Going Digital, Part 1

I have the great good fortune to be the choir director for two really good church choirs at Lancaster Church of the Brethren - the Chancel Choir and the Carillon Bell Choir. These folks are talented, hard-working, good-natured, caring and generous! And their generosity evidenced itself this Christmas with an extravagant gift of cash to me. I was humbled and deeply appreciative. I told them that I would spend the money at the Grant St. Garage.

Part of the money went to upgrade my table saw dado set to a much better one (this one), and the rest went toward this: a set of three iGaging Digital Readouts for my milling machine. Digital Readouts are universally abbreviated DRO, and pronounced "dee-are-oh", not "dro."

The readouts have three parts: the slides (in this set, 8", 12" and 24" long) the head, which measures the distance moved, and the display, where you read the measurements. They measure to 0.0001" - that's one-tenth of one-thousandth of an inch! Guaranteed accuracy is actually a bit less, more like a thousandth per 3 inches - plenty good enough for the work I do. This set was designed for a larger mill than I have, so part of the project is figuring out how to make them fit. The other part is that my mill was not designed to have DRO's, so there would be some figuring involved there too.

The project started with a visit to Grant St. by my friend and metalworking mentor, Phil Oles. We did that important task of "standing and thinking" for about an hour and a half, holding various slides to various parts of the mill. We decided on a configuration that would fit and not compromise the mill's range of motion in any way.

A milling machine moves very precisely in three directions, called "axes" - the plural of "axis." I had to double-check the spelling of that plural, and quora.com noted this: "Interestingly, axes is the only word in English that can be the plural of three different singular noun forms--ax, axe, and axis."  

The axes are labelled X, Y, and Z, just like in coordinate geometry. On a mill, the X-axis moves left and right from the perspective of the operator, the Y-axis moves in and out, and the Z-axis moves up and down. I decided to start with the Y-axis, because later changes for the X-axis will cover part of it. Installing the rack was pretty straightforward - just marking, drilling and tapping a couple of holes. You can see that I had to grind the washers on the bolts holding the mill to the table, for clearance.

I was more focused on the work than on taking pictures, so not only did I get my finger in the picture, I didn't really get all I should have. The point of the picture is that a magnetic base, only partially shown, is holding a dial test indicator to ensure that the track is perfectly level relative to the mill table. I did that to less than 0.0005" before I realized I would have to remove the track (twice, in fact) and re-do it!

The part of installing DROs that is always daunting is figuring how to attach the movable head to the part of the mill that moves. I started by making a plate that mounts to the back of the head, first by carefully marking some hole locations. the L-shaped bit with blue ink is the bracket being made:

After drilling holes and milling the back so the screw heads would be flush, I mounted it on the Y-axis head:

The track was reinstalled and leveled (which machinists call "trammed" for some reason), and then I had to decide how to tie it to the Y-axis part of the mill. In research on YouTube, I found that most implementations used a slightly flexible arm. I made mine out of 1/8" thick, 1/2" wide iron bar. Here it is installed. Yes, that's my finger again.

It's not obvious from that picture, but the mill base is actually slanted top to bottom where I mounted my track. Some implementations go to a lot of trouble to make precise wedges, but I just used a vise, a great big clamp, and a 3-foot length of square iron tubing to bend the arm slightly:

Of course, I bent it the wrong way the first time, but iron bends both ways. After getting that done, it was time to test. The mill has dials that precisely measure movement. They can be tricky to use, because one rotation is 0.050" - 50 thousandths. I wanted to test over 3 inches. That's 60 turns of the wheel. When I did that, here's what the DRO said:

That's off half a thousandth in 3 inches. I can live with that! Even better, when I wound it back 60 turns to the start, the DRO read this:

Being able to get back where you started is called "repeatability," and it's super important for accuracy. Man, I think I'm gonna like these DROs.

I then moved to the Z-axis. Phil and I had decided to use the 24 inch scale there, because there was plenty of clearance. The instructions with the DRO say to cut the track to fit, but I didn't want to do that. Looking at the brackets, I realized that a slight modification would allow them to be used without cutting the track. In the left photo, you can see a bent stop that keeps the bracket at the end of the track. On the right, a bit of quality time on the grinder removed that, so I could slide the bracket down the track.

I drilled, tapped and mounted the Z-axis track. You can see in this photo how the modified bracket works:

Using a similar process to the Y-axis, I made a bracket to tie the DRO head to the mill. The hole on the end is so big because I'm using an existing bolt hole that is 10mm in diameter. Much larger than needed, but it was handy so I used it.

I mounted the bracket on the head and reinstalled on the mill. Then, I used a machinist's trick to figure out how thick a spacer I needed. I tried various drill bits in the gap until I found one that fit perfectly. It was a #42, which measures 0.093" in diameter. I used the lathe to make a custom washer of that exact thickness. On the left, I'm drilling the hole for the bolt, and on the right I'm parting the washer to the exact 0.093" thickness.

I also had to shorten a 10mm bolt to tie it all together.

And here it is all installed:

I didn't go to the pain of checking for parallel on that Z-axis track, because a small angular error on a track that long is negligible. Instead, I checked it against this:

\\

This is called the "poor man's DRO." Made from a digital tire tread depth gauge, it has magnets on the back and nose to put it in place. I have found it to be very accurate over its short range. I compared a 0.750" movement on that vs. the new Z-axis DRO, and they matched to the thousandth. Plenty good enough.

In the next part, I'll tackle the X-axis - by far the most difficult. But the good success of Y and Z gives me plenty of steam to get this done. It's going to make a huge difference in the capabilities of this little mill.