Unscrewing core draft analysis

Boy, that is hard to image what you are doing.

I have designed many "Bottle Cap" thread type things, mostly medical Luer fittings, but nothing which I might call assymmetrical, so I am trying to imagine that.

For short distances (.25 - .50" long threads) & pliant plastics like ABS, PP, PE, I have routinely left zero draft. For things in Polycarboante, Acrylics or Styrenics on Luer Fittings, I tend to leave

0.5 degrees per side draft on the helix cone in the Sweep feature in SolidWorks, and I use a constant thread "Profile", and that has worked well. I would also trust experienced toolmakers who have seen what does and does not work in specific situations.

Beyond that, I would be in experimental territory.

Bo

Do your features have edges or is it highly organic? If you've got edges to play with, then you should be able to generate 3D sketches from some of them and sweep them to make surfaces. Check if the surfaces intersect with your part.

If the part is very organic, you might want to put curves on the surfaces (intersection curves, split lines, or your favorite other method), then sweep them the same way. Careful picking of the curves should give you a pretty good idea how well the core will clear.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"

Ok, when Matt's got a modeling problem, it's a problem...

I think I'd make a dummy screw part, mated to the core, and mated to turn as you drag it back and forth (helical surface mated to sketch point). This would somewhat automate the incremental moves idea you already had. I think I'd also make a surface copy of the cavity to watch for anything poking out as the core moves. Dynamic clearance detection might be useful as the core turns in and out.

If I remember correctly, you're somewhat handy with macros. Drive the position of the core with a distance mate. Write a macro to step through a range of motion with suitably small increments, doing an interference check at each step. Stop at the end of the range, or if a collision is detected. I think this would have to be saved for the last check after manually checking, since it would probably have to run overnight. Perhaps it should record a table of steps where interference was found, so that the whole range gets tested, and then you know where the problem areas are.

Well, those are my ideas, anyway.

matt wrote in news:yDO7g.4959$k%2.3036@trnddc03:

Thanks, guys,

Here's a picture of the part I'm working on.

\images\screwcap.jpg

The part isn't complex, but I'm hoping I can design a part without huge masses of plastic between the finger grips, and I would prefer not to have the separations between grips be as narrow as ribs. I've already considered making the finger grips asymmetrical, but they turn out asymmetrical the wrong direction (they would make it easy to tighten and difficult to loosen).

I know the indentations at the bottom of the core are too tall as shown to be screwed out of the mold, but how will I know when they aren't? I did the bit with the copied surfaces and the move bodies. It's very tedious, but it gives results you can understand. Modeling a solid core and automating movement and interference checking would be useful, but probably a little more tedious, although you could reuse the macro for the next time this problem comes up.

I guess I could do the sweep Jerry suggested and use that to construct faces of the part. That approaches the problem from the other direction, making it a modeling problem instead of the way I've been looking at it as an analysis/verification problem. This is the first time I've ever come up against this. I guess most people just have more common sense and make unscrewed cores completely circular.

I'd need the model to be certain but you should be able to push/blow this part off if it's propylene or PE. The thread would need altering but not to much. What is it going to be molded in?

I'm going to recommend PE. The part is very chunky, lots of material. You're suggesting not unscrewing it, but just pushing it off the core? Yikes. It's a 4-lead screw thread. The OD of the part is about 2.5".

I know and that was my first reaction before I had done one this way on a baby bottle top for Cherubs/Playtex. That stuff was 30 melt PP and unless there is a good reason to use PE I wouldn't. Interesting story behind Cherubs BTW. I think Platex still make the line of Juice cups and bottles but the tools I built in the 80's are definitely toast by now. Little GD hearts 5 color pad printed absolutely every F'ing where on the bottles! Sheesh... The most stubborn part of the project was designing a seal on the juice cup that would seal properly after a couple of months. The baby bottle tops worked like a champ right from the start and I was pleased and relieved not to have ended up with a couple of 8 cavity stainless boat anchors if you know what I mean.

You might also want to consider giving Paul Catalanoti a call at Roehr. This really looks like a good collapsable job if you can't fiddle the thread and he's both a great guy and the King of collapsables and expandables. You can rely on him 100 percent. I've got a number for him here if you want it privately.

John, does Roehr do collapsible tubing per chance.

I designed both the first collapsible tubing used for infant breathing tubing in the NICU (10mm ID) and 22 mm tubing for adult ICU (about 26 mm ID) and a complete extruder corrugator system back in the 80s. We sold that company, and now on a new project, I am looking for the right collapsible tubing supplier again. There were only a few companies doing collaspsible medical tubing I know of so far. I really don't want to even think about building or buying a new extruder corrugator system today, as they have to cost the better part of $300k+ by now, probably from Corma in Toronto

Thanks - Bo Clawson

Could it be a two part solution? The screw out portion with no draft, and a heavy spring loaded internal plunger that goes into the finger area with standard draft. If both portions were made of materials having the same thermal expansion coeffecient, the tolerances between them would preclude flash of any signifigance.

The sprung plunger would only have to move far enough so that when unthreading occurs, there are no clearance issues. Visually, it looks as if about 1/2 the height of the finger knobs would be enough.

Depending on material, I'd definitely look at bumping it off as the easiest first option. I've worked w/ PP, cap diameters ranging from

38mm to 120mm, single thread lead, 1 to 1.5 turns, and thread depth about .040 - .045. Successful bump-offs for the entire product line.

SolidBug, your comment is very interesting and rings true for something I've been trying, though for the opposite of threads.

I'm stripping PP off of undercuts so I can do two parts that snap together with undercuts & be virtually a locked joint that a person can't take apart (think Tupperware that you can't reopen).

Are you able to use a flat stripper plate or ejector sleeve, without a chamfer on the parting line, and still get good parts off without warping and rounding edges?

My first parts were done with a bit of a chamfer to aid "stretching" the part outward to come off an OD undercut without deformation, by stretching about 1.5% (or .0275" in that case) radially, and that worked.

Any help would be appreciated here. I certainly have NOT found much on this technique in print, and have learned by experience to date.

Thanks - Bo Clawson

I should add that in order to keep the product compact, that my joint is "thin" and the undercuts are "sharp" in order to make them very resistant to being pulled apart, as opposed to a normal part with a rounded undercut that is striped off a core with a stripper-sleeve.

Successfully dealing with a thin sharp connection that is stripped is where the tricks lay hidden, particularly when the part is not purely round.

Bo

Matt

I like this idea if the threaded portion can spin or rotate around a post the supports the "Finger Grip" core. How are you going to actuate the threaded core? At 2.5" OD there should be plenty of room inside and it does not look like it needs to be timed to the mold opening.

Interesting problem, I'd like to know how you end up solving it.

Mike Butler Volex Inc.

I think we're going to go with a collapsible and a PP material. It's big enough to do it, the molder suggested it, and it solves our design problem, so everyone's happy.

It was that trying to core the thing out in a helical direction was tough to visualize. Even with a relatively steep thread angle, the angle is too shallow to really do much coring.

Matt, the last Polypropylene (Huntsman 13T25A) part I did with a .0275" undercut on a 3.5" diameter stripped off over a 45 degree angle shoulder without a problem. It was a bit of a gamble, but it worked.

Looking at your part, I can't tell the height of the thread segments. I'll bet if they are below .05" and have a 45 degree ACME thread type side angle on at least the interfering faces, that you ought to be able to strip it fine. Dupont's design manual recommendation is that you should be able to strip a 5% PP undercut. 5% of 2.5" = .0625" on the radius.

Dupont's manual notes the book where they got their recommendations from & the pdf is named "H76838.pdf", and you can find it on Google.

Bo

Dupont did mention that to strip higher % undercuts, you want to keep your mold temperature up to around 105-110 degrees F as opposed to 90 degrees, to keep the elasticity of the part up for the stripping action, as I recall their manual.

I found similar things with my mold.

Bo

Bo,

The thread is about .055 all around, and the angle is far less than 45, more like 15. If it were an outside thread, I'd go for it because it would shrink away from the cavity, but as an inside thread it shrinks onto the core. The molder wasn't crazy about trying to just strip it off. It would introduce some pretty high wear on a surface critical for function. I appreciate the encouragement, but the mold base is too big to use as a door stop or paper weight, we're gonna play it safe with this one and go collapsible.

Thanks for the discussion on this, it has been very useful!

Matt

Bo wrote:

I would agree 15 degrees will not easily slide off. I have succeeded with as low as 30 degrees, but not at 4% undercut.

Given the strength of PP, and its use with threads having a high side angle, like many bottle caps and drum caps, I would bet a hundred bucks that a 45 degree holding face on the thread would hold the cap on the container mating threads just fine with hand tightening and strip just fine off the core, particularly with the multi-lead thread.

Bo

The way to know if this works is to use a single cavity MUD insert set with aluminum plates and just the thread portion, if you want to be quick.

Stick a quicky P20 ejector sleeve in it or buy the stripper plate and just stick on a 35-45 degree chamfer on the threads and shot a few test shots and see what you get.

The savings in mold complexity and cost would be considerable with a simple stripper.

The Cost of an 8.4 x 9 MUD insert set is less than a split core assembly.

Bo