help regarding material selection and tolerance design 1

[IDdesigner]

I'm looking for help regarding material selection and tolerance design.

We have a system that calls for a connector (cast aluminum) and a tube (steel) not DOM tubing, to interact in a male-female relationship. The tube will slide over the connector. I'm looking to achieve a very tight fit between the two, but not to the point that tools are needed to slide the pole onto the connector. Can someone advise me where I can find information on using a shim (maybe nylon because of its self-lubricating qualities) and where I can get ideas on rib designs based on material selection to create a high-tolerance fit. Thanks for your time.

 

[rickfischer]

Material selection will be based mostly on loads, environment, and cost factors. The need for tight tolerances will be addressed more with molding process, and hence part design, and mold design. Having said that, you suggested a poor material. Nylon is hydroscopic, meaning it will absorb water and swell. This will mess up your fit. Also, nylon is a semicrystaline resin, and I've seen significant post mold crystalization in nylon cause shringage to continue well after the part has cooled. It's been my experience that it is easier to produce tight tolerance parts with amorphous resins than semicrystaline resins because of the added shrinkage due to crystalization. This is temperature dependent, which adds a variable that must be controlled. It's not impossible, just easier with an amorphous resin.

What are the loads? Intermittant, long term, fluxuating? Generally, semicrystaline materials are better in fatique. If the load is continuous, creep may be a factor. In such cases, the closer the operating temperature is to the glass transistion temperature, the higher the creep rate.

Ribs usually need to be thinner than the main wall to avoid sink marks, but it sounds like the shim will not be visible, so a sink would not be an issue. Generally, you want the wall thickness to be uniform, with minimal variations from nominal. And you want to gate into the thickest section. Gate location can effect shrinkage and warpage, so it must be considered as the part is designed.

You need to tell us more about the application

 

[IDdesigner]

Thank you for your response. The loads will be constant and long term. The system created from a series of poles and connectors will make cage-like configurations. Each tube at 1.5 O.D. Therefore if you can envision tubes as long as 4 feet, the joints will have to have to sustain a substantial weight, however the load will be distributed throughout the system. Probably 100-600 lbs total. Let me also mention the system is static and will be indoors.

When designing the shim given the dimensions and amorphous resin material, what would be a good wall thickness to start with? Basically I'm having a hard time controlling my suppliers I.D. tube dimensions. They can guarantee around .2mm. The cast aluminum is obviously guaranteed to a closer tolerance.

I'm not familiar with the molding process of this material. If you could recommend a wall thickness/ rib size, perhaps a rib design conducive to the physical interaction of sliding a tube onto the connector and also a design that would lend itself to easy manufacturing and also tight tolerances.

This product should also be fairly cost effective; however quantities for this molded piece will be 5000+.

 

[patprimmer]

I certainly question some of rickfischer's statements.

Nylon is certainly used successfully in similar applications. Polypropylene and ABS are also frequently used and under some circumstances polycarbonate can be used with some precautions.

The biggest problem will be getting consistent steel tubing, and variations in tube size will be somewhere between 10 fold to 100 fold as great as variations in mould shrinkage and variations from different levels of swelling from moisture absorption in nylon.

For clarity I will call mould shrinkage any shrinkage that occurs in the mould and for up to 1 hour after moulding. I will call post moulding shrinkage shrinkage that occurs more than 1 hour after moulding.

In over 30 years of experience in nylon applications, I have never once seen a problem with dimensional stability caused by post moulding shrinkage. Post moulding shrinkage with nylon is more than offset by the real problem of swelling from water absorption. Polyester and polypropylene occasionally have such a problem where tight tolerances are required, but not nylon.

Nylon does have a problem with dimensional stability and moisture absorption over time. Mostly this can be resolved by conditioning unless the relative humidity of the environment of exposure changes considerably on average for a substantial time. The time element is important as nylon is slow to absorb or desorb moisture. If the parts were exposed for several months in water in the tropics, then moved to a heated indoor environment in the Arctic winter or the desert, then swelling and shrinking is a real problem, but if only indoors in temperate climate, then the small short term variations in RH do not create a problem unless you need to retain dimensions within microns.

The nylon I would recommend is nylon 6 as it is relatively cheap for a nylon, is easy to mould and has relatively low shrinkage.

If the connector were made from glass filled nylon 6, I would think it might be possible to eliminate the aluminium casting and make it from 1 part.

I have seen a tube and socket type fitting in many automotive roof rack applications, where the joint is so strong the roof of the car fails before the rack if severely overloaded. These are mainly 25% glass filled nylon 6.

These roof racks also have a problem with variations in tube size. They address this in several ways, these being:-

1) To have a long plug to get good depth of engagement.
2) They have a well designed rib system.
3) They are sized so that the body of the moulding is a snug fit in the smallest likely tube size. They have thin tapered ribs that are an interference fit in the largest size. These ribs shear off on assembly and in essence are custom fitted to the tube.

This will need tools (like a hammer) for the first assembly, but should be a tight fit for hand assembly from then on.

Amorphous resins are more dimensionally stable, but are also more susceptible to fatigue, abrasion and environmental stress cracking. Nylon and PP have good natural lubricity and ductility. They resist cutting oils etc typically used to make and cut the tube.

I do some work for a company that compounds all types of plastics including many semi crystalline and amorphous plastics.

I am a director of the SPE in Sydney Australia, a committee member of PIMA

http://www.pima.asn.au

and I am a past president of The Plastics Pioneers.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[IDdesigner]

Thanks for your insight. Can you reference me to a common rib design in industry you have seen successfully executed in this instance? Possibly also an example of design that uses ribs intended to shear off. I appreciate the materials break down. My background is in product design; therefore I have little experience when it comes to structural design.

 

[patprimmer]

I guess your in the USA.

I will see if the roof racks mentioned are available there.

If not, If you choose, you will be able to find a way to contact me via the PIMA web site. We can then work out how to get some example mouldings to you.

Rib design depends a lot on the geometry of the part, but for round tube, an "X" shape or an "H" shape running parallel to the axis of the tube and with extensions off the cross bar of the "H" to contact the side of the tube.

As the ribs will not be visible, heavy wall sections can be used if required.

Lighter cross ribs or bridgework can be used to stabilise the main beams of the "H" if required.

Say the main ribs or beams of the "H" are 4mm section, the bumps or shear off ribs should be say 0.5mm high and wide and say 3 mm long and near each end of each rib. These can normally be ground into the mould by hand with a small hand held die during trials. Start very small, try it in min and max sized tube. Allow 24 hours immersed in water and another 24 hours exposed to air between size changes. Longer time would be better if possible. Time to maximum dimensional stability and stable material shear strength and hardness will vary with material grade, thickness, RH and temperature.

I have a lot of data on this, but it is to much to publish here.

I know I should set up a web page somewhere, maybe on the PIMA site if time permits.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.