a real metalworking question - what is wrong

ok, some setup - I have a mill (Abene), I just bought a chinese coaxial indicator. I got a large Starrett caliper in the mail with some other stuff - it has a threaded post that is part of a larger (about 3/4 inch) protrusion that forms part of the friction hinge. The threaded post was broken off, hence too short. Thinks I, this is an opportunity to learn to use that coax indicator - I'll just mill the rest of the post off, center with the indicator and drill and tap a new hole, insert threaded rod (cut head off screw) and I'm good to to.

Well, I clamp the part to the mill table firmly, mill off the stub, center with the indicator, insert a center drill in a chuck to drill a little starting hole, and as I raise the table to the drill, I notice the hole is at least 1/8 inch away from the center. I repeat this several times, each time centering so there is less than 1 thousandth movement on the coax indicator, and each time the hole is not centered. So, I set it by eye, measure and drill - it's perfect, and all is well

EXCEPT!!! - what the heck went wrong???

well, maybe the table moved - no, I had the gibs locked. maybe the piece moved (no, it didn't)

well, mabe the indicator is bad - or maybe not.....

I tried the indicator in a 3/8 end mill holder (the shaft is 3/8) - noticed about a thousanth clearance (well, maybe 2, I didnt' measure it) and I also tried it in a drill chuck - both ended up with errors - '

maybe the cat 40 taper is bad, or dirty - clean it, it wasn't dirty.

Put 3/8 drill rod into chuck - it sweeps out a cone - waaay to much error - so there is some angle on the chuck with respect to the spindle - that's not good. put 3/8 drill rod into 3/8 end mill holder - about .100 runout at the tip - not good (note - drill rod about 5 inches long, not bent) put 3/8 drill rod into Ericksson collet holder - TIR less than .002 ---- that's more like it Put 1/2 inch drill rod into 1/2 inch end mill holder - no measurable runout - repeat, mesure .001 - good....

so, I've got at least two problems I figure, maybe 3 -

  1. the 3/8 end mill holder may be damaged since it seems to be oversized
  2. the chuck is obviously not running true so I'll need to separate it from the NMTB40 to jacobs taper adapter (it's a name brand adapter, though I don't remember the name right now) - the runout is more than the chuck should have - so ????
  3. maybe the indicator was defective - how to test?

And, once all this is solved, how can I hold the indicator so I am sure it's doing it's job and not finding me a hole that is offset from the center?

ok, no politics, this is real on topic stuff

Reply to
William Noble
Loading thread data ...

A common problem in the mini mills & mill-drills is that the spindle may not travel parallel to the vertical travel of the mill head or table. First square the spindle to the table - there is usually a provison of squaring about one axis but not the other, while on a BP, you can easily square to both axes by sweeping an indicator across the table. Now it gets tricky: You set up a square (or a 1-2-3 block) on the table and indicate on it as you raise the spindle or lower the table - test in 2 directions, along and across the table long axis - you may find that the spindle travel isn't parallel to the vertical axis travel of the table or mill head. If this is the case, you can indicate the center of a hole, then after vertical travel of the mill head or table, you are no longer centered. How you fix this dependss on the mill design.... Joel in Florida

Reply to
joelblatt

Head's not square to the table, and the two tools are of different lengths. Maybe.

Beat up toolholders, another to watch for. A shitbeat, used, abused, US made toolholder, is more suspect than a new cheapo import, but all deserve a look, with an indicator and some known straight pins. Collets, too!

Cheers Trevor Jones

Reply to
Trevor Jones

What Trevor said. And by Chinese coaxial indicator do you mean a clone of an American made Blake Co-axial indicator? If that is what you have, it seems that it would always indicate the amount of runout present in any specific setup. From what you found, I opine that you have runout with some setups and zero runout with others. Meaning you have some defective combinations of spindle, holder, work and some good combinations of spindle, holder, work. I think the Blake clone will always show non-concentricity from whatever the source may be. If your test with known good components shows different concentricity at different table-to-spindle locations, it is telling you there is an out-of-tram condition.

I have a chink Blake clone. With it revolving slowly in the spindle and "feeling" the inside of a true circular hole, it's indicating pointer will never become totally quiescent, when "zeroed" via table movement. Although the pointer is oscillating over a very small excursion on the dial, say

1/32", I have concluded this is due to inherent runout in the setup.

Bob Swinney

Head's not square to the table, and the two tools are of different lengths. Maybe.

Beat up toolholders, another to watch for. A shitbeat, used, abused, US made toolholder, is more suspect than a new cheapo import, but all deserve a look, with an indicator and some known straight pins. Collets, too!

Cheers Trevor Jones

Reply to
Robert Swinney

A centering indicator will work best in a collet. An end mill holder may force it off-center a bit. An Abene has a pretty fancy articulated head, you need to tram it to make sure it is perfectly normal to the X-Y plane of table motion. I have a CNC routine I use, but it can be done on a manual machine. Probably the easiest way is to take a small end mill (3/8 or 1/2" max) and plane off the top of a piece of scrap aluminum. Don't use a fly cutter, you want it to make a sawtooth if the head is out of tram. Now, put a dial test indicator in the spindle and with the spindle centered over the scrap, sweep around in a circle, and see what is high and low. Retram the head to get as parallel in both front-back and left-right as possible, then repeat. This time there should be very little sawtooth in the surface, but recheck with the indicator. You want to get down into the tenths of a thousandth of an inch over the largest scrap you have, I use a 7" block.

(The other method is to level a block in the vise by moving the table and reading the 4 corners with an indicator, then sweep the indicator in a circle and read and tram as above.)

The idea here is that if the head is not in very good tram, then any length change in the spindle puts the tool off-center from where you set up. Only if the tip of the drill bit was at the same length as the tip of the coax indicator will you be spot-on. Also, if the head is out of tram and you drill by raising the knee, you will bend the drill and can't get a straight hole.

Jon

Reply to
Jon Elson

great suggestion but on my mill (an Abene), there is no vertical travel to the spindle - though I can rotate it, so it is possible that the vertical stop isn't exactly at the vertical position - I guess that could do it, I'll try it.

This particular mill (there's a photo on my web page, wbnoble.com, under tools/hobbies, but it's not very informative) is much larger than the mini mills/drills - 6 hp spindle motor, 2 hp traverse motor -

I'll try the test you suggest tomorrow or Saturday

Reply to
William Noble

Jon - this makes sense - I keep forgetting the potential for the head to not be vertical - but your statement to get it parallel in left right and front back directions has me puzzled. As you know, on this mill, the head can rotate from vertical to horizontal by simply loosening a couple of lock bolts - but there is no adjustment for any other alignment, the only adjustment is the position of the stop.

So, were you suggesting two adjustments? or were you suggesting two measurments, one left/right (which I don't see how I can adjust anyway) and one front/back (which should reveal any out of tram condition of the head)???

thanks. Slowly getting to the bottom of this puzzle.

more followup - removed drill chuck from 40 taper to 33 JT adapter, measured

1/2 thousandth (or a little less) using a last word indicator, but I didn't see if the high spot at each vertical position along the JT taper was along a vertical line - I'll check that tomorrow. The holder is a new Kennametal holder, so my inclination is that it's probably OK, but it never hurts to verify.

so now my list of problems seems to have at least three things on it:

drill chuck/taper holder problem (research on that in process)

3/8 end mill holder problem possible out of tram condition on head - to be checked "soon"

Reply to
William Noble

While being off center isn't the best scenario, it will have little effect on the outcome (assuming it's not a great deal off). Remember, regardless of how far off center it may be held, when the center of rotation (the spindle) is concentric with the hole, the indicator will see the hole equally, regardless of position, be it on center, or not. What ever error is generated will be generated equally.

Harold

Reply to
Harold and Susan Vordos

Make things real easy on yourself. Place a rod, bent @ 90 degrees, in the spindle, be it in a collet, end mill holder or drill chuck. Makes no difference. The rod should accommodate a DTI of sorts. Mount it such that when you turn the spindle (which should be out of gear, so it turns easily), that the pointer of the indicator trails off the slots, so it won't hit them from the side of the contact. Set the indicator so it is just barely in contact with the table (a thou or two) and sweep an arc that is only slightly smaller in diameter than the table is deep, front to back. That will give you the best level of precision possible. When you sweep the table with the indicator, you should find it equal in all positions, NSE&W. If it can be adjusted and is out, do so, then reset the permanent stop, assuming it's out in that direction. It's always a good idea to sweep the head once it's been moved, even when you have a stop.

Be aware, if your gib on the knee is worn, or very loose, the entire knee assembly may be tilting away from the mill, or even sagging side to side as the load move across the saddle. Adjust the gib so it's snug before you do the head adjustment. That will insure that you don't set the head to a sagging knee, which will be a repeat of the same problem you find now.

How about a report on what you find? Could be your machine needs some attention beyond adjusting.

Harold

Reply to
Harold and Susan Vordos

According to William Noble :

Is this a Bridgeport, or some other brand?

I know that there are two separate adjustments on *some* Bridgeports, while others (some of the Series I machines and probably of the Series II as well) have a rigid head.

Those that do have the adjustments have the loosening of the two nuts which then allow the head to be rotated clockwise/counterclockwise as viewed from the front. Note that some (such as for sure the Series I which I have) include a hex projecting to the right and down which turns a worm gear to rotate the head. When nearly vertical, there is enough slop to make it difficult to get the adjustments correct. On these heads (and perhaps on the ones without the worm gear as well) there are pins to keep it from rotating too far to the left or right, and these have to be pulled to allow inverting the head for moving.

But aside from this, there is a knuckle on the end of the ram (I'm presuming a machine new enough to have a dovetailed ram instead of a round ram, and I'm not sure what is present with the round ram.) Anyway -- this has three bolts which can be slacked which allows the head to "nod" so it points more towards the column, or away from it. This is where you adjust the other tilt.

But some Series I machines lack both of these adjustments, and simply have the head bolted onto the end of the ram, set *once* to be right, and then drilled and reamed for taper pins to allow it to be replaced in that specific position if it ever has to be removed for whatever work.

Harold has already given you a description of how to make a tool for checking how far out it is in each direction.

Good Luck, DoN.

Reply to
DoN. Nichols

top posted --

Don - this is an Abene VHF-3 mill (made in 1970) , info on it is here if you are curious

formatting link
- mine does not have the quill downfeed lever, so any vertical motion is via the table - it is very different from a bridgeport.

so, today I spent some time and set the head properly vertical - it was off by a couple of degrees - I've set it pretty close - if I sweep out a 4 inch circle, there is two thousandths total difference between the back and front - to do this, I tilted the head, removed the stop, set the head vertical (and dialed it in), moved the stop back to the head, lossened and rotated the head so I could tightendown the stop (can't reach the bolt with head in vertical position), and then rotated head back to stop - if I want it better than the .002, I can disengage the stop and dial it in more carefully.

so, now I have the mill in alignment, and I've tested that the Co-Ax indicator is not defective - that leaves me with two problems:

  1. my 3/8 end mill holder needs to be replaced - the entry into the holder is oversized by about a thousandth, though it does neck down to close to proper tolerance about 1/2 inch into the holder. the only solution I see is to replace it, which I will do in due time

  1. the drill chuck wobble problem is under investigation - the chuck is mounted on a Kennametal arbor (40 taper to JT33 taper) - I measured the runout of the JT taper - top, and bottom - I measured .001 at each place, but the high spot was at opposite sites of the taper - this could throw the chuck off a little - but I'm more suspicious of the chuck body itself - this remains a homework assignment.

Meanwhile, the next issue with this mill is tightening of the traverse drive belt - I'll start a new thread on that subject since it is probably of more general interest.....

thanks to all

Reply to
William Noble

formatting link
"[L]iberals are afraid to state what they truly believe in, for to do so would result in even less votes than they currently receive. Their methodology is to lie about their real agenda in the hopes of regaining power, at which point they will do whatever they damn well please. The problem is they have concealed and obfuscated for so long that, as a group, they themselves are no longer sure of their goals. They are a collection of wild-eyed splinter groups, all holding a grab-bag of dreams and wishes. Some want a Socialist, secular-humanist state, others the repeal of the Second Amendment. Some want same sex/different species marriage, others want voting rights for trees, fish, coal and bugs. Some want cradle to grave care and complete subservience to the government nanny state, others want a culture that walks in lockstep and speaks only with intonations of political correctness. I view the American liberals in much the same way I view the competing factions of Islamic fundamentalists. The latter hate each other to the core, and only join forces to attack the US or Israel. The former hate themselves to the core, and only join forces to attack George Bush and conservatives." --Ron Marr

Reply to
Gunner Asch

According to William Noble :

Agreed -- very different. They can switch between being a vertical and a horizontal mill -- and do a proper job of each. While I've never seen one in person, a distant friend has one, and I've admired it in photos.

O.K. And this machine has *only* the "nodding" adjustment, not from side to side as far as I know.

Good!

Agreed -- that was probably abused at sometime in the past.

While the body *may* also have problems, 0.001" opposed from top to bottom would make a significant runout at the end of the chuck. Remember that those two points are not that far apart, and by the time you reach the drill's flutes you are up to significant runout. I suspect that that arbor has been crashed sometime -- with that, or another chuck in place, so I think that it should be replaced, too.

I actually use 40 taper tooling too -- in my Nichols horizontal mill -- with a vertical adaptor as part of the set. Also, *my* Series-1 Bridgeport (A CNC -- BOSS-3 one) uses 30-taper tooling in a quick-change spindle. That tooling also fits the Nichols with a 30-taper to 40-taper adaptor.

O.K.

Good Luck, DoN.

Reply to
DoN. Nichols

continuing follow up on this drill chuck friend with lots of tooling and measuring stuff - took the chuck - mounted it on an accurate arbor with no measurable runout, in his lathe. Measured less than .001 runout at the chuck body, but with a 1/2 inch drill blank in the chuck, measured .020 runout (repeatable, no matter which hole used for tightening) at the far end of the rod - so for sure the chuck is a problem - The arbor is Kennametal, and looked new when I got it (cheaply via ebay), and I haven't crashed it. so, will try a better chuck and report results. I'm learning, I guess....

snip ---------------- ">

snip ------------------

Reply to
William Noble

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.