Fixturing Round Stock

[SCzEngrgGroup]

One of the many problems with using V-blocks is properly locating the center, especially after you turn the parts over, as they will shift, based on the actual dimensions of the parts, and the V-blocks themselves. To get any kind of accuracy, you pretty much have to center up to each individual part every time - a LOT of extra work. With collet chucks, or a fixture, your centers will remain in the same place the whole time.

Regards,
Ray L.

[victorofga]

i think making some aluminum jaws and clamping two of them closely to the jaws both end would be the simplest and cheapest..

yet, a simple lathe from harborfreight could make it same time or even faster.. pending on how much experience you have before with lathe..

if you checking on harfreight frequent this lathe sometime priced down to 2000... possible on first serious work it brings it price back..
Geared Head Gap Bed Metalworking Lathe - 12" x 36"

[sansbury]

Take a look through the Mitee-Bite catalog. Jergens also makes a lot of fixturing components. That'll give you ideas for what's out there. Something like this or the Mitee-Bite Uniforce clamps might work.

Hex Clamps | MSCDirect.com

If this is a repeat job then I'd spend some money to build a good fixture. I'd be tempted to do a 10+10 setup with ten top side, ten bottom side on one plate so you get ten finished parts with one setup. Both sides involve a lot of material removal by endmill so you could walk away for a while even without a tool changer.

Someone on the Practical Machinist board (you might try asking over there, too--they're gruff but not entirely hostile) once said that prototyping/job-shopping and production machining were sort of opposite disciplines. In production you are looking for every second of efficiency, minimizing waste, using purpose-specific tooling, etc. With one-off/short run jobs you want to minimize the time spent making fixtures, use whatever tools are set up, spend the least time working the CAM, use oversize stock if it saves you time, etc.

If this is a likely one-time job I wouldn't want to spend a lot of time or money on fixturing unless it was reusable for other jobs. Building good fixtures is at least half of the game of being a good CNC machinist and it would be easy to spend more time making a 20-part fixture than it would to just grunt your way through twenty parts done one or two at a time. Also, I generally don't make a large multi-part fixture without making a small 1-2 piece one first so I can validate that it will work. After a while you start to have a better sense for what's likely to work (and how to work around problems) but it would stink to spend 5-6 hours and hundreds of dollars building a fixture only to figure out it's a dud. So like I said, you build a 1-2 station one and knock out the twenty parts the hard way, you might well be out less time and money than needed to build the 20-piece one and then make the twenty parts. Now, if you are likely to make another twenty every few months, then bite the bullet and build a good fixture.

[hjl4]

Victor, I think he knows a lathe could make it quicker, but certainly not a manual one. He wants to be able to cycle start the mill and be able to walk away from it so he can do other things while the parts are being made.
I still think Ray's idea of making a fixture for rows of them standing up, is the best and most accurate, probably the cheapest.
Hard to say because I don't specialize in small parts or high volume. Everything I do are one offs and mirror images for molds. So high production is not my specialty. I thought maybe Neil or Dan would of chimed in by now.

[Don Clement]

I normally use 5C collet fixtures (e.g.Eaglerock , Bison 5C chuck, 5C fixture hex and square blocks, etc) to hold round stock, however that is usually limited to one part at a time. If you could hold your part internally then these ID expansion clamps could be used to hold many 1" diameter parts on a fixture plate. Mitee-Bite Products Co. - ID Xpansion Clamp I saw these at Westec one year and looked like just the solution for fixturing multiple parts held on the ID. I can also imagine threading the ID expansion clamps to hold the threaded ID.

Don Clement

[Circle Support]

Hi Hirudin,

These are all good ideas. That goes to prove that there is no one perfect way to a project. So I will chime in with my technique for holding round material in a mill. I use a flat back 3 jaw lathe chuck clamped to the table.

Dean

[rowbare]

Why not make a fixture for it?

While I am not advocating something as elaborate as this:

http://www.cnczone.com/forums/work-f...-clamping.html

you could make something similar using clamp bolts instead of the cam. Using 3" bar stock, you could do two rows with the clamp bolts front and back. Each bolt would clamp two parts.

You could set it to do the pocket, flat and internal thread on one row and the outside thread on the other. That way you get finished parts each run.

Assuming the diameter is consistent along a bar, you will only have to tweak your Y and Y' values occasionally. Since it self centres, the X should remain consistent. Depending on how much your bar diameters vary, you might relieve the pockets slightly to clamp on 4 points.

bob

[totalhack]

Thanks everybody for their feedback and information I really appreciate all of the different thoughts and ideas. At this point I think I am going to opt for using 5C collet blocks like the ones mentioned by TurboStep. Initially I was going to use the machinable mitee bite clamps and just make a fixture plate but at nearly $23 per clamp I think I would be better off spending an extra $5 each and making a fixture plate to locate individual collet blocks. My goal was to be able to machine 20+ plus parts per run and I should be able to fit that many collet blocks into the work envelope of the Tormach.

The down side I see to using collet blocks mounted into a fixture plate instead of just clamping the parts in directly is now I will have two steps involved to flip the parts. Instead of just loosening a clamp and flipping the part I will now have to loosen a clamp, remove the collet block then loosen the collet in the block, flip the part and re-tighten the collet then re-clamp the block back into the fixture plate. My thinking is that although it will take longer to flip all the parts and re-clamp the blocks I will save money in the future by having more sizing options available to me.

I quickly made a model in SolidWorks of the general idea I have going in my head. There is a base fixture plate with multiple 5C collet blocks held down using standard non machinable pitbull clamps which I already have and the round protrusions out of the top represents the stock that is to be machined. I'm thinking I could even use emergency collets in the blocks so that I could machine in a set depth just to make setup easier. Am I doing anything wrong if I go with this type of setup?

[rlockwood]

Various companies also sell ganged 5c collet blocks, and even ones that are designed to interlock and be ganged together.. Probably simpler than what you're currently headed towards

Sent from my Nexus 5 using Tapatalk

[SCzEngrgGroup]

What you show does not actually appear to hold the collet blocks - what holds the side opposite the Pitbull clamps, and what restrains them laterally? I also question how well the Pitbull clamps will hold on the hardened collet blocks. As tall as they are, there will be considerable torque on the bottom of the collet blocks due to machining forces. I would be concerned they will not be held nearly solidly enough, and will move around when you start machining. And, the time to turn over the parts will be quite considerable. Finally, how will you control the depth of the parts in the collets? Overall, I think a simpler, cheaper approach will also work better.

Regards,
Ray L.

[johnedward]

Thanks everybody for their feedback and information I really appreciate all of the different thoughts and ideas. At this point I think I am going to opt for using 5C collet blocks like the ones mentioned by TurboStep. Initially I was going to use the machinable mitee bite clamps and just make a fixture plate but at nearly $23 per clamp I think I would be better off spending an extra $5 each and making a fixture plate to locate individual collet blocks. My goal was to be able to machine 20+ plus parts per run and I should be able to fit that many collet blocks into the work envelope of the Tormach.

The down side I see to using collet blocks mounted into a fixture plate instead of just clamping the parts in directly is now I will have two steps involved to flip the parts. Instead of just loosening a clamp and flipping the part I will now have to loosen a clamp, remove the collet block then loosen the collet in the block, flip the part and re-tighten the collet then re-clamp the block back into the fixture plate. My thinking is that although it will take longer to flip all the parts and re-clamp the blocks I will save money in the future by having more sizing options available to me.

I quickly made a model in SolidWorks of the general idea I have going in my head. There is a base fixture plate with multiple 5C collet blocks held down using standard non machinable pitbull clamps which I already have and the round protrusions out of the top represents the stock that is to be machined. I'm thinking I could even use emergency collets in the blocks so that I could machine in a set depth just to make setup easier. Am I doing anything wrong if I go with this type of setup?

I'm guessing that since you say you want to walk away and let the parts run that you have the atc. have you thought about using the mill as a vertical lathe? you could load your stock into 10 er40 holders, put them in the toolrack and walk away while it runs the operations on each tool. come back, flip the parts and reload the tool rack for the 2nd side.

[totalhack]

Various companies also sell ganged 5c collet blocks, and even ones that are designed to interlock and be ganged together.. Probably simpler than what you're currently headed towards

I wish I could afford what you are suggesting but the ones I have seen cost a couple grand each and the biggest I have found only holds 10. That one was like 5 grand, ouch.

What you show does not actually appear to hold the collet blocks - what holds the side opposite the Pitbull clamps, and what restrains them laterally? I also question how well the Pitbull clamps will hold on the hardened collet blocks. As tall as they are, there will be considerable torque on the bottom of the collet blocks due to machining forces. I would be concerned they will not be held nearly solidly enough, and will move around when you start machining. And, the time to turn over the parts will be quite considerable. Finally, how will you control the depth of the parts in the collets? Overall, I think a simpler, cheaper approach will also work better.

That was just a quick mock up, the collet blocks would be recessed into the fixture plate 1/4 to 3/8 of an inch. I was hoping that a tight tolerance around the collet block and the down force provided by the mitee bite would be sufficient to stop the collet blocks from moving around during machining.

I'm guessing that since you say you want to walk away and let the parts run that you have the atc. have you thought about using the mill as a vertical lathe? you could load your stock into 10 er40 holders, put them in the toolrack and walk away while it runs the operations on each tool. come back, flip the parts and reload the tool rack for the 2nd side.

Thanks for the idea but that is outside my skill level at the moment. A vertical lathe setup is something I want to eventually try but right now I that is a little too much to bite off. Up until this point I didn't think threading was possible on the vertical lathe setup so thanks for posting that video.

[Steve Seebold]

Thanks everybody for their feedback and information I really appreciate all of the different thoughts and ideas. At this point I think I am going to opt for using 5C collet blocks like the ones mentioned by TurboStep. Initially I was going to use the machinable mitee bite clamps and just make a fixture plate but at nearly $23 per clamp I think I would be better off spending an extra $5 each and making a fixture plate to locate individual collet blocks. My goal was to be able to machine 20+ plus parts per run and I should be able to fit that many collet blocks into the work envelope of the Tormach.

The down side I see to using collet blocks mounted into a fixture plate instead of just clamping the parts in directly is now I will have two steps involved to flip the parts. Instead of just loosening a clamp and flipping the part I will now have to loosen a clamp, remove the collet block then loosen the collet in the block, flip the part and re-tighten the collet then re-clamp the block back into the fixture plate. My thinking is that although it will take longer to flip all the parts and re-clamp the blocks I will save money in the future by having more sizing options available to me.

I quickly made a model in SolidWorks of the general idea I have going in my head. There is a base fixture plate with multiple 5C collet blocks held down using standard non machinable pitbull clamps which I already have and the round protrusions out of the top represents the stock that is to be machined. I'm thinking I could even use emergency collets in the blocks so that I could machine in a set depth just to make setup easier. Am I doing anything wrong if I go with this type of setup?

Your idea looks like a good one, but I see one or two huge problems with it.

1. You will have to remove the blocks to change parts.

2. Holding your collet blocks with that much above the Pit Bull clamp, leverage will come into play and you could rip the blocks out of the fixture.

I think I would take the time to design and build a fixture that would hold the parts as short as possible.

[totalhack]

Your idea looks like a good one, but I see one or two huge problems with it.

1. You will have to remove the blocks to change parts.

2. Holding your collet blocks with that much above the Pit Bull clamp, leverage will come into play and you could rip the blocks out of the fixture.

I think I would take the time to design and build a fixture that would hold the parts as short as possible.

So then what you're really saying is that is a bad idea. LOL

[victorofga]

hjl4


i know cnc very comfortable.. i was in machine shop from 79-93.... i dont know how many can be made daily on a manual or conventional lathe...
i made setup around 93 for a job , a 1in dia part about 1 1/4 long and a small shoulder with a grove and all edge chamfered... i made between 600 and 700 a day...in 8 hour...
i used stops and front toolholder has a parting tool wit a ""roller stop"" and back toolholder had 3 hss cutter set...

with cnc this knowledge dont need any more ,,,

my concern actually for 20-30 part a conventional lathe still faster..due a cutter with stops can be set faster, and workholding is certain on a lathe..

i seeing your point... while cnc works i can walk away and do other works.. also a cnc never tired and never read wrongly scale...


one thing is sure, if he has less than 100 of these parts, then fixture will be too much for this work...

[Steve Seebold]

So then what you're really saying is that is a bad idea. LOL

YEP!!!

[Hirudin]

How about using basically the same fixture but flip all the collet holders over and cut holes through the fixture to allow access to the stock. That way you would have easier access to the collet nuts and the Mitee-Bite clamps would be closer to the machining.

So the bottom of a single part of the fixture would look like the attached drawing.

[hjl4]

victor,
I agree with you 100%. I too have worked on manual lathes and mills long before CNC's in a machine shop, where every minute count.. So I know where your coming from.
Good day.

[sansbury]

Crazy idea. Use square stock and clamp it with your pitbull clamps. Do the threaded portion, then screw it into whole to do the top. Slightly more material to remove but makes fixturing a lot easier. Mill won't care whether it's making round stock round or square stock round.

[300sniper]

Just thinking out loud here because I haven't done it myself.

Make two sets of soft jaws for your chuck. If the parts are as small as they look, the jaws could probably be quite a bit wider than your vise. Machine one set of jaws smooth to hold the raw stock. In that set of jaws, machine the boy side of the part. After you have finished the whole batch, machine a set of soft jaws by thread milling the jaws to hold the boy side of your parts, half in each side of the jaws. Thread them down and snug up your vise and complete the girl side of the parts. I don't think you are going to get 20 parts per cycle start but depending on the size, you could probably get 10+.