How do I secure a bearing for a concave turning tool holder?

[LightningEngineer]

How can I make this design work? I have absolutely no idea how to use a tapered bearing. I'm pretty sure it is the type I want, however, how can I make the upper tool holder rotate without binding while using a bolt of some sort? I'm thinking if I use a nut of some sort up top it will prevent the entire thing from moving smoothly. Essentially what I think I need to do is isolate the shaft from the upper assembly to it "floats" on the bearing, right? I can CAD it, but I have no idea how to design it!!

Anyone have an idea as to how to make this thing work? I need to cut some concave shapes into the face of some 6" aluminum cylinders. Any help would be greatly appreciated!!!

 

[404]

you need the cup part of the bearing, and then the hole is straight walled.

look up timken cup and cone

 

[404]

also normally you would need 2 complete bearings. taper rollers are not used singly that I have ever seen

 

[ranger302]

^^^^ This. You will need two bearings.

 

[zkling]

What you are trying to build is called a ball turning attachment.
I'd suggest you look up a few plans online.
You really don't need rolling element bearings and actually it could make it hard to control if it had too much freedom of movement about the axis.

Tapered roller bearings do not need to be used in pairs, however each one will only support axial loads in one direction. Hence most are used in pairs (ex wheel bearings) where a bi direction axial load is applied (pushing and pulling), but there are a good number of applications where only one is used.

I can CAD it, but I have no idea how to design it!!

That is a very dangerous, yet common statement. Based on your sig, do you have a BSME or BSEE?

 

[Kevin54]

Press fit you inner race onto your shaft, and have a stepped shaft so you can hold the bearing depth. Then the top piece will have to have threads on the shaft to draw it down to the bearing. BUT, your design does not allow you to tighten down to hold the T-Nut in place unless you add another plate and nut into the design to draw it up tight.

Make sense, or you need a quick sketch?

 

[Kevin54]

After looking at your sketch again, toss it. It won't work because you don't have the means to fasten it to your compound correctly. There are a lot of plans out there on the interweb. Let me see if I can find you a few that you can get some ideas from. I do have an idea for your design, but you need to make your T-Nut longer so you can fasten it to the compound, then work off of that for your ball turner. Also your tool height needs to be where it is exactly centered with your headstock of your lathe, so some precise measurements need to be made.

I can still give you a sketch if needed.

Take a look at this for one idea http://www.machine--tools.com/By-Lo...tool-post-radius-ball-turning-attachment.ASPX

Another type http://www.timsmachines.com/home/hardware/grizzly-10x22-benchtop-lathe/ball-turning-attachment

And this (pdf file) http://www.littlemachineshop.com/Instructions/SmallRadiustoolinstructions.pdf

And here http://www.homemadetools.net/lathe-ball-and-radius-turning-tool

That should get the brain cells thinking a little about the design. The first thing you need to do in the design is figure how to fasten it to the compound of the lathe, then build off of that.

 

[larry_g]

You need to study the forces on the cutting tool and then design a tool holder that will counteract those forces. The bearing you have will take the vertical forces but you have nothing to maintain the forces tangential to the bearing. Then the tool starts to cut depending on the geometry of the cutting tool it may be pushed away or pulled into the material being cut. You have design to handle all the forces in all directions. If you study most tool holders and slides on machining tools you will find plain bearings, and for good reason.

lg
no neat sig line

 

[abk241]

How can I make this design work? I have absolutely no idea how to use a tapered bearing. I'm pretty sure it is the type I want, however, how can I make the upper tool holder rotate without binding while using a bolt of some sort? I'm thinking if I use a nut of some sort up top it will prevent the entire thing from moving smoothly. Essentially what I think I need to do is isolate the shaft from the upper assembly to it "floats" on the bearing, right? I can CAD it, but I have no idea how to design it!!

Anyone have an idea as to how to make this thing work? I need to cut some concave shapes into the face of some 6" aluminum cylinders. Any help would be greatly appreciated!!!

Using just the last 6 words of the highlighted statement in a google search box gives you a good starting point...
https://www.google.com/?gws_rd=ssl#newwindow=1&q=how+to+use+tapered+roller+bearings

Although 404 has definitely steered you in the right direction...you need a cup to go with the cone!

 

[Guster]

I’ll support Kevin on the premise that the bearing is overkill in this situation. If you read many of the ball turner project discussions, the outcome is that a little friction in the setup is actually a good thing to help produce a smoother cut as it allows for a consistent pressure rather than just relying on the cutting tool resistance. Many designs were actually retrofitted with things like fibre washers to achieve this. If all you want to do is reduce the chance of binding then a brass or UHMWPE bush or washer may be more appropriate. Else if it is to allow you to tighten the axle to the circular base without it binding in contact surfaces then a thrust bearing or UHMWPE washer may be more appropriate.

There are several ways to make this and Kevin supplied some excellent examples. A little research and google will find you many many more. There is a lot to be learned from their experiences rather than just the designs too. Tool height and tool geometry is the most important factors to consider. Size of work and clearances for the machine it is fitted to, will dictate the rest. Then the differences in design lend them to specific work ie. the one you show is good for small convex and limited concave work though cannot overhang the saddle in order to turn something like a rope pulley.

Which reminds me… wonder how A_Pmech got on making his Holdridge clone ball turner - http://garagejournal.com/forum/showthread.php?t=88034

 

[sanddan]

I made my ball turner without bearings and it works very well. I used material I had on hand so the base is aluminum and the rotating part is 1018 steel. I just added a dab of grease before assembly and snugged the retaining bolt enough to take out slop but still allow smooth movement. It uses a triangular ceramic insert which for steel and aluminum has worked fine. You could add a thrust bearing (I did design one in but didn't end up using) but it's not needed. For mine, I remove the compound and replace it with the ball turner. This is for a typical 12X36 chicom lathe like Grizzly 4004 or PM or in my case Enco. If you had the room you could mount it in place of the QCTP but I didn't have room for that. Here's a screen shot of the cad layout, not all details are shown.

 

[AndyA]

Skip the bearing.

Just give the minimum clearance between the shaft and hole so that it'll spin freely. It's not like you'll be putting 100,000 miles on this thing. I'd drill and ream the hole, then turn the shaft to size.

Drip a few drops of oil on the shaft, and you're ready for action!

Take a look a the ways on any lathe. No bearings at all. Just smooth surfaces and a little oil.

 

[mikegt4]

I agree that a bearing is waayyy overkill. My experience has been that you don't want it to be too easy to rotate, it will definitely get away from you at some point.

 

[LightningEngineer]

I've been playing around with different ways of attaching it this time with a pivot screw which I guess will be a shoulder screw, taking all of the comments into mind.

The problem is that I don't think this thing will work to make my part anyway since the compound already sits reasonably high with that plate in there to hold it down it will not fit close enough to the workpiece. I think I'd have to throw in a long cutting tool but that would act as a lever and add excessive force to the holder.

I've researched various ball turning tools quite excessively before coming here, but they were all a bit different from what I need. Most of them are turning the circumference whereas I need to turn the face. The one that comes the closest is the one here - http://www.littlemachineshop.com/Instructions/SmallRadiustoolinstructions.pdf that was linked to me. I found a guy that built one larger, but I'm not sure how something like that work work for my application either since I need to create such a large part and with the vertical space required for the tool it may not fit in my lathe either.

I'd really like to figure this one out, do you guys think if I built up this tool and just stuck a tool out 2 inches that it would still work without breaking or bending the tool in half?

That being said I'm certainly not a MechE, nor will I claim to be one. I'm just trying to learn some machining and make some parts I need for projects pertaining to my main area of knowledge -- electronics. I do greatly appreciate all the feedback from each and every one of you. I'm going to figure out how to make this work one way or another and when I do I will make sure to post my findings so that others may benefit from what I have learned.

 

[sanddan]

Could you show what you are trying to make? Are you doing a concave cut or convex? My ball turner will do up to a 2" ball. It can do both concave and convex cuts if I remade the "L" shaped part to hold a diamond shaped insert.

 

[LightningEngineer]

Could you show what you are trying to make? Are you doing a concave cut or convex? My ball turner will do up to a 2" ball. It can do both concave and convex cuts if I remade the "L" shaped part to hold a diamond shaped insert.

Essentially a spherical cut into a rod. ~5" in diameter.

 

[dr_clyde]

Do you have a milling machine? This can be done with a boring head and a rotary table.

 

[LightningEngineer]

Do you have a milling machine? This can be done with a boring head and a rotary table.

You just made my day. Mine does not tilt, but I think I might be able to get my hands on one that does.

I can't seem to find much information as far as how to set it up, though. Do you happen to know of any sources with the formula to calculate it, and maybe some technical drawings? In the mean time I'm going to keep looking for information, but I think the correct search term was "hemispherical" cuts, not just concave which is probably where some of my issues came from.

So far I have found a few videos for those interested:

Thanks again!!!!

 

[dr_clyde]

Tom Lipton does an excellent treatise on this in his books "Metalworking, Sink or Swim" and "Metalworking: Doing It Better".

 

[Guster]

There was also a great discussion on tangential flycutting on http://bbs.homeshopmachinist.net/forums/3-General/page2 a while ago. Complete with the equation to calculate angle and sector circumference of the flycutter. It was used to make large radius convex and concave shapes in aluminium. Much larger than could be produced on a manual lathe with conventional tooling.

In your case you may get away with the rotary table at 90deg as long as the cutter does not interfere with the work at full depth of cut though it is fairly slow going which is OK if you don't need to produce many.

Otherwise the Holdridge design is made for turning objects at near full lathe capacity with the clearances needed to achieve what you want to do as it overhangs the compound rest.