ABS or 30% Glass Filled Nylon feature size

[martin05]

I'm working on a design that requires a ratcheting mechanism to be molded into the parts.

Two parts slide with respect to each other. Each of the parts would have triangular teeth molded on facing surfaces. The parts would click one step at a time as they are slid with respect to each other.

The ratchet is also one way. You can slide (click-click-click) one way but have to press a button to disengage the teeth to be able to slide them the other way.

I would like to make the steps as small as possible. I find myself wondering about what the minimum reasonable feature size might be from a moldability standpoint for something like this. Right now I have them specified as right triangles 0.050in by 0.050in. The pitch or step is also 0.050in.

Strength isn't a huge issue as the assembly probably won't see more than about 5.0lbs applied to the teeth. The ratchet will have about six teeth engaged at all times.

Are there any issues to consider if one were to use ABS vs. 30% GFN.

Any recommendations on a forum that might be more intensely dedicated to plastics engineering and design issues such as this one?

Thanks.

 

[MikeHalloran]

I think you've got the right forum.

The 45 degree pressure angle might be a tad high, but the tooth size seems about right.

I've seen teeth about that size in a medical torque limiting screwdriver, in ABS I think. The teeth were arrayed like holesaw teeth, on the end faces of annular sleeves maybe 10mm OD and 5mm ID. The assembly included a small stainless steel spring to push the teeth together. That device was set up so you could apply essentially unlimited torque to undo the application screws, but the spring, working through the pressure angle of the ratchet teeth, would limit how tight you could make the application screws. The application screws held the electrode wires into a pacemaker, and the screwdriver was set up so a gorilla doctor couldn't overtighten the screws and break something or make the screws impossible to back out; all he had to do was tighten until the ratchet clicked. It only had to work twice, then it was supposed to be thrown away.

I don't think I'd try to make teeth much smaller. For a ratchet and pawl, I'm visualizing a way that maybe you could use two sets of pawl teeth on two pawls to halve the resolution, by offsetting one pawl by half a pitch. ... but I haven't drawn such a thing; it may or may not be possible to actually do.










Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

Does a small cable tie represent your application requirements. They are nylon 6.6 for the best ones, nylon 6 for very good ones ane Pp for cheap crappy ones. ABS would not be my material of choice for a ratchet tooth unless there where other compelling reasons to use it.

Also the glass fibre orientation and distribution to the tips of the teeth might be an issue. Cable ties use no glass fibre.

Other good gear tooth materials are acetal and polyesters.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[MacGyverS2000]

It only had to work twice, then it was supposed to be thrown away.

I'm assuming this is to prevent use on more than one patient... it just seems like it would be more cost-effective to make it a little more sturdy to handle the autoclave and allow more than two uses.

Meh, I suppose someone did the study and those were the numbers that fell out...

I work for a medical device company, and it's staggering the sheer amount of plastic that goes into a landfill for the sake of single use products.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[MikeHalloran]

The thinking on the screwdriver was that it had to work, but it only had to tighten two screws, and maybe remove two, but it absolutely positively had to be there. It was packaged in a bubble with the pacemaker so it wouldn't get lost.

The biggest design challenge to the plastics was that pacemakers are (or were then) predeployed by consignment to surgeons, who were known to store them on the package shelf of a car, possibly for months, so they would always have one ready to implant when it was needed. Car interiors are not a benign environment for plastics, what with high and low temperature cycles and a fair amount of ultraviolet radiation.



Mike Halloran
Pembroke Pines, FL, USA

 

[martin05]

Thanks for all the input so far.

I've attached a few screen shots of a mock-up in order to clarify the question further. For reference, the green part is about 2.5 by 2.9 inches.

You can see the proposed engagement of the teeth on the back of the green cover plate with those on the locking mechanism. Again, the idea is that they would be able to move in one direction but not the other unless the button is pressed.

The comment citing cable ties is an interesting one. My problem is that I don't think that I can mold this out of plain nylon. I need ABS or GFN for structural reasons. I least, that's why I think I need them.

Mike, you mentioned that a 45 degree pressure angle might be too much. Now that you've seen the images does your comment still hold? What would be a more appropriate pressure angle?

Also, the mock-up has the back side of the tooth at 90 degrees to the plate. I would have to tilt this somewhat in order to have a little bit of draft. I was thinking about keeping it to a minimum, like 0.5 degrees. I don't think that would affect the locking effect too badly.

This is not a medical device, BTW.

 

[Pud]

.05 x .05 teeth moulding is trivial. I have seen (and used to have some, but they get easily lost), involute gears 1mm od x 1.5 with 8 teeth and an undercut at one end!

The problem with polymers in thin sections is flow length vs. thickness. Your item will not be a problem. Limiting feature size will be polymer molecule size.

I would suggest that unfilled nylon or acetal might be better for warpage than GF, and ABS, (although the latter being "sort of ok"), as its around the same price but processes slower, so part cost would be the same for improved properties.

H

http://www.tynevalleyplastics.co.uk

Why be happy when you can be normal?

 

[martin05]

Pud, 0.05 x 0.05 is trivial, how much smaller can I get if I went with ABS? How about unfilled nylon? I would love to see the steps in this ratchet be smaller. I think I'd be happy if I could achieve 0.025in steps.

I guess that part of the problem there is that you have to make absolutely certain that the part will not deform in use because a small movement could disengage a tooth that is only 0.025in high.

Thanks!

 

[Pud]

.025in (jeez, that's over .5mm!) will be ok in anything. your problem will be keeping engagement - as you alluded to. A high stiffness material will be the order of the day!

Cheers

H

http://www.tynevalleyplastics.co.uk

Why be happy when you can be normal?

 

[MikeHalloran]

You need to analyze how much force it will take to deflect the spring/hinge via the ratchet ramp using axial force between the red and green parts in order to evaluate the pressure angle. That's sort of a human factors/ interface thing that you have to fuss with to get the feel you want. That should lead you to an effective spring rate that you need to have at the ratchet teeth.


I see some problems with the spring/hinge, by the way.

When the ratchet rides up and clicks, the teeth rotate around some point in the hinge, so they don't all have the same depth of engagement as they are riding up on mating teeth. IOW, all the force to flex the hinge comes from the teeth nearest the hinge, only. The others are just along for the ride. You need to take a side view section in both conditions in order to appreciate what's going on.

I don't think combining the hinge and the spring is necessarily a good idea anyway. A detailed force analysis may lead you in a different direction. Perhaps a completely separate, e.g. steel, spring, would be needed. Perhaps you could mold the spring as a longish cantilever bar that's twisted up over an adjacent hook at assembly to get a preload. Or perhaps you could support the moving tooth plate with four parallel cantilevers leading from its corners to the undeflected plate.

Something like this:
+===========+ +=============+
+---------------------+
| |
| || || || || || |
| |
+---------------------+
+===========+ +============+

So the ratchet plate could remain parallel to its initial plane as the teeth rode up and over their mates, all more or less in unison.



Mike Halloran
Pembroke Pines, FL, USA

 

[martin05]

Yes. I did see the fact that the tooth closes to the hinge will work harder. I just wanted to mock something up quickly in order to get a sense of proportions and where things can be fitted. I'm tempted to try for a topology that might make motion a little more parallel rather than heavily circular. One option is to move the teeth as far as they can go and effectively increase the radius.

 

[MikeHalloran]

My attempted ASCII art conveys my idea better in a monospaced font; copy the text in Notepad or something similar to see it as intended.

Or, let my try tgml:

[tt]
+===========+ +=============+
+---------------------+
| |
| || || || || || |
| |
+---------------------+
+===========+ +============+
[/tt]

Yeah, that's better.


Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

If you intend moulding in a spring action, unfilled acetal is the best thermoplastic material you can possibly get. It also makes good teeth and fills details in the mould fairly well. Other physical properties tend to match about 15%GF nylon.

Nylon is the best for conformance to intricate mould detail and will flow a short distance into an 0.001" gap. This is one of it's problems at times as the mould requires very high precision at parting lines and pin holes to avoid flash. I have seen flash on a pin that is so fine that it is invisible to the naked eye, but it forms a suction cup that causes the moulding to stick quite tightly to the pin. From memory, nylons typically flow about 300:1 re length to thickness, but that of course largely depends on the specific grade and conditions.

Oh and as mentioned and recognised, you hinge design as such really does suck. It needs to be a lot longer at least. At best it would be a third part wit a matching wedge or shafts into angled slots and sliding action.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[martin05]

Thanks for your feedback. I'll look into acetal.

Again, the model I uploaded is one in a series of thinking tools, if you will, that I am quickly throwing together in Solidworks to eventually converge on geometry that will be integrated into an actual design starting from a fresh clean file.

I am taking a look at cable ties as an interesting example of a simple way to do this that seems to work very well.

 

[martin05]

This is interesting:

http://upload.wikimedia.org/wikiped...rossSection.JPG/1280px-ZipTieCrossSection.JPG

 

[MikeHalloran]

Okay, I recognize that it's a temporary talking document.

Do look at cable tie ratchets. Of the ones with plastic molded pawls, note particularly that the pawl cantilever is quite short, and generally deforms inelastically (takes a set) once the pawl teeth have ridden over just a few of the tail teeth. That's probably not what you want. You need a longer cantilever, which will probably need to be thicker in order to generate any ratcheting force.

Basically, you need to select a material, guess at a geometry, and run a cantilever beam calculation, balancing the force and the deflection while keeping the stress below what corresponds to an elastic limit for plastics. You don't need FEA; hand calcs will do, but putting them in a spreadsheet will make it easier to screw around with the parameters, a lot, before you bother to put the geometry into CAD.



Mike Halloran
Pembroke Pines, FL, USA

 

[martin05]

Thanks for the clarification on the design calculation methodology. It's far too easy to fire-up FEA. I've learn to be very careful about the results. I use it mainly for thermal management simulation where it can be useful. I'll go the spreadsheet route and see what kind of trouble I can get into.

I'm wondering if something like this can be 3D printed for verification. Maybe the newest machines can resolve features this small?

 

[patprimmer]

Remember most cable ties are designed for single use, although there are exceptions.

We still have no idea how many use cycles, load, temperature, food/biological contact, UV exposure, dishwasher exposure, sterilisation, freezing etc apply here. Each can have a significant impact on material choice.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[MikeHalloran]

I may have missed a meeting, but I don't think any 3D printing technology can give you a representative part at the geometry you want. ... yet.

The material choices are limited, so the mechanical properties will be different in addition to being nonisotropic because of the laminar deposition, so the forces won't be right, and the Z-axis resolution is still too coarse to get the surface finish, hence the 'feel', right.





Mike Halloran
Pembroke Pines, FL, USA

 

[martin05]

OK, so I guess that the best option I may have is to do a test part that Protomold can handle and run it that way. The "how it feels" part of things is hard to calculate or simulate.