Messages

11

Third party software and hardware support forums. / Need help with c41 board and pwm spindle control with VFD

« on: July 31, 2013, 09:39:44 AM »

Problem #1
I am replacing the BLDC driver and motor on my mill with a Durapulse VFD.  The Durapulse VFD uses a 0-10v analog signal input for spindle RPM.  The proprietary hardware electronics of the mill outputs a PWM signal so I have a C41 board (rev.1) from CNC4PC to convert the PWM signal to an analog voltage signal for the VFD.  Sound simple right?  Well the problem is that the PWM signal from the machine is inverted, meaning 0% duty cycle is max RPM and 100% duty cycle is min RPM or an active low configuration.  This is the way the supplied BLDC driver needs the signal.  The Mach3 interface has the same configuration to make it compatible.  I can change the Mach3 step setting to active high but I can't change the hardware of the machine so it will always output an active low signal.  The C41 board however only reads active high.

I think I need to put a circuit in between the Mill hardware that is outputting the backwards PWM signal and the C41 board to make it work.  The question is how do I invert the PWM signal, I think it should be easy to do but really don't know where to begin.  I'm assuming I have to make some type of op amp circuit.  Please help or offer suggestions.

Problem #2
Even when supplying the correct active high PWM signal directly from Mach3 to the C41 card I get no dynamic response on the output analog voltage.  There is just a constant ~1.3v which I can adjust slightly with the pot.  No matter what the input signal is to the c41 I can't get the output voltage to change or even get close to 10v.  Perhaps it's wired wrong?  I am supplying the external power with +12VDC and -12VDC for ground (I'm not sure if that is the correct way to do it).  I don't have the relays wired yet, I'm just trying to get an analog output voltage first.  Please tell me what I'm doing wrong.

Here is how it's wired:


Also here are some relevant images of the VFD terminals:


I'm banging my head against a wall trying to figure this out.

 

12

General Mach Discussion / Re: Having a few issues with Mach

« on: March 31, 2011, 02:15:38 PM »

Ok I installed the chipset drivers so that alert went away, I still have the unknown device though.  The key to finding out what the problem is is identifying what is different between mach 3. 042. 020 and mach 3.043.022.  All the newer versions destroy my CPU but the old version does not.  How can I find this information?

13

General Mach Discussion / Re: Having a few issues with Mach

« on: March 29, 2011, 10:14:26 PM »

Does mach use these?  Device manager says that they are not working properly and are experiencing problems.

14

General Mach Discussion / Re: Having a few issues with Mach

« on: March 29, 2011, 10:53:00 AM »

Thanks for checking, the xml runs fine on my laptop too.  On the machine I don't have anything extraneous going on, I am using bone stock Mach install, standard default screen, no brains or scripts or anything like that.  It is quite curious.

15

General Mach Discussion / Re: Having a few issues with Mach

« on: March 29, 2011, 08:37:49 AM »

Sure here is the active one I am using

16

General Mach Discussion / Re: Having a few issues with Mach

« on: March 29, 2011, 01:12:56 AM »

I've confirmed the issues are either with Mach or with my xmls.  Something with my xmls are hogging the CPU.  Here is a synopsis:

Here is an update, I've done some diagnostics and discovered some interesting behavior.  I uninstalled Mach completely, deleted the C/mach folder, uninstalled the parallel driver and even deleted all instances of mach in the registry.  I then reinstalled Mach, the problems were still there and I was monitoring the CPU usage.  When either the default xml or the unaltered backup xml was used the CPU usage hovered around 60% when sitting idle.  If I tried to jog it at 100IPM it would spike and flatline at 100%, jogging at 20IPM was smooth and the CPU usage hit a high of about 80-90%, jogging at 30IPM was stuttery and the CPUs went up to around 100%.  When I started mach with just the mach mill default profile the CPU usage dropped significantly and hovered around 10%.  Indicating something about my xmls are hogging all the CPU.  I then uninstalled it completely again and reinstalled earlier versions of Mach beginning with the current lock down 3.043.022 and working backwards repeating the process because I had updated to the current version as the demo version installed on the computer was an earlier version (I can't remember the version number).  I went through about 5 versions going back to 3.043.012 with no luck.  Then by accident I installed 3.042.20 which was a circa 2009 version and all of a sudden it worked.  The CPUs were much lower around 20% with mach running my xmls and the jogging was smooth and responsive, the DROs updated in real time again, and command G0 moves were smooth at 100IPM.

I'm not sure what version came on the computer but I would like to know because although not current it at least worked.  However even the current version only hogs all the CPU and bogs down the system when using my xmls so I'm wondering if they are corrupted somehow (even though I've never directly loaded the backup xml that was on the desktop into mach, I always make a copy and put it in the Mack folder to use it) or perhaps the newer versions of mach just don't like them. 

17

General Mach Discussion / Re: Having a few issues with Mach

« on: March 28, 2011, 08:51:10 AM »

I looked at the CPU load before and I was surprised at how high it was when running Mach.  Without Mach running it was low, around 1-5%, with mach running it bumped up and stayed around 40-50%.  This is on a 1.8Ghz machine with pretty much nothing else on it.  That seems kind of high because when I run Mach on my laptop it hovers around only 3-9% on a 2.6Ghz CPU, that can't be normal can it?

18

General Mach Discussion / Re: Problem with Error Involving Arcs

« on: March 28, 2011, 01:23:51 AM »

So just playing around, using .5" flat mill a I can do a 1.4999 hole pocket with spiral arcs but not a 1.5.  That just seems like a weird cut off point and I would venture that it could be changed somewhere if it is just an arbitrary value.

19

General Mach Discussion / Re: Problem with Error Involving Arcs

« on: March 28, 2011, 01:04:05 AM »

Your right that is what is screwing my spiral arcs.  I have the right format XYZIJ but the values don't work unless I tweak the X and Y values to be lower.  Why is the limit .5"?  That seems so arbitrary.  Is there any way around this .5" limit?

20

General Mach Discussion / Re: Having a few issues with Mach

« on: March 27, 2011, 10:54:03 PM »

Ok that fixed one problem, the X axis homing.  Somehow the x axis homing got disabled, easy fix. 


The stuttering is still there but it is very weird.

I can jog the axes at 100IPM without any stutter, it is very smooth (albeit unpredictable when it will stop).  However if I enter a location in the MDI it stutters like crazy on anything above 20IPM (25 gets a little stutter every couple of seconds, 30 it constantly stutters).  I also notice that sometimes in the commotion of the stutter you can see the axis very briefly reverse direction while it is trying to move to the location like the motor is fighting against itself between trying to go forward and backwards.

The issue with the lagging jog is still there and disconcerting, I almost broke my probe and crashed it a couple of time because it just kept going, I had to hit the E-stop a few times.  The DRO still only update during a jog every few seconds.