Would cams work?

[shootingarts]

I am starting to believe that a fixed gantry design best suits my needs. Although I may need to raise and lower and lock the entire cross bar on my gantry I only need two inches of "Y" and "Z" travel and I could reduce vertical travel to less than an inch if I needed to. My "X" axis should have ten to twelve inches of precision travel and I also need about thirty inches of jog capabilities. Combine that with needing a rotary axis and I am looking at five controls or I think perhaps the better option for precision, four axis including the rotary and hand jogging and locking down the table using a DRO for the long movement.

I was sitting here thinking this morning, always dangerous I know, but why wouldn't a large cam and follower arrangement work for the two axis with only a few inches of travel? It seems like a simple offset circle in the six to twelve inch range could offer great deal of precision and of course fancier profiles could be cut if the first generation machine proved functional.

Thinking a router with ceramic bearings for the spindle but a special spindle may be required. Have a possible idea there too but I'll discuss that sometime in another thread.

This machine would be used for cutting the male and female parts of inlays. The female part would normally be wood but if accuracy is great enough then at times both male and female components may be made of hard and brittle materials.

Am I heading in a reasonable direction or am I trying to reinvent the wheel with the cams? I have a hand turning tool that uses the cam principle and a fixed cutter to make cuts that stretch creditability. The markings are roughly an eighth inch apart and are calibrated in ten-thousandths of an inch. It is easy to split the markings and make repeatable cuts that can't be measured with a ten-thousandths mike. Only the fit assured me that they were consistent. Of course it did take me twelve hours to make 28 tiny pieces.LOL Seems like the cams would allow super accurate positioning of the spindle and then my limiting factor would be spindle runout.

Hu

[Evodyne]

Hi shootingarts,

You seem to have answered your own question: if you have a machine that already uses the principle and it's working for you, then "copy" the design into an automated cutter.

On the gizmo you have does the cutting head follow the cam via spring pressure, or is the cam grooved so that it both pulls and pushes the cutter? The grooved method seems like it might allow some "slop" to enter into the equation; but, depending on what you are cutting, the spring method might require a lot of tension. I guess for inlays the cutting dosn't require that much force.

A cams mechanical advantage will change relative to the follower, so you'll need enough torque for the "worst case" cam position. An interesting solution might be to use a servo motor to drive the cam via a belt (no slip) with a healthy reduction.

The biggest problem is sizing: this wouldn't seen to be something that would scale too well beyond a few inches. Neat idea though.

Lance

[shootingarts]

Lance,

Scroll down the page at the link to see the graduated markings on the handheld tool It is maybe two to two and a half inches in diameter.(Mine is loaned out so I can't look at it right now) Each of those marks is precisely calibrated to move the cutting tool in or out .0001". Looking at how wide apart they are you can see that they are easily split.

http://www.benchrest.com/nielson/

There are ways to get around some of the size problem with either cam profiles or a full cam controlling a partial cam profile but as you say the design is fairly self limiting. I considered using a cam for a rube goldberg mechanical taper device and while my mind was on cams I remembered the Pumpkin tool and realized that cams might have possibilities for precision work. As you see though, the original tool and my idea only have the cam theory in common.

On the Pumpkin tool the cutter head and mandrel are locked in place and the material is placed in a holder and hand fed onto the mandrel both feeding in and turning. No moving parts to the tool itself. The offset only moves the cutting head when you are setting the depth of cut. Likewise the length of cut is set with a few simple set screws and feeler gauges once the setting is close. This is an amazingly simple tool to perform so well. I am always impressed with folks that can simplify things to do a sophisticated job. While turning rifle brass necks is simple, this level of precision is tough.

A servo motor and reduction was in my thoughts. Once I have a machine close enough to present a complete package on the forums I will ask how much overkill I am using and try to reduce some of my components. I tend to build to last a thousand years. Makes for bulky and costly creations that work well but are far more expensive than they needed to be if I just wanted them to last a lifetime.

Thanks for your thoughts. I may start some sketches before long to illustrate what I have in mind. Right now I am still changing my mind at least twice daily though.

Hu

[Evodyne]

Hu,

Thanks for the link-what a neat tool. Like you, I find it interesting how simple something can actually be.

O.K, you want a resolution of 0.0001", but what total travel would the cam-motivated pieceholder need to have? One inch? Five?

Lance

[turmite]

Hi Hu,

Since it seems that we are both interested some of the same things, like shooting and cnc, I feel I must show my ignorance immediately!

In your original post you mentioned male and female parts of inlays, yet you showed an example of a neck turning tool! I apologize for being dense but how do those relate? Ok I went back and read and re-read and think I understand you mean the graduated cam to position one or more of your axis for better resolution??? Right? If that is the case why not simply go to a higher res ballscrew or rack?

When I first read your post and followed the link my first thought was "Oh no! He's gonna start engraving and inlaying the brass! "

Mike

[shootingarts]

Lance,

I don't need .0001 but probably a total tolerance between the two pieces of zero to .0005 or slightly more would be nice. I just used the turning tool as an example of the fine adjustment that can be gained with a cam type device.

If I were to deal with minimums I could actually work with one inch of travel on the vertical axis and two inches on the y axis(across the table). I want ten to twelve inches of accurate cut the length of the table and at that I would have to manually move the table back and forth and lock it down to position it. That should work pretty well with some good stops and a DRO. I actually need to be able to position pieces up to 32" long but my cuts will be much shorter.


Mike,

Sorry for any confusion. I was running out the door as soon as I finished typing this morning and may have been less than clear. The last thing I want to do is prove that we could put ID numbers on brass or bullets!! The turner just happened to be the tool that gave me the idea since it uses an offset like a cam to get super precise positioning. The necks on my brass turned to an interference fit over the base of my bullets was one component used to put together a string of sub-eighth inch five shot groups at 100 yards with my 6PPC. That is under the world record but of course was not shot in competition.

I don't know if a ballscrew or rack can actually work in ten-thousandths. Everything I have found says accuracy to .0002 I believe but then repeatability to .001, a far cry different! I assume the shorter the more accurate and it may come down to the usual. Will it cost me hundreds of dollars each to get the super accurate conventional short linear motion device? If so a simple circle offset one inch from center will give me two inches of total travel. Seems real feasible when I read I can divide a circle into 2000 parts. Of course I would have to calculate and then calibrate the cams but that should be fairly easy as the plain circles should have linear travel increments.

Of course none of this may be feasible, that is the reason for my original post! I have that itch we all get to start cutting metal and I need to layout the basic design of my machine first.

Hu

[lerman]

Well, instead of making a circular cam, make a linear cam twenty inches long.. Take a wedge shaped piece of metal with a slope of 1/10. Then using a cheap linear encoder with 1000 counts per inch, would give .0001 resolution.

If you are concerned with the space this would take, you could bend the wedge into a cylinder. Also bend the encoder.

Whoops. I think we just invented the screw. :-)

Ken

[shootingarts]

After I re-invented the wheel it was rounder and better than ever before!

Hu

[ViperTX]

I have to agree with "turmite"...use a higher resolution ACME or similar screw...precision ground ball screw of that size would be fairly inexpensive.

[turmite]

shootingarts I am quite familar with what you are doing since I am in the business...only in a different arena. I make bench stocks.

I am curious if you are wanting to design a cnc neck turner?? or just a better more accurate one?

Mike

[shootingarts]

Mike,

My current projects have nothing to do with benchrest or shooting although you may be able to give me some good advice on a taper machine. Seems like a duplicator would work.

I only used Don's toy as an example of how accurate a cam type mechanism could be. I had the part of my front rest that I revealed to a couple of people stolen and the last thing I would do is steal somebody else's design and tweak it to put back on the market. With a total lack of modesty I admit I may have designed the best front rest in the world. I'll be damned if I will say any more about the technical aspects unless I put it in production though.

Using my proto-type rest I shot a quarter inch aggregate. Not real impressed? I was using an AR-15 and feeding from the magazine semi-auto. Wasn't even using precision ammo. RVO once fired, primed, mil-spec brass. I chamfered the inside of the neck and stuffed the brass full of N-133 and put some great 52 grain bullets I got from Jef Fowler on top. Seated with an inline die and was good to go. That is when I decided to sell my 6PPC that was only about .05" better. Using the same rest I shot six twelve gauge shells someone had thrown or ejected onto the hundred yard berm in about 5-6 seconds on another day. There was one miss involved, I was running a speed test and my crosshairs drifted high as I broke the trigger. Missed about an inch as the crosshairs indicated and my next shot was on the way well under a second later.(Yes, I shot pistol speed competition too so I was used to stroking a trigger but not that one!)

Which stock do you make? Send me a private message if you prefer to not mention it here please. Idle curiosity at the moment since I have two on the shelf with no plans for them but I'm a bit of a stock junkie. Do you get on Go Go Varmint Go? An extremely busy live varmint board and I think that they buy more stocks than the competition guys do. The owners are friends of mine and I can introduce you to them if you like if you aren't already there. My user name is "Hu" there and everywhere on the net that will accept just two letters. I haven't shot in a few years now but the Louisiana gang benchrest shooters are likely to remember me.

Hu

[skippy]

EDITED