Nomad work holding

I’m looking for ideas on a pallets or mass workholding or fixture plate or something, that fits on the nomad and can handle 38mm diameter rounds 1mm thick for diamond drag engraving. I’m making them one at a time now, but I want to make a production run of it.

My first experiment was with pockets milled out of aluminum plate and eccentric nuts that partly intrudes in to the pocket. Unfortunately it is proving to be extremely difficult to use that. The rounds i bought or not precision and can be out of round by as much as as 0.48mm which means I need an over sized pocket and a lot of force on the eccentric nut. I’m having the rounds spin and occasionally pop out.

Is there a better way to do this? I make about 150 a month so doing them one by one is excruciating.

1 Like

Are you engraving the entire surface of the round stock?

1 Like

an idea of where you can hold on the round stock would be helpful. are the lengths consistent?

thinking you can compress the ends if there is room for the workholding and the stock.

perhaps in combination with some sort of rubber backed straps that prevent spinning and marring up the surface.

Lay out as many circles as will reasonably fit — cut them out slightly undersize using a dovetail cutter.

Run a channel across each of them which is deep enough to make the part flex enough to insert the rounds.

machine holes in-between the circles which match up with holes to secure this in place.

these are disks 1mm thick or I guess “long”.

Not sure I follow you, I get the under cut circles, but will the channel run across the diameter? Not at all sure where the holes are going or what they are for.

Again if this wasn’t clear these are round disks with edge to edge engraving.

ah gotcha, sorry for some reason i was picturing tubes from initial description. haha.

i’d do a full bed size plate with as many circles as Will mentions. id go with cheapest mic6 plate or comparable to start with a nice flat stock fixture plate.

step 1:
do all the reliefs to allow as much of the stock (discs) to rest inside as feasible.

step 2:
3 or 4 arced slots surrounding the relief. evenly spaced from each other and deep as possible.

step 3:
in the center of these slots bore for either a countersunk bolt or a tapered plug (pipe fitting). thread and tap or helicoil (for bonus points)

the slots become flexures surrounding the stock. as you thread in the pipe fitting or countersunk bolt they expand/deform the plate into the discs. I would trial this out on a small bit of stock on a single part setup and once you have it dialed in expand it into the full bed plate. adjusting the offset distance and number of slots should help with the variance in stock you’re dealing with. quick and dirty model to reference. beauty is at most youre into basic bolts and a tap aside from the plate.

3 Likes

Well it’s pretty brilliant, any chance I could do this with regular 6160? Mic6 in 8x8 is over a hundred dollars.

there’s some brilliant guys on Youtube and IG under the #instamachinist tag that have brought this design to my attention

yes, any ol’ stock will suffice. i find it nice to have minimal cleanup/facing needed on high production fixtures. mic6 is gummy so you would save some grief going otherwise.

On round stock, doing a circular cutout in your workholding will almost always lead to the part coming out. If you constrain a circle in a larger circle, there will only ever be two points of contact (the clamping point and the opposite wall between workholding and stock.) Two points of contact is not stable and highly likely stock will shift loose. If the stock sides and fixture sides aren’t parallel to each other, this workholding situation gets dramatically worse.

For round stock, aim for 3 points of contact, so think of doing a v-block or better yet, two pins.
120 degree v block design, which is going no where. 3 points of contact always which prevents the parts from sliding or rolling. (Well, ‘always’ as long as it’s within the throw of the clamp, these are Mitee-Bite clamps shown.)

The downside of machining your own workholding is endmills have rounded corners, whether by design or worn down my usage. Those rounded corners can make the actual contact point between stock and workholding very questionable. To get around that, bore two holes for pins and now there is a sharp corner between fixture plate and vertical wall of the pin! Second advantage is there is a steel wall now which will take decades to wear down at 150 parts a month. Magical pins shown below in black.

5 Likes

I’ve noticed. So in comparison to the shrinking circle method this one has less fixturing and I can fit more parts per pallet. But Mitee-bite clamps cost more than anything I’ve ever seen which is that small. :slight_smile: They should be used in place of diamonds for elicit transactions.

I like the pin idea if I can get a clamp with enough force. I thought about pins way back but was worried I wouldn’t have enough force pressing into them to constrain the piece. Which is why I suppose you used the mitee-bites. I suppose I can cut a recessed radius or angle in to the pins, sort of making them soft Jaws.

1 Like

Haha, ya they are frustratingly expensive but once you bite the bullet and buy a set, you will use them on all your projects. I got a set of 10 and then just get installed on whatever I have on the table.

Ok but seriously, thin flat stock, the right approach is vacuum fixture. Someone needs to do this on a Nomad.

As mentioned, vacuum would be very useful here.

I was wondering if you are engraving the whole surface of the ‘coin’?

If not, is it possible to cut an overlay sheet of (say) 36mm circles that sandwiches the coins between it and bed of 38mm recessed circles? The overlay could be aluminium with a rubber/silicon mat on the coin-facing side. There should be enough pressure there, if you bolt the top down firmly enough, to stop the coins rotating.

I actually tried that, but did not get too far. The amount of vacuum is unsurprisingly limited by part fit and surface finish. I made a model out of HDPE and never got it hooked up correctly. I’d leak vacuum something awful. I never got past the demo phase with that. I’d be happy to share my set up if people think that it would be helpful.

Pretty near yes. I can’t seem to eliminate all my repeatability errors of over about 11inches (the diagonal distance between the opposite corner parts) I’m getting about 1.5 mm drift. I can compensate for that by making and the engraved surface smaller than the coin.

Honestly, this would be my preferred solution, but I need to chase out the repeatability errors before it could realistically work.

Hmmm… do you get the same drift if you run your job with a solid piece of something covering the whole nomad bed? Maybe try using a 8x8 bit of MDF and use a pen instead of the (spring-loaded?) drag engraver.

1.5mm drift?

Could you elaborate? Do you mean random/unpredictable variation? or is it repeatably incorrect?

I’d love to see it it you’re willing to share! Did your fixture contain an o-ring gasket inside the extent of the part profile? That will make a huge difference in the ability to seal as well as increasing friction between part and fixture. About a buck each, even on Mcmaster for this size.

Since you are drag engraving, are you sure the side loads aren’t being significantly impacted by stock thickness variation?

Yeah… I thought this too - or perhaps there’s an problem with the source design when it was tessellated.

My Nomad doesn’t make this sort of error. Perhaps engraving a solid sheet of material that’s bolted to the bed would add a useful datapoint?

It’s all user error :expressionless:.
I made a fixture plate? Jig? That takes 16 disks at a time and the first 4 are very good but by the time I make it to the opposite corner the error is magnified.

I think part of the issue is the rapids when zeroing the nomad using carbide motion somehow operate on a different increment than when using the 1mm advancement. The options are “fast” which if I’m doing my math right uses inches?, 1mm, .25mm, and .025mm. After a “fast” move I’m often as much as .2mm off of where I want to be. With the movements set at 1mm, .25mm, .025mm I can’t reach the machine coordinates I need to. That might be repeatable enough to compensate for in the gcode but I have other projects to do and taking on and off the fixture plate always seems to move things around.