The ER32 Spin Indexer

Wow, if you even made it past the title of this geeky machinist post, you must be pretty interested in machine work! A spin indexer is a useful tool that lets you precisely locate round stock in a milling machine, and rotate it very accurately in one degree increments.  This photo shows the indexing wheel that has holes every 10 degrees, and the 10 offset holes that bring it down to one-degree increments. There's a pin, not shown, that locks the rotation at the desired setting.

OK, but what's the "ER32" part? Well, almost all spin indexers use a collet to hold the round stock securely in place, and nearly all of them use a 5C collet. It doesn't much matter the exact specs for that, but it's important to know that with a 5C collet, your stock has to be very precisely sized, within just a thousandth of an inch or two for it to hold tightly. I had used 5C when I first began doing machine work, and found them very frustrating because so often you would find stock you couldn't clamp.

So, a few years ago I switched everything in my shop to ER32 collets. The ER system is more modern, and allows you to clamp from 3/32" to 3/4" with no gaps in coverage. Ever since, I have been looking for a spin indexer with an ER32 collet system in the nose instead of 5C. Finally, one came on the market. Here's a view of that "better" nose:

I told my machinist buddies, Phil and Jake, about my find, and they both suggested a good way to test the new piece. It must be able to rotate through all 360 degrees without the stock wobbling - otherwise, you can't guarantee accuracy. A quick check with a piece of drill rod and a dial indicator showed it to be quite good.

The photo above also shows the fly in the ointment. See the edges of the base at the bottom? No way those rough, painted edges would be parallel to the stock in the spindle. I checked it just for fun, and sure enough, it flunked. My advisers recommended mounting the indexer on another plate that could be mounted in the vise, and adjusted to be perfectly square.

I cast around in the shop, and found the perfect thing: a ground table that was designed to be part of a mill vise. These can be bought very inexpensively, and I had bought this one to be part of a custom vise in a band saw. You can read that story at this link.

I remembered that I had actually done precise measurements of the table for that project and stored them in the blog post. Great! I didn't have to do it again.

The only problem was that the indexer was just a bit wider than the table. Regular readers of this blog know that I'm pretty bold about modifying my own stuff. I bought it, I own it, and if I ruin it trying to make it better, so be it. In this case, I figured out a secure way to clamp the indexer in the mill vise, and milled 0.100" off each side.

Those edges were still not parallel to the spindle, so I wrote a warning on both sides to remind me never to trust them. Alignment will be on the ground edges of the table instead.

Using those precise measurements above, I figured out a hole layout that would align with the T-slots in the table, and set up to drill. The little post between the mill vise and the base is called a "machinist's jack" and this proved I need more of them, since using just one allowed a little wiggle at just the wrong time - I broke a drill bit. But I had another, and soon the holes were drilled and the base mounted.

The next step was to align the indexer precisely to the motion of the mill vise on the milling machine, and to do that I clamped a very straight piece of drill rod in the indexer, adjusted the position until the dial indicator showed no movement across about four inches of travel, and then locked it down. "Perfect" doesn't really exist in machine work, but this is plenty close enough for me.

The gratifying thing is that after this photo was taken, I removed the drill rod from the indexer, and removed the indexer from the vise. Next time I was in the shop, I put it back in place and checked it again, and it was still zeroed out. That means I can very quickly set this up for use when needed.

Finally, time to do some real work on it. I was continuing the project begun in Beck TD: Part 51, where I needed to drill some precisely located holes in a shaft for a transmission. First, I mounted the old piece in the indexer so I could determine the correct angle. Using less accurate means, I had gotten a measurement of 43.5 degrees from horizontal, and Phil, Jake and I all agreed that the actual spec was much more likely to be 45 degrees. Sure enough, after setting the shaft to zero, I rotated it until a test pin fit in the hole, and the indexer's lock pin dropped right in at 45 degrees.

Here's how I set the shaft to zero in the first place: there are two notches on the shaft that are used to capture ball bearings as a detent mechanism. I clamped a straight drill rod in one of the notches, and then used a digital edge finder to precisely locate the shaft to level, measuring equal distances from the centerline:

Finally, I felt I had all the information I needed, and I drilled the hole in the replacement shaft:

And it looks like I have a nice match. I can't test it fully until the transmission is back together, but I'm pretty confident that it's going to be OK.

Jake tells me that I really need a matching tailstock to hold the other end of the work, and I'm thinking I can make a custom one using a spare tailstock from my Logan lathe. But I'm very pleased for now - a new tool that is working well!