Ajax CNC Mach3 3-axis mill conversion thread

[skyg]

Here's some screen shots, let me know if you need any others.

-Sky

[cncsnw]

My mistake, I momentarily forgot you were using Mach and not CNC11.

Since you are using a separate home switch instead of homing off the limit switch, limit switch settings and wiring would not be affecting your homing process.

[skyg]

Do you think it might be a problem in the script file? Also, I should mention that when I try to home all three axis at once, it does not work. The x and z which normally work fine when homed alone, don't back off the limits.

-Sky

[diycncscott]

Sky,

The Mach brains (and the M92.m1s script) are coded to assume that the first 6 inputs are homing AND limits for X, Y & Z. If you are using something other than those inputs you will need to modify your m92.m1s to read different inputs.

If you are using "Ref All" in Mach, it just moves in the direction you have specified in homing/limits in your Mach configuration until it hits the switch (also specified in homing/limits) It doesn't use the index pulse to home to - and hence doesn't back up. (Being open loop, Mach doesn't know what an index pulse is)

[skyg]

Scott -

I'll try disabling my home switches and see how it homes with just the limits.

-Sky

[skyg]

Haven't updated this in a while, but I'm very happy to report that I have successfully upgraded the control setup to tach mode, with really excellent results. As Scott suggested, the tuning parameters are FAR less sensitive, and far easier to dial in. I'm now getting errors of less than 10 steps when moving at 300 ipm. I have great success machining medium density foams at 200 ipm, which is about 3X faster than I was before. Overall I'm thrilled with how it's working. Now that it's all up and working, I want to actually make some improvements to the mill setup. Right now for a spindle we just have a Bosch router. We have a nice mill spindle, and a Hitachi inverter to drive it. It would be really nice to get the "proper" spindle all hooked up, so I'm going to work on that in the coming weeks.

Only one issue I'm having, not sure what can be done about it. Our z axis wants to slide down once the E-stop has been hit, or an error triggers Mach3. Sometimes this has happened over the work, which of course is not good. Is there any way to make the z axis hold position, even when Mach3 needs to be reset?

-Sky

[diycncscott]

Hello Sky,

While not recommended because of safety reasons, you could program the Mach Brains to hold power (keep motors enabled) under certain reset and estop conditions but you would probably be better off installing a brake on the Z axis. The GPIO4D includes a brake output for each axis that turns on to release the brake anytime that the motor is enabled. The brake is on a spring return and engages when power is removed from the system and/or the enables are taken away from the axis.

You can either purchase a Servo motor with an integrated brake or put the brake on the ballscrew itself (usually on the end of the screw at or near the support bearing)

Regards

Scott

[skyg]

Scott -

I believe my motors do have a brake, but I'll have to double check.

One other thing I've been having trouble with is homing. I think we realized that the homing script for Mach3 was not setup for setups with both home and limit switches. What would it take (cost) to have you revise the script so that it would work?

-Sky

[skyg]

Scott -

My z axis slides fairly easily, which makes me think that I should adjust the Kg setting, but I'm not sure how. Can you give me some pointers?

-Sky

[diycncscott]

Sky,

What are you trying to accomplish? Just because your axis moves easily would not be a reason to add to -or subtract from- Kg

Kg is used to compensate for a load imbalance in 1 direction - for instance a vertical axis without a counter balance.

Running CNC11, slow jog the axis in both directions and watch the load meter. If one direction requires noticeably more
power to move in one direction compared to the other, try adding 5 or subtracting 5 from the current Kg value and then observe the results. rinse and repeat