Design options with plastics

[jmw]

At the risk of being shot down for not googling properly, I have a query way outside my field.
I am trying to discover the penalties of poor product design.

I have a fairly simple component that is to be moulded and the final component should meet very precise dimensional properties and should not distort.
But when designing the part I have two options.
For simplicity suppose we have a hollow cylinder where the target internal and external diameters must be met with very precise tolerances. The cylinder also has slot through the wall extending along the entire length.
It is a mass produced item.

The question is to know if there are any penalties if I make the part with a constant diameters along its length or if I should introduce a slight taper to both the internal and external dimensions.
Is there a cost penalty?
Does it affect the choice of plastics and/or fillers?
Does it affect any aspect of post moulding management?

JMW

http://www.ViscoAnalyser.com

 

[jmw]

Let me add this which I found on a web site which is pretty much my understanding of what is best but not what the consequences are from not observing this recommendation:

Design Considerations

Part design should include draft features (angled surfaces) to facilitate removal from the mold. Depending surface length draft angles down to half a degree are reasonable. Typical draft angles should be about 1 to 2 degrees for part surfaces not exceeding 5 inches.
Dimensional tolerance specification will govern the part cost and manufacturability. If you have a small region of the part that needs higher tolerances, say the location of a critical feature used for alignment. DO NOT specify tight tolerance, instead design and plan for post molding processes such as machining using "assembly intent" fixturing.

This nicely reflects the issues I have.
The whole purpose of the shift to plastics is to eliminate any form of post moulding machining.
Tight tolerances are essential.
The part must be capable of being simply assembled to another part within which it must move with minimal working clearances. The other part will also be moulded without any final machining and it too has internal and external surfaces that are critical.

JMW

http://www.ViscoAnalyser.com

 

[MacGyverS2000]

I'm no expert, but it seems to me that whenever the terms "precise tolerance" and "mass produced" are spoken in the same sentence, it should be followed by the term "arm and a leg".

While it's possible one such machine exists, I cannot image a tight-tolerance mold of a long tube without some amount of draft. I can imagine an expanding mold, but that removes either "high-tolerance" or "mass-produced with the same mold" as options.

I would think post-process machining is the best option here, like drilling out the tube, or in the case of the slit running the length, a properly shaped router bit.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[KENAT]

You don't say what you mean by 'tight tolerance' but there are places that do molding for medical applications etc. that claim some pretty tight tolerances.

If you're outside of the typical tolerance range then you probably need to find a 'partner' to develop it with. Generally I'd expect draft to be required, though depending on process there is sometimes some limited leeway.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MikeHalloran]

A slot through the wall for the entire length removes much of the rigidity from the part, so ID and OD become theoretical concepts, not so much measurable features.

E.g., the OD is meaningful only while the ID is supported by a precise mandrel, and conversely.

The slot also makes it possible to wrap the part from sheet, which will have a more uniform thickness over its unwrapped area than would a molded cylinder.

As for draft, and other issues, the best thing you can do is make a nice drawing of what you actually want, and spend a couple of hours marking up a print with a good molder. Bring a lot of red pencils/markers.




Mike Halloran
Pembroke Pines, FL, USA

 

[jmw]

The real question relates to the consequences of including or not including a draft angle.

How significantly does this affect the quality of the final component or the tolerances achievable against target values? Costs?
Choice of and cost of materials and fillers?

Can we make this a comparative response?

This is a real world application.

The original design was in the late 19th century using cast and machined brass/ bronze for component 1 and moulded and machined silvonite/ebonite (mineral filled hard rubbers) for component 2 (the primary component here) with a lot of final craft skills to fit them together.
In the 1960's the switch to plastics enabled a major manufacturing change based on no machining and no craft skills. It meant simple assembly using unskilled labour. It also meant they were able to do away with virtually all inspection and use the final test and some innovative methods to fine tune component design.

The trouble is I suspect that it was manufacturing who wanted to revolutionise the product using plastics but could only go so far battling against the product design engineers who probably were by now the guardians of the sacred form (i.e. the product design was "mature" but matured a generation or two before they were born).

Some of the original 19th Century design compromises necessitated by the available materials and manufacturing methods are still evident in the 21st century - which plastics would have resolved but didn't, so hence I am assuming the product design engineers fought to protect the original design and manufacturing did what the could (it was enough to sweep the field of most of their competitors).
But it is also why I suspect there were no drafting angles allowed which would compromise the presumed purity/perfection of the original design.

Unless you eliminate all machining and craft skills, plastic only delivers some possible material cost advantages.... virtually nil.
So the lack of ability to hit some tight tolerances in some aspects of the design would explain why they have changed about the only thing the design engineers would allow - the size.

Dimensions?
think of a cylinder about 5cm across, height 2-2.5 cms with a wall thickness of around 3mm fitting inside another previously machined component.

A competing technology has plenty of opportunity for drafting angles though it's design is also subject to the same problems of design purity (it is a late 19th century design) but is less good in some performance aspects and better in others.

But its this use of the drafting angle that I need to get my head around.


I'm trying to assess if this would be cheaper and deliver better tolerances in the finished component.

JMW

http://www.ViscoAnalyser.com

 

[MikeHalloran]

Absent many unrevealed details that may contraindicate...

I forgot to mention that the axial slit, if molded in, coincidentally reduces the circumferential stiffness enough that it should be possible to extract the part from the mold despite zero draft.

In fact, in a slotted tube type part, it should be possible to mold threads or annular features of limited depth in either ID or OD (not both) with zero draft on the major portions of either. ... because the part can be temporarily deformed while still hot, to get it out of the mold, and cooled on/in a fixture to get back its design shape.



Mike Halloran
Pembroke Pines, FL, USA

 

[MacGyverS2000]

The slot also makes it possible to wrap the part from sheet

Duuuhhhhh... forest for the trees, Dan, forest for the trees


Dan - Owner

http://www.Hi-TecDesigns.com

 

[jmw]

Yes, sorry. Absent many details. The design is a bit more complicated in that there is a web across the centre of the tube which also has a chunk missing where the slot is but it stops you making from sheet or deforming to remove.
OK, let's see if I can draw up the part with no drafting angles.

JMW

http://www.ViscoAnalyser.com

 

[jmw]

OK, here is a quick sketch:

JMW

http://www.ViscoAnalyser.com

 

[jmw]

And a section and plan view.

JMW

http://www.ViscoAnalyser.com

 

[MikeHalloran]

Don't put any draft on the drawing unless the molder demands it.
It shouldn't be necessary. ... depending on the resin.

Do specify where ejector pin marks are not acceptable.

It may distort from perfectly round as it cools.
... depending a lot on the resin.



Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

The second drawing helped a lot.

As the slot is full depth it reduces the strength of the wall of the cylinder, and in the absence of restraint from a cavity, the cores should be easy to extract as the part springs a bit. Kinda like putting piston rings on an engine. The centre diaphragm complicates it a bit but in my opinion it is still possible from that point of view.

Could you mould it with the axis of the cylinder across the parting line instead of along the line of draw and with hydraulic side cores to extract the cores after the mould is just cracked open.

I believe you should be able to extract the cores with no draft in that design if there is no constraint from the outside. The equivalent to the piston ring gap in the design should make it flexible enough from that perspective.

A lot will also depend on exact materials used and your definition of precision and on how much time you are prepared to invest into fine tuning the process to maintain precision.

Several possible not yet mentioned hick ups I can see are:-

1) There most likely will be a sink in the outer surface where the diaphragm meets the cylinder.

2) There will be a tendency to pull into a reversed draft shape due to higher shrinkage in the area where the diaphragm meets the cylinder inside wall. Details of part shape and mould cooling will be critical.

It looks like part of a water meter. If so, continuous exposure to water will influence material choice.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[jmw]

Perceptive Pat.
It is part of a rotary piston meter.
The competing component is the nutating disc.
This is what I am attempting to establish, if one meter is more expensive/difficult to manufacture than the other.
Here is a sketch of the disc. The edge of the disc is theoretically part of the surface of a sphere but possibly, for ease of moulding is whatever profile suits best.

So what I am trying to ascertain is to what extent the different geometry of these components affects the difficulty of producing finished moulded components with no machining and as good a tolerance and finished dimensions as possible.
And is there a cost benefit?
The disc has pretty well no restrictions on drafts. Even the notch, which appears here as flat faced is not flat faced but bevelled....


JMW

http://www.ViscoAnalyser.com

 

[jmw]

OK, here is a modified view of the disc to show the slot.

JMW

http://www.ViscoAnalyser.com

 

[Pud]

Looks all fairly trivial* to me...unless I've missed something!

My choice of material: nice, very low shrinkage, great in cold water, mPPO. (e.g. Noryl.)

For close fitting parts it is usual to leave metal on the tool, run it and adjust to get desired fit.

I see it as being an open and shut tool, and for ~25mm draw length, draft can be zero.

*Trivial as in low complexity.

Cheers

Harry

http://www.tynevalleyplastics.co.uk

Why be happy when you can be normal?

 

[jmw]

Thanks Pud,
So as I understand you, both the piston and disc are equally easy to make and I don't need a draft angle on the piston despite the straight parallel internal and external geometry (the disc has no need for special draft angles)?
Equal costs?
Equal manufacturing complexity?
Equal control of finished product tolerances?
No special materials?
(I must check what materials are actually used.)
We are talking about hundreds of thousands annually and ultimately every home will have one.

JMW

http://www.ViscoAnalyser.com

 

[Pud]

As depicted, nothing untoward - the ball section is not a good design for plastics, but not too difficult.

imho, the whole assembly could probably usefully be redesigned for modern materials.

I've no idea regarding costs - (I have really, but not for here!), moulders need "real" pictures to get that!!

H

http://www.tynevalleyplastics.co.uk

Why be happy when you can be normal?

 

[patprimmer]

If it's a water meter I would expect PPO to be the material as it is already used extensively and successfully for many years and has excellent hydrolysis resistance and dimensional stability in contact with water.

Pud

I was a bit concerned with it sticking in the fixed half if it was straight open and shut.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[jmw]

As I said, the original designs were developed to suit 19th century production and materials.
The change to plastic came in the 1960's.

Reading around it seems the Xtel XK (a PPS alloy) is the coming material for compatibility with potable water.
This site -for some reason I couldn't get into the Chevron Petroleum site - seems to have some good moulding guides.



JMW

http://www.ViscoAnalyser.com