Drill Press Accuracy (Lack Off...).

One thing I haven't seen in this thread is how much drift are you seeing. As has been pointed in other threads, drill presses are not precision devices. Perhaps the single biggest source of error is our ability to visually align the work with the drill.

If you have a mill, try this experiment. Scribe and center punch a piec of plate and clamo it to the table. Mount a small drill and visually center it and note the x and y table position. Then move the table off center and repeat the process, each time recording the position. Do this ten times. This will give you a good idea of your ability to visually locate a drill. Extra credit: now mount a 3/8" drill and repeat the experiment.

My eyesight is nothing like what is was twenty years ago. I can usually get within .01" with careful alignment from two viewing positions. More casual alignment, is more like .02".
 
I am getting some mixed messages here.

1. We seem agreed that a drill press with worn out bearings and run-out is a "bad thing", and that the experience is immediately better if the bearings are replaced.

2. We take it for granted that the rod in the chuck had better have much the same run-out as the chuck itself.

3. We have the contribution from @RJSakowski that the progress of the drill bit in making the hole can allow some variation in where it started from overhead, at the quill and bearings. Moreover, that some amount of play up there can actually give the drill bit a better chance of getting going at the punch mark.

4. We then have, into the mix, good hints from @NortonDommi about keeping the top of the punch in good order, and introduces the concept of having the work underneath free to move a bit, and "allowing the drill to find it's mark". RJ's point of the bow drill by ClickSpring still makes holes where he wants them to happen, despite the hand at the top wandering all over the place, is telling!

Given the drill bit is always a lot longer than it's diameter, so not acting insistently like a rigid milling tool, something else is going on. The small straight edge across the central core of a drill bit helix, at the pointy end, is clearly usually wider than the punch pop mark. It bites somewhere, and does a little squiggle, either by being allowed movement in slop from above, or by the workpiece under sliding about a bit, or by the drill bit bending a little, or some of all three. Where a center point starter is used first may be a separate situation, somewhat more forced, but still may allow the first two movements.

Given that we all aim the bit at the pop-mark as best we can, and it kind of "pulls itself into order", and except for the state at @Nogoingback 's machine , the holes usually end up close enough to accurate (do they?), we don't much care to wring out the real mechanism of what goes on in the first instants of cutting. There may even be some of the same effect that gets a mill align tell-tale to settle up spinning true.

OK - I admit what goes on may be extremely well known, and I still have to catch up some.
 
Last edited:
graham-xrf said:
I am getting some mixed messages here.

1. We seem agreed that a drill press with worn out bearings and run-out is a "bad thing", and that the experience is immediately better if the bearings are replaced.

2. We take it for granted that the rod in the chuck had better have much the same run-out as the chuck itself.

3. We have the contribution from @RJSakowski that the progress of the drill bit in making the hole can allow some variation in where it started from overhead, at the quill and bearings. Moreover, that some amount of play up there can actually give the drill bit a better chance of getting going at the punch mark.

4. We then have, into the mix, good hints from @NortonDommi about keeping the top of the punch in good order, and introduces the concept of having the work underneath free to move a bit, and "allowing the drill to find it's mark". RJ's point of the bow drill by ClickSpring still makes holes where he wants them to happen, despite the hand at the top wandering all over the place, is telling!

Given the drill bit is always a lot longer than it's diameter, so not acting insistently like a rigid milling tool, something else is going on. The small straight edge across the central core of a drill bit helix, at the pointy end, is clearly usually wider than the punch pop mark. It bites somewhere, and does a little squiggle, either by being allowed movement in slop from above, or by the workpiece under sliding about a bit, or by the drill bit bending a little, or some of all three. Where a center point starter is used first may be a separate situation, somewhat more forced, but still may allow the first two movements.

Given that we all aim the bit at the pop-mark as best we can, and it kind of "pulls itself into order", and except for the state at @Nogoingback 's machine , the holes usually end up close enough to accurate (do they?), we don't much care to wring out the real mechanism of what goes on in the first instants of cutting. There may even be some of the same effect that gets a mill align tell-tale to settle up spinning true.

OK - I admit what goes on may be extremely well known, and I still have to catch up some.
Click to expand...
If the work is clamped down when the drill "pulls itself into order", the drill will either flex and start to drill at a tangent or it will start a new center off the punch mark. Most likely a combination oif both. This can be seen on a lathe when the drill is not correctlt aligned with the spindle axi. The drill mat find the center but as the drilling proceeds, the hole deviates further and further from the spindle axis. Allowing the work to float permits the work to align with the DP spindle axis. In drilling, I will sometimes start the drill and back off on the pressure to permit the work to move slightly. If I need to clamp the work, I do so after I have started the hole.

I mostly use the drill press for punching holes where a .02" location error is inconsequential. I use the mill/drill or the CNC mill when I want better accuracy. If I want a truly accurate hole, I will punch a pilot hole and finish with an end mill or an end mill using a pocketing routine on the CNC or a boring bar.
 
Nogoingback said:
If there's play between the quill and the housing, there isn't much to be done about it I think. I've adjusted the setscrew on the side
to minimize play, but that's about all there is. That pretty much leaves bearings as my only option if I understand correctly.

Yes, I center punch before drilling, and am very careful to line up the drill with the center punch "divot".
Click to expand...

If you need an accurate hole location, you do need to spot drill first. A center punch is not needed for a spot drill. A spot drills point is ground to be side cutting so that even if something tries to deflect it at first contact it will quickly cut its way back to center. A center drill or most other drills can not do this.
By center punching you are eyeballing a location, then when you try to align the part to the drill you are eyeballing it again. This doubles the inaccuracy of your eyeballing, whatever that may be. I know people that can eyeball within a couple thou and others that cant hold 1/16. Even on a milling machine or a lathe where everything is held solid and accurate, for accurate holes you still need to spot drill before drilling. A center drill is not the same as a spot drill but is still better than than not spot drilling. I like to spot drill to slightly bigger dia than the final hole, this also acts like a slight chamfer on the finished hole and insures that the drill is cutting into a clean surface with nothing left to deflect it.
 
Hi graham-xrf,
graham-xrf said:
from @NortonDommi about keeping the top of the punch in good order
Click to expand...
Allow me to clarify, any punch you use should have the struck face in good order but I was actually referring to the punch mark itself. Next time you centre punch something take a file and draw it over the surface. You will be left with a bright ring around the dimple you have created.
graham-xrf said:
work underneath free to move a bit, and "allowing the drill to find it's mark".
Click to expand...
I 'spot' the face with a centre drill,(or a spot drill if I had one),and drill only a very short depth to provide a guide for the bit to follow. I like to use a fixed stop as it is quick and convenient especially if drilling a series of holes. Depending on the job I will clamp securely when drilling and even make a jig if necessary. Those vice-grip type locking clamps are my favourite thing on a pedestal drill.

Flyinfool said:
A center drill is not the same as a spot drill but is still better than than not spot drilling.
Click to expand...
Absolutely but needs must and some of us po' people have a lot of centre drills and only one spot drill and that lives elsewhere. I must pull finger and correct this.
 
mikey said:
On almost every Taiwan or Chinese drill press I have ever seen, they use cheap shielded bearings. All of them have similar construction. There are two spindle bearings inside the quill and two drive sleeve bearings at the top. These are the only bearings in there and they are simple to change. I did my Sears Taiwanese drill press some time ago. You can see the write up here.
Click to expand...
Nice write-up, Mikey.
 
Nogoingback said:
So, my DP is a Taiwanese Jet (17") that I bought new about 26-27 years ago. I've repowered it, installed a quality chuck,
and done some work to align and balance the pulleys.
I got about .025 of play, so I'm thinking bearings could at least be part of my problem.
They aren't noisy or rough, but they are 25+ years old and probably not the best bearings to start with. Ideas?
Click to expand...

I'm late to this thread, so not sure where you're at with your drill press but .025" of runout is massively way off. An ok cheapie, not the best drill press should measure under .008", You're three times worse than that. A good drill press should be under .004 but more like .002". One of my DP's is a precision drill press that measures .0007" runout and leaps and bounds my all time favorite. Run-out makes a huge difference for several reasons despite what some say that it doesn't really matter. I can't tell how much nicer it is to use a very low runout drill press and your drill bits will thank you.

Back on topic, you have something bad going on somewhere. I think you mentioned you have a dial indicator and ability to measure runout?

You didn't post a pic of your whole drill press except for a close up of the chuck and arbor. I think I'm seeing your new chuck mounted onto a Jt tapered arbor that has a male MT taper on the other end going into a female MT taper (I'm assuming MT3)?

Do me a favor and please dial indicate the two surface areas above your chuck, measure the arbor that the chuck is mounted to (going into) and the spindle that the arbor is going into. Then the outside of the chuck, on the non satin milled angled surface below the knurling. See red arrows in pic where to indicate. Before you go into full blown rebuild and replace bearings, I wouldn't be surprised if one of the tapered mountings could be having an issue, they may not but always good to check and make sure. Your indicator will tell you.

I have rebuilt countless Albrecht's and Jacobs and replaced countless arbors. This area is very key to getting good runout measurements below the chuck jaws.
Any junk or crud caught in one of those tapers can really effect runout. Looking at your pic, your drill press looks quite fresh and low wear compared to most drill presses, even the rubber gasket on the quill still looks minty fresh. I don't see any grime, patina or rust on the column, its been well cared for.

Pick good clean spots on the areas I mentioned and let us know how all three measure up? You might be surprised on what you find. After that, your options and fixes will be more clear.

Here is a pic of you DP with arrows on where to measure.
IMG_1118.jpeg

FYI: Jacobs gets their runout measurement number when measured 1" below the chuck jaws, using a pin gauge at half of max capacity (yours was measured with a 3/8").
 
Thanks for the suggestions silverchips. I think we might be talking about two different problems. If I put an indicator on the chuck, or on a pin in the chuck and rotate it by hand the runout is quite good. But if I apply pressure sideways on the chuck and measure, I get .025. That suggests play in the bearings to me (as well as the quill). Given their age and uncertain origins, this doesn't surprise me.

Yes, the chuck is on a MT3 to JT arbor. The chuck is not new however: I've had it for a number of years and have been
very pleased with it's accuracy. So, my problem isn't runout, it's play in the spindle. Since I can't do anything about play in the quill, the bearings seem the obvious choice.

Thanks for the comments on condition. It's located in a warm, dry basement and I do try to take care of my
machines. No arc of shame so far! :)
 
Last edited:
Sorry, I completely missed that important tidbit and yes, applying side loads and getting bad runout is typically a problem above in the spindle and bearings. I will say crud or a metal chip in the MT3 taper can induce quite a bit of play and looseness when not seated in correctly making it wonky. I would just for giggles pop it out, stick your finger in there to feel around for crud or marring bumps and then use some rubbing alcohol or acetone to clean out the finger oil and re-seat.

You may also want to check the (round) nuts holding the spindle together (If its the type of design I think it is), they can loosen up inducing play.

After that, you're correct the bearings will need replacing which is not a hard thing to do and lets hope its that or more hopefully the above.

Good luck
 
You must log in or register to reply here.
Back
Top