Dividing/Indexing questions

[jgarrett]

I am new to the group and a wannabe novice machinist. I have a EMCO V10-P and want to be able to make small (<3") gears. I am totally confused on whether I need a rotary table with indexing or a indexing head. I have my eye on a 4" rotary table with index plates that LittleMachineShop has. Can someone fill me in on what I need or point me to a book or article that explains what the difference is in the two. I have the Sherline book but their method of indexing is not what I want. ( I think).
Any help would be appreciated.
Thanks,
Julian

[seal killer]

Julian--

I am a wannabe as well. So, I can't answer your question with any authority but my GUESS is that you will need the rotary table with the dividing plate to make your gears.

Grizzly's H5940 4" Rotary Table w/ Indexing is less expensive than the LittleMachineShop Rotary Table, 4" with Tailstock & Dividing Plate. However, although I have what I think is the same Grizzly table, I don't have the dividing plate.

I will watch with interest the replies you get . . . we will both learn something!

--Bill

[ALCOSTEAM]

You will need a dividing head. The rotary tables with the masking plates are not precision enough and do not have the multi hole plates that allow for just about any gear spacing you can do. Looking at enco, travers, and acouple of the other suppliers the dividing heads are not even as expensive as the rotary tables. I have a tiawian built one that came with my first milling machine (an enco mill-drill) and while I got rid of the mill-drill I kept the dividing head set.
Right now I have my dividing head set up on the bridgeport turning out gears.

tim

[JimGlass]

Julian;

A dividing head has sector arms that are adjustable for making partial turns of the indexing crank handle. Rotary tables do not always have indexing plates or sector arms but some do. Often indexing requires so many turns of the crank handle plus so many holes in a certain hole circle. Without the indexing plates and sector arms you will then need to keep track of the exact angle for each gear tooth within the 360* circle.
I have made spread sheets that add each angular division for each gear tooth in a circle and made the gear successfully. But, the dividing head is the best way to go.

I'll assume you need a dividing head with a horizontal axix then mill the gear teeth with the gear cutter mounted to an arbor.

I have made a number of custom gears in my career. They appear complex at first to make. But, knowing how to use a few formulas and understanding indexing is really all there is to it.

Good luck
Jim

[jgarrett]

Thanks for the replies...Now I just have to find a good deal!!!
Julian

[toastydeath]

I would like to argue that rotary tables used for indexing are much, much more accurate than it would appear. Using one example, a Moore rotary table is accurate to +/- 2 arcseconds at any point in the rotation. They're a bit pricier than other tables, however. I am not aware of any model of dividing head that is manufactured to the same standards of accuracy as the high-end rotary tables. That does not mean you can't have one made, or that one doesn't exist.

The only functional differences between dividing heads and rotary tables:

Rotary tables are best suited for arc milling. They're a rotating table, t-slots or threaded holes. They have fine angle positioning and are not limited to indexing plates.

True dividing heads are best suited for chuck work, helical milling, and drilling holes in long shafts. They can work on longer objects with a tailstock, have a variety of options to mount things with chucks, collets, or centers, usually have some provision for compound angles, and most importantly, a real dividing head will come with a gear train to do helical milling and hobbing. They operate with index plates, a plate with holes in it to provide fixed angular increments to rotate the work.

There is no fundamental reason why a rotary table would be more, or less, accurate than a dividing head of the same quality. The only real difference is the positioning features and mounting options that each have. The only way to determine which of two arbitrary rotary fixtures is more accurate, regardless of type, is to test the indexing accuracy of each fixture. You can buy accurate rotary tables, and you can buy accurate dividing heads.

I would note that neither rotary tables nor dividing heads are the most accurate form of angular location.

[seal killer]

toastydeath--

Re: "I would note that neither rotary tables nor dividing heads are the most accurate form of angular location."

I am a newbie and may not understand your answer. But, if you care to give it, what is a better form of angular location?

Thanks.

--Bill

[toastydeath]

Bill: No sweat.

A quick preface. In a machining sense, nothing is going to top a microsine table, dividing head, or rotary table. So, if your interest is, "What's the best way to machine angular relationships?", than any of those items can be had to about the same accuracy. The rotary table will probably win, in terms of absolute limit of accuracy, if you have a jig grinder and are trying to do something ridiculous.

For a guy in a shop, even a professional shop, you don't need something past those devices. There's just so much other crap that will cause error that once you have a decent fixture that fits the kind of work you do, you don't have to worry about that as a source of error. And if a person DOES have a need for something beyond that, they certainly already know what that is, and how to use it properly.

So, all my rambling aside, and with the understanding that none of this applies to a home shop under any circumstances: Off machine in the inspection and metrology laboratory, there are a couple other options for measuring angle.

Probably, the fastest, most convenient, and very adequate would be a master angle index of some sort. They're like a rotary table, except they only index. Some geometric mechanism such as interlocking gears, very precision ball bearings, et cetera, establishes angle to fractional arcseconds. Again using Moore as an example (because they widely publish the tolerances of their devices, and I can remember them), their Master Index Table is accurate to +/- .1 arcsecond, which is 0.000027777.. degrees. It is repeatable far, far beyond that. Most other master indexes are of similar quality. They're not cheap; last time I saw a used Master Index, the guy wanted (and I believe got) 30k USD for it. There are some other "rotary indexes" that can use multiple mechanisms for angular location, and can subdivide whatever index spacing is given (to a lesser degree of accuracy). There are also prismatic angle standards, which can't be read directly, but can be used with some of the other instrumentation to detect angle.

Better yet, is an autocollimator. This is an optical instrument that can usually get fractional arcseconds at considerable distance. Some of the better electronic autocollimators can get .01 arcsecond resolution (maybe .001, I haven't checked recently). Autocollimator accuracy is affected by air currents through the optical path, so some kind of shielding and environmental control is necessary at high resolution.

Best, in terms of accuracy, but slowest and most sensitive to error, is interferometry. By using the interference fringes of monochromatic light there are a whole bunch of ways to detect ridiculously minuscule changes in angle, each with its own set of advantages and disadvantages.

[seal killer]

toastydeath--

I've spent the last thirty years in higher education. You write as if you are no stranger to academia. I understood every bit of what you wrote. Including your unstated simile to the "last mile" problem experienced by the telecommunications industry . . . and usually everyone else!

Thanks!

--Bill

[toastydeath]

Haha, you guessed it.

I'm a 21 year old college student and career machinist.

I'm studying mechanical and precision engineering, trying to go into precision machining engineering in the long term. My employer is quite generous, and provides plenty of resources and opportunities to pursue (and work on!) my passion for precision machining and meterology, and to talk with some of the big names in the field. They also have some cool toys that I wouldn't get exposed to at the average job shop.

[seal killer]

toastydeath--

Take advantage of EVERY opportunity your employer GIVES you to further your education and experience in your chosen field(s). At 21, even if you decide to become a chef, those opportunities will pay off big time down the road.

--Bill

[Jose Rivera]

Hey toastydeath, you are sure going over my head.

I worked for seven years on the Space Shuttle Program doing final machined parts inspection.
The place was the best tooled inspection department that I have ever worked at.

Also worked for eight years at the U.C. Berkeley Micro Electronics Research Lab where the last job I did there was the first prototype microstepper that made the first .3 micron wide lines using deep ultra violet lasers.
After the success of this equipment, all CPUs and memory chips are made now with this technology.

To complete this job I had to make things with no more of 1/10 of a micron error, otherwise the lens will be out of focus and will render any prints useless.

I surely would love to have one of those instruments that you're mentioning at hand. Never saw of heard of one at either place.

Working with 1/10 of a micron tolerances was like working blind. There was nothing available that could help me measure what I was building. Everything I did was checked with gage blocks and .0005 resolution finger indicator.

Anything bellow one micron the eye cannot see it anymore. Lines created a sub-micron sizes cannot be seeing even with the most powerful optical microscope, only with an electron microscope.

Once installed, the adjustments built to allowed for any miss alignment where never used. It worked right out the first time.

With 21 years on you back you sure have impressed me with the knowledge you have. Goes beyond anyone that I have had to work with !!

If interested send me a personal e-mail address and I will send you a CD with a video of this equipment and others you may find interesting done for E.E.C.S Micro Lab.

Jose Rivera