Should spindle turn while M19?

[chipwitch1]

I'm having a couple problems with my VMC15RT involving the spindle lock. First of all, strange as this may sound, I'm turning small parts in the spindle like a cnc lathe with single point tools mounted to the table. I know it requires a shift in thinking, but if you can get post the concept, the principles are exactly the same. I simply rewrote my post processor.

The first problem I'm having is with tapping. Since this is a single point tool, I'm having to "re-thread" the same thread as I'm only taking about 5 thou per pass. One day, my program was working perfectly... on multiple parts. The next day, same program, same tools, same set up and the threads aren't tracking identical with each pass. In other words, the threads are no good. I'm using G84.2 rigid tapping for the procedure.

The second problem I'm having is likely connected to the first. I was replacing a tool in an ER32 collet chuck while it was still in the spindle. In manual mode, I set M19. The spindle oriented and locked. I used the collet wrench to remove the old tool. No problem. When I retightened the collet nut on the new tool, the spindle slipped! I assume the belt is what slipped? How much torque should the spindle take before M19 slips? Is there any possibility of damage? Will this affect how the spindle orients or does something need to be re-homed? I'm guessing the belt is loose and that may be why the threads aren't coming out very nice.

A likely unrelated problem, number three, is when I first booted up my machine today, and ran my program, the machine kept "waiting" between tool changes (M1's are coded into the program there). But, the option stop was off. I checked the "options table" (typing MU during program execution). It too indicated the option stop was "off." Subsequent runs of the same program skipped the M1's as intended. Strange?

Thanks

[Neal]

Doing a lathe operation is not unheard off. Although it is not a straight forward thing to do, with a little thought it is not that difficult to do.
G84.2 is a calibration cycle and NOT a rigid tap cycle. Unless your software is over 12 years old DO NOT USE G84.2! The rigid tapping code is G84.1 and the cycle includes the calibration cycle.
As far as the M19 locking the spindle issue the spindle arm has a roller on it that drops into a dent on a cam ring at the top of the spindle. It is NOT a solid locking device. Given enough rotational force the spindle can pop the roller out of the dent.
Tightening the tool in the spindle is sometime necessary for tool orientation purposes but only tighten the tool enough to hold it in place. Always finish tightening the tool on the bench.
BTW, using a collet for rigid tapping in any tap larger that 3/16 dia is asking for trouble. There can be enough rotation resistance to slip the tool in the holder. Now you have lost the thread lead orientation. Do yourself a favor and invest in the proper rigid tapping holder!

Neal

[chipwitch1]

Thanks for the info, Neal. Any idea what may be causing the bad re-threading?

As for the 84.2. Yes, my machine is a '95. The software is old. I'm only using the 84.2 for its calibration purposes. Subsequent code blocks do the actual threading. As I said, the threading was working fine one day, bad the next with no changes to anything.

I'll probably be taking the cover off the spindle head this morning to take a looksee at the belt. I just got the machine less than a year ago. No idea if or when the belt may have been changed.

Thanks.

[edit] I've taken the cover off the head. The belt looks good and is tight. It's the poly chain type. I can't imagine it slipped. Single point threading in steel at 5 thou per pass isn't going to produce enough friction for that. I don't have a test bar to check calibration, but am I wrong in thinking that wouldn't cause the problem I'm having? I mean, I can see the thread not being accurate, but what might cause re-threading passes to not track the same as previous passes? In other words, the threads aren't uniform.

[DareBee]

You did not clarify - you ARE using G84.1 for the threading?

[Neal]

CHIP--
If the Z axis amp drifts it signal settings it can affect the tracking ability. Also excess backlash or other mechanical changes in the system will affect the tracking results.
FYI-- you can buy the thread testing bar. They are not very expensive.

Neal

[chipwitch1]

Thanks Neal... I'll see if I can get one ordered. Interesting discovery. I tried running the program again today after a cold start. Worked good the first time, then got worse with subsequent parts to the point they're unusable.

@DareBee... no, I'm not using G84.1.... I'm using G84.2

[Neal]

Chip--
The fact that you are using the G84.2 tell me that you are NOT, repeat NOT, using rigid tapping. You are only synchronizing the spindle hall effects switch and the spindle motor encoder. It is only preparing the control for the G84.1 rigid tapping call. It is gaining you NOTHING in the manner that you are using it.
Unless you use the G84.1 rigid tapping is NOT initiated. Using the G84.2 on the tapping line instead of the G84.1 causes the control to ignore the .2 and defaults to the G84 compression/tension tapping cycle. This fact has zero to do with whether the machine is brand new or the original 1983 vintage machine softwares.
If this is confusing you, call me if you still have my cell number and we can get it squared around. :=))

Neal

[chipwitch1]

Originally Posted by Neal Chip-- The fact that you are using the G84.2 tell me that you are NOT, repeat NOT, using rigid tapping. You are only synchronizing the spindle hall effects switch and the spindle motor encoder. It is only preparing the control for the G84.1 rigid tapping call. It is gaining you NOTHING in the manner that you are using it.

You mean it there's another way to sync the spindle and z-axis? The G84.2 permits me controlling the C-axis (rather than trying to control rpm via the S-word). I know of no other way to do that. I'm not actually using any of the canned rigid tapping cycles, true. But, you're telling me I can't manually block the code to "rigid tap" myself? I seem to be doing it.

Unless you use the G84.1 rigid tapping is NOT initiated. Using the G84.2 on the tapping line instead of the G84.1 causes the control to ignore the .2 and defaults to the G84 compression/tension tapping cycle. This fact has zero to do with whether the machine is brand new or the original 1983 vintage machine softwares.

I'm not actually tapping with the G84.2. The G84.2 sits alone on it's own block. What follows are my hand coded blocks to do what I assume the G84.1 does. If you recall, when we spoke a few months ago, the canned cycles wouldn't allow me the flexibility I needed to single point threads with the part in the spindle. The way I have my program working, I'm able to program thread "pull-off." That's not something you can do with the canned routines that I'm aware of.

If this is confusing you, call me if you still have my cell number and we can get it squared around. :=)) Neal

[Neal]

OK, you are using the coding in a manner that was not originally intended. This is a special application, obviously. Remember when computer code is used in a non-proven manner there is no guarantee that it will ALWAYS function the same each time. There is always a potential for a mis-step.

Neal