Spit balling fixturing ideas

Zahnrad Kopf · Aug 6, 2017

We've been tasked with the possibility of modifying some parts for a customer and after much contemplation am looking for more ideas on fixturing a part.

The part is basically a sheet metal stamping, and not flat, having offsets and perpendicular features that prevent simply laying them down and clamping.

Plus,... there might be 3,000 of them to do. It could even be as much as 6,000. We do not know yet. Discussions have me thinking it will be at least 3,000 though.

The modification itself is rather straight forward, and simple. Take .010"-ish off the highlighted face in the picture. So, obviously, I would like to be able to stack them and whack them. But their geometry is not very friendly for that. this leaves me thinking about how to best get the biggest bang for my machine-time buck, and considering a FAST method of fixturing as many as possible ( within reason ) to optimize our efforts.

Ideally, I would like to stack them side by side, so that the surface to be machined is parallel to the table, and an end mill can walk right over and by, hitting as many as possible in a pass. However, I am also open to laying 30 or 50 or so down, clamping, and using a small end mill to side cut them. Obviously, some ways are going to be more efficient, but I"m not ruling anything out so far.

Machine will likely be a machining center. ( 4ax capable if someone has some wild idea )

Any thoughts from out there in PM-land?
( for idea of scale, the thickness of the parts are ~ .100" and they are ~ 4.400" long )

Erich · Aug 6, 2017

Can you hold tolerance if the parts reference on the bottom surface below the central hole?

petersen · Aug 6, 2017

Are the holes precise enough that you can stack them on a rod with spacers between
them to allow the formed end to clear??

Stack as many as you can hold in a vise and another rod thru one of the other
holes for alignment.

WILLEO6709 · Aug 6, 2017

I would wire edm contour some jaws, put some pins in, slide the part on, clamping only on what you cut so you don't fubar the rest of it.... have an extended jaw with the other pins just for angularity. In the end it would be cheaper to fix the die, scrap these, and run new ones.

PeteM · Aug 6, 2017

Locating from the holes could speed things up; but it seems to me the dog leg will present stacking problems and require a spacer if you thread a number of them up on rods for a single pass in your machining center..

Another option would be to stack one part vertically on two locating holes then have a sliding fixture to run it into a mill, grinding wheel, sanding disk etc. with a stop set to take .010" off. I'd think the cycle time would be under 6 seconds each, 10 a minute, 600 an hour, five hours for the 3000? I doubt you could stack them in a fixture a whole lot faster than that.

Monarchist · Aug 6, 2017

Zahnrad Kopf said:
We've been tasked with the possibility of modifying some parts for a customer and after much contemplation am looking for more ideas on fixturing a part.

The part is basically a sheet metal stamping, and not flat, having offsets and perpendicular features that prevent simply laying them down and clamping.

Plus,... there might be 3,000 of them to do. It could even be as much as 6,000. We do not know yet. Discussions have me thinking it will be at least 3,000 though.

The modification itself is rather straight forward, and simple. Take .010"-ish off the highlighted face in the picture. So, obviously, I would like to be able to stack them and whack them. But their geometry is not very friendly for that. this leaves me thinking about how to best get the biggest bang for my machine-time buck, and considering a FAST method of fixturing as many as possible ( within reason ) to optimize our efforts.

Ideally, I would like to stack them side by side, so that the surface to be machined is parallel to the table, and an end mill can walk right over and by, hitting as many as possible in a pass. However, I am also open to laying 30 or 50 or so down, clamping, and using a small end mill to side cut them. Obviously, some ways are going to be more efficient, but I"m not ruling anything out so far.

Machine will likely be a machining center. ( 4ax capable if someone has some wild idea )

Any thoughts from out there in PM-land?
( for idea of scale, the thickness of the parts are ~ .100" and they are ~ 4.400" long )

A stack of bespoke spacers. That simple. Rectangles on a bar or rod, Abacus-bead style, resilient openers between to speed loading and unloading when clamping force is released. Complex features left with nought but ignorant air between.

Low-stress at one end. That for alignment, only. Tight clamping force of the stack only where the cutter(s) have to apply force.

Even so.. I'd bet a GOOD jig and a small manual, face mill cutter, lever operated, ELSE air/hydraulic cycling and a foot-pedal, could do onesies, hand loaded, fast and cheap nearly as rapidly as one could load and unload a batch for the CNC rig.

Small, light, one hand pick-up and aside, not "big numbers" a few thousand at a go, once the set-up time is moved into a jig or fixture with respectable repeatability & Idjut-resistance.

Just boring as all Hell for the operator.

Part that small, some among us us might even have sheared or shaved 'em on a foot-press back in the day.

"When the only tool you have is a CNC mill, the whole world looks like.."

Zahnrad Kopf · Aug 6, 2017

Erich said:
Can you hold tolerance if the parts reference on the bottom surface below the central hole?

Thank you for asking, and for pointing out my omission.

The surface does indeed get referenced from the holes. So, locating from them ( or at least referencing them ) will be a necessity.

petersen said:
Are the holes precise enough that you can stack them on a rod with spacers between them to allow the formed end to clear??
Stack as many as you can hold in a vise and another rod thru one of the other
holes for alignment.

... and ...

PeteM said:
Locating from the holes could speed things up; but it seems to me the dog leg will present stacking problems and require a spacer if you thread a number of them up on rods for a single pass in your machining center..

... and ...

Monarchist said:
A stack of bespoke spacers. That simple. Rectangles on a bar or rod, Abacus-bead style, resilient openers between to speed loading and unloading when clamping force is released. Complex features left with nought but ignorant air between.

Low-stress at one end. That for alignment, only. Tight clamping force of the stack only where the cutter(s) have to apply force.

Yes, this is exactly what I have been leaning toward. My reservation being that one does not really get a great many in attendance per cut and I would like to be able to minimize the time these are here, running.

Ideally, I would LOVE to run 50 at a shot. :eek:

PeteM said:
Another option would be to stack one part vertically on two locating holes then have a sliding fixture to run it into a mill, grinding wheel, sanding disk etc. with a stop set to take .010" off. I'd think the cycle time would be under 6 seconds each, 10 a minute, 600 an hour, five hours for the 3000? I doubt you could stack them in a fixture a whole lot faster than that.

Hmmmm... That is interesting. I will have to give that some more thought. Thank you.

WILLEO6709 said:
I would wire edm contour some jaws, put some pins in, slide the part on, clamping only on what you cut so you don't fubar the rest of it.... have an extended jaw with the other pins just for angularity. In the end it would be cheaper to fix the die, scrap these, and run new ones.

I agree wholeheartedly. And they are. But I still have to fix these for them. We do what we are told. Eh? I can't always make gears, Surgical prototypes, and Nuke research parts... :cool:

Monarchist said:
Even so.. I'd bet a GOOD jig and a small manual, face mill cutter, lever operated, ELSE air/hydraulic cycling and a foot-pedal, could do onesies, hand loaded, fast and cheap nearly as rapidly as one could load and unload a batch for the CNC rig.
Small, light, one hand pick-up and aside, not "big numbers" a few thousand at a go, once the set-up time is moved into a jig or fixture with respectable repeatability & Idjut-resistance.
Just boring as all Hell for the operator.

Well, that IS the issue... One man band and all that...

Monarchist said:
"When the only tool you have is a CNC mill, the whole world looks like.."

When you're right; you're right.

sophtayl · Aug 6, 2017

Zahnrad

Can you stack them 180 degrees (plan view) so that the surfaced to be machined are aligned. Then make a pin assembly to locate them where you load them in alternate configuration and a small spacer to make up the difference?

Similar to above but may get a bit more density

Sent from my SM-G930F using Tapatalk

Davis In SC · Aug 6, 2017

WILLEO6709 said:
I would wire edm contour some jaws, put some pins in, slide the part on, clamping only on what you cut so you don't fubar the rest of it.... have an extended jaw with the other pins just for angularity. In the end it would be cheaper to fix the die, scrap these, and run new ones.

But less profitable for the shop.. A few years ago, I had a customer needing a few thou trimmed off a few hundred plastic parts while the mold was being modified. I machined a fixture in place and used a pin and rubber band to hold 4 parts in.. Emergency price was a dollar each, had to finish that night. Over the next few weeks, I did over a thousand of them.... No complaints about price...

Finegrain · Aug 6, 2017

1. Are the stampings accurate and burr-free enough to allow the outside perimeter to position the part accurately enough to get that .01" cut, which presumes a tolerance of only a couple thousandths?
2. If not, are the holes the features that dictate where exactly that face needs to be?

Those question, IMO, are what determine how they are to be fixture.

I sympathize with your desire to "Stack and whack," but the dogleg prevents it I should think. Trying to stack those doglegs up will inevitably throw off the precise positioning of the subject face.

Regards.

Mike

David_M · Aug 6, 2017

PeteM said:
...Snipped...

Another option would be to stack one part vertically on two locating holes then have a sliding fixture to run it into a mill, grinding wheel, sanding disk etc. with a stop set to take .010" off. I'd think the cycle time would be under 6 seconds each, 10 a minute, 600 an hour, five hours for the 3000? I doubt you could stack them in a fixture a whole lot faster than that.

I would especially like this approach if a Grob Band Filing Machine were available. (This would be accomplished horizontally instead of vertically, though.) And with a single (pivot) pin and a stop, you could arc it into the file using the length of the part for leverage. It would get it done very quickly.

Monarchist · Aug 6, 2017

Finegrain said:
1.Trying to stack those doglegs up will inevitably throw off the precise positioning of the subject face.

?? Only if they touch each other. They need not be in full contact at all to be "stacked".

Thats's why a fixture entered the room with the "opportunity" (ain't no such thing as a "problem") in the first place.

Zahnrad Kopf · Aug 6, 2017

So I'm thinking of making a plate to act as the platform for operations, with blocks positioned at the nose ( where the machining will occur ) having pins that the hole will locate on protruding from the side and recessed leaving a radius ledge that the nose cane "hook" under to lock the part in. Then, a pin inserted through one of the more rearward holes, run through them all, and held with downward pressure into clamping, split pillow blocks to accept the pin and secured with a screw to close down the each of the blocks upon it.

Should be able to get 18-ish at a shot, making it only 15"-ish long, overall.

( rough model for illustration )

WILLEO6709 · Aug 6, 2017

that looks like a lot of work for a 1 time shot

Monarchist · Aug 6, 2017

WILLEO6709 said:
that looks like a lot of work for a 1 time shot

Seconded.

A third the count, triple the cycles, lower risk. If problem is addressed at source, need goes away.

If not? Raise the price until it is corrected at the source.
Or just goes away.

This is not going to fund anyone's retirement.

Zahnrad Kopf · Aug 6, 2017

WILLEO6709 said:
that looks like a lot of work for a 1 time shot

Maybe. But I look at it like this - probably take me 3 hours to make the fixture. 4 tops. There are 3,000 for sure. Possibly 6,000. Could go as high as 10,000. ( found out today ) I hate repetition. So anything I can do to speed these through and move on to the mext job is worth it to me. And, it's not like we don't still get paid for the efforts. Win-win.

No? I am open to alternative views and options.

Zahnrad Kopf · Aug 6, 2017

Monarchist said:
Seconded.

A third the count, triple the cycles, lower risk. If problem is addressed at source, need goes away.
If not? Raise the price until it is corrected at the source. Or just goes away. This is not going to fund anyone's retirement.

Like I wrote earlier, they are correcting the issue. The simply are tasking us with remedying these. And, they are a good customer. They push a lot of work through here. Who am I to tell them how to operate? If they want to pay for the efforts, why shouldn't we let them? Plus, the main Engineer and I have worked together for a decade. I like to help him. ( he inherited this one, it's not his error )

Monarchist · Aug 6, 2017

Zahnrad Kopf said:
Like I wrote earlier, they are correcting the issue. The simply are tasking us with remedying these. And, they are a good customer. They push a lot of work through here. Who am I to tell them how to operate? If they want to pay for the efforts, why shouldn't we let them? Plus, the main Engineer and I have worked together for a decade. I like to help him. ( he inherited this one, it's not his error )

No - what I meant was balancing the cost of the fixture against time saved - because they ARE correcting it, for one. Smaller fixture is faster to do, easier to do WELL, easier to do more than one of for extra hands on remote loading/unloading as 'mini-pallet'.

And because a longer traversal makes holding a(ny) tolerance harder in the larger batch, and scrap losses multiply over any loading or tool-fail FUBAR, any machine-tool most of us can afford.

Better to design for "worst case" than to expect perfection, each batch, on a known few-time project. R&D time and budgets go into NEW work, not salvage jobs. This was not "in their plan", there is scant funding for it.

You cannot get too exotic and keep trying alternatives before that is all used-up.

KISS

Finegrain · Aug 6, 2017

Monarchist said:
?? Only if they touch each other. They need not be in full contact at all to be "stacked".

Thats's why a fixture entered the room with the "opportunity" (ain't no such thing as a "problem") in the first place.

If they aren't touching, they're not stacked

. High-density fixture with parts aligned is fine, but not technically "stacked". IMO.

But even so, 10 seconds part swap time if done 1x at a time x 6,000 = 60,000 seconds = 1000 minutes = 17 hours. If you can make your high-density fixture (2 of them so you can swap parts while the machine is working) in that time, then you are ahead with the fixture. You said 4 hours. That seems pretty optimistic :scratchchin:

. I would be pleasantly surprised if I could make 2 of those in less than a couple days. Maybe I'm just slow

.

OTOH if faced with this mind-numbing task, I'd make the fixtures just to make the job minimally entertaining

.

Regards.

TDegenhart · Aug 6, 2017

I don't see how these can be stacked because of the right angle top form on the left. Stacking will cause the parts to staircase unless the spacer is an inch or so in length.

I think David N has the best approach, sliding fixture on belt sander, spring pins for locating and foot operated clamp. Cylinder to move fixture into the belt and spring loaded to retract, timed with clamp cylinder.

Operator loads fixture, stomps on foot pedal, fixture clamps, goes in, stops, retracts and opens. Repeat.

Tom

Edit: Just saw proposed fixture. Left hand block will cause the blades to fan plus area of work will bow up unless top clamped. If you top clamp, you don't need the left hand block.

T:-

Spit balling fixturing ideas

Diamond

Hot Rolled

Cast Iron

Diamond

Diamond

Diamond

Diamond

Cast Iron

Diamond

Diamond

Hot Rolled

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Similar threads