Low volume suggestions workholding / pallet for round stock on mill

[Fal Grunt]

I have a customer who wants me to quote 1000, 2500, and 5000 pcs for a part that I am currently making for them. The parts are a simple 3" long, 1/2" dia cylinder that gets milled on both ends.

This part is on the bottom end of the spectrum I hate to play, that is, cost is more important than quality.

Parts are turned, and while the range of diameters is pretty consistent, they may vary batch to batch. So some way needs to be incorporated to securely clamp, with varying diameters (to some extent).

I have a swiss shop quoting them as well, and hopefully the price comes back so cheap it is not worth my time even attempting this job. I suspect based on what I know about swiss shops, the Brother will be able to significantly outpace them.

I have a 20" Orange Vise, and I was thinking about building two pallets to be swapping parts while the machine is running and cramming as many on a 8"x20" as I can.

What have you found that is an effective way to hold quantities of round parts for milling? Suggestions? Recommendations?

 

[crossthread82]

What kind of milled features? Side of cylinder? End of cylinder? How much diameter variation, 0.005? 0.015?

 

[DanASM]

If you want me to machine you some blanks let me know. I can have a few swiss shops quote it locally too if you want.

 

[DanielG]

Depending on the complexity of the milling, a twin spindle lathe might be the way to go. While it might be slower at milling than the Brother, done in one with a barfeeder is hard to beat.

If you do them on a mill, this sounds like it might be a good application for one of those spindle mounted grippers and a 5C air-actuated dead length collet. Round parts are usually really easy to grab.

 

[Fal Grunt]

What kind of milled features? Side of cylinder? End of cylinder? How much diameter variation, 0.005? 0.015?

There are several milled features, both side(on the end) and end milling.

I think I put plus or minus .003" on diameters, the current batch seems to run -.003 to -.004.

If you want me to machine you some blanks let me know. I can have a few swiss shops quote it locally too if you want.

I'll see what the local Swiss shop comes up with and if it is outlandish, I'll have you do up a quote.

Depending on the complexity of the milling, a twin spindle lathe might be the way to go. While it might be slower at milling than the Brother, done in one with a barfeeder is hard to beat.

If you do them on a mill, this sounds like it might be a good application for one of those spindle mounted grippers and a 5C air-actuated dead length collet. Round parts are usually really easy to grab.

Twin Spindle lathe would be a good way to go. I may have to shop around and see if I can find a shop that has one available.

I'm thinking that an air actuated collet would be too expensive and too time consuming per part. Maybe I am wrong on that I have never used them. Even setting up 5 parts, so 10 chucks, at $800-$2500 per chuck (quick google prices), I can't see it as a very good option compared to making a pallet that might hold 10+ pieces.

 

[crossthread82]

There are several milled features, both side(on the end) and end milling.

So can you get to everything if the parts are laid over on their side? If so rows of parts held with machined uniforce clamps could be a good option, could get probably at least 24 per 8x20 pallet...orange also has 12x20 pallets too just as an fyi. If they need to be stood up on end that's a bit more challenging...you could maybe use some of the concentric OD clamps from mitee bite, have a bunch like 16 of them mounted on a pallet with the parts sticking up out of them. Or if its both you could do half and half on one pallet or have one of each kind of pallet.

 

[mhajicek]

If you need to get at the ends and sides, you could do strips of uniforce along a 4th axis fixture. Unless you need to get at more sides than that allows.

 

[crossthread82]

If you have a rotary what @mhajicek said would more than likely be best. I have a pierson rotovise so I can put the pallets on all four sides, put rows of parts along the pallet pointing in the "Y" direction with uniforce clamps between them.

 

[2outof3]

Loading from a tray using a spindle gripper could be an option.

 

[GiroDyno]

Load the parts kinda like a machine gun ammo belt along a 4th axis, tilt up and get at one end, flip 180 and get the other?
If you don't have a 4th axis and project budget doesn't justify one you could get a 180* indexer for a couple hundred dollars and roll your own indexer to do the flipping.

EDIT
I meant a 180* pneumatic rotary indexer like this

 

[wmpy]

If I had to make these in a VMC, I'd be looking for a used multi-collet fixture like the ones made by Beere.

https://pawsworkholding.com/workholding-collet-systems/

 

[Fal Grunt]

So can you get to everything if the parts are laid over on their side? If so rows of parts held with machined uniforce clamps could be a good option, could get probably at least 24 per 8x20 pallet...orange also has 12x20 pallets too just as an fyi. If they need to be stood up on end that's a bit more challenging...you could maybe use some of the concentric OD clamps from mitee bite, have a bunch like 16 of them mounted on a pallet with the parts sticking up out of them. Or if its both you could do half and half on one pallet or have one of each kind of pallet.

No they need to be stood on end for one operation. The other end can be done from the sides.

I'm thinking a row of op one up one side, and a row of op two up the other. If I can cram 2 sets I might do that.

If you need to get at the ends and sides, you could do strips of uniforce along a 4th axis fixture. Unless you need to get at more sides than that allows.

This is a good suggestion, and I thought about that, but I don't have a 4th as of yet, and this job wouldn't pay for or justify it.

If you have a rotary what @mhajicek said would more than likely be best. I have a pierson rotovise so I can put the pallets on all four sides, put rows of parts along the pallet pointing in the "Y" direction with uniforce clamps between them.

I had considered this, and index my dividing head each operation, but I am hoping to let this one run for as long as possible without me attending.

Loading from a tray using a spindle gripper could be an option.

Maybe... I'm not sure how you would do the flip to get at the other end? I might ask a friend of mine if he has any ideas. He used to build the feeder bowls at a company we used to work at, and he could make parts do amazing things.

As Load the parts kinda like a machine gun ammo belt along a 4th axis, tilt up and get at one end, flip 180 and get the other?
If you don't have a 4th axis and project budget doesn't justify one you could get a 180* indexer for a couple hundred dollars and roll your own indexer to do the flipping.

I'd thought about this, I do have a super nice dividing head, and for the amount of indexing work I do, it is fantastic and incredibly accurate. I might take a quick look at doing a setup with it since so many have mentioned it.

If I had to make these in a VMC, I'd be looking for a used multi-collet fixture like the ones made by Beere.

https://pawsworkholding.com/workholding-collet-systems/

If it was a long term production job, that might be feasible. I've looked at PAWS and while they make great stuff, they are beaucoup bucks. Based on pricing I have seen, those collet systems are probably way out of the range of feasible for this job.

Some more thoughts I had:

The only other argument I have against the 4th option is the speed at which I can replace parts. I don't have any "quick" change setup for the 4th where I could pop a load of parts in and out. So that would be something I would need to look at. With the Orange vise, I can quickly pop a pallet in and out and be changing parts while the machine runs the next batch.

Additionally, I have to factor in the cost of work holding in comparing to any swiss or twin spindle machine, and in this case, unless the work holding is "cheap" they may be a better option. Spending $10k on workholding simply isn't an option on these parts.

 

[GiroDyno]

There's no reason your rotary/indexer can't also function like a pallet. Heck, just make a trunnion for the Orange vise and use their pallets...

 

[Mtndew]

Spending $10k on workholding simply isn't an option on these parts.

If you get an order for 5,000 of these, and find a "cheap" 4 axis rotary, you could potentially make a fixture to grip the O.D. with the ends exposed and make the entire part in 1 push of the button. If you the OD work can be done from the side like with a woodruff cutter or whatnot.
Even if you can't get the ROI back on this 1 job, adding a 4th will recoup money back very quickly because you are more versatile. I would seriously consider looking into it. Check with Koma Precision, they have used/refurbished models (or used to). You might be surprised at what you find.
The RWE series is their economy line.
They sell plain indexers too.
https://komaprecision.com/products/rotary-tables/

 

[wheelieking71]

davidn posted these wonderful pics a while back:

 

[DavidScott]

My vote would be to palletize them holding with strap clamps and hold the pallet on it's side so you can flip the pallet for ops 1 and 2 so you only have to load the parts on the pallets once. I would use strap clamps, like davidn in wheelie's post, since they don't put any stress on the pallet. A bunch of uniforce clamps will want to bow your pallet, I use them a lot and know this very well. IMO a rotary is about as necessary as tool holders, I can't imagine getting a machine without one. If you do ever get a rotary DO NOT dick around with anything BUT a Brother ball drive! You don't want to hobble your racehorse with a slow rotary.

Here is some eye candy to see how I would make them on a Brother with a 4th, or two. I am holding these parts with my version of uniforce clamps but do hold parts on both sides to even out the stress on the fixture. I do love those pneumatic bit drivers! With only one 4th I have a valve on the pneumatic tailstock so I can swap the fixtures out in about 15 seconds of spindle down time.

 

[Mtndew]

My vote would be to palletize them holding with strap clamps and hold the pallet on it's side so you can flip the pallet for ops 1 and 2 so you only have to load the parts on the pallets once. I would use strap clamps, like davidn in wheelie's post, since they don't put any stress on the pallet. A bunch of uniforce clamps will want to bow your pallet, I use them a lot and know this very well. IMO a rotary is about as necessary as tool holders, I can't imagine getting a machine without one. If you do ever get a rotary DO NOT dick around with anything BUT a Brother ball drive! You don't want to hobble your racehorse with a slow rotary.

Here is some eye candy to see how I would make them on a Brother with a 4th, or two. I am holding these parts with my version of uniforce clamps but do hold parts on both sides to even out the stress on the fixture. I do love those pneumatic bit drivers! With only one 4th I have a valve on the pneumatic tailstock so I can swap the fixtures out in about 15 seconds of spindle down time.

In the 2nd video, you're only putting new parts on 1 side and not both? Can you not index while the machine is running? I don't have a 4th on my R450x1 so I wouldn't know.
I love the shelf you put up there, genius!

 

[DavidScott]

In the 2nd video, you're only putting new parts on 1 side and not both? Can you not index while the machine is running? I don't have a 4th on my R450x1 so I wouldn't know.
I love the shelf you put up there, genius!

I'm putting parts on both sides. You can rotate the outboard rotary with an optional switch and by how many degrees in the control. While the shelf is pretty obvious once you try to figure out the new workstation I did see it in a video some years ago. I want everything needed to swap parts in the machine to save time/movement.

 

[Fal Grunt]

I've put some thought into this and I can't see how to make the index work. If I put the round parts in to index to work from both sides, I don't see a good way to put them against stops for length.

The current batch I am running standing up, 5 per jaw, in a 2 station orange vise, similar to the davidn pictures. I could do a gang block, closer to how his is setup, and run 10-12 or even more in each station.

The primary concern right now is faster changeover and chip management. Running them like how I have them takes almost 3 minutes to pull the 5 complete parts out, clean second op, load 5 from the first op, clean first op, load 5 blanks.

So I'm back to running pallets on the Orange vise I think. It's also the lowest investment cost to see if these really continue or not. If I run 4 rows, 2 of each operation, approximately 20 if the spacing works out, while the per part time won't decrease much, if it takes me 5 minutes to swap a pallet, it shaves 19 minutes of changeover time. 25 cycles to make 1000 pcs saves 475 minutes, so 7.92 hrs.

 

[DavidScott]

What is your tolerance for the end stop?