Questions about barbed leg snap fit

[kjoiner]

Hello,

I don't do a lot of plastics design and I'm working on a design that will involve a snap fit and before I get too far down the design path and especially before I commit to tooling I wanted to get some input from other users. I've looked over previous posts regarding snap fits and have some design guides and material data from plastics companies (DuPont etc.) but still have some questions.

Application:

A round ring - a torus - that looks a lot like a shower curtain ring. The ring will split into two semicircles so that two identical pieces make up the ring - one end will have a barb and the other will have an undercut hole that passes through the cut face and exits tangengially out the other side. The cross section is Ø.25" and the ring itself is Ø1.00 on the ID. Hopefully this paints an adequate picture.

I want to take the two halves and snap them together permanently so I'm planning to use a 90° on the barbs. Also I need some location and good bending strength from the barb (not all of it can be split) so the ring won't break in half like a cookie. I also want at least a 25 pound pull apart load ( 12.5 lb per side of the ring or 6.25lb per barb leg).

I'm planning to use black Acetal for its low friction and good mechanical properties.

Question

1. I'm starting with a .02 undercut and my "beam length" of the barb is about .160" long. The OD of the shank of the barb is Ø.155. This leads to a stress of around 14,000 psi which sounds like trouble, but when I use other examples for barbs and calculate the strain, I get around 6%. The beam cross section is semicircular and about .05 thick (I can' thin it down too much or I'll lose the tensile strength for the pull apart force). My question is what is an allowable strain for Acetal? I've seen anywhere from 1% to 10%.

My main limit right now is that I'm limited to the overall length of the barb since I don't want it sticking out the back of the ring where someone could possibly unlatch it. I want the ring to require cutting to be removed. Therefore I'm juggling between the length of barb and the length of the slit in barb which detracts from the loctaional requirements I mentioned previously.

Hopefully I've provided enough information to provide background for my question. Thank in advance for your help.

Kyle

 

[kjoiner]

Here's another question:

On a part where I have limited depth, should I consider using a cylindrical or annualar fit and then the deflection would be handled by the female side and I could make the "pin" solid. Could the hole and undercut be blind and created by a collapsible pin?

Kyle

 

[patprimmer]

Collapsible pins are expensive.

Acetal is variable re some physical properties depending on grade and moulding conditions.

A homopolymer extrusion or low melt-flow grade probably suits your application best.

Use a hot tool (80 deg C min) a large gate and plenty of hold up time and pressure if you want to approach the maximum numbers. Annealing the mouldings can also help improve the numbers.

I think you need to do some FEA on this.

Regards

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]

The collapsible cores I've seen won't fit through your assembled ring. I doubt that anyone can make them reliable at the size you propose.

You might be able to make a not too visible cavity with a squarish pin extending radially from the i.d. of the assembled ring, intercepting a core pin to form the smaller part of the barb receiver. It still requires more moving parts than I care for; a tangential stepped pin with visible hole is much simpler.

WRT what you can do strain-wise to the barb and/or socket, you can get away with a lot of abuse if the parts are assembled immediately after they come out of the mold. If they're allowed to cool first, you have to be more conservative.

If the assembly is to be done by a consumer, you have a whole new set of constraints regarding assembly force and simply grasping the parts. I have assembled some shower curtain rings that join exactly as you have described, and just getting hold of them with my arthritic hands is not fun.

In similar parts, the 'classical' solution is a barb that's hollow, or split, or both ... and not very strong.

One alternative to that is a partially frangible feature that's known in woodworking as a 'foxtail wedge', wherein some feature at the base of the socket dislodges a wedge- shaped piece molded with the barb, which forces the barb apart as it's assembled. It would be an award- winner for your molder at your scale. That's not usually an earmark of a good design from your standpoint.



Mike Halloran
Pembroke Pines, FL, USA

 

[israelkk]

https://ew20.asp.dupont.com/plastics/pdflit/americas/general/H76838.pdf

 

[israelkk]

https://ew20.asp.dupont.com/plastics/pdflit/europe/design/L12565_9.pdf

 

[kjoiner]

Pat & Mike,

I realized as soon as I posted my second message that a collapsible core would not be a good idea. Maybe an expanding one that would form the barb for the annular fit would work, but then I would have that plus at least another core or maybe two to create the socket.

As far as assembly, we would be performing the assembly, but the rings may sit for a while after they arrive at our facility so assembling them right out of the mold would not be possible. We have considered doing some FEA on the parts, I just need to come up with a path for the design and then start the iteration process.

Kyle

 

[Pud]

Hi,


As you already have a second operation to press/snap the halves together, have you considered ultrasound welding the two halves?

Welding will probably be as quick as snapping together as the parts would be much simpler to locate. With a careful joint design you will be able to get any "overflow" into the inside of the part. Also, impossible to get apart without destroying the assembly.

Mould tooling would be much simpler so money saved there could go towards a welder and horn assembly....

Rgds

Harry

 

[kjoiner]

Harry,

That's an intersting idea. My only hesitation to go with that method is that we may need to replace rings in the field and would want the rings to be installed minimal tools. Also, there will be around 32 rings on this machine so I want to keep the process as simple as possible.

On another note, by offsetting the split line of the ring, I can buy some more length of my barb arm and can gain some unsplit length to serve as a locator.

I'm still a little unclear about whether snaps are designed around acceptable stress levels, strain levels or both?

Kyle

 

[Pud]

Hi,

Misread post - assumed it was torus split along plane.

Ticona, for their "standard" grade of co-polymer acetal give a figure of 8% strain for barbed fits and 4% for cylindrical snap fits.

Here is the book, but I suppose you already have it.

http://www.ticona.com/tools/search/lit_details.cfm?docid=13257

imho, I would use a copolymer acetal - they are easier to process (more thermally stable), and there are many producers of like products to enable shopping around for best prices.

Rgds

Harry

 

[patprimmer]

But homopolymer has significantly better mechanical properties, especially flex mod and tensile. That seemed to be an issue here. Modern homopolymers are really no problem to mould on modern equipment.

Regards

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.

 

[kjoiner]

Everyone,

I was able to pick up some extra length for the barbs by ofsetting the split line for the ring. That should keep the strain down to about 5-6%. My next step is to machine a barb and hole out of some delrin stock to test the barb.

I had downloaded the Ticona design guide and followed along with the examples. It's a nice and concise guide, I just had to keep converting back and forth between metric and english dimensions.

The discussion about exactly which type of resin to specify has me interested. Any suggestions? I can also check with my mold maker. Thanks again for the help so far.

Kyle

 

[MikeHalloran]

You will get mechanical properties more representative of a molded part if you avoid the center of Delrin machining stock. I.e., along the axis of each round, and in the central plane of each sheet, Delrin is basically a foam. So buy stock that's more than twice as thick as you need, and don't let your sense of symmetry govern.





Mike Halloran
Pembroke Pines, FL, USA

 

[kjoiner]

Hello,

I've made a few design iterations with my moldmaker and was able to tweak the length of the barb legs using the offset split line. This will allow extra length to start with and then I can shorten the legs if necessary. I've also left plenty of room on the other aspects of the snap per my moldmaker and we can tighten up the fit after we run the first parts. Thanks again for all the help and I'll let you know how the first run of parts turns out in a few weeks.

Pat,

I saw some of your posts in a the engine forum regarding the Leroi engine. There are some intersting topics that come up in that forum. Have you ever seen the web site of the gentleman that has built a miniature Deltic engine?

Kyle