THIN-WALLED PLASTIC MOLDING 1

[bicolexpress]

Hi! I'm a mold designer and is currently managing
a design team in plastic injection molding.

For the past 15 years, we are designing, fabricating and injecting industrial products such as automotive battery plastic components, automotive lenses and numerous plastic components. But we are new in the THIN-WALLED PLASTIC
MOLDING. We tried our first thin-walled bucket mold with 0.65mm thick, 200mm top dia, 175mm height but it didn't last long. The problems usually are flashing and uneven thickness (offset core/cavity). We tried for the second time using bigger plates and improved interlocks but the same problems occurred though it lasts longer than the first one...

Need somebody that can help me about DO's and DONT's in thin-walled molding... Thanks in advance.

 

[MikeHalloran]

Uh, don't do it? ;-)

Clearly you've reached the limits of what your shop can do now, so you've got a real data point that you didn't have before.

ISTR that most of the super- thin walled parts I've seen were done with liquid crystal polymers, or more plebeian resins that exhibit really good mold flow.

Speaking of which, mold flow modeling software might be a good investment if you wish to proceed farther in the thin- walled direction. Adjusting computer models is easier than adjusting molds.

Tightening up every other part of your process will probably be necessary, too. Air condition the shop, record and track every temperature you can measure, and work to keep them consistent, control the moisture content of your resin, all that good stuff.

ISTR also some activity along the lines of molds that are relatively open during fill, but some features close up just before the shot completes. I can see how to do it for something like a tapered cup, and maybe in principle how to do it for a straight part with square corners, but it has to add lots of complexity to the mold and the controls. I don't know if anyone's actually done it in quantity, it's just an interesting idea.

I don't think the molds are that fancy, but I'm impressed by some of the pump-up toothpaste tubes I've seen. They have really thin, straight walls, but note that both the blind end where the paste comes out, and the flange that the assembly rests on are relatively thick. I suspect that has a lot to do with making them moldable.

Do a little disassembly and reverse engineering on whatever cheap consumer products come to mind. Given the quantities they shoot, those guys have a big enough budget to get it right, and an incentive to not waste a microgram of resin on extra thickness.

Oh. I make no claim to expertise in thin- wall molding or any other kind of molding. I've designed a few interesting parts and worked closely with some amazing moldmakers.





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

 

[MikeHalloran]

I'dd like to add one more thing.

If the mold core is a solid of revolution, as is often the case, knit lines opposite the gate have been a perennial problem.

No more.

A recent "Design News" carries a real forehead- slapper of an idea. Spin the core while the resin is entering and setting, and the knit line's plane of weakness gets smeared out into a spiral surface. Better, if the resin carries fibers, the shear from core rotation tends to orient them in a circumferential direction, which is usually a good thing.

There's just one bit of bad news; there's a patent on the process, held by Solvay. I assume that if you buy Solvay resin, you get a license to use the technology. The idea is so good that it might be worth buying a few carloads of Solvay resin, even if you don't like the color or otherwise can't actually use it.



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

 

[unclesyd]

Here are two website that you might want to checkout. We used the Fluent product and when I left they were talking about the MoldFlow software.

http://www.moldflow.com/stp/english/products/des_opt_sol.htm

http://www.fluent.com/solutions/polymer/applications.htm

 

[patprimmer]

As said above, do flow analysis.

Also do finite element analysis on the mould structure using the worst possible case for uneven fill.

Remember that once core shift starts, it is compounded by the extra fill on the now thicker side.

Remember that when you do the analysis and presume uniform surface temperature of the mould, you might not actually get that in practise, unless you carefully design the coolant passages and the coolant flow control. Also, the best design fails if the the die setter hooks the water lines up in the wrong sequence.

Very small variations in dimensions when making the mould have a more significant effect on thin sections. e.g. a 0.1mm variation is 10% of 1.0mm, but just over 3% of 3mm.

To keep cores centred, they need to be inserted deep into the base plate, and the best hard rigid steel is required for both, as is the optimum interference fit. The old example is set a 7' post 1' into soft soil, then set a 9' post 3' into soft soil, then repeat the exercise with both in concrete, then try to move each post at 6' above ground level.

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.

 

[strokersix]

Pat,

With my apologies for picking details: Which steel should you choose to be more rigid? To my knowledge all steels have essentially the same elastic modulus.

 

[bicolexpress]

Dear all,

Thank you for the inputs... Right now, our existing mold is running in 350T Toshiba Injection Machine but with 0.3mm shim between Machine's fixed platen and molds top plate to compensate product offset. Set machine parameters are as follows; 95% of Injection pressure and speed and 99% of clamping force. We are using PP with 40 M.I.

We do have moldflow software but for the cylindrical parts, the flow analysis will normally results to uniform/even plastic flow. The software can give us the required Injection pressure, Speed and Clamping force.

We are now in the process of designing the 3rd mold. We are already considering the reduction of number of plates, i.e., redesigning to 2-pc main block mold; 1-pc cavity block and 1-pc core block (product contours will be machined directly to each block), to minimize/eliminate core shifting. We are using M300 prehardenned mold steel.

Again, thank you for sparing your time for my query.

Regards...

 

[SteveAf]

Look at the X melt system developed by Engel. It will give you the abitly to fill this part so extreamley fast that the cavity pressure gained during a normal injection filling process will not come into effect.

It works very well for thin walled parts.

 

[bradphillips]

You may get better results in a 500 ton machine or bigger. You stated that you are using 99% of the injection speed and pressure but it may not be fast enough to fill properly.

 

[Pud]

Hi,

This walled mouldings require two specialist products:

1. A tool built by toolmakers who specialise in such products. As Patprimmer said, rigidity of cores is paramount. I have seen tooling to produce a one pint polypropylene beer glass with .5 -.6mm wall having the core(s) set into a 150mm thick backplate by means of a taper fitting so the core can be pulled up tight. To give some idea of the degree of specialization, there are only two sub-contract toolmakers that I know of in Europe (one UK, one Switzerland) who do this kind of work. The major packaging manufacturers have their own toolmaking facilities. Correct venting is the other requirement.

2. Injection moulding machines with very high speed and accurate injection. A "normal", general purpose machine will not have sufficient pressure or oil delivery rate to achieve required injection rates. Usually this is achieved with small diameter screws and large hydraulic accumulators. Injection times are in the fractional seconds range. (A Dutch company called Stork make high speed packaging machines - google should produce a result - no connection, just know of them)

In addition, when the above tools are put in the above machines, the tools are measured and set up in-situ for alignment as the usual pin/bush clearances are too great.

Hope this hasn't put you off, but I know from experience the problems you are having!!! (fortunately on a prototype tool!)

Rgds to all

Harry

 

[bicolexpress]

Hi Guys!

I posted this thread 1 year ago... and the 3rd mold that I'm talking about is already mass producing the product in a 200T machine. This machine can't be considered as high speed but this is on the high-end of the conventional type.
The cycle time is 13 sec and we have to control the cooling to prevent plastic from freezing before it fills the cavity.
Using a high speed machine will definitly improve the cycle time to at least 6 sec but our present volume requirement can't justify the cost of a new high speed injection machine.

As per experience, the mold design really does matter when it comes to rigidity. Pure air ejection instead of stripper
also helps improve cycle time and at the same time minimizes wears and flashes.

At present, I'm designing a 2-cavity thin walled mold for a cup and considering my previous experiences, I'm confident that this will already work as expected.

Just sharing my experience,
bicolexpress

 

[MechNorth]

There are a number of things that you can do to control the fill rate; venting (need lots!), thermal profile control (using insulative / conductive materials in the right segments, and maybe even heat pipes). There are a number of good injection mold design textbooks out there, and that's a good place to start. Another option is to consult with your machine supplier. They cater to these markets and have a wealth of experience that you can tap into (if they're a decent supplier). An even wiser option would be to hire a thinwall mold designer that is retired or near retirement; they like to keep their minds and hands busy, and pass along lots of industry knowledge. Shots in the dark are very expensive, and you may find that some of the lessons you learn are not applicable across the board.

 

[Drazen]

Did you ever consider thermoforming instead of injection molding for that particular product?

Sometimes you can get much better results with thermoforming large thin-walled parts, not to mention possiblity to redo many scrap items (if parts are not for some very critical application, and thin-walled bucket doesn't seem as very critical stuff).

Some injection molding presses can be pretty easily rebuilt to forming presses.