There is room for improvement in my RT setups. I would not mind spending the money on a centering indicator (if it really works), but round column mill-drill limits (correct me if I am wrong) appear to make it pointless. However, I find that by mounting a DTI with the shaft near the end of its dovetail, it works fairly well.
I would have an easier time given a spud of some type, but once it is close, I seem to be able to align it, but there must be a smarter way. I end up zeroing the dial to what I think is the correct position; it helps me note deflections, but it probably causes trouble too. When should I _not_ do that?
In case it helps, for indicating a vise, I find that I do best to always bump during one direction of travel. Taking advice from various sources (here and James Harvey's "Machine Shop Trade Secrets"), I align by feel against the slots, then get the left bolt reasonably tight and the right bolt just slightly past removing slack. Then I bump during translation, but only with the table moving to the left. I do not try to gain ground during the return pass because I seem to do more harm than good.
Is there a similar trick for locating the RT axis, such as making changes only from one direction (left/front)??
I think that from a conceptual standpoint, there isn't much else you can do for RT alignment. Getting the spindle center over the RT center of rotation is the key for sure, and I don't know of any other mystery techniques to do this other than careful use of an indicator in the mill spindle. Your comments bring up the whole question of the sources of uncertainty in any kind of measurement. The point about dialing in the vise jaws, for instance: Check everything and figure out all the sources of error, such as the vise jaw flatness, the travel straightness of the mill (gibs tight?), whether the parts of the vise might be actually moving relative to each other when you give the alignment taps, etc. There is a whole lot of stuff that can move under those conditions. Your specific issue about travel direction difference might be loose gib in X-axis creating different clearance in one direction vs. the other. I don't know a lot about mill-drill construction, though. Maybe they don't even HAVE adjustable gibs. :-) The same comments apply as well to the RT alignment.
My hope is that I can make it a little more systematic. For example, any time I gain by tightening this bolt and loosening that one to bump in the reverse direction is more than lost by messing with the bolts and correcting the mistakes I make :( The return trip is quick, and by feeling the vise vs. slots, my initial alignment is typically quite good, so it goes fairly well, and if not, I know to start over. Since I figured out the "tighten left, just remove slack on the right" trick, I have had very good results indicating vices.
Adjusting the gibs is still a little hit and miss for me. Rather than constantly fiddle with them, I take note of whether the machine is "listening to me". If the dials are moving on their own, it is probably time to check resistance. It is difficult to feel in the x direction because of the power feed. When the gibs are set correctly (or so I think), the backlash announces itself with a very subtle sound in addition to the change in feel. I have tried to control to that, but I always seem to overshoot and end up backing off. Suppressing rattles seems to work best???
With the RT, the thing that is moving is the loose screw in my head :) I will give myself some credit for having gotten the hang of handling backlash. I hate to think what this setup would have been like for me a year ago =:0
Only I can say bad things about my mill :) As I have said, most days I find it difficult to justify even thinking about spending money on another mill. I suspect that I will do so after some more mulling about weight and capacity.
You could easily be correct about the gib, but I suspect the real difference is that simplification reduces operator error. In truth, by the time I reverse the tight/loosened bolts, I can be in position for the next one-way pass. With the setup I just broke down for the RT, I got it on the second pass, and that is becoming (almost) common. I can center the RT, but it is not nearly so elegant a procedure :(
What's the possibility of getting a Morse taper shank or whatever fits the center hole of your RT and boring it to some good known diameter, let's say .500 =/-.005. Then, when you want to center the table under the spindle, stick a piece of .500 drill rod in the mill spindle and move dials until the rod slips nicely into the hole. It may not be diesel injector accurate, but do you do diesel injectors?
Not diesel injectors, but this one is going to need to be pretty tight (a medical/scientific instrument package). That said, I fully agree that what you describe is a great starting point.
I'm centering with an edge finder. The RT has an hole in the middle and that's where the edge-finder goes in. With an DRO and the "center between 2 points"-function done on the X-axis and the Y-axis it is quite accurate (few 1/100 mm).
If you need it more accurate then proceed with an indicator. A simple gadget:
TMBR (number unknown) has plans for a nice DIY centricator.
I have gotten quite comfortable with dials and scales. Every so often, I grumbles about the clamp or bolt in my way, but it is usually not a problem to pick up locations. Some of my early lathe work will probably be turning a set of spuds/pins in common sizes to make it more obvious in some situations (why not?). I think I'd rather spend the price of a DRO on basic moving equipment for the iron I keep buying. Think storm damage, with the mill in harm's way and/or blocking repairs =:0 A well-placed $500 for a crane or palette jack could end up saving me thousands of dollars and LOTS of hassle.
Nice, but it probably should have a screw for fine adjustment?? when I realized the zero set clone was $70 instead of $150 (that includes an indicator), I bought one. It's ok, but has a few rough spots. Some of that might be fixed by a trip to Lowe's for new set screws, and maybe a shim or two, but it also (IMHO) has some design flaws.
Sad to say I had not seen those books. Are they any good? No offense to author, I just want to know whether I would be hunting it down solely for the plans, or it's something I should already have read.
Even better, in my opinion, is to turn the end of that Morse spud to, say, .500" and put a piece of the same size in the mill spindle. Both with flat ends. Bring the two pieces together, and you can easily feel with your fingers whether they are aligned. Probably to within a couple of thousandths or less. Clamp the table and lock the spindle and you can see if things moved any.
Oh, and Bill - it's pallet jack, not palette jack.
=========== I have found all three of Lautard's books to be good reading, with the plans a bonus. click on
formatting link
formatting link
also see Lindsays books for a number of older machining books [reprints] with plans.
formatting link
Only danger is spending all your "tool" money on books..... :-)
Unka' George [George McDuffee] ============ Merchants have no country. The mere spot they stand on does not constitute so strong an attachment as that from which they draw their gains.
Thomas Jefferson (1743-1826), U.S. president. Letter, 17 March 1814.
To answer your specific question see pages 100-101 of John Edwards "Turret Mill Operation." This section also covers alignment of a RT when used in the horizontal position.
Book is highly recommended as there are many other turret mill hints, tips, and dodges included.
for publisher click on
formatting link
see
formatting link
Edwards has a companion book for lathes which is also very good.
formatting link
A useful dodge on round column mills is to attach a laser pointer to the head before you more it, so that the spot is projected on a far wall. A magnetic base is good for this if you have a flat spot on the head. Before moving the head up or down turn the laser on and put a piece of tape on the wall for a target. Move the head as required and "bump" the head back into alignment as shown by the laser spot on the [edge of the] tape. Because of the long distance the beam travels you will have very close alignment, even if the tape/column are not exactly vertical and/or aligned.
for some examples to get you started click on
formatting link
for the cheap-screw special (my kinda' deal)
formatting link
I can see a magnet, a laser and some hot melt glue, epoxy, or rtv being adequate.
Unka' George [George McDuffee] ============ Merchants have no country. The mere spot they stand on does not constitute so strong an attachment as that from which they draw their gains.
Thomas Jefferson (1743-1826), U.S. president. Letter, 17 March 1814.
It is indeed a great book, but AFAICT, it does not answer my question. I am asking for more detailed advice on *how* to "move the x- and y-axis until a zero reading is obtained." The part about backlash is well-taken (and I'm somewhat surprised to say becoming second nature to me). In particular, I am interested in any opinions you and others have on when to re-zero the indicator (which I find helpful when looking at it upside-down and backwards), and when _not_ to mess with it. It is probably as simple as always doing it with or against backlash, but I do not see it yet. I am almost certain I am wasting steps.
formatting link
I had not seen that one - thanks!
I have gotten reasonably adept at seeing trouble coming, and an ER collet chuck helps a great deal. I have also been known to swap parallels in a pinch. On a vise setup, a couple of edge finds and I'm back in business. However, I have not found a good solution for using a centering indicator, but there is relative room for a DTI with or without a zero set holder - at least so far. The first time I ruin an RT setup, that will be a different story though.
One man's wasted time is another man's doublechecking....
What you may want to try is inserting a "good" #2MT [most of the smaller RTs now sold in the US have a #2MT -- YMMV] and indicating off the external/end. Because the table will [or should be] locked when using the RT, x/y back lash should have no effect if you are using a test indicator. I would not try to rezero the indicator, rather just keep it at the same reading, i.e. no needle movement as you rotate the spindle. Indicate the x [side to side] first and note the needle position. Then rotate the spindle so the indicator is to the front, and adjust the y until the reading is the same. You may have to go through this cycle several times as the x readings will change as you run the y in and out. On the last adjustment cycle the reading should remain constant, and then check with the indicator to the back, possibly with a mirror that the needle is not moving or is moving within the accuracy you desire [or will accept].
formatting link
possibility is a high quality center that you use just for set-up
formatting link
imports click on
formatting link
found the 2394 to be ground as close as the Jacobs, but your part may be different.
The problem with a co-ax style indicator is these are loooooong and you may not be able to get one into position, although using a #2MT/3/8 collet will gain you some space over a drill chuck or ER adapter. for a clone click on
formatting link
?PARTPG=INSRAR2&PMAKA=607-4907&PMPXNO=952232
formatting link
the real thing click on
formatting link
Unka' George [George McDuffee] ============ Merchants have no country. The mere spot they stand on does not constitute so strong an attachment as that from which they draw their gains.
Thomas Jefferson (1743-1826), U.S. president. Letter, 17 March 1814.
I do not have a 3MT shank yet, but clearly need to get one.
I must respectfully disagree. I need to move the mill table to get away from the axis, and I need to understand the backlash in order to make use of the dials. Please set me straight if I'm spouting nonsense.
Ok. Just survived pretty much that process.
formatting link
I found the 2394 to be ground as close as the Jacobs, but your
Thanks.
I ran the numbers and thought better of it. The only prayer of making it work would be to get the feeler to contact near the bottom of the taper. That's if it would clear the other side, and probably using an R8 collet as you suggest.
I spent some time looking for a better way, and here is what I am starting to think works. Please add/disagree/correct freely:
(1) Use a taper, spud, anything in the spindle to get good initial alignment. My slitting saw arbor turns out to be ok for this, but I plan to do better for future setups.
(2) Move DTI "forward on its dovetail" or use a zero-set/clone. At least with a tapered hole, one can simply lower the spindle to control offset. Some fiddling with the feeler is necessary, but the taper helps, and the initial alignment takes much of the guesswork out of it.
(3) Accounting for backlash in the table screws, compare compass point readings and move to correct alternating x-y axis until a constant reading is obtained. I still find setting zero to be helpful, but admit that it causes some problems too, as the indicator is not very rigid.
No nonsense, just the use of the dials. I suggest moving the table bases on the movement of the test indicator movement and not the table dials for exactly this reason. When you do this any backlash affect is eliminated.
Of course this refers only to centering the table, and if you are offsetting the table after centering, for example to mill a face cam, or drill holes based on radius and angular specification, then you will need to [accurately] measure the table movement.
On many mills it is possible to adjust the nuts on the adjusting screws, but the problem is that the adjusting screws generally have considerable wear in the middle and not much on the ends, so when the nuts are adjusted to fit the middle, the nuts are too tight at the ends, and when right at the ends, too tight in the middle.
There was a thread on lathe cross slides and most of the comments/suggestions there will apply to your mill. This tends to be a major operation involving either replacement of the table adjusting screws, or turning the adjusting screws down to a consistent [but not stock] thread pitch diameter and replacing/fabricating/modifying the adjusting nuts.
A major problem is the use of ACME [price an one inch acme tap] or nonstandard thread forms or specifications [e.g. left hand.]
There are some ways around this. You can "slick up" the screws to a consistent pitch diameter and finish, and them mold a liner in the existing nuts to the screws. The old-world-craftsmanship way is to use babbitt, and the new get-it-done way is to use a teflon/epoxy compound such as moglice.
for information on babbitt see
formatting link
and for moglice see
formatting link
The cheap-screw way [my preferred process] is to mount long stroke drop indicators on the x/y/z axis as required and use these to measure table/head as required rather than the dials. You can easily fabricate bolt on holders for the indicators, but my experience is that either a magnetic back or a magnetic holder such as a "mightymag" works just as well and allows you to more easily reposition the indicator or use on another machine, such as your lathe, as a tail stock depth indicator, cross slide indicator or slide stop.
Most of the mill supplies will have suitable indicators and backs some sources
formatting link
?ProductID=401-0005-5
formatting link
type of holder will depend on your mill/lathe configuration and your personal preference.
Note if you get the screw on magnetic back with a plastic rear cover for the indicator that these are prone to breakage if you let these "snap" onto the machine.
Note that you can make up indicator extensions from 3/16 drill rod [or aluminum] and use long 4X48 set screws loctited in holes in the ends of the rods with 20-30% depth of thread to avoid dealing with ultra small [for me at least] threads.
If it would be helpful, I can suggest a cheap-screw sensitive drilling/tapping set-up for not too much money.
Good luck and let the group know how you make out.
Unka' George [George McDuffee] ============ Merchants have no country. The mere spot they stand on does not constitute so strong an attachment as that from which they draw their gains.
Thomas Jefferson (1743-1826), U.S. president. Letter, 17 March 1814.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.