Stuck getting a G83 to work.

[knobhead]

I'm trying to get a parametric program running on our Mazak Integrex 200Y, and I'm struggling to get a G83 cycle to run on it.

Here's the code I'm using:

T0600.0
M169
M212 (LOCK C AXIS)
M13 (SWITCH TO MILLING MODE)
G00 X0.0 Z5.0 (START POSITION)
G83X0.H0.Z[#7+5]R5.Q1400.P0F150L1M210
G00 X0.0 Z5.0 (BACK TO SAFE POSITION)
G80 (CANCEL HOLE CYCLE - RESETS MODAL VALUES)


It's a mill-turn machine, so the 'H' in the G83 controls the C axis. The #7 value is -38.42, and the M210 locks and unlocks the C axis at the beginning and end of the cycle. (The M212 in in there as that's what the operators recognise as spindle lock)

What happens is the tool comes in as far as the G00 line, starts rotating, and begins the G83. The position screen comes up with how far the tool has to go, but it just sits there without moving. I'm fairly new to using canned cycles as our post processors are set up to throw everything out 'long hand' so I'm a bit stumped why this doesn't work. Any suggestions?



Excessive accuaracy is a sign of poor breeding. -Socrates.

 

[mrainey]

I know nothing about your control, but tools "just sitting there" often aren't getting a feedrate they can use. Could the F150 need a decimal? Does the control know the tool is turning? Feed per inch versus feed per revolution properly spelled out?



Manufacturing Freeware and Shareware

http://mrainey.freeservers.com

 

[fastasleep]

Another possibility: on the G83 line you are giving the C axis a position (H0) after you locked the C axis (M212). Even if the C axis is already at zero, the controller may still be trying to process the command.

 

[knobhead]

ah, those are both worth trying, thanks loads! I've been thinking it might be something to do with the feed, as when I play the cycle back on the machines own graphics, it works perfectly. The first peck was taking 5mins according to the time display, so I thought aha, I've somehow got a super slow feed in there. I altered it so that the peck now only takes one second according to the graphics, but it still just stalls when i try it in 'real life'.

Right, off for another play...




Excessive accuaracy is a sign of poor breeding. -Socrates.

 

[tstanley]

I agree with the decimal for the feed rate.

Isn't Q the amount you want to cut in with each peck? If so, the value of 1400 doesn't seem to relate to the depth of the hole at -38.42 to me. That might also explain why the feed seemed so slow (1400/150=9.3minutes).

Tom Stanley

 

[knobhead]

Cracked it.

One of those 'doh' moments. I didn't have a spindle speed in there: added M203 S180 at the top, and off it went quite happily.

Surpised it worked on the machined backplotter though, and just hung (with the tool rotating? ) rather than errored out - still, just another Mazak quirk to remember..

Thanks everyone.



Excessive accuaracy is a sign of poor breeding. -Socrates.