threaded inserts and residual stress in Polycarbonate

[actaris]

I have a polycarbonate part with threaded inserts (brass inserts, 4-40 thread, ultrasonically inserted into PC about 7mm or 1/4" thick). I am having a problem with cracks appearing around the inserts. Some of the parts appear to have some shrinkage or voids that occur due to the thicker wall section that is necessary around the inserts. Significant cracks occured during a 2000-hour salt spray test. Are the cracks coming from residual stress, the inserts, or a combination of both? Does anyone have any resources or advice concerning threaded inserts and/or residual stresses in polycarbonate? Also, since not all of the parts have problems with shrinkage or voids, is it safe to assume that this is a process problem and not a design problem?
See image:

 

[gabriel7aaron]

I would suggest the followings:
1. Check if the crack is always at more or less the same area. It could possibily be due to weak weld line (knit line due to material wrapping around the core pin which forms the hole for the insert), or weak sections due to voids.
2. Check with your ultrasonic inserting experts or machine supplier or plastic material supplier if your ultrosonic inserting process is done probably. If the insert is pushed in faster than the plastic would heat up during inserting, it would create internal stress and final failure.
3. Most of the time design and process work together. It is hard to say at this point whether the design or the process is the culprit.

Good luck.

 

[Like2Strike]

Judging from your picture (by the way, very clever inserting it into the forum) it definitly shows that a problem exists. I can't tell if the design of the part is pushing the envelope of nominal wall thickness, or could be influenced by gate size or location , processing parameters, or improper technique to install after molded inserts.

What I can suggest; is that you should seriously consider molding the inserts into part.

From looking at the picture, it appears to me that the cracks are in result of the inserts expanding the material to the point of fracturing.

Live Aloha

Frank M.
Tradewind Resources

http://www.tr-usa.com

 

[CoryPad]

The part cracked due to hoop/expansion stresses at the insert site. The fact that it cracked on one side suggests that the part is weakened due to the small thickness and groove on one side of the insert. Molding the insert into the part has its own problems, and may exacerbate this one, although you could do trials with this method and compare. I don't have any good information sites, although a knowledgeable company is PSM International. More information is available at:

http://www.psminternational.com

 

[patprimmer]

Ge wiz
The picture makes it a hell of a lot easier
It is obviously a stress crack, which is the real archilies heel of polycarbonate.
It might be on a weld line which doesn't help.
It is obviously at a point where the section thickness has changed dramatically, thereby inducing some moulded in stress.
Also, there is a shiny lump of plastic at the edge of the flange to the righ of the insert. This looks like moulten plastic breaking out of its skin shortly after moulding. This might mean that it was not held in the mould and packed until the part was fully set. This can also result in stresses. It might also be that the material was a little bit wet when moulded, which will significantly reduce the elongation at break, making it more susceptible to stress crack.
Another likely cause is that the inserts still have some cutting oil on them, which also causes stress cracking
Another posibility as mentioned earlier is incorrect technique installing the inserst

 

[PrentP]

The stress, hence the crack, is probably the result of the different coefficients of thermal expansion.

Brass - 19x10^-7 C^-1. Polycarbonate - 70x10^-7 C^-1
Regards,

Peter P

 

[patprimmer]

Many people successfully place brass inserts into PC without stress cracks. There must be some stress, but that does not automaticaly mean that the stress is releived by crack formation.
With ultrasonic insertion (if done correctly), the moulding and the insert will remain at about ambient, with only a very small amount of PC being melted. This melting should only occur at the brass Plastic interface, so different C of E is normally not a problem in this regard.
Regards
pat

 

[actaris]

Thanks everybody for your advice!

Actually I was incorrect when I said that the insert was ultrasonically inserted. It is thermally inserted.

CoryPad, we are actually getting the inserts from PSM. Thanks.

Patprimmer, you are right, there is a much thicker wall section around the insert which is probably responsible for much of the problem. The "shiny lump of plastic at the edge of the flange to the right of the insert" is actually a water drop - the part was still wet when I took the picture - sorry. We had our molding house sample the part again in clear PC, and they were unable to eliminate all of the voids - this is the best they could do.

 

[patprimmer]

What can I say
I guess the gate is a long way from the void, so extra pack won't help as the part will be frozen between the gate and the molten plastic.
Is there any way you can get extrapack to the part of the moulding that has the void.
I guess that really cold pins where the inserts go in and a really hot mould would achieve a little, but not nearly enough.
You need a thiner section around that boss.
You might also try about 10% glass fibre in the PC, or try nylon if it's other properties are acceptable. Nylon is tough under a very wide range of circumstances. Thats why they make Auto underbonnet parts and hammer heads from it, but it does have some very different properties to PC, especially shrinkage, dimensional stability and stiffness. Regards
pat

 

[Soupcat]

Sonic Lok inserts can cause continual radial stress in plastic because of the nature of how they are held in the material. That is not to say they can't be used in many applications. I recommend that ultrasonically-installed inserts be considered as an alternative, the part will need to be annealed.
Alternately, the diameter of the hole might be a little too small? I assume you it has been designed to PSM's spec?

 

[actaris]

Thanks, Soupcat
PSM recommends a hole size of 0.157" +.004".
the hole is 4mm or 0.15748", plus draft. So if anything, the hole would be probably slightly too large, but I don't think it's too small.

 

[dugudugu]

Are the parts clean? Sounds funny, but while parts may look clean, you may have some residual hydrocarbons left over after cleaning which we have found cracks PC material. We aleviated this problem by going with a company that does not use such lubricants in their process. The name of the company is Emhart Teknologies. You will see a noticeable difference between their parts and that of any other insert manufacturer.

dugu