No announcement yet.

Deep Hole Drilling 101

  • Filter
  • Time
  • Show
Clear All
new posts

  • Deep Hole Drilling 101

    In the past year, there have been several inquires into drilling deep holes
    Lots of fellows have their own method and I for one will not foresake anyones pet way of doing things.
    I did want to point out several important things to consider when you undertake such a project.
    Us old timers have been around the shop long enough to see things that make us shake our head, but hey, they work, so you don't argue
    We have seen 'perfect drills" not work and those not perfect do a good job...Why ?
    Perhaps it's because we don't fully understand what we are engaged in ? or are lucky ? been there a few times !

    Well, maybe the following will help some in this discussion.
    First drill points
    Why 118 degrees ? why 135 degrees ?
    You know that 118 degree drills have sharper points , which is good for guiding ( in some cases), BUT they have longer cutting edges.
    135 Drill bits have shorter cutting edges and that means a smaller drill face OR in simple terms more HP available per drill size to cut into tough steels.
    Anyway, back to deep holes, except for one very important fact.
    Note that a 118 or 135 degree drill does NOT match the same angle as a 60 degree center drill. This is important for precison considerations.
    Does a centerdrill work...absolutely..but don't trust it !

    How you approach a hole where you want accuracy of centerline is critical. I am not talking hole size, or finish, but strickly keeping the drill bit on center
    PART 1
    Here is a sketch of a hole being drilled in a drill press or Mill

    Now I have exaggerated the bend but please "know" the concept.
    I was taught this years ago by a expert in gundrilling. Any drill will deflect.
    Causes are incorrect sharpening, poor starting location/error, spindle/ chuck issues, or hard spots/inclusions in the material.
    The thing to realise is that on a drill, the tips or outer edges of the bit see the most load. They cut the most material...obviously
    When the drill is rotating however, the tips both travel at the same speed ( say 1000 inches per minute) as they revolve around the centerline of the drill.
    This means if the drill drifts, it does not know it and continues of its merry total error. Note the centerline of Rotation always coincides with the flute tips ! This drill will follow the line of least resistance

    PART 2
    Now lets look at rotating the work piece and keeping the drill stationary.
    You normally see this in a Lathe setup. We will use the same drill here.
    The physics is now different. The center of rotation ( COR) never changes !
    This causes the drill bit tips to try to stay balanced

    Neither one wants to do more work than the other.
    Note as the distance increases from the COR the loads get greater
    As you go outward the distance for the tip to travel increases, so the 1000 IPM may go to 1001, and the other tip drops to 999. This causes the drill bit to self correct, as the drill will bend back literally to keep the loads equal.
    ( By the way, this is why some poorly sharpened drills still cut OK, even though the angles are not equal, the cutting loads are balanced, so the drill is happy .. do not do this however )

    Part 3

    Now we look at Gundrilling.
    The thin sheet metal drill, becomes hard as a rock when the oil flows through it, because the discharge hole is smaller (restrictive)
    The drill bit is retracted to show what the hole looks like .
    It is quite different from a normally seen hole

    The gundrill has a single cutting edge (flute ) with two seperate angles seen here. I have flattened out the drill so you can see it easier.
    Like the lathe in #2, Gundrillers generally rotate the work piece for accuracy.
    The two angles on the carbide fight each other, with the outer tip trying to win, by forcing itself inward, however all that force is repelled by the Heel or wearpad of the carbide on the opposite side. The pad is riding on the fresh smooth surface of the hole. This is similar to a car with lots of toe-in on the front wheels . the angles are critical because the inner angle takes as much load off the outer tip as possible, and yet keep the tip from twisting off and going straight.

    Remember the discussion on centerdrilling the start hole ?
    Now you can see why a prepped hole made with a Ball endmill is so important.
    The hole is to size, and guiding the drill flutes, which will always give better accuracy for deep drilling as it is a larger surface than a small hole!!
    Hope this helps fellows

    Green Bay, WI

  • #2
    Thanks, Rich. Very nicely written, easily understandable, valuable information!

    I would add another technique: say you need a 5/16" hole through a 5" block. That's 16 diameters, a deep hole. Lay out the hole on both ends, and drill undersize from one end halfway through. Then flip the workpiece and drill from the other end. Now switch to your 5/16" bit (or reamer) and drill straight through.

    Maybe I got lucky, but it worked for me!



    • #3
      Great post Rich. Thanks for such a clear and concise description of the factors involved in deep (for that matter, any) hole drilling.

      I nominate this post for inclusion in the Favorite Posts thread as a primer on drilling.
      Jim H.


      • #4

        I thought about your explanation, but I think there is something wrong. In the case where the hole is off-center, and the workpiece rotates, there is NO change in the sfm from one part of the hole to another. Consider the case where the hole has deviated by half a diameter, so that one edge of the hole is on the rotation axis. The sfm of the spinning workpiece is zero here, but the sfm of the drill is not, even though the drill is not spinning. The drill face is translating in a circle, and so flutes still scrape by the rotation axis at the usual SFM. Put another way, it is not possible for the sfm to vary across the circumference of a hole because that would somehow involve the flutes of the drill compressing together and expanding apart.

        I can't see any difference between a spinning workpiece and a spinning drill, assuming the mill is not out of tram. Comparing both setups and assuming the drill has deviated, in both cases the natural springiness of the drill tends to push it back to center, even though this effect is small for deep holes.

        Neglecting the springiness of the drill, there is also a very very small inertial effect, where in the case of the spinning drill, a crooked drill stays in place. If the drill is crooked and the workpiece is spinning, the centrifugal force tends to throw the crooked drill even further off center.

        So therefore I am still wondering why it is better to spin the workpiece. I know that's how everybody does it, so I must be missing out on something. When I did deep hole drilling in a lathe, I was able to drill a 80 * D hole with nearly no deviation. I somehow feel this must be due to careful setup and not the physics of rotating coordinate systems.


        • #5
          No discussion of pilot drills and/or solid carbide oil-fed drills?

          I've been doing some holes requiring 20xd and 25xd depth of hole. Using a 150؛ pilot drill to go 2xd, then switching to the 140؛ long drill, I get very straight holes and 2x faster than gundrills because they've got two cutting lips to the gundrill's one.

          My experience in doing this kind of work on CNC swiss screw machines teaches me that it works best when the drill and the work are both rotating. I get the least runout this way, and I have no idea why. Doesn't matter that I don't understand it, it works.
          Last edited by PixMan; 09-07-2010, 08:32 PM.


          • #6
            Thank You for your questions Beanbag

            First, I didn't address a severe off center hole start so your example can be discussed but it is impossible to replicate in the real world. No drill would allow such an error before breaking. Poorly placed starting holes prevent any accuracy for hole location and I really don't want to go there as you have maybe 10 MORE variables, but here is your question :
            ...". In the case where the hole is off-center, and the workpiece rotates, there is NO change in the sfm from one part of the hole to another.".....
            I disagree Beanbag , and tried to show that in photo #2
            The portion of the hole furthest from the C/L of rotation inscribes a larger circle ,or circumference relative to the C/L and thus has a faster speed.
            The reason I was showing speed was to show that it had more material to remove (per revolution), compared to the inner flute. The more material, the more relative load ( look at a light boring bar, taking a heavy cut, compared to a light cut)
            We know that drill bits dislike unbalanced loads
            When you enlarge a hole, you cannot make the drill go to one side, because the drill always wants to see a balanced load on the flutes, thus it follows the pre-drilled hole.( By the way, we are only talking about 2 flute cutters)
            I have to admit, I don't understand what the springiness or inertia of the drill has to do with maintaining accuracy. I would say they are detrements.
            Look at an aircraft drill which is solid, or a Gun drill which is as near a solid as possible. A drill not spinning has no inertia, other than static inertia, in which case, if the load presented to the flute is less than the yield point of the drill design and material, the drill will cut it. If the load is greater , the drill deflects

            I was hoping this would help understand the physics of deep hole drilling.
            While we did gundrilling at work, our gundrilling vendor in Paramont CA would quote us with this accuracy ratio.
            Round work drilled had .0001 per inch runout (think lathe)
            Large stationary work .0015 per inch runout ( think Mill)

            Others can be better or worse

            thanks for the questions
            I hope I clarified them
            Green Bay, WI


            • #7
              Pix Man
              You have several things working in your favor
              First, is the 2 x starting depth, made with a starting drill (rigid) ( and more important, controlling the OD of the hole to match the flutes !)
              (and the accurate center location found on a Swiss)
              The other is using a solid carbide drill, which is far stiffer than a gundrill and that means bigger chip load. The two flutes help .
              Turning the drill AND the workpiece is not unusual and gives you the benefits you mentioned. The turning work give accuacy, and the drill bit gives production.
              You are in the perfect place to run a test by not turning the work piece and seeing what happens
              Green Bay, WI


              • #8
                Beanbag: SFM (Surface feet per min, a linear movement measurement) is RPM * Diamiter * PI / 12"
                Meaning, it gets lower the smaller the diamiter.. As in, the closer you get to the center. So a drill is exposed to a wide range of SFM.

                Also why a large drill works soo much better with even a tiny pilot hole: The center of the drill has very low SFM, and the max RPM is much lower with a large drill or you'll burn out the outside lip of the drill with high SFM. a small drill can do much higher RPM, hence the inside SFM is higher.

                Question: How stiff is a gundrill anyway? Stiffer then HSS? carbide? I assume it depends on pressure used, but lets talk about typical applications (maybe with a subnote for some exceptional high/low pressure applications)
                How much pressure is typicaly used inside a gundrill anyway?
                Last edited by Black_Moons; 09-07-2010, 10:40 PM.
                Play Brutal Nature, Black Moons free to play highly realistic voxel sandbox game.


                • #9

                  My example was for a hole that started straight, but then deviated at some depth. But my point is still the same. Even if the SFM of the spinning workpiece is lower the closer you go to the rotation axis (no argument there) the important points are these:

                  1. A wobbling drill (from a hole that went off-center) also has a SFM, even if it is not spinning around its own axis.
                  You know how those little hand held electric wood sanders move the pad in a circle by translating but not rotating?

                  2. The material removal SFM is the workpiece SFM PLUS the wobbling drill SFM.
                  (Technically, it is the difference between the vector quantities)
                  In the case of an off center hole on a lathe, the SFM around the circumference of the hole is still constant. I think that due to the properties of geometry and such, this must be true.

                  You are right that the other points I mentioned are minor distractions, but those are the only differences I can think of between using a lathe and a well-trammed mill.
                  Last edited by beanbag; 09-08-2010, 02:45 AM.


                  • #10
                    I made a picture to demonstrate my point. The math also works out if the hole is only slightly off center. The main thing to notice is that the entire face of the drill (flutes, web, center) translates in a circle of radius r2 and thus has a SFM of 2 pi *RPM *r2.
                    What I call the total SFM is the speed at which the tool scrapes against the stock, which is the relevant material removal SFM.

                    Last edited by beanbag; 09-08-2010, 02:42 AM.


                    • #11
                      Originally posted by beanbag
                      I made a picture to demonstrate my point. The math also works out if the hole is only slightly off center. The main thing to notice is that the entire face of the drill (flutes, web, center) translates in a circle of radius r2 and thus has a SFM of 2 pi *RPM *r2.
                      What I call the total SFM is the speed at which the tool scrapes against the stock, which is the relevant material removal SFM.

                      I don't even pretend to be a math whiz but in your diagram I can't see how the material would not be moving faster past the the upper flute of the drill than the lower. Unless you are saying that the drill has to be following a wobbling circular path if it gets off-center? If that were the case then wouldn't the result of the drill deviating be a conical shaped hole thats on-center?


                      • #12
                        I do pretend to be a math wiz, and I really don't see it beanbag. Id argue with you, but after 3 days argueing over trig/math with my housemate, im sick and tired of convinceing people of thier incorrect math. Everyone here has been disagreeing with your view of it, can't you just accept that everyone might be right and you could be wrong?
                        Play Brutal Nature, Black Moons free to play highly realistic voxel sandbox game.


                        • #13
                          Beanbag is correct in that each and every point on the drill is describing a circular path of radius r2, but in space. This might indeed imply that the average SFM IS the same for each point.

                          The error lies in the fact that none of these paths (except the drill centre) is concentric with the rotating workpiece, so the instantaneous SFM is different for each point, and each varies sinusoidally. The two flutes are 180 degrees apart so when one flute is travelling at its fastest relative to the work piece the other is at its slowest and vice versa.

                          HOWEVER...., all this being said, I'm confused... It seems to me that the difference in cutting speeds/forces of the two flutes would produce a force tangential to the workpiece rotation, and NOT the force acting toward the centre of rotation required to self centre.

                          The situation in the diagram above is rather extreme because only the upper flute is cutting - the lower is rubbing backwards, but this does however raise an interesting point - even when the offset is small, there will be some small part of the lower flute which is rubbing backwards...

                          So - If any part of a flute is rubbing can another part of the same flute be cutting ???

                          If the answer is no, could it be the reaction force at the rubbing flute be a part of the correcting action ???




                          • #14

                            The lower flute is not rubbing backwards. It is cutting as usual. The negative sign for the SFM is due to vector mathematics and stuff. I can make another diagram later tonight.


                            • #15

                              Many apologies, you are quite correct. If the workpiece rotates 'n' degrees, then the bit rotates 'n' degrees in the hole, irrespective of the hole's translation.

                              Cheers !