PathPilot on NM200

[PCW_MESA]

Peter's settings did not work.

Umm not true, they do work if what your steprate --> Analog converter needs is a 0 to 25 KHZ variable frequency squarewave for 0 to 10V -->
0 to 4000 RPM spindle and as I mentioned, PP (well LinuxCNCs) motion.spindle-speed-out is netted to the stepgens velocity command
(no other scaling applied to spindle RPM)

Verified with Scope/ frequency counter and Gcode

[kvom]

What I meant by "not works" is that my mill is not working with those parameters and no other changes. The tormachspindle routine has other params that may or may not be screwing this up, and I'm going by guesswork here. These are pretty much those from the Tormach HAL. Is there any known public documentation of tormachspindle?

The 25KHz max figure is right from the documentation of the C11 BoB. Scott may be correct about PWM but that is wishful thinking since the BoB doesn't have that.

[kvom]

More clues from playing around with the HAL spindle config.

With the scale set to 6.25, the spindle will not run until the RPM DRO is set to 200. At this point the VFD indicates 4045 RPM. That indication doesn't change until the DRO is lowered to 108. At 107 the RPMs drop to 3970, and drop a further 170 for each 1 RPM indication in the DRO until the DRO is at 100 for 2765. Can't command it lower because of configured minimum.

Taking a simple-minded aopproach I took the ratio of 108 to 4045 and multiplied that to 6.25, giving 1.67. I set the scale to this value and restarted. Now I get 2141 RPM with the DRO 550 and the spindle won't run at a higher setting. Actual RPM drops 500 for every 100 drop in the DRO value.

The only conclusion I can draw is that if the stepgen is at a frequency higher than the BOB's 25Khz it either doesn't generate any voltage or else the VFD refuses to run over 10V.

As to how to tie commanded to actual voltage I'm still seeing through a glass darkly.

May not be able to work on this much more until the Hurricane goes by.

[photomankc]

Have you tried looking at the voltage output without the VFD attached. Is the BoB producing 10V out when you hit max RPM? Is it moving up and down in proportion to the commanded speed?

Ultimately the BoB is doing PWM, but it looks like the interface to you is step and direction with 24Khz steps being 100% output from the PWM. I know my PWM (different BoB) was not anywhere close to linear with the output voltage. Most of the range was used up in the first few hundred commanded RPM so I had to straighten out the nasty curve it was producing. First step was to see what the actual output voltage was doing at various RPM settings. I'd do that here and make sure that you are seeing an output voltage that makes sense. If the BoB is being driven by step/dir frequency then it should be roughly proportional as you increase RPM. If you have a meter with a frequency counter, that could be used to look at what step rate is actually being produced at pin 14 on the DB25 at various RPM settings.


Stay dry!

[brianbonedoc]

Kirk if you get close with the spindle speed (within a couple hundred rpms) you can use the command lincurve to smooth out the response. It's pretty easy to use. I know Scott just did that on his Torus pro.

[kvom]

I can measure the voltage at the VFD, but given that Mach3 drives the spindle within +/- 50 RPM through the range I'm more suspicious of PP parameter settings.

I was trying to register on linuxcnc forum but am not receiving the confirmation email.

[kvom]

One problem solved. For the each axis config in the INI file I set the HOME value to 0 (so referencing doesn't move the axis, and I set MIN_LIMIT to -n and Max_LIMIT to n, where n is the amount the table can move along that axis. Now the soft limits are set beyond the ends of the table no matter where the table is at startup when I reference.

Note that my limit switches don't work, so settings has the ignore box clicked.

[kvom]

I loaded a g-code file today that I'd generated in CamBam and run on Mach3. I commented out the M3 command and ran it. No issues at all with any of the generated code, but one thing was different. PP apparently assumes that the initial tool is already loaded, so it skips past the first M6 without stopping. A subsequent M6 is obeyed.

Is this normal behavior, and if so is there a way force a stop at the first M6?

[brianbonedoc]

CamBam probably isn't the best to run without editing. Many/most use Fusion 360 as there are post processors that are great for PP.

U need a T# G43 H# M6 before any run if you want to make sure you have the right tool.

[LeeWay]

And if you are going to run Path Pilot, you really should get in the habit of homing the machine. Things just start working like they should when you home first and use the correct Post Processor.

[brianbonedoc]

And if you are going to run Path Pilot, you really should get in the habit of homing the machine. Things just start working like they should when you home first and use the correct Post Processor.

^^^^^^^^^^^ This.

If youre limit switches aren't working just by some cheap inductive proximity sensors from amazon (like this: https://www.amazon.com/Uxcell-a12091...624K0E7NF3K8H4)

[sagreen]

Umm not true, they do work if what your steprate --> Analog converter needs is a 0 to 25 KHZ variable frequency squarewave for 0 to 10V -->
0 to 4000 RPM spindle and as I mentioned, PP (well LinuxCNCs) motion.spindle-speed-out is netted to the stepgens velocity command
(no other scaling applied to spindle RPM)

Verified with Scope/ frequency counter and Gcode

PCW, could you create a config for kvom that uses PWM instead of STEPGEN similar to the firmware you built for me?

Not sure why he is fighting this problem still and trying to fake a PWM signal with StepGens. Unless I dont understand his BOB he would have had this solved a long time ago by just using PWMGENs and copying my spindle parms from my .HAL file.

Scott...

[kvom]

Unless I dont understand his BOB

BoB doesn't have a PWM.

And if you are going to run Path Pilot, you really should get in the habit of homing the machine. Things just start working like they should when you home first and use the correct Post Processor.

I have been running the machine without homing for years, so I'm not sure what advantage it has now under PP.

I can modify the CB post if needed, but as of now I can't see any issues other than the initial M6 stoppage. The G43 H# is unusable since I don't have tool holders or use the tool table. It's an R8 spindle, and I have to zero Z for each tool. I can try it using a dummy tool table entry with the same H# for every tool. That's an easy fix to the post.

This is all moot until I figure out the spindle issue.

[brianbonedoc]

Kirk - pp requires all the tool length offsets preset before the program is run (as far as I know). So you may just have to break ea part of your CAM into a separate program. Unlike Mach there is no changing offsets during a program run. I would strongly encourage you to change over to TTS tooling as it will save countless hours and headaches.

[kvom]

I just verified that I cannot jog once program starts, even if paused at an M0 or M6. Breaking a job into separate files per tool is the likely solution I'll employ. Given that I use a large number of different tools for different parts, if I used TTS I still have to measure each one before a job. If I'm drilling holes with different size drill bits, that could get out of hand.

[brianbonedoc]

Kirk - what i do w drilling is use the same chuck and load each drill bit all the way deep in the chuck. Then measure the length off line. That way I can load each drill in a repeatable manner during the program.

[sagreen]

BoB doesn't have a PWM.



I have been running the machine without homing for years, so I'm not sure what advantage it has now under PP.

I can modify the CB post if needed, but as of now I can't see any issues other than the initial M6 stoppage. The G43 H# is unusable since I don't have tool holders or use the tool table. It's an R8 spindle, and I have to zero Z for each tool. I can try it using a dummy tool table entry with the same H# for every tool. That's an easy fix to the post.

This is all moot until I figure out the spindle issue.

Any chance you have access to an oscilloscope?

Scott...

[PCW_MESA]

OK so a closer look at the C11 REV8 manual shows this for setting up Machs stepgen:

"Go to Config / Motor Tuning / Spindle. On Steps per unit put 1,000, set velocity to
maximum. For Acceleration, choose the acceleration that you feel comfortable
with. Start slow, increase acceleration as you test your system. Under Step Pulse
length, use a number from 3 to 5, but start with 3. This number is directly
proportional to the final voltage you will get in the analog output. Use this number
and the fine tuning pot to adjust the voltage you want to get at max speed."

This indicates that the analog converter circuit is not really a frequency to voltage converter
but in fact is probably just a filter with odd scaling for converting the stepgens
variable frequency fixed width pulse (PDM) to analog.

So the stepgen setup for this would be (normal step/dir with 3 usec step length)

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.dirsetup 100
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.dirhold 100

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.steplen 3000 # 3 usec
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.stepspace 100

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.position-scale 6.25

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.maxvel 4000
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.maxaccel 4000

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.step_type 0 # normal step/dir
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.control-type 1

[kvom]

OK so a closer look at the C11 REV8 manual shows this for setting up Machs stepgen:

"Go to Config / Motor Tuning / Spindle. On Steps per unit put 1,000, set velocity to
maximum. For Acceleration, choose the acceleration that you feel comfortable
with. Start slow, increase acceleration as you test your system. Under Step Pulse
length, use a number from 3 to 5, but start with 3. This number is directly
proportional to the final voltage you will get in the analog output. Use this number
and the fine tuning pot to adjust the voltage you want to get at max speed."

This indicates that the analog converter circuit is not really a frequency to voltage converter
but in fact is probably just a filter with odd scaling for converting the stepgens
variable frequency fixed width pulse (PDM) to analog.

So the stepgen setup for this would be (normal step/dir with 3 usec step length)

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.dirsetup 100
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.dirhold 100

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.steplen 3000 # 3 usec
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.stepspace 100

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.position-scale 6.25

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.maxvel 4000
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.maxaccel 4000

setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.step_type 0 # normal step/dir
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.control-type 1

I'll give it a try after the Falcons game is over.

[kvom]

These parameters gave a more reliable start/stop operation of the spindle, plus full range of speed, but the commanded to actual is still way off.

At S100 VFD reports 106 RPM
At S200 906
At S300 1812

Reaches max of 4045 at about S555.

Each increment of 100 increases RPMs by ~900. After shutting down for the evening, I noticed that the MAX_MCB_FREQUENCY in the INI file is set to 10,000. I'll increment that to 25,000 tomorrow to see if that has an effect.

Read up on lincurve, so if push comes to shove I think that can translate between command and actual speeds.