Scott - try changing these parameters in the ini file:
steplen = 425000
stepspace= 0
After some awesome help from Brian in the other thread, I've gotten PathPilot to move all the Axis's and turn on the spindle. I am using most of his .ini and .hal files. Brian got PathPilot working great on a pulsar.
Problem is that the Torus and the Pulsar use different types of spindles. The Torus uses a VFD controlled spindle, and the Pulsar has a servo. This requires a ton of different settings to get the spindle speed to work for the Torus. Currently my spindle turns on, reverses and goes forward correctly but the speed goes to maximum.
Anyone solved this for a Torus Pro? i.e. you have installed PathPilot and have gotten your Spindle to correctly react to speed changes? If so, I could sure use your .ini and .hal files!
Thanks!
Scott...
Similar Threads:
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Scott - try changing these parameters in the ini file:
steplen = 425000
stepspace= 0
I would look at what those parameters are in an original Path Pilot ini file. I think the basic Tormach and the Torus Pro have similar spindle setups. Hardware wise that is.
Sounds like you are getting close.
Lee
The Pulsar has an AC servo spindle, so control is via a step/dir output from the controller SW. The Torus Pro has a 3-Phase AC spindle, so control is via analog voltage, which is controlled by a PWM output from the SW. I would assume this would require a different bitfile for the Mesa, to implement the PWM logic. I don't see how this can be resolved by a simple HAL edit. That said, I'm pretty sure people here HAVE gotten PP working on the Torus Pro. Look in the original PathPilot thread in this forum. I thought the first guy here that got it working was running on a Pro.
Regards,
Ray L.
Ray - you are correct to some degree. The AC spindle can be controlled by StepGen outputs which mimics a PWM output. If you think about it, all the MESA is doing is telling the BOB what to send to the spindle.
Last night, Scott got the spindle running, yeah! But it seems to be responding oddly to speed control. He would input 4500rpm and get 2900 at the spindle, and then type in 250rpm and get 5900 at the spindle.
So he's getting closer. . .
Yes, you _can_ use a stepgen with an averaging filter (most analog output BOBs use such a filter)
to get a analog voltage (it would generate Pulse Density Modulation (PDM) not PWM)
Compared to PWM it will be more awkward to scale
(PWM is trivial to scale, you just set the PWM scale parameter to spindle RPM at 100% PWM)
You would need to set the step pulse width long enough so that its passed without much distortion through
the isolation circuitry.
There may well be an existing bitfile that matches the Torus Pro's BOB.
So you could look through the various .pin files in the 5i25.zip from Mesa
If you had a pinout of the Torus Pro BOB I could make a specific bitfile for it
but it may well already exist
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
The Torus Pro BOB has a 0-10V PWM to voltage circuit on PP pin 16. Feeding it a Step/Dir signal will produce SOME variation in output voltage, but not much, and certainly not enough to be of value for spindle control. Scott needs to either get a bit file with a PWM output (on the order of 12kHz, IIRC - I could check what I output on my KFlop), or he needs to bypass the logic on the stock BOB with a suitable PWM-to-voltage converter, like one of the Homann DigiSpeed boards,
The VFD also needs FWD and REV logic-level control signals which are normally high, with FWD driven low when the spindle is needed to turn CW, and both low when the spindle is needed to turn CCW.
Regards,
Ray L.
http://freeby.mesanet.com/novatp.zip
PWM_OUT on pin 16, PWM_DIR on pin 14
Note that you may need to invert either or both of these signals by
setting the invert_output parameter of the corresponding GPIO pin
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Thanks. I was able to decypher the file to understand the pins.
Im having another problem now...
This line in my .hal file fails saying "Pin hm2_5i25.0.pwmgen.00.value not found"..
"net spindle-speed tormachspindle.mcb-pwm-out hm2_5i25.0.pwmgen.00.value"
This line works fine...
"net motor.03.command pid.3.output hm2_5i25.0.stepgen.03.velocity-cmd"
I believe that its the fully qualified hm2_5i25.0.pwmgen.00.value. Like it doesnt see pwmgen in the hm2_5i25.0 name space... In other words it finds stepgen but not pwmgen
This is my definition at the top for pwmgen
loadrt hm2_pci config=" num_encoders=0 num_pwmgens=1 num_stepgens=14" #SAG 7/31/17 - was num_pwmgens=0
Any ideas or does this sound foreign?
Thanks,
Scott...
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Ok, false alarm. I got the .bit file to work. Oddly enough I flashed 3 times and on the 4th it worked. The only thing I changed was I renamed your bit file something else and reflashed. Then it came up with the proper number of PWMs. Brian asked me like 3 times if I had reflashed so its likely a goof on my part.
I dont have the spindle turning yet, but at least I can reference hm2_5i25.0.pwmgen's now. So I just have to figure out the logic to get the spindle to spin.
Thanks, and stay tuned
Scott...
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Ray,
You mention above:
The VFD also needs FWD and REV logic-level control signals which are normally high, with FWD driven low when the spindle is needed to turn CW, and both low when the spindle is needed to turn CCW.
These 2 signals come out of the novakon BOB. I believe that the output of PIN 14 is interpreted by the BOB to create the FWD and REV signals. How should pin 14 be treated (HIGH/LOW CW/CCW) to produce the desired FWD and REV outputs?
Other than pin 16 (PWM) and Pin 14 (DIR) is there anything else that has to happen on the parallel port of the BOB to cause the spindle to move in the proper direction, at the desired speed?
Thanks,
Scott...
Instructional Videos for CNC Guitar Building
http://www.rmgvideos.com
Scott,
On the Novakon BOB, the FWD signal is controlled by the MCU, based on presence of both the charge pump signal and a valid PWM signal. The REV signal is controlled by PP pin 14.
Regards,
Ray L.