Plastic Knobs Cracking 1

[Hiker79]

Hi,

I have a molded plastic knob whose application is on the end of an encoder. I see about 10% come back from the field with the hollow shaft (it's a D-shaft with a D-ring) having a crack in it. How do I determine if the knob shaft has a thick enough wall to contain the hoop stress?

Thanks!

 

[MouseTrap]

Hiker,
I don't have hoop stress info available, but it is my guess that this crack is probably propagating itself at
(A.)The flow front weld line (usually opposite the gate)
or
(B). at the 'corner' of the D inside the knob hole,
and if either A or B are true, then this would not be a hoop stress issue, even though it appears to act like one.

(please come back and tell us exactly what the material is and where the crack is)

If the crack is at (A), a processing mod might help the weld line knit together more effeciently. If a tool mod is in order to increase wall on account of your pereived hoop stress problem, then I would also look into adding a rib (where the crack appears) so as to give the flow front a larger area in which to knit.

One other small bit - I don't understand your term 'D-Ring' - is this a molded-in metal insert at the knob hub to receive the shaft?

Mousetrap

 

[Hiker79]

Hello Mousetrap, thank you for your helpful reply! I'll see if i can add some more detail here. The material is a polycarbonate blend. It appears that the cracks are random; not propogating from the corner of the D or any place in particular. The D-ring is a small off-the-shelf ring that fits in the knob and receives the shaft of the encoder. It is basically a spring that hold the knob to the shaft of the encoder by expanding slightly and applying force to the inside of the knob shaft. The knob shaft wall thickness is .040" and the ID of the shaft is about .175". The plastic material is inherently flexible to a degree, however it appears that the fractures act more like that of a brittle material. Could the .040 thickness of the wall be acting to make the material more brittle in that area?

 

[patprimmer]

If the knob is PC or a PC alloy, check for oil or residual solvent on the metal parts in contacr. PC has lousy stress crack resistance in the presence of any aromatic or chlorinated hydrocarbons. Many so called aliphatic hydrocarbons contain aromatics as impurities.

A solvent to cause a crack can come from an inoculous source like cosmetics, sun tan oil, insect repelant, hand cream etcetc etc

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[MouseTrap]

Hiker,
You speak of a .040" (hub?) wall? - Sounds small (real small) - more dimensions (and a couple more questions)
(I've searched for a few web sites and built this response as I progressed, but you might want to look at my third referenced website down below.

Some questions that I'd like to know anyway:
A. Knob 'handle' OD?
B. Knob 'handle' ID?
C. Knob 'handle' Length?
D. Thickness of knob 'handle'
(or if the knob handle is not round, then nominal wall of the knob portion)
E. Knob hub OD?
F. Knob hub OD?
G. Knob hub length?
H. Knob hub hole depth?
J. How many cavities is the mold?
K. Is each cavity in the mold 'marked' so as to differentiate (identify) specific cavities?



N. Is it possible to misuse the knob so as to force failure - you might be gently using the knog, but there are gorillas out there in the field - some of which may be turning 'forcing' the knob, and maybe not even be turning it in the right direction to begin with.

P. Does the hub appear to be fully molded?

Been searching the web for hoop stress and came up with this - doesn't give an 'equation/calculation' but it may help. Surf this website - lots of basic stuff, but nothing specifically about hoop stress. I am wondering if you might be trapping gas in the hub at molding (check out mold design 'venting' at this website:

http://www.dow.co.jp/styron/design/guide/fits.htm

Another good read (no specific 'hoop strees' stuff at first glance, but good none-the-less:

http://my.execpc.com/~drer/phil2.htm

And !!!!!!:

http://www.delphi.com/pdf/techpapers/2000-01-0316.pdf

(This one looks like it's right down your ally)

Do I get a star today ?

 

[phoward]

Did this just recently start after not having the problem?

Check/ask the molder for any change in the mold release agent formulation. T

he EPA ban on a prominent solvent for release agents, 141b, has caused universal re-formulation this year. I have one customer who discovered embrittlement after a week after using the competitor re-formulated silicone mold release.

The first suspect would be a halogenated (chlorine etc.) family element in the new solvent as well as some of aformentioned hydrocarbons. The harder the resin/plastic, the more likely the embrittlement.

 

[AG62]

I will raccomand to check the cooling on the mold. If the area of the knob is not well&uniformely cooled may couse the problem that you described.