Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: Jubei on April 28, 2010, 05:51:57 PM

Title: Rotary Axis Blues
Post by: Jubei on April 28, 2010, 05:51:57 PM
I've spent several hours reading through a lot of posts concerning the Rotary A-axis and CV. 
My A-axis moves in jerky pulses. Any movement in X or Z while rotating the A-axis seems to make it worse.

I read that including G64 in my code as well as making sure CV is set in the general config should help, but it hasn't. Is this normal?

This does not occur when jogging/homing the A-axis. There is no difference in the amount of jerky'ness regardless of whether
CV or Exact Stop is selected.

It also makes no difference whether G64 is in the beginning code to force CV. I did a search through the gcode and there are no G61's anywhere.

I've tried different accelerations, and increased the CV feed rate but that doesn't seem to help. The A-axis still moves in odd slow pulses.

Any ideas?

As an aside, what I am trying to do is mill wax ring using the rotary a-axis. I am mainly cutting the top half of a ring for testing using rotary positions between -90 to 90.
Title: Re: Rotary Axis Blues
Post by: Jubei on April 29, 2010, 12:11:21 AM
Never-mind, It seems that the g-code that was generated was incorrect.  It doesn't appear to be related to Mach.  I haven't found out yet why the code was causing problems, but when I do, I will reply to this post in case someone else has the same problem.  :-\
Title: Re: Rotary Axis Blues
Post by: simpson36 on April 29, 2010, 02:36:24 PM
I think most people who have tried this with Mach have had one problem or another at some point.

Perhaps you could elaborate on what you do know.

What generated the g-code. Why do you say it seems that it was incorrect?

Title: Re: Rotary Axis Blues
Post by: Jubei on April 30, 2010, 09:56:15 PM
The code is generated by RhinoCam Pro.  After a bit of testing it seems that I have narrowed the problem down to the ROT 360 Rollover command. Apparently having Mach set to Rollover to 0 after 360 seems to cause the rotary and x axis to not move in sync with the RhinoCam code if the code uses negative rotary numbers.  

Note: Mach will still work with the 360 DRO Rollover, which is what made it hard to troubleshoot. It just runs very very slowly and in odd pulse behavior. Allowing negative rollover numbers in the DRO causes Mach to run at normal rotary speed.

Apparently my Mach was setup with the Rollover and I never really bothered to see what the effects were. I thought the DRO was more for information. At least that seems to be the problem. I will update as I go.
Title: Re: Rotary Axis Blues
Post by: Jubei on May 01, 2010, 02:55:28 AM
Ok, just finished some more rotary milling and while it works...Mach seems to sputter some when rotating the A-axis and moving in the X-axis.  Does anyone know if this is a Mach issue?  It does not sputter when doing regular 3 axis (xyz) work.
Title: Re: Rotary Axis Blues
Post by: Jubei on May 01, 2010, 04:20:57 PM
Final Resolution:
Two issues were causing the problems I was having with Mach.

1) Make sure that if your code generates neg. rotary degrees select it the general config. to not allow 360 rollover
2) The A-axis sputtering while moving in the x-axis was caused by my old computer not being powerful enough. It seems the rotary calculations with the diameter, feed speeds etc. require a bit more calculating. If your computer is close to the bare minimum (Mine was an old freebie but within specs), it may have problems doing rotary work.

I used a more powerful computer and all rotary sputtering went away.
Title: Re: Rotary Axis Blues
Post by: Sam on May 01, 2010, 08:21:41 PM
Thank you for posting back with your findings so that other may benefit as well. Glad you got it sorted. Happy cutting.
Title: Re: Rotary Axis Blues
Post by: derekbpcnc on May 02, 2010, 04:31:13 AM
Final Resolution:
Two issues were causing the problems I was having with Mach.

1) Make sure that if your code generates neg. rotary degrees select it the general config. to not allow 360 rollover
2) The A-axis sputtering while moving in the x-axis was caused by my old computer not being powerful enough. It seems the rotary calculations with the diameter, feed speeds etc. require a bit more calculating. If your computer is close to the bare minimum (Mine was an old freebie but within specs), it may have problems doing rotary work.

I used a more powerful computer and all rotary sputtering went away.

Hi,

Thanks for sharing your info.
I am putting together a 4th axix for my Bridgeport - a 10" rotary table, 8Nm stepper belt drive.
What spec was you old PC that caused the stutering problem? I think my pc is a 1Ghz machine - hope that will be enough.
best regards
derek.
Title: Re: Rotary Axis Blues
Post by: Jubei on May 12, 2010, 05:37:40 PM
Old PC was 1GHz P4 and 1.5G ram...performance with 3 axis was acceptable...4th axis was a struggle.
Title: Re: Rotary Axis Blues
Post by: simpson36 on May 12, 2010, 05:52:33 PM
Just an FYI about computers running Mach3.

I have discovered that the built in PP can cause symptoms that look as if the CPU is not fast enough when actually it is the PP chip, (sometimes built into the bridge).

My current computer misbehaves if I use the built in port, so I disabled it and use PP port one and two from an add-in PCI card. Works fine that way.

You can check this out easily by watching the task manager. If the CPU is not pegged, but the machine starts acting up, it may be the PP, which is semi-autonomous and can cause problems even if the CPU is humming along fine.

Also don;t assume that the drivers that come with an add-in card will work with Mach3. My card has the Mos chip and the drivers that come with the card do not work. The drivers from the Mos site, however, work fine.
Title: Re: Rotary Axis Blues
Post by: Jubei on May 12, 2010, 06:05:59 PM
This testing was done using a third party parallel I/O card. I would probably not recommend using the built in ones for the reason you suggested. Third party/dedicated cards are not that expensive at all.