constant for every run of the disc. However, errors in the accuracy of the injection machine's sensors will affect warp. For instance, the melt temperature of the resin has a range of acceptability, so the barrel temperature sensors on many older presses aren't super precise in their measurements since it will have a negligible effect on the final warp of a part (but it's still a variable we should consider).

-When a mold is first fired up from idle, it needs to run for a few cycles before the mold reaches its transient (steady-state) temperature. If the mold has to be paused for a while in the middle of a run, it will affect the next few discs until the transient temp is reached again.

-The time a part spends cooling in the mold will affect the final warp (although it is negligible for discs).
Manufacturers want to eject a disc as soon as the material solidifies, since time is money and a couple wasted seconds per cycle add up to days over the life of the tool. Plastic will continue to shrink for up to 24 hours after it leaves the press, so how it is handled during that time can affect the final shape of the disc.

Top is volumetric shrink at ejection, bottom is the final shrink after full solidification
https://i.redd.it/4j03pfezotx01.jpg

-How discs are stacked, stored and shipped can affect the final shape.

Intentional design changes:

PDGA Technical Standards Section I(E) states "Retesting is only required if a mold includes the addition or removal of a new structural feature such as a bead, or results in a measurement that may violate any of the technical standards." This means that the dome of the flight plate can be controlled by tweaking the transition from the rim to flight plate or the dome radius of the flight plate itself in the cavity steel as long as the wing profile remains unchanged.

Quality control:

Mold cavities need to be capable of venting all of the air contained in the cavity while the plastic is being injected, or else the cavity won't fill completely. In our case, the last portion to fill happens to be the parting line, so that's where the vents reside. Vent gaps are only .0003-.001" deep to allow air through but not plastic, but they will gradually get larger over time as the steel wears down, resulting in flash (extra plastic) at the parting line. How each company chooses to deal with it determines how consistent their runs are. How often companies refurbish/replace their cavities will also determine the consistency between new and old runs.

Any manufactured part that needs to meet technical requirements must be inspected frequently to ensure it's within the set tolerance. Tighter tolerances mean more consistency, but it also means more failed parts. Companies need to find the balance that will provide them with the most efficiency and least amount of acceptable waste.

*********

Injection molding is a complicated process with a lot of finicky variables, so temper your expectations a bit when complaining about a disc. If you have any questions, I will answer them to the best of my ability.

 

[jakebake91]

As a toolmaker myself...this was heaven sent!! Thanks for sharing!!!

 

[swhite]

Good reading. Very informative.:thmbup:

 

[RFrance]

☑ Learn something new today.

 

[Countchunkula]

Very cool, thanks for the cross post, Sidewinder22.

The company that I work for does plastic injection molding for some components. It's a pretty cool process to watch. A lot of what the OP said jibes with what I've been told by manufacturing and engineering regarding the challenges of molding parts that consistently are within acceptable tolerances. They usually don't go into as much detail as the OP did though (or when they do most of it goes right over my head).

One point that the OP brought up that really stood out to me was the cooling aspect. I've often heard (read on dg message boards) that plh inconsistencies are due to lousy climate control in the factory, which affects the cooling process. If I understand the OP right, the vast majority of the cooling process happens while the part/disc is still in the mold and is achieved by running cold water through channels in the mold. From reading disc golf message boards, I was under the impression that discs came out of the mold hot and cooled at room temperature. At our factory, I never saw parts set out to cool after they were molded. The operators would stage bins under the molding machines and exchange them every x number of cycles. As far as I could tell, once a bin was full, it was emptied into a box that was then taped shut and labeled with part #, lot #, and quantity. With this process, if the parts came out of the mold hot, there would be major inconsistencies in cooling time. We primarily mold gears, so I have to believe our tolerances are way tighter than those that a DG manufacturer would be dealing with. If the gears aren't within tolerances, they won't properly mesh. If the majority of the cooling didn't happen in the mold, our failure rate would be through the roof. I think we can put to bed the image of stacks of hot, freshly molded discs cooling at room temperature.

 

[Grippenripp]

This was pretty friggin neat! Thanks.

 

[DiscJunkie]

Thanks for the post, SW22!!

 

[ShadowHunter]

I think we found the driving factor for warpage. But why is this? It all has to do with the hoop stress caused by the resin shrinking to its nominal size as it cools. When the part cools, it wants to shrink in towards the central axis. This is how a disc without a flight plate will deform:
https://i.redd.it/qmhcit6eotx01.jpg
https://i.redd.it/rdr8ggscotx01.jpg

There's no change in the parting line height. For a ring-shaped object, this deformation will be even. The flight plate of a disc, however, prevents even deformation. The bottom rim of the disc is free to shrink in towards the center, but the top is constrained by the plate which forces the middle of the plate to pop up and the parting line to drop down.
As shrink rate increases, so does the hoop stress and consequently the force that the flight plate exerts on the top rim of the disc, causing a greater shift in parting line movement.

Does this suggest that over mold discs are less susceptible to variations in parting line height?

 

[AdenoidHynkel]

One point that the OP brought up that really stood out to me was the cooling aspect. I've often heard (read on dg message boards) that plh inconsistencies are due to lousy climate control in the factory, which affects the cooling process. If I understand the OP right, the vast majority of the cooling process happens while the part/disc is still in the mold and is achieved by running cold water through channels in the mold. From reading disc golf message boards, I was under the impression that discs came out of the mold hot and cooled at room temperature.

OP here. You're right that most of the cooling occurs in the mold. Climate does have an effect on the final shape of the disc since it will continue to shrink for a few hours after it leaves the tool, but it's not as major of an effect as some of the other things I listed. I'd reckon that how they store the discs immediately after they cut the sprues off has a larger effect. If they stack them, the ones on the bottom are going to have a different shape than the ones on top.

 

[Muddyboots]

Fantastic posting. Thanks to all for contributing.

 

[AdenoidHynkel]

Does this suggest that over mold discs are less susceptible to variations in parting line height?

Absolutely. As long as the plates finish their shrink before being overmolded, they will not shift the parting line as much.

 

[BlueMonk]

I love getting the real-deal info on how things work. This is really interesting.

One question about metal flake: It's not really metal, right? I think I saw someone get an answer directly from a manufacturer confirming that.

 

[eman6854]

Absolutely. As long as the plates finish their shrink before being overmolded, they will not shift the parting line as much.

Do you think discs would be much more consistent if they tried putting some glass or carbon fiber in them? I'm sure it would jack up prices too though?

 

[GMcAtee]

Very cool, thanks for the cross post, Sidewinder22.

The company that I work for does plastic injection molding for some components. It's a pretty cool process to watch. A lot of what the OP said jibes with what I've been told by manufacturing and engineering regarding the challenges of molding parts that consistently are within acceptable tolerances. They usually don't go into as much detail as the OP did though (or when they do most of it goes right over my head).

One point that the OP brought up that really stood out to me was the cooling aspect. I've often heard (read on dg message boards) that plh inconsistencies are due to lousy climate control in the factory, which affects the cooling process. If I understand the OP right, the vast majority of the cooling process happens while the part/disc is still in the mold and is achieved by running cold water through channels in the mold. From reading disc golf message boards, I was under the impression that discs came out of the mold hot and cooled at room temperature. At our factory, I never saw parts set out to cool after they were molded. The operators would stage bins under the molding machines and exchange them every x number of cycles. As far as I could tell, once a bin was full, it was emptied into a box that was then taped shut and labeled with part #, lot #, and quantity. With this process, if the parts came out of the mold hot, there would be major inconsistencies in cooling time. We primarily mold gears, so I have to believe our tolerances are way tighter than those that a DG manufacturer would be dealing with. If the gears aren't within tolerances, they won't properly mesh. If the majority of the cooling didn't happen in the mold, our failure rate would be through the roof. I think we can put to bed the image of stacks of hot, freshly molded discs cooling at room temperature.

I'm an engineer for an automotive supplier. How you cool the part often depends on the resin and how thin or how large the part is. Warp is a big issue. Some parts go straight to a tote after the molding process. The travel of the conveyor belt to the operator is the only cooling time needed for those. Other parts need longer to cool and have processes designed to allow that to happen.

 

[armiller]

Maybe this isn't the place, but here goes.

My question relates to mold size and item weight. When companies make a mold (i.e. disc) a certain weight, I assume they control density of the injected material (cuz aren't the Star, Champ, Blizzard, Pro, and DX often from the same dye or cast?). So if I want a max weight Firebird, or Nuke, or Aviar, I'm designing that before I inject the plastic, right?

Now my question. If that disc was designed to be 175 grams but weighs 172g or 177g on a scale, then what happened? Was some of the mold underfilled? Or was the density just off?

And if I'm looking for discs that fly a certain way, doesn't the intended weight matter as much or more when compared to the actual disc weight? Would a "heavier than intended" disc fly differently from "just as intended" from "lighter than intended?" Or is that just random with no bearing on PLH or final performance?

I've been working a night shift. You can tell I've had too much time awake at night thinking about such things.

 

[Halcón]

Couldn't help read the OP and think of Three Putt's stories about Dave McCormack and this video.



Something tells me there's a lot of eye balling when Dave mixes those pellets.

 

[wims]

Maybe this isn't the place, but here goes.

My question relates to mold size and item weight. When companies make a mold (i.e. disc) a certain weight, I assume they control density of the injected material (cuz aren't the Star, Champ, Blizzard, Pro, and DX often from the same dye or cast?). So if I want a max weight Firebird, or Nuke, or Aviar, I'm designing that before I inject the plastic, right?

Now my question. If that disc was designed to be 175 grams but weighs 172g or 177g on a scale, then what happened? Was some of the mold underfilled? Or was the density just off?

And if I'm looking for discs that fly a certain way, doesn't the intended weight matter as much or more when compared to the actual disc weight? Would a "heavier than intended" disc fly differently from "just as intended" from "lighter than intended?" Or is that just random with no bearing on PLH or final performance?

I've been working a night shift. You can tell I've had too much time awake at night thinking about such things.

I believe they use a weighting agent to make heavy discs. They add a material that's heavier than the polymers that make up the base weight of the disc itself

 

[AdenoidHynkel]

One question about metal flake: It's not really metal, right? I think I saw someone get an answer directly from a manufacturer confirming that.

I haven't been exposed to them in person, so don't take this as gospel, but they are most likely large pieces of glitter. Actual flakes of metal would quickly wear out the press's barrel, screw, and nozzle as well as the mold cavity.

Do you think discs would be much more consistent if they tried putting some glass or carbon fiber in them? I'm sure it would jack up prices too though?

Yes, but it would also make them really stiff. They wouldn't pass PDGA flexibility testing.

Maybe this isn't the place, but here goes.

My question relates to mold size and item weight. When companies make a mold (i.e. disc) a certain weight, I assume they control density of the injected material (cuz aren't the Star, Champ, Blizzard, Pro, and DX often from the same dye or cast?). So if I want a max weight Firebird, or Nuke, or Aviar, I'm designing that before I inject the plastic, right?

Now my question. If that disc was designed to be 175 grams but weighs 172g or 177g on a scale, then what happened? Was some of the mold underfilled? Or was the density just off?

And if I'm looking for discs that fly a certain way, doesn't the intended weight matter as much or more when compared to the actual disc weight? Would a "heavier than intended" disc fly differently from "just as intended" from "lighter than intended?" Or is that just random with no bearing on PLH or final performance?

I've been working a night shift. You can tell I've had too much time awake at night thinking about such things.

The weight of the disc isn't controlled by filling parameters, but rather using additives of differing densities (from what I've been told by multiple sources). So the volume of the disc is going to be the same for any given weight. There might be slightl variances in weight from flukes in the filling/packing process or resin. But the whole appeal of injection molding is it's repeatability.

 

[Nubes]

Just basic common sense all disc golfers should know.....


lol I'm kidding. That was cool to read. I'm a closet nerd and love learning how and why things do what they do. Seems like the breakthrough could happen with completely redesigning the machines that make the disc? DuPont is inventing a new chemical every other day so that makes you wonder what plastics and what their tolerances will be like in the future? Carbon Fiber discs lined with lead tape just to get them to weight? Might cost a few hundred a piece but it would be neat to see one. I think someone will reinvent the wheel at some point and discs and different tolerances will be next to nothing but then again, who knows? 50 years from now this may still be the process but I doubt it.

That was a cool read. I'd love to tour a plant and witness the entire process. That would be cool to see

 

[Nubes]

Couldn't help read the OP and think of Three Putt's stories about Dave McCormack and this video.



Something tells me there's a lot of eye balling when Dave mixes those pellets.

Dont get me wrong. I love Gateway and have to have several Wizards in my bag. But its almost like he whittles his disc by hand with a dull pocket knife from a sold piece of plastic?? LOL!

Ive never seen discs with so many inconsistencies. I usually have to "clean" them up a little after I buy one. His inconsistencies is what makes them great too. Sometimes you find one that "one"