Beginner questions

Discussions and file drops for Auggie
MarkW
Old Timer
Posts: 31
Joined: Tue Mar 12, 2019 12:40 am

Re: Beginner questions

Post by MarkW »

Well that was too easy.  I just edited the SpindleOn and SpindleOff functions to include SetRelay_1 parameters to turn on/off my air solenoid. 

Watch out.  I'm getting cocky!

Mark
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Beginner questions

Post by ArtF »

Mark:

:) , yeah, its actually all quite easy to do, just so many options even I forget how to do them.
In the end even I look back at the library functions as examples of how to do things.
  Just be aware, inside the scripter is a set of library functions and ways to set scripts to buttons.
Id stay away from there.. I dont use it anymore, and though its functional, its much better to simply
add buttons with names, and then add scripts of those names to library files.
  The scripters method of saving scripts in the buttons themselves is an advanced process, and
at this point its too advanced even for me, Ive forgotten many of the ways it interacts, so Id stay
away from it unless your very brave. Other than that Auggies interface is a very simple one for the
most part.

Art
MarkW
Old Timer
Posts: 31
Joined: Tue Mar 12, 2019 12:40 am

Re: Beginner questions

Post by MarkW »

Art,  I have another observation that I think I need some help understanding.

I have been doing some experiments to try to get an understanding of the depth of cut (DOC) achieved by my 10W laser in various wood materials.  to do end, I wrote some simple gcode that would let me see/measure the DOC at different feedrates.  Here is where I observed some unexpected results. 

The bottom line is this.  When I execute a file with slow feed rates that don't change throughout the code, I achieve Max (97%) laser power (measured as ~4A at 11.9V input power). It doesn't matter if the feedrate is 3ipm or 10ipm.  As long as it is consistent throughout the file, I get 97% power.  When I execute a gcode file where the feedrate varies from line to line, it shows minimum power at the slowest feed rate and the power builds as I reach the highest feed rate.  Ex. F3 = .03A @12V,  F5 0.5A @12V, .... F10 4A @11.9V.  It also does the same gradual buildup of power when I go from F10 - F70.  The slow feed is always min power building to the max power at the fastest rate.

I have changed PWM freq, MinPower setting, and Lookahead parameters.  No noticeable difference.

I don't think I would ever craft gcode (by hand or via postprocessor) to do real work with this type of feed rate variation within it so I don't think I will be concerned about it.  But it does make me wonder.

I've attached 4 small examples to this post.  DOC#s 2&3 use consistent feed and they result in full power.  DOC#1&4 vary the feed and result in "modulated" power.

Mark
Attachments

[The extension txt has been deactivated and can no longer be displayed.]

[The extension txt has been deactivated and can no longer be displayed.]

[The extension txt has been deactivated and can no longer be displayed.]

[The extension txt has been deactivated and can no longer be displayed.]

User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Beginner questions

Post by ArtF »

Mark:

  This is the function of the power function of Auggie. It takes your commanded feedrate, say F600, and then
sets the power word as Actual/requested * MasterPower = Output.

  The reason this is done is to stop CornerBurn. As most lasers have to accelerate to speed, and slow in the corners
its not unusual for a laser cut square to be burnt in the corners. Auggie will  make the output linear
to speed with 100% requested power only when its up to feedrate. This keeps things from burning in areas
where it must run slower. It is important when setting the feedrate to make sure its achievable to mitigate any
non linearity in the materials properties. Its better to go at a speed which stays as constant as you can make it,
and let Auggie control it from there. 

  You can turn off this distance correction but youll find many results more burnt in corners or slow smaller radius
circles which by rules of Anti-jerk formulas must be traveled slower than the straight areas. Pick an actual feedrate
command in the code for a balance between full out capabilities and slow corners and circles..

  Where the feedrate varies on a line by line basis, augige may average some of it in the queue depending on queue
lookahead as it will attempt to find the highest possible speed for all the lines in the queue while respecting feedrate
commanded. For example consider these two samples..

1)
G1X100 F100
M30
2)
G1X100F100
G1X200F200

  In the first example you may only reach 100 in the center of the move,if at all, because it wil begin to decelerate
at some point to hit zero speed at 100 on the X.. depending on accel requirements.

In the second example Auggie knows that the second move may be entered at 100IPM, but that youd like
200 IPM for the second, so the first move will actually run faster as it only needs to decelerate to 100 , not to zero.
Laser output will be very different between the two depending on accel settings.

  If you have a queue of 100 moves, the entire run is replanned as each line is entered to the queue to see if
it made a difference in how fast it might go in every preceding move not yet in travel.  Setting the queue to 200
for lookahead allows it to add more moves to the queue and figure out the best speed on each of them after
their all added. By trying to make a smooth transition from one speed to another in each line, it actually allows for
faster motion through some areas than with fewer lines. It gets pretty complex to figure out the actual blend speeds
youll get so I typically will run in the air to see how fast a particular feedrate is, if it varies too much, the result will
be better with lower commanded feedrates, or with higher lookahead to help blend them.

  All of this of course is clipped by the 6th order jerk limitations, so raising jerk also smooths motion as the system
wont try so hard to keep things within smaller limits of jerk.. Its all a tradeoff between these settings.

  As youve tried jerk settings and such its probably mostly a function of acceleration time vs length of move
and the match between commanded and achievable feedrates.

Art
MarkW
Old Timer
Posts: 31
Joined: Tue Mar 12, 2019 12:40 am

Re: Beginner questions

Post by MarkW »

I think I got it.  Thank you Art.

Mark
Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests