Part Roundness | PP

[chrisprocess]

Hi all -

Relatively new/intermediate to processing and injection molding (at least comparatively to some)..

My question is how to improve part roundness. ID of Pipette Tip needs to be in 0.0005" roundness range, or less.
Initial thought was cooling rate, pack pressure, pack time. But so far nothing seems to work. Current part measures in the 0.002" roundness range. Material is a PP with 35 MI.

Any seasoned injection molders have experience with part roundness? Any tips on how to proceed?

Thanks in advance.

[brentb]

Where is it gated?
How is core supported?
Is wallstock super consistent all around part?

KOM

brent

[chrisprocess]

Where is it gated?
How is core supported?
Is wallstock super consistent all around part?

KOM

brent

Hi Brent,

See image below.
Dimension 4 and 5 is where I have a roundness issue. The red dot is the approximate gate location. Part is dual gated - cold runner

175



Mold is bored out to support cores.
If there was 'play' on the cores it would show a shift in our part wall thickness. Wall thickness is approx. 0.0175" on the low side and 0.0182" on the high side. In my experience this is OK but I suppose it could 'warp' during cooling resulting in roundness issues?

Roundness on the steel checks out approx. 0.0002" in these locations.

-Chris

[brentb]

that wall stock % looks nice,but 5 tenths is a toughie!
How is the part gated?
Single cavity?

KOM
brent

[chrisprocess]

that wall stock % looks nice,but 5 tenths is a toughie!
How is the part gated?
Single cavity?

KOM
brent


Yes 0.0005" can be difficult especially with older machine/mold - granted this mold runs on a 90's Arburg. Machine does have eDart system so we're not just turning nobs thankfully. Unfortunately the validation process to move to another machine would be strenuous, given customer requirements. Also if it's moved I had better be darn sure roundness improves!

This mold is 4 cavity cold runner, dual gated. Pictures attached176177

[chrisprocess]

And really any info on types of gating/ runner designs is very helpful. Some of these molds are old and out-dated. I was looking into melt flipper technology, any experience with this sort of thing?

[rickbatey]

How much of the core tip is supported? There may not be enough to stop the core pin from deflecting which I think is your root cause.
Rick.

[brentb]

How much of the core tip is supported? There may not be enough to stop the core pin from deflecting which I think is your root cause.
Rick.

YES!

KOM

brent

[chrisprocess]

How much of the core tip is supported? There may not be enough to stop the core pin from deflecting which I think is your root cause.
Rick.

Thanks Rick.

Core tip is only 0.019" in diameter and supported by a small button. There certainly is some deflection in my mind. Unfortunately there's not much more we can do to support the core given part design.

However the reason I don't believe this is root-cause is because core deflection would affect the concentricity of the ID dimensions, not the roundness. In other words even with the deflection the core would still be round, but the concentricity (of two diameters) would be off. Right ? Though we may be splitting hairs.

Leads me to believe roundness happens in the melt and/or cooling rate. Or worse just properties of this specific material - heat transfer etc. In which case I'm screwed.

I've taken following steps so far:

1) Achieving homogenized melt throughout dosage as much as possible
2) Adding wells to cold-runner to reduce any slugs into melt
3) Any other?


I'll let you know what I find

Thanks,
Chris

[chrisprocess]

Better yet root-cause of part failure may be combination of both roundness and concentricity. (possibly one more than the other) Rick and Brent you may be right after all.

[brentb]

If you measure right out of the mold,and compare to "cool" parts,is there much difference?

KOM

brent

[brutusjax]

What are the steel temps? This seems to be cooling and or gate seal related.

[rickbatey]

So I bet the core pin deflects and the cooling holds the part in the out of tolerance condition. I fear you'll need either more core pin support and maybe even a design that offsets the deflection prior to molding. A core pin that small hasn't got internal cooling and is dependent on cavity cooling.
Rick.

[chrisprocess]

So I bet the core pin deflects and the cooling holds the part in the out of tolerance condition. I fear you'll need either more core pin support and maybe even a design that offsets the deflection prior to molding. A core pin that small hasn't got internal cooling and is dependent on cavity cooling.
Rick.

Certainly some deflection but it's dual gated so in 'theory' it shouldn't deflect. I'd be skeptical to offset the core prior to molding given this particular tip has 0 draft in the critical area. An option could be to open up one of the gates (dual gated) to help influence the offset during fill if deflection is the root cause. But I think you're right about the cooling. Most if not all the cooling is from the cavity. Here's a picture of critical dimension probed down from .197" depth to .297" depth in a spiral. 179
Looks to me like it just is warped. Leads me to believe uniform cooling may be root-cause. And I still think cold slugs are getting into the melt.



What are the steel temps? This seems to be cooling and or gate seal related. Approx. 130 F Not sure how uniform the cooling is though.. To Rick's point cores probably aren't getting good flow. Gate seal seems okay. I tried .2s of pack and 2s of pack - very similar results. Gate seals around .5s


If you measure right out of the mold,and compare to "cool" parts,is there much difference?

KOM

brent

Good question - Would need to study this more. In the past I've seen roundness unchanged from 20 minutes of cooling to 24 hours of cooling.
But I've also seen roundness improve on the same parts months later. At least with Polypropylene

[rickbatey]

Chris, two gates may give you the potential for equal fill but a cold slug could significantly change peak pressure and timing of WHEN the two ends are packed. Have you tried blocking one gate off and shooting parts?
Wait a minute. You have two gates but they have to be cut into each half of the mold, right? Maybe you have a steel temp variation between the mold halves resulting in peak psi, gate freeze, melt front advancement in the cavities!? I think more gates might make this a work but it might take 3-4 of them. Could you use a diaphragm gate over one end of the parts?
Rick

[Joel JS]

Not sure where you are on this.... have you tried a short shot to verify both gates are filling evenly?

Joel

[chrisprocess]

Haven't forgot about this.

-Added some cold wells in runner and replaced heater bands in the manifold. As well as new stripper rings.

Haven't had a chance to run mold yet as priority has been directed elsewhere. (typical in our manufacturing environment)

[JayDub]

ID of Pipette Tip needs to be in 0.0005" roundness range, or less.
Current part measures in the 0.002" roundness range.
Material is a PP .

Any tips on how to proceed?



Just some semi-random observations:

Roundness is tricky. Parts whose roundness is critical, such as gears, generally have more than 2 gates – 3 or 4 is typical. Stresses are always concentrated at the gate, and those stresses will pull the part out of round.

PP is not a good material to hold tight tolerances on – shrinkage is quite high and anisotropic.

Even if you mold it round, it may not stay round: Tg of PP is well below room temperature, so parts will creep on the warehouse shelf. [My experience of molding PP parts with close tolerances involved post-mold annealing twice – and those parts were made to order and sold with a 90 day warranty on dimensions].

As for how to proceed: Does the roundness really need to be .0005”? Just because the designer put a number on the print and got an engineer to sign off on it doesn’t mean that number wasn’t just extricated from a very dark location. My suggestion would be to go back to the customer and ask if they can live with .003” roundness.

[chrisprocess]

Just some semi-random observations:

Roundness is tricky. Parts whose roundness is critical, such as gears, generally have more than 2 gates – 3 or 4 is typical. Stresses are always concentrated at the gate, and those stresses will pull the part out of round.

PP is not a good material to hold tight tolerances on – shrinkage is quite high and anisotropic.

Even if you mold it round, it may not stay round: Tg of PP is well below room temperature, so parts will creep on the warehouse shelf. [My experience of molding PP parts with close tolerances involved post-mold annealing twice – and those parts were made to order and sold with a 90 day warranty on dimensions].

As for how to proceed: Does the roundness really need to be .0005”? Just because the designer put a number on the print and got an engineer to sign off on it doesn’t mean that number wasn’t just extricated from a very dark location. My suggestion would be to go back to the customer and ask if they can live with .003” roundness.

Thanks for the input. From what you and rickbatey said it sounds like additional gates is the best solution for getting parts round.

In our case the main reason these parts need to be in the 0.0005" range is because, historically, they'll fail functionally on the robot if they get much higher. Though we've seen parts pass as high as 0.003" and parts fail as low as 0.0007".. I think it's safe to say the roundness of the part is only as "achievable" as the roundness of the robot tip in which it interfaces. Almost as if when the roundness on pipette tip and robot tip is just right in relation to each other it can fail. In which case it would be difficult to qualify pipette tips with multiple robots.

In my mind it's clearly a design issue and part and/or drawing need to be updated. Unfortunately we tend to inherit old molds from other sites and are left to reconcile these problems. (as you could imagine is a quality nightmare) And getting R&D or product development to investigate this would be a lost cause so in many cases we're left to do it ourselves!

That being said, just yesterday we checked roundness of exact same parts I made a month ago and they decreased from 0.002" roundness to around 0.0006"...
Which leads me to believe 1) Material continues to shrink even after a month 2) Fixturing ( or fixturing process) with the CMM needs to be investigated. All functional testing PASS

[chrisprocess]

UPDATE - 5/22/17

We installed new stripper rings, new manifold heats, and added a cold-well to the runner.
We also had to slightly modify mold opening profile and timing to help reduce gate freeze from getting in the melt.

We measured parts approx. 0.0011" roundness on the CMM but was able to go back and adjust the part fixture with a minor "touch" and bring same parts down to 0.0006" roundness .. Probably room for improvement on the metrology side.

Other than that parts are working functionally and dimensionally.