Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: Davek0974 on May 04, 2016, 02:42:22 PM

Title: Odd behaviour?
Post by: Davek0974 on May 04, 2016, 02:42:22 PM
I'm tweaking the post-processor for Vectric Aspire for use on my mini-mill, by default it puts a tool change at the start of each file, I don't really want this.

This is the start of a file...

N240 G00 G21 G17 G90 G40 G49 G80
N250 G71 G91.1
N260 T2 M06
N270 (T2 - 5mm Carbide YG)
N280 G00 G43 Z20.000 H2
N290 S19000 M03
N300 (Toolpath:- Apertures Roughing)
N310 ()
N320 G94
N330 X0.000 Y0.000 F1000.0

When edited I get this...

N240 G00 G21 G17 G90 G40 G49 G80
N250 G71 G91.1
N260 G00 Z20.000
N270 S19000 M03
N280 (Toolpath:- Apertures Roughing)
N290 ()
N300 G94
N310 X0.000 Y0.000 F1000.0

Seems ok to me, I tried running it but the mill started running the file with the spindle off, luckily I e-stopped early enough to stop damage.

I then reset and tried an M03 in the MDI box and I got "Compiler error in M03.m1s" message, could not get spindle to start, rebooted and suspecting the code was wrong I returned the file to it's original state and tried again to get the job done - it worked ok this time.

So, I don't know until i try again if the code change was wrong or it was just some sort of duff boot-up first time round??
Title: Re: Odd behaviour?
Post by: Hood on May 04, 2016, 06:58:59 PM
That is a problem that  a few people have seen, I have never seen it myself but a friend in Glasgow suffered with it for a while.
Don't know why it happens but doesn't seem to be related to whether you use the PP or an external controller.

One thing you can try is open your m3 macro in notepad (not Mach VB editor) and see if there are any extra characters  after the DoSpin(CW), if there are  delete them and hopefully that will be the end to your problem, it worked for him.
Sad thing is however it seems to run fine lots of times then suddenly one time it wont, so you may not know for a while whether that cured it or not :(

Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 05, 2016, 02:15:44 AM
Ok, good to know i'm not going nuts;)

I will check the M03 macro.

My removal of the tool change looks correct though??
Title: Re: Odd behaviour?
Post by: robertspark on May 05, 2016, 02:11:31 PM
Why do you have to remove the toolchange?

On the general configuration tab you can tell mach3 to ignore m06 toolchange commands (top left hand side from memory)

I have found with vbscript (macropump etc) that you need one carriage return after the last command (press keyboard "enter" after the dospin(cw) )
Title: Re: Odd behaviour?
Post by: Davek0974 on May 05, 2016, 02:28:32 PM
I want to remove it because my job sequence is -

mount plate on mill bed
fit first tool
zero work coordinates
zero Z axis to plate surface
run job

When i start the job running, having a tool change wastes time as it just runs through what i have done already, I prefer to know that I have the first tool correctly set before running, my jobs always start with the first tool loaded and set.

I can't simply ignore tool changes as most jobs have multiple tools in use. The excellent 2010 screen-set with its semi-auto tool change macros gets a good workout on my stuff;)
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 08:49:49 AM
Happened again today - started the run without starting the spindle, not so lucky this time and the tool took one for the team :(

I had a look in the M03 macro and there were no characters after the do-spin bit but there was also no carriage return and I seem to remember reading that the macros like to have a return on the last line?

I put one in and after a reboot it run the job fine, will keep an eye on this one.
Title: Re: Odd behaviour?
Post by: Hood on May 14, 2016, 11:27:07 AM
Do you have spindle feedback to Mach?

Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 11:52:47 AM
No. ;)
Title: Re: Odd behaviour?
Post by: Hood on May 14, 2016, 11:56:35 AM
Ah ok, shame as I had a macropump a long time ago that would look at the commanded speed and the true speed and if not within a certain percentage it would Halt the code.

It worked well but only when I was not using Modbus, sadly I needed Modbus for my PLC so couldnt use it.
Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 12:09:36 PM
Ah, but, yes, but no, but, I could maybe get a run signal from the VFD back into Mach?

Then it would halt if no run feedback was seen before issuing the first Gxx code??

I'm using the VFD relay configured as "VFD Output On" to control the cooling pump, it would only take a relay in parallel to do this, I think there is a soft output as well on the VFD.

Maybe worthwhile?
Title: Re: Odd behaviour?
Post by: Hood on May 14, 2016, 12:38:07 PM
If you could get the run signal to Mach then you could have the M3 macro look at it before allowing code to run, problem likely would be when you get the problem with the macro not being read Mach just continues anyway I think.
So that would not work :(


Ok so maybe you  could add a custom macro after each M3 and have that custom macro look for the signal and if not present within a set time do an E-Stop.

Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 12:50:58 PM
I'm sure there is a way forwards here, will look further, odd that the plasma has never done it though.
Title: Re: Odd behaviour?
Post by: Hood on May 14, 2016, 01:29:43 PM
I don't know why some people get it and others, like me, never see it.
I have used Mach since the Mach2 days and have had a manual Bridgeport conversion, a CNC Bridgeport, Beaver NC5 mill, Computurn lathe, Conect lathe,  Chiron VMC and a coil winder and I have never once had that problem.

Might be worthwhile making a new profile from scratch and seeing if that helps, or even deleting the macro folder and starting Mach up and it will create new macros for you.

Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 01:38:06 PM
Hmm, odd.

I'm using the 2010 screenset so the macros came with that, not sure about the M03 though I think thats standard.

I'll give her a bit longer before reinstalling as its all working well apart from this very intermittent issue.
Title: Re: Odd behaviour?
Post by: Hood on May 14, 2016, 01:44:47 PM
Well if you can get a spindle running signal from the VFD to Mach then you can make a macro and you wont have to worry.
Hood
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 01:56:00 PM
would seem to be a reasonable addition, it is pretty important that the spindle is running i guess ;)
Title: Re: Odd behaviour?
Post by: ger21 on May 14, 2016, 01:58:26 PM
Delete the M3 completely, and create a new one from scratch in Notepad or whatever editor you use.
Just type
DoSpinCW()





and save as "m3.m1s"

From what I've seen, if you create a new one, it usually stays OK.

And no, I don't supply that macro with the screenset. only the ones unique to the screenset.
Title: Re: Odd behaviour?
Post by: ger21 on May 14, 2016, 01:59:12 PM
And fwiw, it's never happened to me either.
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 02:05:10 PM
Weird init ;)

Will try with my edited version tomorrow.
Title: Re: Odd behaviour?
Post by: mc on May 14, 2016, 03:27:34 PM
I had it randomly on my lathe. I also had it once where it turned the spindle off instead of reducing speed. That one wouldn't of been so bad, but it knocked both the turret and spindle out of alignment, as well as shattering a relatively new drill  >:(

I've got to admit, I definitely don't miss Mach on my lathe!
Title: Re: Odd behaviour?
Post by: robertspark on May 14, 2016, 04:02:05 PM
I had a problem with a macro that would not complete the function, and tracked it down to not having a carriage return after the last function.

Try it with a number of dry cuts and see if it now works, presume you have the two cypress basic mach3 manuals, can't remember if it was in there or on here or a known issue with basic...

(It was a THC macro not a dospin() function though)
Title: Re: Odd behaviour?
Post by: Davek0974 on May 14, 2016, 05:17:27 PM
As i said, i did read it somewhere, just buggered if i can remember where - it needed a carriage return at the end of the do-spin line.

I have added one as it was missing and will be running again tomorrow, see what happens.