Fixturing/Workholding/etc for Plastic Enclosures

[forrestc]

For various cost and inventory reasons I need to bring the machining (hole cutting) of the stock plastic enclosures we use in-house.  I have all of the pieces in place (mostly a CNC mill), except for production workholding/fixturing.

I've been testing just by using a machining vise but that's a pain in the rear and doesn't hold these vertically all that well, and isn't that fast to change things over.    I'm thinking about some pneumatic or mechanically driven clamp, but don't really know where to start.   I've seen these types of jigs in use, but haven't built one myself.   My goal is to have something which we can drop an enclosure into and the fixture takes care of ensuring that it is positioned and held correctly on the X and Y axis.

I've been perusing various workholding/fixturing sites but can't seem to come across anything which feels right.   I've also been looking for various books/instructions to give hints, along with perusing youtube videos - but most people are milling things out of blocks of metal which isn't really the same unfortunately.    So far, I just don't feel like I've found enough information about what options I have.   I'm hoping that someone on here has been through this and can point me at least toward the resources I need to figure this out.

I'm pretty good at building something out of available resources, but in this case I just don't seem to have the right set of resources to even get started.

Ideas?

[Nauris]

Try vacuum. It is fast and strong enought for plastic.

Depending on the shape of enclosure you may be able to cast suitable fixture from epoxy or gypsum using your enclosure as mold. Or mill something from a block of aluminium and drill vacuum channels there.

[reboots]

I faced the same situation. I made a fixture from three aluminum plates, screwed together to form a corner with three reference surfaces. I mounted toggle clamps positioned to hold the enclosure in that corner, clamping horizontally and/or vertically as needed to counteract the cutting forces of a given operation. Toggle clamps are available from a number of manufacturers, with generic versions all over eBay. The clamps can be adjusted to fine-tune a compromise between sufficient holding force, and distorting the enclosure.

https://www.destaco.com/manual-clamps.html

I added a zero mark to the fixture, and offset the CAM job from that mark to match the enclosure clamped in the fixture. That way the fixture can be removed and reinstalled on the table without concern for repeatability: just re-zero the machine to the fixture.

One complicating factor was that the polycarbonate enclosures I was machining had poor manufacturing tolerances, and dimensions were slightly inconsistent. I addressed this by locating the job features with enough clearance that a couple mm of offset wouldn't scrap the part, and offsetting machine zero to split the variation in geometry I encountered. You'll want to watch for the same issue.

Once you have a working process for one workpiece, you may want to consider building a gang fixture so multiple pieces can be machined at once.

[Edison]

Plastic parts with flat surface - handling, gripping, piece replacement, easy operation = vacuum   

[forrestc]

I'm going to have to give this some thought.

This particular enclosure ( https://www.polycase.com/ek-50mbt - the side we're machining is opposite of the clip ) is likely to be a challenge, although there is definitely one side we could vacuum to...  would need to have a vertical fixture with vacuum holes in it, but I think that would be possible.

Oh, Now I think about it, this probably  isn't going to work, since there are two sides of this case which need to be held together during the machining process.  The one side would be fine for the vacuum, the other wouldn't be held at all, so we still need clamping.

However, I do have another case which vacuum will likely work just fine for.....  Thanks for the reminder.

[forrestc]

I faced the same situation. I made a fixture from three aluminum plates, screwed together to form a corner with three reference surfaces. I mounted toggle clamps positioned to hold the enclosure in that corner, clamping horizontally and/or vertically as needed to counteract the cutting forces of a given operation. Toggle clamps are available from a number of manufacturers, with generic versions all over eBay. The clamps can be adjusted to fine-tune a compromise between sufficient holding force, and distorting the enclosure.

Did you just use aluminum stock for this, or did you use some sort of pre-machined plates?   I'm envisioning a couple of chunks of aluminum drilled and tapped as appropriate with screws...

Did you by chance use clamps on two axis at the same time, and if so, did you have any problems with the enclosure moving out of the corner (or not moving into the corner) as each was applied?

[Edison]

Again, it can be solved with a vacuum - or pressure - controlled clamp - Easy grip pressure adjustment without deforming the workpiece 

[forrestc]

Again, it can be solved with a vacuum - or pressure - controlled clamp - Easy grip pressure adjustment without deforming the workpiece 

Would you care to give me a hint on how to build such a device to hold the enclosure shown at  https://www.polycase.com/ek-50mbt such that the clip is down, the opposite face is up and the two sides are held together?  (See the 2d .pdf if you need more details).   

Vacuum just plain isn't going to work that I can see.  I could hold one side in place, but not the other.  When you originally said vacuum (without mentioning pressure), that is what I meant by it wouldn't work.   If I had a single piece case or was machining only one half this would probably work.   But since I want to machine both pieces together as one, I'm going to need that pressure portion.  And no, I'm not going to machine them separately as that is just asking for misalignment issues of each half to cause cosmetic issues where they join (I'm machining across the joint).

I understand the ability to use an air cylinder to supply the holding force.   That isn't my problem.  That's easy.  I can go to any one of  dozens of different sources and get one of a dozen different types of cylinders, air bags, etc.   I can get air controlled clamps, or if I want to dispense with air, the simple mechanical ones as well.   What I'm having issues with is how to take that force and set up a fixture to apply the force in a way which consistently shoves the enclosure into a known position (i.e. up against a mechanical stop in X and Y and ensures that both the enclosure is held together and is also parallel to the X axis).

I've been considering an pneumatic vise/double cylinder with largish jaws (wide and tall), but that only helps in one of the two other axis.  I can set up an end stop on the other axis but that requires staff to consistently push the enclosure up against the stop (something I can't ensure they'll do consistently based on past experiences, and lean tells us we should idiot proof things where possible, aka poka-yoke).   

I considered adding some additional guides to create a pocket so that the enclosure has to fit between them before clamping, but the cases vary enough that this is problematic - the tolerances are such that many of the cases will be rather loose in that direction, creating potential issues when machining.

I briefly considered using a cylinder on a corner where you have two L shaped Jaws and the cylinder pushes the second one diagonally toward the first one so that both sides are clamped.  But this seems to require more engineering than I'm prepared to exert on this problem (the non-fixed jaw has to be able to float around a bit to be able to clamp on the slight range of sizes for the enclosures).  This floating creates a range of issues including how to hold the jaw such that it can slide and float around in the range it needs to slide around in, but still constrain it such that it doesn't flap about. 

And I keep coming back to the thought that there really should be an 'erector set' type of parts for doing this type of work, which would allow me to try out things until they work.  But I haven't found them yet, so I'm faced with expensive trial and error.

So, please, if you have a magic bullet which just allows me to somehow use 'vacuum or pressure' to hold this case let me know more details...

[Cerebus]

As this is a DIN rail mount case, the immediate thing that comes to mind for me is to make a jig based on a length of DIN rail. Find or machine something that will slide onto the DIN rail either side of the case to act as clamps. Fix one of those in place permanently to give you your datum and slide the other on after your work piece as a clamp and hold it in place with 1 or 2 quick clamps.

BTW, something to always look out for when planning jigs to hold any cast or moulded parts - draft angle. Almost always the sides of moulded parts are not square to the faces, but have a draft angle on the order of single digit degrees. It's small enough that you may miss it in a casual, sometimes in a careful, visual examination and find that parts don't fit your jig because you assumed some side would be square (i.e. at 90°, orthogonal), when in fact it is not. I've made this mistake myself in the past.

If you've got a really awkward shaped part to hold you can make a caul out of polycaprolactone beads. These melt at 60°C and can be moulded around your part with gloved hands to form a solid, form fitting caul. You can then cut, drill or tap the caul to fit onto the jaws of a clamp or to some other surface. When you're done with it, melt it down and reuse it.

[tpowell1830]

For plastic cases as you've shown, wooden jigs. They are easy to make and reusable many times.

Hope this helps...

[forrestc]

BTW, something to always look out for when planning jigs to hold any cast or moulded parts - draft angle. Almost always the sides of moulded parts are not square to the faces, but have a draft angle on the order of single digit degrees. It's small enough that you may miss it in a casual, sometimes in a careful, visual examination and find that parts don't fit your jig because you assumed some side would be square (i.e. at 90°, orthogonal), when in fact it is not. I've made this mistake myself in the past.

These cases have a bit of this on the 'short edges' - i.e. the perimeter that I'm machining one side of.  It causes them to not be exactly flat along this side which complicates things a bit and makes it a bit more likely for a misalignment along one axis.

I'll think about using the din clips/rail, that might be useful.   And I'll have to get some polycaprolactone beads to play with.  I'm always looking for some sort of molding compound which will work for this type of stuff, and somehow didn't know about this.

[forrestc]

For plastic cases as you've shown, wooden jigs. They are easy to make and reusable many times.

I don't know why I don't fall back on wood or wood-like jigs.  I'll have to experiment with some since that's a lot easier to work with than cutting a block of aluminum.

[forrestc]

BTW, something to always look out for when planning jigs to hold any cast or moulded parts - draft angle. Almost always the sides of moulded parts are not square to the faces, but have a draft angle on the order of single digit degrees. It's small enough that you may miss it in a casual, sometimes in a careful, visual examination and find that parts don't fit your jig because you assumed some side would be square (i.e. at 90°, orthogonal), when in fact it is not. I've made this mistake myself in the past.

These cases have a bit of this on the 'short edges' - i.e. the perimeter that I'm machining one side of.  It causes them to not be exactly flat along this side which complicates things a bit and makes it a bit more likely for a misalignment along one axis.

Hey, replying to my own post, it seems to me that I might actually be able to exploit this draft angle to create a 'self centering jig'...  basically cutting/molding/3d printing a pocket that the draft angle slides into.  I'll have to do some experiments to see if this is workable.

Oh wait, there's also the screw holes I could use with pins to align the case more precisely and prevent 'upside down cutting'......  Hmmmm.....   Yeah,  This might just work...

[Cerebus]

For plastic cases as you've shown, wooden jigs. They are easy to make and reusable many times.

I don't know why I don't fall back on wood or wood-like jigs.  I'll have to experiment with some since that's a lot easier to work with than cutting a block of aluminum.

The problem with wooden jigs for some tasks is dimensional stability. Most wood is about 14% water in its 'dry' state at 50% RH and is, of course, an anisotropic material so will change more or less on different axes as it dries out or gets wetter. Sometimes it's not a problem, other times it is. A hot machine shop with cutting fluid splashing about would be the least favoured environment.

The  most dimensional stable wooden material in common use is birch plywood and is actually pretty damn good, thankfully cheap, and easy to work. Generally softwoods bad, hardwoods good. Most of the best, most stable hardwoods are so rare nowadays that they are subject to CITES rules (e.g. Mahogany, which once was the classic pattern maker's wood in use in foundries world wide).

[Cerebus]

These cases have a bit of this on the 'short edges' - i.e. the perimeter that I'm machining one side of.  It causes them to not be exactly flat along this side which complicates things a bit and makes it a bit more likely for a misalignment along one axis.

I spotted the draft on those cases, which is exactly what prompted me to mention it. I'm probably telling Grandma how to suck eggs here, but the sloping faces where you're machining it are exactly the kind of thing that makes a drill wander off unless it's a short stubby tool like a centre drill; it you're drilling them it's a good idea to create a flat spot with an end mill first before drilling. All depend on your tolerances if you can get away without doing this. Machining 101, I know, but you never know how much of that kind of stuff gets passed on to the EE types.

[forrestc]

I spotted the draft on those cases, which is exactly what prompted me to mention it. I'm probably telling Grandma how to suck eggs here, but the sloping faces where you're machining it are exactly the kind of thing that makes a drill wander off unless it's a short stubby tool like a centre drill; it you're drilling them it's a good idea to create a flat spot with an end mill first before drilling. All depend on your tolerances if you can get away without doing this. Machining 101, I know, but you never know how much of that kind of stuff gets passed on to the EE types.

Thanks for mentioning this, as I'm definitely not a machining expert,  just have done it enough here and there that I've picked up things along the way, so it's little things like this that tend to trip me up here.  Well, that and workholding once you get beyond 'clamp it in a vise'.

Fortunately, what isn't shown in the picture is the flat area on the side  that I'm milling.  The 'front' of the enclosure is actually flat and smooth.   Actually now I look at an actual enclosure the draft angles are only on two sides, the other 4 are square - and one of them doesn't have any texture which is where the connectors go. 

After sleeping on this, I think what I'm going to try is machining a plate (probably a 3d printer or laser cut prototype first) which has tapped holes for alignment pins which fit into the screw holes on the bottom of the case which hopefully are relatively tight toleranced (they seem so with some plug gauge tests I've done on the half-dozen cases I have floating around here).   That should handle x and z alignment and solve the 'staff put the case in backwards issue'.   I'll then apply some sort of pressure on the case (toggle clamp, half of an open bed vise, dual rod pneumatic cylinder, etc) to push it up against the machined plate and as a side effect gain Y axis alignment and hold the case together.    I also think I've got a couple of extra inputs on the machine controller that I should be able to use to do automatic jig alignment so I'll keep that in mind while I'm working on this.

Thank you everyone for discussing.   I'll try to remember to report back when I've either been successful or ready to try something else....

[Cerebus]

Fortunately, what isn't shown in the picture is the flat area on the side  that I'm milling.  The 'front' of the enclosure is actually flat and smooth.   Actually now I look at an actual enclosure the draft angles are only on two sides, the other 4 are square - and one of them doesn't have any texture which is where the connectors go. 

Ah, I think I misunderstood which bit you're machining, but never mind, it looks like I inadvertently passed on some good advice anyway.

[mikeselectricstuff]

How about filling a case with resin to make a close-fitting mould to place cases on for machining.

[Edison]

gripping jaws need to install pressure sensors
ensures timely shutdown, no deformation of the workpiece
by sensor options: piezoelectric - continuous adjustment of gripping force
: mechanical (microswitch) - limited setup options - cheaper design

[reboots]

I faced the same situation. I made a fixture from three aluminum plates, screwed together to form a corner with three reference surfaces. I mounted toggle clamps positioned to hold the enclosure in that corner, clamping horizontally and/or vertically as needed to counteract the cutting forces of a given operation. Toggle clamps are available from a number of manufacturers, with generic versions all over eBay. The clamps can be adjusted to fine-tune a compromise between sufficient holding force, and distorting the enclosure.

Did you just use aluminum stock for this, or did you use some sort of pre-machined plates?   I'm envisioning a couple of chunks of aluminum drilled and tapped as appropriate with screws...

Did you by chance use clamps on two axis at the same time, and if so, did you have any problems with the enclosure moving out of the corner (or not moving into the corner) as each was applied?

I used scrap aluminum stock. Thick acrylic or other plastic would also have worked. But I prefer your idea of threaded pins, actually.

I clamped horizontally only in the X direction. Certain milling-through operations would sometimes pull the enclosure up out of the jig, so I added a vertical toggle clamp for those.

FWIW, vacuum fixturing would likely not have worked for me. The enclosures have a number of mounting and knockout features, and not much flat area.

Beware of using the draft angle for self-centering. If your enclosures aren't completely dimensionally consistent (not unlikely for cast plastic), any given enclosure may ride higher or lower in the jig and you won't have a good Z reference. Might not be a problem for drilling or milling through.

[rcourtney]

Edison's suggestion is what I used. 

Clamps can damage the surface of the plastic box.   

A block of aluminium is milled to form a guide.  Recessed cavities, top to align the plastic box, bottom for a sealing plate.
Holes drilled between the two cavities allows vacuum.  A silicon gasket seals the bottom plate.   A barbed hose nipple is screwed
into the side of the block.

Dove tails were milled into the block to align and secure the holder to the mill table.

Once secured and zeroed in, the operator simply places the plastic box into the jig and applies vacuum.

[Edison]

I like to help, it's good that it works