Sorry for the metal content, but there might be someone here who knows the answer!
I would like to find the center of a 6" metal disk (center the mill spindle precisely over it).
Here is my plan...
Use an edge finder and find 3 points at roughly 45 degrees to the center by moving only the X axis, then only the Y axis (forming a right triangle) with the hypoteneuse over the disk. Go halfway on both X and Y between those pairs of X and Y numbers.
Does this sound like a reasonable plan?
Yes, but...
Why not start by eyeballing one axis -- say the Y axis. Then use the edge finder to find the X dimension each way, and split that. Then set your X axis to that center point, and do the same with the Y axis. If you really want to be anal, repeat with the X axis, just to assure yourself that any error is lost in the noise.
That way you'll be finding an edge that's more or less perpendicular to the travel of the edge finder, which should lead to more accuracy.
I do this a lot. You're over thinking it.
put the spindle about at the cneter by eye. Zero X axis by going to one edge with edgefinder Find to other ege with edge finder. record reading and move exactly 1/2 way back. Zero Y axis at one edge with edgefinder Find other edge with edgefinder, record reading and move exatly 1/2 way back.
You're at the center of your circle.
Karl
Thanks, Karl and Tim!
Best regards, Spehro Pefhany
Another option might be to pinch a straightedge between a blank in the spindle and the edge of the disk. Adjust until the straight edge is perpendicular to the x-axis. Then move to the center of the disk on the x-axis (allowing for the diameter of the blank). (Similar to the trick of setting a tool bit when turning a round on a lathe.)
Nice trick, thanks!
Best regards, Spehro Pefhany
Clamp the disk to a rotary table. Eyeball the center. Then use a dial indicator as you turn the table to measure the error. Correct the X and Y axis as needed.
Sort of analogous to centering a part in a 4 jaw chuck on a lathe.
I guess if I had a setup that would hold a dial indicator in the spindle and would reach the outer rim, I could just rotate the spindle and adjust X/Y. Might be easier since there is no spindle lock on my X3.
Best regards, Spehro Pefhany
Whoops. I screwed that up. You've got to center the rotary table with the X and Y feeds. Then mount the disc on it and center the disc on the rotary table.
Its accurate, but may not be worth all the trouble.
If he has a DRO. If you have to allow for backlash it gets tricky...
mattathayde had written this in response to
------------------------------------- Spehro Pefhany wrote:
why not use a center square (combination square) to get the center, maybe its not precise enough for you but it should get you what ever the width of your scribe line is
-matt
##-----------------------------------------------## Delivered via
True. I am DRO-less, and compensate for that by either pulling the stage in on the back side as I turn the wheel, or by doing something like measuring the diameter of the thing before hand.
Both are a PITA and prone to error, for both the error can be overcome with careful work.
The simple version of that setup is an L shaped piece of stiff wire. Chuck the short end, cut the long one to just clear the column. A Starrett Last Word will clamp directly on the wire, otherwise make a little block to fit the test indicator's dovetail. I would size the hole in the block for 1/4" rod. While you are set up to cut the dovetail you could make a lathe holder too.
To roughly center something into the range of an indicator I center a
6" ruler on it and move a point in the chuck to the 3" line. Usually this gets it within 0.010".jsw
Quick way to get a center (close) Use a postcard or other rectangular cardstock with crisp edges. Place the card on the round workpiece such that one corner is at an edge. Mark where the two adjacent sides intersect the edge of the disc. Connect those two marks with a line, which intersects the center Move the card approx 90 degrees and repeat for an intersecting line. The intersection is the center.
My goodness! What's wrong with using a machinist center square. This thing should be in everyone's tool box or chuck it up in a 4 jaw, eliminate runout and mark with your tailstock. I can't believe you are asking the question. Steve
You are a embedded control guy. How darn retro do you want to be? Break open that wallet in a dim light so the Benjamins don't get blinded and get yourself a DRO.
If it was a lathe, that would be different for some strange reason IMHO.
Wes
-- "Additionally as a security officer, I carry a gun to protect government officials but my life isn't worth protecting at home in their eyes." Dick Anthony Heller
Take a trip to
OSBORNE.ZIP
That plus the milltable and an edge finder will do the job.
HTH rgentry at oz dot net
reak open that wallet
I don't have one on my mill at home, and don't really miss it.
At work making new parts to drawings the DRO is an enormous help. At home I more likely fit a repair part to an older one by cut and try, and it's easier to incrementally determine how much more to remove. The resettable dials work fine for this.
jsw
should be in everyone's tool box or chuck it up in a 4
asking the question.
Junior was so impressed with the centre head the the next time I was in HF (Port Huron, Mi) I invested $6.99 in their three head square, took the protractor head for me and gave him the rest! Gerry :-)} London, Canada
Spehro,
Probes are kinda crappy for finding circles (CMMs aren't even good with curved surfaces). Not sure about your tolerances, but indicating the disk would likely be the best (indicator mounted to the spindle).
Why?
-Well, the edge finder prob may roll oddly along the curved edge of the disk as you likely won't be on center the first try. I mean, you have to have Y centered to make an accurate X measurement, and the reverse. When you start, you'll have neither. If I had to do this, I'd center one axis, center the other, then center the 1st again. You will likely end up with a slightly different result.
-The disk may not be round. Depending on how it was made (turned, punched, laser cut, plasma cut, waterjet, cylindrical ground, jig ground, milled, etc, etc) your edge finder prob won't tell you anything about its circularity and may touch a high or low point. An indicator allows you to average highs and lows and find out the location with the lowest deviation.
Just my $0.02 worth.
Regards,
Robin
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.