I need to cut an external thread (I'm using G76) but the thread has to
start in the same place for every part. The thread start is in
relation to a number of drilled features on the periphery and on the
front face of the part. This fixed relationship is a requirement for
the part's functionality.
Anyone know how to do it?
Lathe has a C axis and live tools.
Thanks for any and all suggestions!
On Fri, 24 Jul 2009 18:29:16 -0700 (PDT), "Robin S."
What kind of holding fixture are you using? Does this index the
part in the chuck to the same radial position every time? If
not, can you add a pin or something to do so?
Unka' George [George McDuffee]
He that will not apply new remedies,
must expect new evils:
for Time is the greatest innovator: and
if Time, of course, alter things to the worse,
and wisdom and counsel shall not alter them to the better,
what shall be the end?
Francis Bacon (1561-1626), English philosopher, essayist, statesman.
Essays, "Of Innovations" (1597-1625).
On Jul 24, 11:06 pm, F. George McDuffee <gmcduf...@mcduffee-
associates.us> wrote:> -------
The part is run out of bar stock and is complete after cut off. I
guess I should have mentioned that.
Does the machine naturally always start the thread cycle from the same
angle, given the same program? I didn't want to assume that the
machine would exhibit this type of behavour, but I suppose it makes
sense to some degree.
On Fri, 24 Jul 2009 18:29:16 -0700 (PDT), "Robin S."
Is the thread and all of these related drilled features being machined
on the lathe in one chucking?
What is the tolerance and how are you inspecting it?
================Assuming all features are being machined in one chucking and not
knowing your machine and controller I suggest;
1) Thread milling is the easiest and most precise method to control a
thread start, IMO.
2) (G76/G92 thread cycle) If thread start position is not too tight
run a sacrificial setup part and adjust thread tool start position Z
3) (G76/G92) If start position is not too tight you might run a
sacrificial setup part and rotate the positions for the drilled
If the related features are machined in a different machine or a
different setup then you need a fixture/Stop/Pin to orientate your
On Jul 24, 11:19 pm, firstname.lastname@example.org wrote:
Yes. The part starts as bar stock and is cut off as a complete part in
I should have mentioned that. Approx. +/-5º (pretty loose, given the
process). Inspection happens during assembly. Misalignment is obvious
during assembly as some holes on a mating part have to (roughly)
That's not a bad suggestion at all actually, and it hadn't occured to
me at all. The thread is 1.0mm pitch and only about four or five
threads long (about 1.3" diameter - I know inch AND metric thread
dimensions!). Maybe faster to threadmill than to take multiple passes
with a single point tool.
I do have some clearance to change the Z position of the start. Would
a Fanuc 18T always start the thread in the same angular location given
the same program? I wasn't sure if the control did this, or just
started the first feed "whenever".
I'd likely just change the C offset using G50 or something as I don't
want to reprogram all the drilled angles. This again begs the question
as to whether the machine will always start the thread at the same
Thanks for your help Tom. Great suggestions.
On Sat, 25 Jul 2009 05:29:36 -0700 (PDT), "Robin S."
Sounds like your solution as long as it's within your work envelope.
All I can say is it has worked for me where thread milling was not an
available option. If it is the same program, same machine it should
repeat. If it is the same program, different machine then Z start may
need adjustment on setup part but all subsequent parts should repeat.
I suggest you run a couple test parts on scrap or inexpensive material
to verify it works on your machine.
You can thread a part using G76 then run a deburring routine to break
the thread burrs using another tool and return with same threading
tool using G92 as a single spring pass to make sure you didn't push
the burr back into the thread. As long as the Z start is exactly the
same in both threading cycles the tool will pick up the same lead so
it is not random ("whenever").
It should but verify by testing it out on scrap or inexpensive
Too bad it's not a 31i control. With it you can specify an angular
relation to the start of the thread.
But you can hold the angular relationship in any case. The zero pulse
marker for threading is the same zero for the C-axis plus or minus any
grid shift value set in a parameter.
The easiest way to do this is to adjust you Z-axis start point for the
thread. You didn't say what the thread lead is, but lets take a 16 pitch
thread as an example. 1/16=.0625 So every 0.0625 movement in Z60
degrees rotation on the spindle. So every 0.00017" is equal to one
degree on the spindle.
The easiest way to adjust your start point in Z without editing the
program is to use a macro variable. Just add it to the Z-value on your
Or something like that.
As far as getting the first part to come out right, that would be
difficult using this method. If possible cut a test part, measure and
adjust the variable.
In a situation where you have an expensive part that needs to be right
the first time, I would cut the thread first, then engage the C-axis and
shift it using a coordinate setting command. If possible have a screw on
gage made up beforehand that can be indicated after the thread is cut.
Indicate it to zero then set the C-axis coordinate to zero or the
appropriate dimension to be in relation to your milling.
You could also thread mill if you have a live holder available and the
part lends itself to being thread milled.
Ran some parts today. Seems that changing the Z start position is the
most straight forward way to adjust where the thread starts. (Z linear
offset = pitch(angle/360)) The thread does indeed always start at the
same angle, given the same program.
Thanks very much for the suggestions guys. Worked out very well.
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.