Struggling with lathe tool auto touch off with tt not at centerline

[suntravel]

Next time in the shop, I will evaluate this.

At the moment I have a back injury and can only play with fusion.

Uwe

[Muzzer]

Many thanks - would be very grateful if you could.

Hope your back mends soon!

[Allin1Chris]

Thinking about this the next day, these inserts have a 0.2mm nose radius, so the Z offsets will end up being a total of 0.4mm different than would be measured by a ball nosed DTI probing the X axis end of the insert tip. So I guess this makes the Z offset problem closer to 9.60mm rather than 10.00mm, which sounds like the width of the probe tip (3/8"), as I previously suggested as a root cause. This effect wouldn't apply to X axis offsets though.

Hello Muzzer,

This does appear to be a small bug in the software. Typically a TT would be put into the chuck like we recommend with our TT-2 or our newer TT-4, so only the X dimension can be "touched" on the "Negative or Positive" side of the TT disk. This is compensated for by the software, and should be for Z too in the case such as yours, however does not appear to be doing this correctly.

For now, you can likely get around it, by adding the "diameter" of your probe tip to the Z Offset value after the auto measurement by hand. (As the offset it will put in, assumes that its touching the same place as your reference tool, which was on the other side). Let me know if doing so corrects it.

Will work on this and have it corrected for future releases, appreciate for bringing it to light.

[cnckeith]

thanks Chris!
and thanks Muzzer,
i think you are the first Acorn user to offset the TT and be approaching it away from the chuck.
easy fix.

[cncsnw]

Maybe this customer will finally be able to use his machine the way it was intended?

[Muzzer]

Many thanks for looking into this. Glad to have unearthed a hidden bug!

It seems to me that I have a similar but smaller problem in X but I will look at that again more closely now and post what I find. At last look, I seemed to have an offset error of 2.5mm in diameter units when touching off in the +X direction with internal tools.

I guess I've been testing some of the features in ways that most people don't. This is largely down to having a rear mounted turret and tools that can only be used in +X ie all on the same side of the spindle. It all started with the internal boring bar, which needs to be mounted upside down with the spindle in reverse, due to the way the machine is constructed.

[Muzzer]

Maybe this customer will finally be able to use his machine the way it was intended?

That's interesting. That TT probe uses pretty much the same scheme as I am using. The main difference is that his spindle spins CCW for external tools (like conventional lathes), whereas mine spins CW. I guess if he retrofitted an Acorn, he'd encounter the same issues currently.

[suntravel]

My first CNC lathe was a Mazak QT8 with such TT-Arm

It had 4 Parameters for the Arm:

Size TT in X and Z
Offset middle of TT X and Z

Uwe

[cnckeith]

yep, will have to lay down some guidelines for support of this.
marc do you have any input on this?
gang tooling implications? reference tool approach ground rules?

[cncsnw]

I think it should be implicit that X and Z Reference positions are set, with the Reference Tool, on the X+ and Z+ faces of the tool setter.

Parameters 281 and 282 should specify the X and Z dimensions of the tool setter. The documentation should make it clear that Parameter 281 is an actual dimension ("radius" value); not the difference in DRO diameter reading from one side to the other.

Auto setting X or Z Reference will always move in the X- or Z- direction. This should be made emphatically clear in the "Press Cycle Start to move to detector" dialog box.

Auto measurement should choose its direction based on the tool nose vector:
Measuring X Offset: vector 3, 4 and 8 move X-; vector 1, 2 and 6 move X+; vector 5, 7, 9 & 0 cannot auto measure X
Measuring Z Offset: vector 2, 3 and 7 move Z-; vector 1, 4 and 5 move Z+; vector 6, 8, 9 & 0 cannot auto measure Z

Whenever the control auto-measures a tool against the non-reference side (X- side, moving X+; or Z- side, moving Z+) then it should include Parameter 281 or 282 in the computation.

The tool library "Tool Orient", "Tool Type" and "Approach" fields should have no bearing whatsoever on the direction of movement or the calculated values for tool measurement. Only the nose vector is needed to determine the direction of touch, and the offset formula.