Material selection for injection molded part

[futureguy]

Hello engi-tips plastics experts. Long-time reader, second time poster.

I am working on possible improvements to an injection molded plastic piece that is used in a riveted caster wheel assembly. The function of the piece is as follows:

The piece is essentially a cam with a collar. Here, the cam is being loaded by a steel spring (maybe 10-20 lbs) to give the caster wheel a directional bias:

http://postimg.org/image/blmy0gawp/

^ the split in the piece is a failure due to compression of the rivet. This happens to about 4% of cams during production and almost always happens along the "knit line" opposite of the injection gate. Here is a cross section of the assembly:

http://postimg.org/image/e1oreaszd/

^ the orbital rivet machine is quite powerful, and the cam smashed in the gaps between bearings, rivet head/shaft. The cam is usually the weak link in the assembly, eventually deforming under load and impacts, creating slop and allowing the rivet to bend.

Here we see a similar product from Taiwan that uses a different material (the impetus for us focusing on material change)

http://postimg.org/image/ghqks5b21/

The first difference noticed was the lubricity of the white cam. Our cams need to be greased up to provide smooth frictionless action against the spring's contact point. This white cam requires no such lube but remains smooth against the spring. I assume this has to do with the lubricity of the plastic, or possibly the rockwell hardness (reducing the contact point between plastic/metal)?

Our original product design specified Acetal Delrin. Currently the parts are poured with Dupont Zytel ST801. Our supplier has suggested Zytel ST811 or Crastin ST820 as possible upgrades to test, although the toughness indicators (impact strength, tensile strength, density) in the spec sheets don't seem to be appreciably different, and the zytel811 and crastin do not indicate rockwell hardness.

Questions:
Do you think that Zytel nylon was used as a cost-saving measure? Supplier claims delrin would be "too soft" - I understand delrin to be more of a "performance" plastic, often used as a metal substitute, but again I can't find appreciable differences in the specs there, either.

What other plastics might you suggest for this application? (integrity of the rivet stack being top priority, and lubricity against the spring being second priority) What would you guess is being used in the white cam, or is this possible to determine?

Thanks in advance for any and all input. Spec sheets for the various material can be found here:

http://catalog.ides.com/datasheet.aspx?I=55947&E=89177&U=1&CULTURE=en-US

http://catalog.ides.com/datasheet.aspx?I=55947&E=72052&U=1&CULTURE=en-US

http://catalog.ides.com/Datasheet.aspx?I=55275&U=1&FMT=PDF&E=85264

 

[Pud]

ST801 is an impact modified nylon. Due to the impact modifiers, it is not an optimal bearing material.
The Taiwanese cam looks like acetal. Acetal is quite a bit harder than nylon, and also a good bearing material - certainly better than ST801. PTFE and silicon oil filled grades are available from most suppliers.
Weld line failures are a processing issue - tool too cold, injection speed too slow, etc., etc. Worst case s is glass filled material where you can expect a significant loss of tensile strength. This I guess is not GF so weld line "worst case" should be a witness line on the surface which should be hardly discernible.
If you have a sample of the Taiwanese one, apply a flame to an edge. If it burns with a smokeless blueish flame, and when you sniff it it takes out your sinuses with an eye-watering acrid smell reminiscent of rotting fish, it's acetal. (gives off formaldehyde) Nylon smells like burning bone...

Cheers
H


www.tynevalleyplastics.co.uk

It's ok to soar like an eagle, but weasels don't get sucked into jet engines.

 

[btrueblood]

Why not a metal sleeve down the i.d. of the bushing, to limit compression caused by riveting, etc.? Sleeve could be molded-in to the plastic bushing.

 

[futureguy]

Thanks for the replies. Yes, btrueblood, we will eventually replace the bearing component with a metal collar. Or replace the whole cam piece with metal and plasticize the spring's contact point...The hope is that in the meantime, we can use a simple material change as a temporary upgrade.

Pud, the general product descriptions of acetal delrin seem to bear out what you're saying..."it is designed for highly stressed
parts where outstanding toughness is essential" but the tensile strength, impact strength and rockwell hardness don't seem to be appreciably greater than the ST801 (the ST801 is r107 while the acetal is r105)...are those the wrong indicators for this application? The major difference I see is in density, the delrin being much more dense. This is the delrin I'm looking at:

http://catalog.ides.com/Datasheet.aspx?I=95947&U=1&FMT=PDF&E=36988

Can you speak to the Crastin or make any other suggestions? Is delrin much more expensive, harder to process, or is there any other reason our supplier would avoid pouring it?

 

[futureguy]

then there's this, indicating Zytel as a higher performance polymer:

http://www2.dupont.com/Plastics/en_...pps/DPM13_19015-Polymer-Pyramid_580x350_1.jpg

 

[Pud]

Of course Du Pont are going to say ST801 is high performance polymer! What would you expect any manufacturer to say? It's rubbish? It's sold for it's high impact strength - and at low temperatures. It's not a bearing material. Plain Nylon 6 or 6.6 would likely be better.

Have look here: heading on centre of page 1. Delrin

Acetal is about the same price as nylon, but higher density. Processing a bit simpler - lower temperature and no drying required. Cycle time around the same. Note there's plenty of good acetal manufacturers apart from Du Pont.

Your moulder might process PVC (vinyl in USA) - when melted with acetal violent decomposition occurs, so some will not have it on site.

Also note that one is assuming that materials are processed correctly - especially tool temperatures. +60ºC for nylon, +90ºC for acetal for maximum properties.

www.tynevalleyplastics.co.uk

It's ok to soar like an eagle, but weasels don't get sucked into jet engines.

 

[Pud]

Oh, and btrueblood's compression limiter is a splendid idea.

www.tynevalleyplastics.co.uk

It's ok to soar like an eagle, but weasels don't get sucked into jet engines.

 

[RogersPancreas]

I don't know if it's been mentioned yet, or if the original poster is even aware, but Acetal can be a lethal polymer to process if the party in question isn't aware of what they're doing. Just type "Acetal" and "Explosion" into Google for all the background information you can handle.

Because there's a pin to mold around, you'll always have some kind of knit line. There's no getting around that. You can do things like turn up the injection velocity, heat the mold, etc. It's all leading to an increase increase in the material's viscosity, to better enable polymerization; in the end, It'll still be there.

What I am curious to know is whether or not you've tried changing gate location. With such a low rejection rate, your design is probably right on the edge. It's good enough under most conditions, but certain variables you don't have any control over lead to these failures. By relocating the gate, maybe you can form the knit line along the thicker portion. You've already proven the thin wall area can handle the stress (most of the time). More material can only help you.

One last question: Do you have any FEA data to show if that thinner wall is subject to more stress (than the thicker wall) under normal operating conditions, due to the nature of the cam? If there are, it only gives you more incentive to change it.

 

[loki3000]

change gate locations and/ or use more gates so you don't have any weldlines.

where is the current gate location? is it visible on any of the pictures (i don't see it though)