Re: [CAD_CAM_EDM_DRO] Which to trust, home-built DROs or hash marks?

----- Original Message -----
From: "Jon Elson" <elson@...>
To: <CAD_CAM_EDM_DRO@yahoogroups.com>
Sent: Tuesday, August 05, 2003 1:43 AM
Subject: Re: [CAD_CAM_EDM_DRO] Which to trust, home-built DROs or hash
marks?
[extra stuff cut]
> OK, your calipers are probably the most accurate, although a 1-2"
micrometer
> would be better. Why does the part measure larger? Because of tool
> deflection!
> The reaction to the cutting forces deflected the tool probably 1 - 2
> thousandths
> of an inch. Did you compare making the same cuts with both conventional
> and climb milling?
No, I'm a machining newbie and didnt think about that.
> How about with removing .010" stock on the finish pass,
> and .001" for the finish?
Nope, newbie thing again
>Have you miked the cutter itself? They are often
> slightly off size, especially if it is a lo-cost import.
I dont have a micrometer, but I bet that's the case
>
> (now the embarrasing part) Did you make sure that both positions were
made
> approaching from the same direction, to eliminate backlash?
Yes, I came in from the right with the X axis, did the first cut, pulled
back to the right, moved 2" in the Y axis, then moved back to the left with
the X axis to make the cut (I didnt care how far in the X direction the cuts
went, as long as they were approximately the same length)
>And, of course,
> you have to lock the axis after making the setting, as the cutting
> forces will
> push the table back.
I normally remember to do that, not sure if I did this time
>Since you have a DRO, now, you can measure the screw
> backlash directly. I'm guessing from the above measurements that you did
> NOT compensate for backlash.
I never reversed direction in the Y axis during this test, I only moved the
table towards me, never towards the column
>You have to make a move of several inches
> and compare the dials to the DRO. If the error appears relatively
constant,
> that most likely is your backlash. if it is cumulative with distance,
that
> is probably a thread accuracy issue.
Over almost the entire travel of the table (moving in one direction, and
after doing a short starting run of 0.100" to insure any backlash was taken
out), I got something like 7" of travel from the dials, while the DROs said
it was 7.007xxx"
>
> >Now the question comes up, what to trust? Should I just accept the 1%
error
> >in the hashmarks on the mill and only use the DROs (the X and Z axis
parts
> >should be arriving on Thursday)? Or should I investigate further to find
out
> >if there's any other error in the system?
> >
> >
> You should always investigate fully until you have a good understanding of
> the errors, and where they come from. A set of gauge blocks is
> invaluable in
> making these measurements. A straight edge clamped to the table, and a
> dial test indicator in the spindle makes a very good accuracy testing rig.
> You put in a gauge block against a stop, and move the table until the dial
> indicator touches the gauge block, and you get some particular reading on
> the indicator. then, zero your screw dial or DRO, and move away. Drop in
> a gauge block, and bring back into contact and the same mark on the
> indicator. If you put in a 1" gauge block, you should see a 1" change in
> the handle dial or the DRO. If not, there is an error to be found. If
you
> use gauge blocks, they are more accurate than anything else you can get
> your hands on, so you know all other errors are in the machine.
>
> Jon
I'll see if I can beg or borrow a set of gauge blocks, my dad knows a shop
manager at Lockheed Martin, so I might be able to borrow some for a day.
John