Sherline 8-axis Mill Setup for Morons
(or, “How you can be a total mechanical neophyte and still get it together without breaking it (too much)”)
This page documents my adventures in putting together the Sherline 8-Axis mini-mill. Here you see all the parts, laid out in their lubricated glory on what was (but is no longer) one of my wife's favorite fluffy white towels. I have already assembled the headstock and motor, as this is relatively straightforward. Also in the pic (click any image to see a larger version) you can see the various tools and measuring instruments needed to align the mill. In particular, you will need a precision machinist's square and a test indicator.
The Sherline instructions currently make some assumptions about the level of knowledge of the poor sap (that would be me) putting it together. As I am a total mechanical moron (otherwise known as a software geek), this makes me an interesting test case; if I can put it together, anyone can. Also, I'll probably make all the mistakes it is possible to make.
One assumption that I will make is that you have the 8-direction mill assembly instructions; I won't replicate them here, just annotate them.
Attaching the round column base (step 1) is pretty trivial, as is attaching the hold down bolt (step 2).
However, at this point, I made a mistake; I added what I now know is the spacer block and extension post. The instructions don't mention them, but the drawing is a little ambiguous (and the exploded drawing shows them), so I attached them. This turns out to be a mistake, since you can't align the mill for the first time with them on. I was scratching my head about how those calibration marks were going to work, because all the round elements except the base rotate; this finally became clear much later in the instructions when you're told to make a mark on the base to match the (aligned) position to the column top (which has all the rule marks on it). This mark that you make, along with the two marks on the spacer block, let you line everything up when the spacer block is in use. Note that all the lines and marks are on the right side of the column as seen from the front; the side with the left/right x-axis handwheel.
In the next step, you loosen up the flange bolt (the big bolt on the side) that lets the elbow of the bed & swing arm rotate, open it to about 90 degrees, and tighten it up again. Then you attach it to the column.
Next, put on the column adjustment block. This locks the column so that it cannot tilt front/back, and lets you make a fine adjustment to its orientation. You'll also want to use the 4 outer hex bolts to tighten up the column rotation axis so the whole thing doesn't flop around, and the 4 inner bolts to lock the vertical column bed to the rotary column (the big red dial).
Finally, assemble the motor mount onto the saddle (that rides on the column bed). This is actually kind of a waste of time, because if you read the instructions on how to square up the column, you'll realize you have got to take it off again…
Aligning the Mill
Before we begin, let's define names for some of the parts. Given that I am a pilot, I'm familiar with aviation descriptions for the various axes. If you imagine the mill as being an airplane, with the nose at the front and the tail at the rear, then the big red rotary column dial at the front of the mill (locked with the 4 bolts on its faceplate) is used to set our position on the roll axis (axis 8 in sherline's figure 1), the knuckle right behind it (locked with the flange bolt on the right [as seen from the front] and fine adjusted with the center bolt of the column adjustment block sets the pitch (axis 7), and the big flange nut at the back of the mill (that locks down the round column) locks both the yaw (axis 6) and the forward/backward “thrust” (axis 5). The first step is to get the roll and pitch (axes 8 and 7) axes roughly aligned using a machinist's square. You will note that at this point, I (a) hadn't realized I still had the extension post in, and (b) missed the instruction on removing the headstock & motor. Using the square against the front of the column bed, you can see if it's aligned correctly (a strong light shining from behind helps). Then you loosen the knuckle's flange bolt, adjust the central bolt in the column adjustment block to get it nice and straight, and tighten the flange bolt again. This sets the pitch.
In a similar way, you adjust the rotary column. I aligned against the sharp edge of the column bed, as it was hard to see if things were correct any other place (but, see below). To adjust, you loosen the 4 outer bolts on the front plate, and rotate to things line up. I found it easiest to actually do it by feel, using the machinist's square to hold the column bed in the right orientation while I tightened the bolts down.
It was at this point that I realized that I should have taken the motor off, as well as the extension block. After I did this, I had to align against the saddle, not the column bed, because my square was too big to do otherwise; I couldn't raise the saddle high enough to get out of the way.
The next step is to measure the built-in error of the y-axis. Frustratingly, Sherline doesn't explain why this is done, or how to use the results, but here is how to do it. You will need a dial indicator, which allows you to measure differences in position in units of 1/10,000th of an inch or less. Mount it in the spindle of the mill using a chuck. Move the little measuring tip (you can reposition it by hand) until it is at a 45 degree angle to vertical, then lower it onto the back of the saddle until the reading changes (I moved it until I got a full dial's rotation, which was 80/10000ths of an inch, as far as I can tell.
Then, using the Y axis knob at the front of the mill, move the saddle towards the back until the dial indicator tip is almost at the front of the saddle, and record the variation in the readiing. I got a reading of +16 at the center of the saddle and +10 (1/1000th of an inch) at the front, as compared to the back. This is at the low end of the range Sherline mentions.
The next step is to square up the ram (the yaw axis (axis 6) and the thrust (axis 5)). Both of these are clamped down by the big flange bolt on the round column at the back. You set the thrust to where you want it (usually, so the center point of the spindle is about in the middle of the Y axis travel), mount the test indicator in the vice (tip: move the Y axis so that the bed is roughly centered under the spindle, mount the vice, mount the test indicator so that it's as close as possible to the vertical column, and do the fine adjustment to get the test indicator to touch; as you'll see, after you finish this alignment, you should NOT move the Y axis wheel at all), and move it left and right on the vertical column. You want to get identical readings on each side of the column; that means the column is square. If one side shows a higher reading (and thus is closer to the front than the other side), a gentle tap on the horizontal swing arm on the opposite side will rotate the entire assembly a bit in the correct direction. Tap, measure, repeat until you've got them as close to perfect as possible, then clamp down the flange bolt.
I found it helpful to tighten the flange bolt 1/8 turn at a time, then remeasure to make sure I hadn't disturbed things, and noticed that if I used a 1/8 turn, wait a few seconds, 1/8 turn, wait a few seconds, sequence, I wouldn't disturb the setting at all – but if I just tightened it quickly as one would usually tighten a bolt, things would get thrown off.
The final step is to scribe a vertical line below the zero mark on the lower part of the round column (at the back-right of the mill) as a rough guide should you need to realign the mill later.
What Sherline calls “Squaring the column with the X-axis” is more precisely “setting the roll axis”, and what you are adjusting is the rotary column. Confusingly, Sherline states “Because the axis that allows you to tilt the column in and out hasn't been squared yet, you should only read the indicator at the same Y-axis location on the worktable that you used before.” But they don't say which previous operation they are talking about, why you need to do this, or mention in the previous step you shouldn't move the Y-axis. My guess is they're talking about the previous step, thus my tip before.
I won't go into as much detail about this operation (you should be getting the idea), but here are some tips:
- The only measurements that count are those on the extreme left and right as you turn the spindle. As long as they are equal, you're OK.
- If your test indicator doesn't have a swivel mount (like mine!), the Sherline boring tool can be used to offset the indicator.
- To get the clamping bolts to the proper tension for adjustment, loosen each one, then tighten it gently until you feel just a little resistance.
- When you've got things set, gently tighten each bolt just a bit, remeasure, tighten a bit more, etc. Gently does it.
Next we'll adjust the pitch axis, freeing the knuckle by loosening the flange bolt behind the rotary column dial, and doing fine adjustment with that central bolt. Read the instructions carefully, and realize that because the front and back points you are measuring are both in front of the pivot, their positions will change in the same direction but different amounts. All you care about is getting them to the same value, so:
- If the front reading is greater than the back reading, you need to tilt the column back, by tightening the bolt.
- If the front reading is lower than the back reading, you need to tilt the colum forward, by loosening the bolt.
- Because of the funky way the numbers will change, make small adjustments, and then even smaller ones as you get closer.
- It helps to reposition the test indicator vertically between measurements to get full benefit of its range.