deflash methods

[brianpaul]

My company is using a Chinese manufacturer to produce the small polycarbonate parts we use in our prototype mechanical devices.

The parts are intended to work closely with each other, where tolerances in the micrometer range are important.

The production quality parts made by the our American molder are perfect, but we use China to lessen the cost multiple design revisions.

We're finding the flash left on the Chinese parts are causing interference problems, leading to misoperation, when doing prototype evaluation/testing. We're sure there are other root causes for misoperation, but we're trying to eliminate flash from the picture.

Is anyone aware of a good method/company for removing flash from small (~0.5"x~0.5"x~0.25") polycarbonate parts?

We require the parts remain undamaged with respect to:

mechanical properties
dimensions without flash
surface quality for contact/friction interaction
corner sharpness

Any tips?

 

[patprimmer]

Carefully trim with a sharp knife or scalpel.

Regards
pat pprimmer@acay.com.au
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.

 

[brianpaul]

We're talking 1000's of parts.

Some of the techniques I've come accross have been tumbling a batch in a basket under near cryogenic temperature, blasting with a softer material... Something that would lead to the thin flash breaking off, but not the thicker regions.

 

[patprimmer]

Tumbling or blasting are popular techniques. Chilling to below the embitterment temp is a technique that works well, however you say you don't want any effect on the corners or the surface finish, that's why I ruled out blasting or tumbling.

Your Chinese supplier might be able to trim 1000's of parts at a cost that is below what you could imagine. Semi skilled and unskilled labour is cheap and in abundance there.

Regards
pat pprimmer@acay.com.au
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.

 

[MikeHalloran]

Right now, you're comparing finished, acceptable American parts against Chinese scrap, and you're >>buying the scrap<<.

Somebody is ignoring or hiding the extra cost of recovering the scrap and making it >>appear to be<< acceptable, so the scrap >>appears to be<< a better deal.

Presence of flash is itself a symptom of a process that's not working right, and is >>never<< going to be repeatable. You will have recurring quality and fit and in-service warpage problems with that cheap source, even if you can get the deflashing cost under control.

Have you met your CFO? They can be a powerful ally against lunacy. That's the first thing you have to get under control.



Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[patprimmer]

Mike

Are you implying that all Chinese mouldings are scrap, and all American mouldings are acceptable.

You are entitled to your opinion, but the presence of flash is not necessarily a bad thing. It can even be considered desirable in some parts as it can indicate the cavity is fully packed.

Unacceptable flash can be caused by a number of things, some indicative of poor quality tooling, some of poor materials, some of inadequate machinery and some of poor process control.

Over the years, I have seen American, Chinese, Japanese, Australian, German, French, Indian etc etc parts with flash on them.

Regards
pat pprimmer@acay.com.au
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.

 

[MikeHalloran]

I did not mean to imply, nor should you infer, that all Chinese molded parts are automatically scrap. Brian could as easily have said 'company A' vs 'company B', and we would remain ignorant of the parts' country of origin.

Okay, >in my opinion<, flash _is_ a bad thing. It _never_ means the mold is properly packed. It means the mold is _over_packed, and the molding machine is opening up in response to excessive injection pressure. That's not good for the molding machine.

It's not good for the parts, either. Packing a poorly designed mold by means of excess pressure produces a part with poor mechanical properties, internal flaws, and locked-in stresses.

Over the years, I too have seen many flashed parts. Anyone can make scrap.

My point, perhaps stated too strongly for your taste, was that I am getting tired of seeing sourcing decisions made on the wrong basis.

E.g., if Company A makes a perfect part, and charges you say $.80 for it, too many people will jump at the chance to buy imperfect parts from Company B for $.05, and ignore or obfuscate the fact that they have to pay Company C $1.00 to make Company B's parts meet the specifications. That concern does not apply just to molded parts.

In this application, overpacking is a special concern because a material like polycarbonate, often selected for high stress applications, is capable of quirky behaviors.

Brian's employer is in mortal danger here.

If Brian can find a way to deflash the Chinese parts cheaply, there will be very strong internal pressure to use only the Chinese parts for production.

They will, after all, meet the dimensional specifications when they are inspected after deflashing. In service, the locked in stresses will relieve themselves by distorting the parts, moving them out of specification and causing the assembly to malfunction, and/or the internal flaws will turn into fractures, causing the assembly to cease functioning altogether. All of this will happen within the warranty period, and the resulting warranty costs will drive Brian's company out of business.

My paranoid jingoistic side says that, just as Brian's outfit is circling the drain for the last time, the Chinese source will suddenly learn how to make a proper mold and a decent part, and will enter Brian's then former employer's former business with a ripoff of his design.

I really hope I'm wrong.



Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[patprimmer]

Mike

I now pretty much agree with what you say, especially when it comes to the total cost of goods considering the full cost of rejects, and the full costs of corrective actions.

I have seem good and bad moulders and good and bad toolmakers in many regions. Some regions have a greater number of good and some a greater number of bad, and it is a dynamic situation as people learn. Also some less developed regions pay expatriate workers from developed regions a very high salary to manage and teach their workers.

I also have seen people put out of business by their suppliers, and I have seen suppliers put out of business by their customers. I think America's anti trust laws were implemented to counter such unfair practises, in fact, I think that Henry Ford was once deliberately bankrupted by the banks that financed him, allegedly because they had stronger ties to his competitors.

I agree that a disproportionate number of such incidence have occurred more recently in China, whereby developed countries teach the Chinese how to make their product, only to create a competitor rather than a supplier, but this does not mean that all Chinese companies behave in this manner, nor that developed countries suppliers never behave in this way, however I would advise Brian's employer to be extra careful of this if using Chinese suppliers.

I agree polycarbonate is a bad material to over pack because of it's solvent stress crack vulnerability, but I do reiterate that flash does not necessarily indicate over pack.

Flash can indicate any one of the following:-
Over pack.
Damaged parting line on mould.
Melt temperature to hot.
Poor tolerance control in mould making.
To high injection speed.
Insufficient clamp pressure.
To high melt flow index.
Wet material.
Worn tie bars on machine.
Material flow into vents.

That's all I can think of off the top of my head.

Many highly respected moulders that I know, especially when moulding semi crystalline materials will look for the slightest sign of flash as proof of sufficient pack to develop full mechanical properties. One such moulder moulds rock bolts for underground mining. He moulds them out of 45% glass filled nylon, and they must pass regular tensile tests. He consistently passes, and he moulds with a whisker of flash. He has plenty of clamp. The moulds are in good condition. The melt viscosity is normal injection moulding standard for glass filled nylon. The parts are 5' or so long and about 3/4" dia. The flash shows as a flash down the vents and a just noticeable sharp edge on the parting line of the split blocks that form the threads.




Regards
pat pprimmer@acay.com.au
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.