Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SSS148 on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Molding/Forming Techniques For A Light Diffuser 1

Status
Not open for further replies.

leonhart88

Electrical
Jan 2, 2011
14
Hi All,

I have part that is around 4.5 cubic inches and acts as a cover and light diffuser for a LED assembly. The current prototype is made of glued frosted polycarbonate, but if we are looking to manufacture hundreds of these I was hoping injection molding or some type of forming would be worth the effort (plus glueing leads to unsightly seams). The plastic type is not all that important, as long as it is fairly strong to occasionally impacts from consumers and it diffuses light well.

I was pondering what the most cost effective manufacturing technique is for this. I've attached pictures of my current conceptual design.

Also, I have been trying to figure out what the best way to fasten the top to the base is without any noticeable fasteners or tabs. If anyone has suggestions here, that would be great too. It would be great to have a top that snaps into place like the product in the links below. My conceptual design is copying that kind of fit, which I assume is some sort of light interference fit.


 
Replies continue below

Recommended for you

Injection mold it with a snap/bump fit for securement. Hundreds of ways to accomplish the snap fit.. devil is in the details.

 
This looks like a candidate for plastic welding. Does the assembly need to be hermetically sealed (waterproof)? I think you have lots of options here. The method you should choose is dependent on the requirements of the product after it is assembled. Can you provide any more details? I'm happy to help.
 
Unless I am mistaken, molding it with a snap fit may require side action in the mold which would increase cost. I've read that you can have slight undercuts with "bump-offs" but I'm not sure those would hold strongly enough. There are also aesthetic reasons why some of these designs may not work...I want the exterior to be entirely flush and when the box illuminates, I do not want to see any shadows, etc. cast by other internal features. The top does not need to be removable.

I was thinking a press fit would work best. Or simply glue the part into the aluminum groove. See the attached photo for my idea. The reason why I put the bosses on the plastic cover instead of on the aluminum and having holes on the plastic, is because the latter would probably be noticeable when the box is illuminated. Not sure if this would work well, as I'm aware press fits are usually hard shafts into soft material and not the other way around. There may also be a lot of shear stress on the bosses if the cover is impacted. The material I'm looking at is polycarbonate. Any further suggestions on how to fasten to two would greatly be appreciated.

Jason, the assembly does not need to be sealed. How would plastic welding work? Would you have laser cut panels and then weld them together? I looked at injection molding because I know it would result in clean, easy to assemble parts that would be repeatable and good quality every time. That being said, I have no experience with either plastic welding or injection molding, so all of this is new to me.

Ideally, I want a slick looking part that is super easy to assemble. Let me know if you need more info from me.

 
 http://files.engineering.com/getfile.aspx?folder=8ffabcc7-9d7a-4dc3-affc-5d10f04290b1&file=LightBox_Assembly_5.JPG
Sorry, I didn't realize that the base was made out of aluminum. I thought it was plastic like the lens. mcgyver is probably right, a snap fit is likely your best option even though it would probably require action in the injection molding tool. If this wasn't an aesthetic assembly, I would suggest extending the length of the bosses in your latest concept so that they extend beyond the aluminum base. You could then heat stake the bosses to retain the lens to the base.
 
That's an interesting idea...almost like riveting. However, it still requires some extra assembly processes afterwards which I'd like to avoid in order to make assembly cheaper.

I've read that you can have slight undercuts called "bump-offs" that can be used without requiring extra mold action. I may look into that. Hopefully the injection molding suppliers I chat with can give me some more suggestions and perhaps even if there is side action, it won't be much more expensive.
 
Lining up such a large number of plastic studs to corresponding (punched?) holes in an aluminum part will cause tolerance headaches of its own.

In a more producible design, there would be smaller number of studs, and they would be molded in such a way as to be cantilevered from the bulk of the plastic part, to allow some lateral flex of each stud relative to the others. Such a feature would probably be too visible for your tastes, so maybe adhesives are an option, in which case you need some bond area.

Studs of this sort can be molded with fairly substantial snap fits, by providing a biconical taper to a relatively large diameter that ends up beyond the far surface of the mating metal part, and usually by making the studs hollow and splitting the extended hollow part to make it more compressible. The core that makes the stud hollow also allows the stud to be withdrawn from a re-entrant cavity by withdrawing the core first; side draws are not necessarily needed. You need to work closely with your molder and your mold builder to arrive at producible details for this sort of thing, because the mechanical properties of the plastic immediately after molding are quite different from the properties once the part has cooled. In general, you can get away with straining hot plastic a lot more than cold plastic. ... depending on the resin, and especially on the filler.



Mike Halloran
Pembroke Pines, FL, USA
 
Hi Mike,

Yes I was thinking about the tolerances before, but drew that up fairly quickly as a concept. I think if I were to go with the glueing method, I would have two studs purely for alignment, and then have a couple of tabs rather than round studs that fit into loose clearance slots for glueing.

Are you talking about a snap fit similar to the following but hollow and with a slot down the middle? I'm having a little trouble visualizing what you mean.
I'm in the process of finding an injection molding supplier and hope that they'll be able to help me with some suggestions too.

My current idea is to have bumpoffs on some internal tabs of the cover which then snap into some grooves on the sides of the base. I'm thinking of designing it such that I can always glue the part as a last resort without having to modify the mold. However, I will continue looking at the literature to find suitable methods.
 

Non-removable (the part) catches will require "up and aways" - maybe slides.

Removable (rounded features) may be possible just bumping it off. A stripper plate for ejection might be required.

It's all pretty simple really (in my opinion). Depends on your requirements really.

Cheers

Harry

www.tynevalleyplastics.co.uk

It's ok to soar like an eagle, but weasels don't get sucked into jet engines.
 
Right. The part doesn't need to be removable but it also doesn't have to be permanent. If I can design a removable feature such that the part can be bumped off without extra tooling, and it provides enough holding force to prevent having the part pop off undesirably (like if the user picks it up, drops it, or it comes loose during shipping/packaging), then I think I am fine. Probably something like the second feature in this picture:
It may be simple and fairly ubiquitous in plastics design, but I only have experience machining and am entirely new to molding. I have definitely been reading the literature but it still makes me slightly anxious when I don't have experience doing this, and a mold is going to cost me thousands of dollars.
 
Probably something like the second feature in this picture:

Yes, but rounded edges. A small "bead" say 5mm long in two places per side of your cover. Groove in base to match.
Your best bet is talk to a moulder/toolmaker. They undoubtedly would have "been there, done that" already. A good toolmaker should hold your hand during the design process. (Not literally, of course....[wink])

Cheers

Harry

www.tynevalleyplastics.co.uk

It's ok to soar like an eagle, but weasels don't get sucked into jet engines.
 
Harry,

Thanks for your advice & help, really appreciate it. I'll have to calculate out what kind of holding forces I want, but I'm glad to have the idea revalidated by someone with more experience than me.

I will definitely work closely with my supplier...I also know a toolmaker or two so I will be in touch with them as well. Thanks again.
 
You need to talk to a toolmaker who specifically makes injection molds; not just any toolmaker will do, no offense intended to anyone.

If you have a Supply Chain Manager onboard, you are pretty much screwed.
Otherwise, it would be a good idea to pick a molder and work closely with that molder only, for the entire project.

It might even be a good idea to do invest some R&D money into making a sample part that has just the features you need, but is otherwise useless, just as a learning experience for you, and to develop a detail that works as you wish. Put a logo on it and give away a few hundred at a trade show.

It's easier to select a molding partner if you already have a history, so ask around if anyone there has had anything molded in the past, and start from there if possible.

I learned injection molded design from an old German who was pretty much self-taught, and had built his first molding machine himself. Most of his tooling was of bolt-laminated aluminum. He bled all over my first drawing with half-holes and sizable draft.

You might bring prints on erasable sepia paper, and bring a bunch of erasers, or bring a bunch of CAD prints. You need to visit your selected molder in person, get acquainted, and eventually produce two identical markups; one for him to use as a working print, and one from which to regenerate your original.

I'd love to refer you to that particular molder, but he died not long ago, and his wife and son-in-law couldn't keep the business going without him. RIP, Siggy.


Mike Halloran
Pembroke Pines, FL, USA
 
Thanks for all the suggestions everyone. I do know some tool makers who have injection molding experience so I'll be consulting with them.

I've also started looking a little into heat staking as was suggested previously. One added benefit of this is that I can create round geometries on the bottom which are basically "free" bumpers or feet on the bottom of the assembly.

Does anyone have experience with that assembly process and how costly it is? I've inquired with some manufacturers about it but curious what others think.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor