Brother TC-225 / TC-229 adventure!

so, i got me a pair of brother tapping centres. they get here in a few days. i ordered up a mesa 7i77 card for the TC-225 and that should arrive monday.

So, ill be askin a lot of questions soon!

The TC-225 uses analogue AC servos for the X, Y, Z and spindle. the tool changer uses a motor running a geneva mechanism to turn it in either direction with an encoder of some sort to report its position. aside from a few limit switches and sensors and a coolant pulp solenoid, there's literally nothing else to the machine.

so, being preemptive, i have a few questions on setting up the atc, spindle orientation, rigid tapping, etc.

1: when using an analogue ac servo for the spindle, does it run just like an axis servo? or does it run like a normal vfd with encoder only used to calculate speed, and not actual position?

the original control was quite sophisticated at saving non cutting time. the spindle would orient as it was retracting and move directly into the tool eject position with no pause. then, as the tool magazine spins, the table would move into its next XY position. finally, the z comes back down, picking up the new tool, and with no pause in z motion spins it up as it moves down to its commanded z position. in this way it could achieve a chip to chip time of under 3 seconds. i really want to try and make it behave like this under linuxcnc if its possible. im imagining if the atc and spindle were treated as coordinated axes, it should work. so any insight as to how this works, or if it works, would be appreciated much.

2: this machine requires two upper Z limits. one limit if for machining travel. lets call that "10 inches" (with zero being the lower limit). the second limit at 13 inches, is the position where the tool is released from the spindle. is it possible to set a soft limit at 10, and still have the machine travel past that to 13 to change tools, before hitting an e-stop hard limit?

thanks!

thanks. that seems to make sense on the spindle. basically stop spindle quickly, send orient command, and while that's doing its thing were moving up to z13. the change initiates as we pass z10 (its all mechanically driven). so we can just check on orientation at z9.5 and if its not done, feed hold. i suspect this hold would only come into play if we were retracting for an already very high up position where there's not a lot of time to orient. i suspect we could also stop the spindle as we move up as well and start it on the way down, if those commands dont have to wait til the z move is complete. i guess we shall find out soon enough.

for the z limits, we'll just have to see when we try it. hopefully it is as simple as disabling soft limits during the tool change - and relying on the hard stop switch if anything goes wrong.

I like to solve difficult problems with hardware instead of software, because hardware is easier to see when it works right or when it doesn't.

How would you feel about using a pneumatic cylinder to move the limit switch out of the way during a tool change? I imagine when M6 is called an output would switch a solenoid valve that would cause the cylinder to swing the limit switch out of the way then the z axis could be raised beyond the normal z max height without an over-travel. The necessary configuration to activate the cylinder might be equal to the amount of configuration needed to temporarily ignore the limit switch, but like I said if it doesn't work you can see something not working.

problem with that is were not using a physical limit switch in this case. were using the soft limits for the z "home".

if you did not use soft limits though, and used the actual triggers, your idea could work. but i don't like the air and the idea of a switch that moves for repeatability reasons. but, it gives me the thought that you could use a "switch" to swap between 2 different limit switches.

that's definitely a possible way of doing things. i've never used linuxcnc without soft limits though, so i'm not sure if there would be any road blocks.

thanks

the only hitch to that i think is that the z cants stop at the top of the cutting travel. it needs to go smooth to the tool release position. no pause.

hmm

ah, so g64 type stuff would apply. well, i think were just going to have to find all this stuff out next week.