Conection between two plastic parts

[drodrig]

Hi there,

We are 3D printing some plastic capsules with laser technology in Shapeways (Polyamide 2200 material). We want to make a hole of 1 mm and plug 4 small optical fibers.

The idea is making a ferrule for the fibers. We glue it and then we polish everything. Then we plug it in the hole of the capsule.

How can we make a system that makes sure the ferrule will stay in the hole? The ferrule should also be made in plastic (maybe PEEK?).

The quality of the printed ferrules (down here some pictures) doesn't look so good. We machine the hole ourselves (after printing).

Thank you

 

[Pud]

Machine a groove on OD of your ferrule.

Add some snap fit features to your printed part - by 3d printing you will be able to add snap fits wherever you like.

I don't think that PEEK is necessarily a good choice - unless you are up to +200C or a bearing. Overkill. If the product is in ambient then POM (i.e. Delrin / Hostaform) would be fine.


www.tynevalleyplastics.co.uk

Politicians like to panic, they need activity. It is their substitute for achievement.

 

[drodrig]

Snap fits.
That's the key. I didn't know about them and I found lots of information on internet about calculations how to do it.
How reliable is with 3D printing? It doesn't seem as solid as plastic injection or machined.
thanks

 

[mcgyvr]

In general 3d printing still has a long way to go..
And even more so when your vendor of choice (Shapeways) is targeted towards hobbyists..

The actual 3d printing processes are important too.. A FDM process won't be the same as SLS or SLA,etc...
FDM is basically weed wacker string squeezed onto a plate.. tolerances/accuracies are fairly low compared to hard tooling/injection molding,etc..


There are many ways to secure that ferrule too.. How about a simple setscrew through the top of your "tophat" pushing down on the ferrule?
What about a toothed insert pressed into the tophat hole to bite into the ferrule?
Snap fits..
adhesives..
many..many more..
a simple hook coming out of your tophat could secure the ferrule at the end,etc..

But don't think 3d printing is going to compare with real injection molded parts/materials.. Its similar but not identical by any means..

 

[drodrig]

mcgyvr,

Sorry I haven't mentioned: This part will be used in a physics experiment where particles come in the direction of the hat-cylinder-axis. So we must use as little material as possible and with low density (i.e. no metal)

Also this will work at -25ºC (-13F).

The ferrule could be machined with lathe, but the hat will be, in principle, 3D printed. We will first make tests with a in-house FDM printer and then turn to Shapeways for mass production (of course we will first test)

The ferrule must be easily plug and unplugged, and many times (I say this because of the adhesives).

Do you think there is a better way than snap fit? I like it, but with the 3D printing looks also difficult
thanks!

 

[MacGyverS2000]

The plastic shape looks so simple, I imagine a local CNC shop could mill them out of solid blocks of plastic... probably cheaper than 3D printing, too.


Dan - Owner

http://www.Hi-TecDesigns.com

 

[drodrig]

We will check prices for both options
thanks

 

[Pud]

MaGuyver's suggestion using a nylon, off the shelf screw is about the simplest.

www.tynevalleyplastics.co.uk

Politicians like to panic, they need activity. It is their substitute for achievement.

 

[hendersdc]

You may consider getting the parts made by these guys as machined plastic. ProtoLabs (formerly FirstCut) I don't work for them but we use them all the time for quick prototypes when the parts are simple and tolerance isn't critical. They will have a lead time of a few days and cost around $100 for each part. Usually more expensive than your average machine shop but you can't beat the lead time.

 

[MikeHalloran]

Print the hockey puck with several long-ish, thin-ish question mark shaped fingers extending parallel to each ferrule at a distance, then looping around behind the back surface of the ferrule and bearing on that back surface, arranged so that you have to spring the fingers away from the ferrule a little in order to engage it.

You can use the equations for a snap fit, but you have to keep the bending stresses really low because of the inherent stress raisers in the printed part, and you want the spring rate that you work against to open the retention arms to be pretty low.

Injection molded snap fits can use short, stiff beams that get near the plastic's yield point when riding over the snap ramp, so you might expect to see something rather like a six-jaw 3C collet in such an application. In your case, you can't afford to stress the beams very much, so they have to be long and delicate, and the material is not all that predictable or reliable at the micro level, so you probably need more than two arms per ferrule, if you can work out the geometry.

OR,

Print the hockey puck with a cylindrical sleeve of the same outside diameter as the ferrule's larger barrel section, and retain the ferrules with a piece of silicone or similar tubing forced over the sleeve and the barrel. Mold/machine/print external annular ridges in both pieces for better retention. Replace the tubing when it gets too easy to disconnect.



Mike Halloran
Pembroke Pines, FL, USA

 

[MacGyverS2000]

Jam one of these in the plastic and screw in the cables. If the ferrule is sized properly and the plastic is right, those puppies aren't coming out.

Dan - Owner

http://www.Hi-TecDesigns.com