gack!!! still more problems with photovcarve

Dan Noren

Moderator
Staff member
Messages
7,984
Location
falcon heights, minnesota
this has been nothing but fun. tried doing a 6x6 photo, came out 5/8x1/2, solved that lil problem with the new post processor, except that the new one would not be accepted by the program. new version of the post processor. it seemed to work, but, almost half way done, it makes a bee line for about a foot down the y axis, comes back, does about an inch or so, goes to the right and left about 2 inches on the x axis, then plants the bit right up to the collet nut, prompting an emergency shutdown. so, yet another email sent off to the vectric people, with pics of the 7 pages of error messages generated, along with a pic of the end result. am i having fun, or aren't i? :rolleyes:
 

Attachments

  • end result.jpg
    end result.jpg
    113.1 KB · Views: 20
  • error page 01.jpg
    error page 01.jpg
    89.1 KB · Views: 17
  • error page 02.jpg
    error page 02.jpg
    73.7 KB · Views: 16
  • error page 03.jpg
    error page 03.jpg
    89.6 KB · Views: 13
  • error page 04.jpg
    error page 04.jpg
    83.6 KB · Views: 13
  • error page 05.jpg
    error page 05.jpg
    83.2 KB · Views: 12
  • error page 06.jpg
    error page 06.jpg
    87.6 KB · Views: 13
  • error page 07.jpg
    error page 07.jpg
    85.9 KB · Views: 13
OK - I got both.

I opened the PVC file and looked through all the settings. There is nothing wrong in PVC. It works fine all the way to the end.

WHAT - post processor did you use?
I would recommend G Code Inch (*.tap)

It does not really matter very much as long as you get good g-code.

The G-Code code you sent look fine and should not generate an error.

The error messages you posted earlier - I DO understand them - I do NOT see what would cause those errors in your G-Code program.

Vectric is going to want the PVC file and maybe the G-Code file. They will tell you the problem is not Vectric - I agree. It is VERY VERY rare that there is a problem with Vectric software and PVC has stood the test of time. I am sure I can run your code in my machine and it will work fine. I can read g-code very fluently and I can tell just by looking at it.

If you machine is generating the errors - there is an issue with how your machine is reading the code -- OR --- the code is getting corrupted in the transfer.

I run my machine directly from a computer and use Mach3 as a machine control software. In Mach3, I simply load the G-Code directly and the G-Codes runs. There is NO transfer.

I know you are using something different and that you mention some sort of "sender" or something like that. I don't know what that is, but it may be corrupting your program.

Can you open the G-Code file directly from your machine operating software? Bypass that sender thing?
 
not too sure about that, i open the file in universal g code sender, and launch it from there. the error was generated when i ran the g code generated under the grbl inch g code that the vectric people sent to me. funny thing is, is that when i generate the g code in photovcarve, there is no entry for the shapeoko 2, so i was using the generic g code inch, which saves it as a .tap file. so there could be something cockeyed with the grbl inch g code setting they sent me. on all other items, it works without a hitch.
 
You could try the other post I mentioned -- you may need to enter the G20 near the top of the program.

There should be a setting in your machine to set the machine to inches - then you won't need the G20.

I didn't see anything wrong with your g-code file. Did you send me the one you got errors with?
 
I'm with Leo - though I have't seen the code. I'd be more suspect of the sender than anything else in the chain. I've seen issues with some gcode that's generated with too many decimal places for the coordinates - you know, something like:

G0X1.003030334342288Y0.3432583823323

Some programs don't work with that many decimal places, so they get a little confused. It could just be one single line doing it, too, like if all but one lines have coordinates that end with a few zeroes, the app might just do fine with the truncating (not rounding) there and don't get upset. I saw it once in mach3 when testing out my gcode script for guitar fret spacing. The math took me to something like the 15th or 16th decimal place and mach didn't like that so much.

But in that case, I could see it in the gcode preview of mach3, so i didn't risk any cutters or material. Does your control chain offer any sort of preview of the gcode?
 
... I've seen issues with some gcode that's generated with too many decimal places for the coordinates - you know, something like:
G0X1.003030334342288Y0.3432583823323 ...

The default parameters for decimal places in grbl config for arduino limit to 4 for inch and 3 for mm, so that shouldn't be a problem, unless the Arduino Uno hasn't been updated.

I've been using UGS for most of my cuts and haven't had an issue unless the software generating the gcode doesn't behave properly.
 
The Error codes are stating a few different things.

1) it is suggesting that too many of the same code group on the same line. That would be like commanding a G01 and also a G02 on the same line. I didn't see any evidence of that.

2) Another error was stating that there was an address with no position. Like there was an "X" with no number like X2.375

3) There were some errors that stated a number without an axis designation.


I didn't see any such errors in the G-Code that Dan sent me - yet the errors were generated.


These reasons, plus my 8 years of experience with Vectric, make me believe the G-code was generated error free from VCP and then corrupted in the trqansfer. I could be wrong - it would not be the first time.
 
this is the latest i got back from the vectric people:

Hello Dan,

I've had a look at the G-code in a 3rd party G-Code viewer and it looks fine(see attached screenshot: apologies that it is not more clear, engraving photographs requires a lot of movement): this would suggest that the problem is on the machine side: it may be losing steps, so you could try lowering the speed down to see if that helps.

I am afraid that since I am not familiar with the control software you are using, I cannot say where it is going with with regards to the error messages: you would need to contact your machine provider.

Regards,

Adam Moran

Vectric Ltd | The Coach House | Upper Skilts Farm
Gorcott Hill, Beoley | Redditch | Worcestershire | B98 9ET | United Kingdom

could that be part of the problem, pushing the machine too hard?
 
Since I haven't seen the code, I can't say - but I trust Leo's assessment. I am more inclined put vectric at the bottom of the list of possible culprits.

If you send me the code, I can try to run it through mach3 and see if it runs fine. If so, and i'm sure it will, we can eliminate that it's the gcode pretty soundly. The problem is more likely something in the driver/sender/control software. I doubt speeds would be the cause - it could be, for all we know - I don't know much about the shapeoko but I have heard of very unpredictable results if electrical noise is at play - any kind of noise could drive the machine all sorts of crazy, but let's make sure it's not software first ...

I'm gonna ask once more - does your sender/controller software offer any kind of preview of the gcode once its loaded? I ask, because mach3 does and you can really catch a lot of weirdness with that preview. That's how I found the bug i was talking about earlier with the decimal places.

Since I haven't seen the code, I can only speculate. If you want to send it to me, please do. beamer@beamerweb.com
 
Top