Rigid Tapping Help

NCPatrol wrote: The hal meter initially shows "true" when the window pops up and changes to "false" when I rotate the spindle by hand. It stays false after that
....
Is it safe to assume that my encoder is shot?.

On the contrary, it shows that everything is working properly.

(You should find that if you set the index-enable to true that it sets to zero again at exactly the same spot in the spindle rotation, and that is the point of it. LinuxCNC motion looks for the index-enable resetting to false to start the spindle-synched move.
)

andypugh wrote:

NCPatrol wrote: The hal meter initially shows "true" when the window pops up and changes to "false" when I rotate the spindle by hand. It stays false after that
....
Is it safe to assume that my encoder is shot?.

On the contrary, it shows that everything is working properly.

(You should find that if you set the index-enable to true that it sets to zero again at exactly the same spot in the spindle rotation, and that is the point of it. LinuxCNC motion looks for the index-enable resetting to false to start the spindle-synched move.
)

So why no - change on the pins when rotating?


And I guess, if the hal indicates the sensor itself is working, where do I go from here?

The encoder index pulse might just be too short to see in halmeter (it might b a fixed-length pulse).

Now we know that the hardware is good, can you remind us what the problem is?

Can you check the question I asked earlier?

Does the encoder count up in M3 and down in M4 mode?

andypugh wrote: Can you check the question I asked earlier?

Does the encoder count up in M3 and down in M4 mode?

Basically, tried a simple tap program. The spindle will move in to position above the hole, spin up and then sit there. Initial presumption was it was not seeing index, but now I'm not sure.

Basically, I'm a little behind the curve here. I bought this machine with little linux experience and I'm playing catch up. The previous owner did use it for rigid tapping, however he also did a servo conversion on it not too long ago and I'm not sure if he used it for tapping after the conversion....it's possible that the config is not currently correct.

I posted the gcode and the hal/ini files on the first page of the thread for reference.

Can you please just do the experiment?
Set halmeter to look at the spindle encoder counts and then run the spindle in both directions. Also look at motion.spindle-revs

The behaviour you see is typical of an encoder counting backwards.

I'll give it another shot in a couple hours, though I'm pretty sure I already did that and it wasn't showing any count.

andypugh wrote: Can you please just do the experiment?
Set halmeter to look at the spindle encoder counts and then run the spindle in both directions. Also look at motion.spindle-revs

The behaviour you see is typical of an encoder counting backwards.

OK, so I think you're on to something!

Both spindle encoder count and spindle revs go down when the spindle is turning forward (clockwise when looking down on the table) and, of course, they co up when it's spinning in reverse.

What's the fix? Is it as easy as swapping the A + B wires at the board?

Yes, you can swap A snd B wires, or you can set the encoder scale to a negative number in the HAL (or INI, if it comes in to the HAL from the iNI in your setup)

andypugh wrote: Yes, you can swap A snd B wires, or you can set the encoder scale to a negative number in the HAL (or INI, if it comes in to the HAL from the iNI in your setup)

You da man!!!

Swapped the wires (didn't want to have to go edit all my backup configs) and ran my test program and it appears to be working just as I'd expect.

Thanks for all the help!!