Hello Guest it is March 28, 2024, 06:48:34 AM

Author Topic: Mach3 stability issues concerning G31 floating head on my plasma cutter.  (Read 2421 times)

0 Members and 1 Guest are viewing this topic.


Hello everyone: I have been working on the stability of Mach 3 for awhile. I started out with windows 7 and a laptop without good results. I had problems with inconsistent Z moves. mach wanted to reset zero all over the place one time and then cut fine the next. This was very frustrating. I now have a dual core 3.2 GHZ dell with 4 GB of ram. Standard on board video running on XP Pro. I am also running a Novuson NVEM Ethernet controller. (Mach version 3.043) I ran the first part with multiple profiles (letters) and it ran the first 3 letters fine and failed to stop on the G31 command (top of plate routine) on the 4th letter. This was odd. I went to the diagnostic screen and the probe light was on. I just rewound the program and hit start and it ran with out a hitch 5 times. The 6th time I over-road the feed bumped it to 70 inches a minute and it ran past the probe switch on the 8th letter. Again I just rewound the program, restarted and it ran flawlessly several times. I then bumped the feed down again and it ran past the probe switch again. After rewinding and restarting again it was ok. I have gone in and changed the start up to open nothing upon windows startup and erased several non essential programs. What am I missing? I can include some G code but this must run deeper than the code. should I create a macro with wait states in it to keep the computer in check? Can anyone help me here I am stumped. and very frustrated not being able to run a part reliably.  Is Mach 3 simply an unreliable program!!!!!  :-\

Dave
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #1 on: October 28, 2018, 03:43:38 AM »
Hi,
Mach3 does have its bugs.....however if had a fault with G31 there would be complaints left, right and center....but there is not.
I suspect that the inconsistancies you  have note are those of the controller rather than Mach.

Craig
'I enjoy sex at 73.....I live at 71 so its not too far to walk.'
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #2 on: October 28, 2018, 09:15:15 PM »
Hello Craig,

Thank you for your reply, I have been suspect of this controller. However, the controller has only so much control. No Pun intended. If Mach receives the probe signal then Mach should stop and set to zero etc etc... Or am I oversimplifying the process. I could definitely see it being the controller if the position is inconsistent. Mach is receiving the signal and keeps driving past the probe input. I have switched to Mach 3.042.40 and there is some degree of stability. One more question, I observe the G92 offsets in the diagnostic screen and there are all over the place. Where does the G92 offset number come from? My code reads G92 Z0.0 but I never see it go to zero on the diagnostics screen. If I can not figure this out I will be migrating to the smooth stepper.
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #3 on: October 29, 2018, 01:46:19 AM »
Hi,

Quote
However, the controller has only so much control. No Pun intended. If Mach receives the probe signal then Mach should stop and set to zero etc etc.

That interpretation is incorrect. Probing is a realtime operation, that is to say that the probe contact event has to be recognized and operated
on (cessation of all axes movement) within 10us or so. If you waited for the signal to be propagated back to Mach it would take 10ms or so
and another 10ms before the axes would stop....way way WAY to slow.

As a consequence probing is enacted by the motion controller, in your case the NVEM.  Note that Machs parallel port is a motion controller that tries
to emulate genuine realtime performance.

Mach is not and cannot be a realtime computing system.....Windows all but precludes it. For this reason Machs parallel port driver which attempts
to be realtime is such an amazing piece of code, I don't know any other that can produce near realtime performance with a Windows OS.

The bottom line is 'probing is a realtime operation.....if probing is going cranky look in the first instance to the realtime controller'

Craig
'I enjoy sex at 73.....I live at 71 so its not too far to walk.'
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #4 on: October 29, 2018, 11:04:53 PM »
So, If I go to a straight Parallel port with a standard BOB I will see probing improvement? I really do not need accuracy. My THC will take care of .010". It is frustrating because I will have to destroy my nice neat wiring job to change to the BOB. That is perhaps why I have drug my feet for so long. I will rewire this thing some day and report back. Thanks for your reply.
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #5 on: October 30, 2018, 01:58:40 AM »
Hi,
yes Mach3s parallel port is a realtime, or at least quasi realtime controller and supports both probing and THC
and so should do the job. The parallel port has quirks of it own of course...which I assume you are  familiar.

There are a number of external motion controllers (USB and Ethernet) which do a good job....mostly European and US made.
They are fully functional and the manufacturers support them. The Chinese manufacturers are overall very VERY poor with their after
sales support and even worse about fixing bugs. Chinese controllers are very patchy when it comes to probing, THC, lathe threading
and backlash compensation...in fact all the realtime operations are patchy.

Craig
'I enjoy sex at 73.....I live at 71 so its not too far to walk.'
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #6 on: October 30, 2018, 09:06:06 PM »
Craig, I see you on this forum a lot but I do not know what you run. I have read thousands of posts and get a lot of mixed information. Some people can sound so confident in their answers and are dead wrong. This is in no way an accusation. I am just simply stating some of my frustration with some of thees forums. I have learned through this build that there are many different components out there and you have to be very careful with compatibility. It would be great if someone would compile a list of cnc controls that all played together nicely. I may do this when all said and done. The mechanical side was a breeze. The THC side was a nightmare.  oh I have not given up on that china crap yet. I have it now where I can E stop it raise the Z and start the cut again. I may add a switch the floating head to automatically do this if it goes past the probe switch. It does not happen that often now.   In your opinion what is the best hobby controller? I have been looking at the uc400eth and the smooth stepper. I like the smooth stepper option because I can wire the BOB in ahead of time and test and then slide the Smooth stepper in the middle. Would you mind sharing what you run on your machine?
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #7 on: October 31, 2018, 01:57:31 AM »
Hi,
until about three years ago I ran Mach3 on one, then later two parallel ports to control my three axis homemade benchtop
bed mill. I used a dual core Atom base MiniITX board as PC. It worked really well.

Three years ago I stepped up to Mach4 and an Ethernet SmoothStepper using the same wee Atom PC. The PC being so underpowered
is slow to do the first toolpath draw of a largish Gcode file but fine for milling files. It would be to painfully slow to draw 50Mb 3D engraving
files.

There are a number of controllers that work with Mach4, and only one Chinese one and its absolute rubbish.

At the current time the external controllers, which are Mach4 'native', have a mix of supported features. Currently Vital Systems HiCon board is
the only manufacturer to have developed THC support in a Mach4 ready board.

Warp9 TD's Ethernet SmoothStepper Mach4 plugin does not do THC but has recently added backlash compensation and single point lathe threading.

You will have to do some research on what features are supported in Mach3. If you were considering an external controller go for an Ethernet
connected unit like the UC300, the Ethernet SmoothStepper or a 57CNC by PoKeys. The noise resilience is so much better than USB.
Further I think all of these boards have full realtime function support, ie THC, probing, lathe threading etc. If you want one of these buy direct from
the manufacturer or their nominated distributor, if you buy off EBay or Amazon you'll get ripped with some Chinese copy.....don't do it!!!

If you buy Chines s*********t then don't expect it to work. I've gotten so jaded with people buying Chinese rubbish and then turning up on the
forum to get help to get them to work......and China might as well be on another planet when it comes to getting help, that I don't
bother responding anymore.

Craig

'I enjoy sex at 73.....I live at 71 so its not too far to walk.'
Re: Mach3 stability issues concerning G31 floating head on my plasma cutter.
« Reply #8 on: October 31, 2018, 10:37:19 PM »
Craig,

I have learned my lesson with the china Crap. It has given me way to much grief. Between this NVEM controller and a torch height controller I wasted a lot of money that I thought I was saving. I did not think of the 57CNC. I will investigate that tonight.  I ended up using a Proma SD THC that works well on its own but not perfect. I was looking at a small atom type computer for a small package. That should work great for Plasma. (maybe on My second Build.) My mill is a full servo unit running Centroid controls and it works well with no issues. Thank you for sharing.