Advice On Inspecting a used Milling Machine

Copyright 1998 by Meridian Machinery

I have been dreading writing this section for quite a while. At issue has been how to deal with inspecting a mill without dragging along a lot of reference gages. Hopefully, you have read my advice on inspecting a lathe section. I will refer to some of the terminology and principles again here, so it might be a good idea to start there.

One thing I've said before is that I never bring indicators or any other devices with me when inspecting a machine. I try to be as unobtrusive as possible when evaluating a machine. A friendly relationship with the seller is important when negotiating the price. I believe you risk that when overly evaluating a machine (ie showing up with an indicator, straight edge etc.)

Secondly, if you rely too heavily on having an indicator with you, what are you going to do when Murphy's law catches up with you and you accidentally stumble upon a possibly great deal when you least expected it and your indicator is back home where you left it?

Next, I believe most readers of this article are novices. I'm not sure how competent a novice will be at taking readings with an indicator and interpreting exactly what they mean. Heck, I'm not sure I'm that competent either. To wit, I start with the following:

The (Potential) Folly Of Indicator Readings

**Note: If you are not familiar with terms like Quill, Knee, saddle, table, head as they relate to milling machines, a bit of homework on your part may be in order.

Allow me to digress for a moment with a short story. Some time ago, when I was employed as an Engineer on a Merchant Ship I came upon two of my fellow Engineers in the middle of a debate. Engineer No. 1 was bashing a huge check valve for the cooling water system with a sledge hammer to get it "unstuck" (believe it or not, they can get stuck and this method can help if applied with caution.) Engineer No.2 was arguing that there was no way the valve could be stuck in the closed position because the pressure gage in front of the check valve showed no pressure. My opinion was called upon. I thought for a few seconds. I knew that the text-book answer was that a centrifugal pump would not create pressure with it's discharge valve shut, but it occurred to me that there was a far simpler way to settle this matter whose results would be undeniable. I started the standby pump, observed it coming up to pressure, and shut its discharge valve briefly. The pressure gage dropped to near zero, no doubt about it. End of lesson One.
"Aha,"said no. 1, "You've proven that the check valve is stuck!"

"No," I said "I've only proven that it is possible the check valve is stuck. Equally possible is a problem with the pump itself."
How does this relate to the use of indicators and other reference gages? When evaluating readings, you must analyze what it is that the data can confirm as well as what it cannot confirm or rule out. It would have been a mistake in my story to have implicated the check valve based solely on the reading of the pressure gage because that reading could not rule out a pump failure. The argument continued long after I left the scene with each of the other two guys trying to prove his point. Unlike my two coworkers, I didn't have any particular axe to grind in terms of proving an opinion. An opinion can steer you off course when it comes to trouble shooting. In my mind there were several possible causes of the problem, none of which could be ruled out with the data presently in hand. To make a long story short, the problem ended up being a sheared key in the impeller. All detectives love to find a piece of evidence that is a "smoking gun." Unfortunately, more often than not, a smoking gun is not found. It takes the careful piecing together of various clues to reach a definite answer. Beware when you think you've found that one clue that proves everything. There is usually more to the story than meets the eye.

With this lesson in mind, let's look at some real life examples of the Potential Folly Of Indicators. I'll bet after you read this section, you may rethink some of the conclusions you've made in the past based on indicator readings.

When I was trying to figure out what to write about the use of indicators, I just started taking readings and pondering what if anything those readings meant. Like the example of my valve story, if you are in doubt, grab an indicator and try it yourself. I plan to try some more checks (relating to Lathe alignment) in the future. I have no idea what the results will be, but I'm sure I'll stumble onto some more of the potential folly of indicator readings in the process.

The first scenario:

You are inspecting a milling machine and you have with you a dial indicator and a magnetic base. You'd like to know if the Quill comes down exactly at a 90 degree angle with respect to the table. This would certainly be a requirement for accurate boring and drilling operations.

  • Your Set up (humor me here a bit if this wouldn't be your choice of setups):

    You mount the mag base and indicator on the mill table with the indicator touching the quill. You zero the indicator and proceed to move the quill downward. The indicator never moves off zero. Perfect, right???.......Wrong, you've proven nothing so far.

  • The folly:

    What if I told you that the quill in the above scenario was tilted (the whole head was tilted) 40 degrees to one side when this measurement was taken. It would not have changed the reading at all (unless the surface of the quill is irregular.)
    If you don't believe me try it some time.

  • The correct way:

    Ideally, you'd have to have a 90 degree angle plate of known accuracy and the head would have to be in tram (setup so an indicator mounted in the spindle on an arm would read the same front-to-back and side-to-side when swung in an arc while contacting the table.) The indicator would be mounted in the spindle and would contact the angle plate as the quill was moved up and down. We are no longer measuring from a fixed point in space.

    The second scenario:

    I show you a dial indicator setup in a fixed location (like in the spindle) to indicate a surface that is passing beneath it for flatness. You are not allowed to know what that surface looks like, you may only watch the indicator.

  • The results:

    The indicator doesn't move more than a few tenths (.0002" for arguments sake).

  • The likely conclusion:

    The surface is flat to within a few tenths.

  • The folly:

    What if I told you that you just measured a bicycle rim that was rotating about a fixed axis. Is that a flat surface? No, but it is passing under the indicator at a point fixed in space while it rotates about its axis. About all we know is that it is not out of round.
    If you don't believe me try it some time.

    Hopefully these examples will make you think next time about what it is you are actually reading and what if any conclusions you may draw from the results.

    So now we are faced with the realization that an indicator and holder will not get us very far without some other references like a reference square. Let us apply some of the knowledge gleamed from the previous examples.

  • Mission:

    To determine that the Knee and Column are square with respect to one another.

  • Problem:

    Looking back to the example where we measured the quill travel relative to the table, we realize that if we run an indicator mounted on the knee that is contacting the column (the vertical ways the knee travels on) our indicator is unable to give us any reliable data on the squareness of the knee relative to the column as we crank the knee up and down. Suppose someone rescraped the ways on the column improperly and they are nice and flat, but at an 86 degree angle (rather than 90) with respect to the horizontal ways on the knee? Our indicator would not really move from zero due to this angle, just like in the example of the quill. Picture driving a car up a uniformly inclined ramp that has no surface irregularities would an indicator mounted on the front bumper that was contacting the road really change with the angle?

  • What to do:

    The indicator reading described above is not without merit. My assumption in the above scenario was that someone accidentally scraped at the wrong angle. Assuming this not to be the case (not too unreasonable an assumption) we might collect some useful data from an indicator. The important thing to remember is that the needle not moving does not automatically make things perfect, however an erratic reading would be a danger sign. If you have nothing more than an indicator and base, I'd say that is about as much as we can conclude without laying a square between the knee and column.

  • Next Mission:

    To determine that the Table is straight, flat, and perpendicular to the axis of the spindle.

  • Problem:

    Remember the example involving the bicycle rim? It is possible that some warpage or curvature of the table will pass by the indicator undetected if it tends to travel in an arc. Unlike the rim in the example, it won't all travel around the same axis or in a perfect arc, but a near zero reading can not be used to rule out warpage or curvature with any certainty. As for the perpendicular part, remember the example of the quill. The table could pass by the single fixed point occupied by our indicator at any uniform angle and not register on the indicator. Our indicator won't help us draw a conclusion about perpendicularity either.

  • What to do: When we run the table back and forth and side to side with an indicator mounted in the spindle, we can see the tendency of the table to dip (move away from the spindle) at certain wear points. At the extremes of end to end travel on a big mill, we will also see the weight of the overhanging table cause the opposite end to rise a bit. The amount of dip can give us an indication of wear to the sliding surfaces. A few thousandths (.001-.0025") would be normal, more would be excessive. I said before that if the table passed under the indicator at a uniform angle, it would go undetected. How then is this dip possible? By uniform angle, I refer again to the example of the car driving up the ramp. If for argument's sake, someone cut several thousandths off one side of the saddle so the table angled to the floor slightly in one direction, that would cause a uniform angular displacement of the table. Wear tends not to be uniform and would only cause local displacements which would register on an indicator.

  • In Summary

    I hope that by now you will agree with the statement that it will be tough to draw any conclusions from any readings you take in the field and that owning and dragging along things like reference squares and 4 ft. straight edges might be beyond the scope of someone going to look at a milling machine for the first time.

  • An alternative:

    There is another way! We can hone your judgement skills and hope that you get a general feel for the machines condition that expert measurements (once you get the machine home) will corroborate.

    Advice on Inspecting a Lathe

    In (modest) Praise Of Klunkers NEW addition to advice on inspecting lathes.

    Back Home