Friday, November 1, 2013

5C Spindexer Indexer

A few weeks ago I proposed a follow along project on my YouTube channel oxtools. A viewer suggested an interesting problem involving indexing. His problem was how to accurately index 127 divisions for making a special gear. This might sound trivial at first glance but it quickly becomes difficult with standard tools and indexing devices.

The basic problem is 127 divisions is a prime number. To use standard circular dividing techniques requires compound indexing to achieve 127 divisions. Very few readily available and affordable dividing heads are equipped to do compound indexing without some modification. So what to do? The need for 127 divisions crops its head up when changing the gearing on a manual lathe to switch between inch and metric threading. As it turns out a factor of 127 happens to be 25.4 Starting to sound familiar now?

So now its actually an interesting project which is why were talking about it here. What I proposed in a video is to modify a standard cheap 5C collet indexer in a simple way to handle the 127 problem and make quick work out of typical spacing operations in the average shop.
These 5C Spin indexers are readily available and cheap at around $50 plus shipping. Many folks already have one of these indexers for the very same reasons. The project involves making some additional parts that allow the indexer to quickly index common divisions much like the common but much more expensive super spacer.

These units can already index 360 divisions with the help of a one degree pin vernier. Our modifications will make it quicker and simpler for small numbers of divisions while allowing us to directly index odd numbers like 127 and fractional angular indexes such as 7 or 11. Its limitation will be the ability to create the needed index plates and fit the desired number of hole on it.

Here is an example of an index plate with several patterns of equally spaced holes in it. The outermost circle of small holes has 127 equally spaces holes in it of 1/8 inch (3.17mm) diameter with approximately the same amount between the holes. As you can see in the drawing the disc needed to contain all these holes is pretty large. So large in fact that the stock indexer will not sit on a milling machine table without elevating it with this disc mounted. The maximum diameter that the stock indexer will support without a riser is roughly 5.6 inches.
In this shot you can see the maximum sized disc that can be used on the standard spin indexer when it sits on the machine table. So how do we get the desired 127 holes in a disc that will fit on the indexer without raising it off the table?
This is how we do it. By changing the radius that some of the 127 holes are on we can now put additional circles of holes and reduce the diameter of the disc to something that fits. The trick is that the angular spacing between any adjacent holes is exactly 1/127 of 360 degrees. The radius of the circle that the index hole lies on makes no difference as long as it lies on a radial line from the center of rotation of 1/127 of 360.
Here we see the needed parts to convert the stock 5C spin indexer to a spindexer indexer. In this example we show the 127 hole plate which is a very special case. For most folks a couple of different plates with some common divisions 2, 4, 6, 8, 10, 12 etc will make the most sense. The example of 127 holes is just to illustrate a method for making higher number or difficult number index plates of a small diameter. Obviously some of the common divisions can be indexed using the original plate and pin. The additional index plate speeds repetitive indexes and eliminated any possible errors of reading the angular scale.

The tapered arm is the index detent pin for plates with rows of holes on a single radius. It acts as a flat spring and engages the hole in the index plate with a tapered pin. To do higher order indexes with multiple index hole radii  such as 127 holes it is necessary to use the half index block part number 16. This slotted block allows you to disengage the tapered pivoting arm and re-engage it in another set of index holes on a different radius. The half index block is accurately located on the spindle centerline to avoid any errors caused by angular shifting.

To create the index plates the array of holes is put it using ordinate positioning in X and Y instead of angular indexing. Following the established jig boring practice of locating circular patterns accurately with this method is well understood and documented in books such as "Holes Contours and Surfaces" by the Moore Special Tool company.

The ability of the average shop to position accurately in X and Y rectilinear coordinates is higher than the typical ability to discriminate small angles. A simple milling machine equipped with a digital readout can make short work of making any desired index plate.

What follows are the drawings of the parts needed to modify the stock indexer and make it a spindexer indexer. Builders are encouraged to modify the design to suit their available materials and abilities and desired levels of embellishment and artistic style.









Thanks for looking.

Tom Lipton

13 comments:

  1. Tom enjoying the videos and these drawings are great. When I get some $ saved up I think I'll give the project a shot. Thanks for sharing the knowledge.

    ReplyDelete
  2. Tom,

    Since discovering your YouTube channel and Blog several months ago, I have really enjoyed reading your posts and watching the videos. I particular enjoyed this series on the Spindexer.

    I am an amateur machinist and wondered if you would consider a project that involves machining close tolerance and interference fits. Your recent video of the tailstock bushing touched on a number of questions I have regarding how to achieve close tolerances. I frequently struggle with trying to creep up on the final dimension and frequently overshoot resulting in my interference fit turning into a slip fit!

    Thank you again for sharing your remarkable expertise in such a professional manner.

    Blair Tyson
    Durham, NC

    ReplyDelete
    Replies
    1. Hi Blair,

      Thanks for the nice comment and excellent suggestion for a video. There is much hand wringing and confusion about press and interference fits. The charts and tables in the handbooks don't tell the whole story. I agree this is an area that needs some rules of thumb to help newer folks. Its not a simple magic bullet recipe. I'll think about how it could be presented in a video format and be meaningful.

      Kind regards,

      Tom

      Delete
  3. Tom,
    I recently found you on youtube, and from there your blog here. I enjoy your work. I really liked the "Tour of the Dial Indicator", I also have accumulated quite a few of them over the years. My favorite every day use test indictor is an Interrapid 312b.

    ReplyDelete
    Replies
    1. Hi Louis,

      Thanks for the comment. The Interapid is also one of my favorites. The long range and ultra smooth movement make is a pleasure to use.

      Regards,

      Tom

      Delete
  4. Tom,

    Thanks for sharing this, very nice mod. I had already planned on modifying mine to make it more useful and a quick and easy replacement to using a rotab, this will just get me to do it sooner. Btw, check your spelling on "industries" in your short bio at the top of your blog.

    ReplyDelete
  5. Hello Tom,


    No news from this blog since the 1st of November.
    All your knowledge must be collected in your videos ?
    --
    St├ęphane

    ReplyDelete
    Replies
    1. Hi Stephane,

      Sorry for the lack of attention on the blog. I have been focusing on the YouTube channel as it get quite a bit more traction than the blog. That said the blog is better for some subjects so I promise not to abandon it.

      All the best,

      Tom

      Delete
  6. Tom, I stumbled upon your myriad U Tube posts a few days ago and drat, I'm losing sleep. They are addictive! Thanks much for taking the time to help so many others.

    In viewing the spindexer indexer project, I may have a simpler approach to the design. I would machine the 63 holes on 2*360/127 or 5.66929 degree spacing as you propose. Rather than machine the second circle of holes, I would use an indexing plate similar to the original vernier plate with a hole located at an arbitrary 0 degrees and a second hole offset by 3*360/127 or 8.50394 degrees and a third hole offset by 6*360/127 or 11.33858 degrees. The first hole will give you the first 63 odd holes, the second will give you the 63 even holes, and the third will give you the last odd hole.

    I was also thinking about how to best make the dividing plate. My Grizzly DRO has a bolt circle capability which makes it fairly easy. If I had to machine it by the numbers, I would calculate the coordinates in an Excel spreadsheet and print out the list so I could check each hole off when completed. One point to consider if not using a DRO is backlash; always approach the positions from the same direction.

    Another way would be to borrow someone's 127 tooth gear. Make a jig to hold a temporary axle and machine the indexing plate to fit. Fasten the 127 tooth gear to the plate.. Make a dog to engage a pair of gear teeth and index the new plate in this manner. The same jig can be used to make the stationary plate, ensuring accurate radial alignment of the set. Ream the holes for tapered dowel pins for best accuracy.

    Bob

    ReplyDelete
    Replies
    1. Hi Bob,

      Thanks for the detailed comment. I like your idea for a vernier plate. However the holes would have to be placed very accurately to avoid an accumulation of error. When you have all the holes it has a averaging effect on the error. Think of it like using a pair of dividers to set a particular fraction of a circumference. It all looks pretty good until you try to close the circle. The accumulation of error for high number divisions is significant. Not to say I don't like your elegant solution to the 127 division problem.

      Kind regards,

      Tom

      Delete