Variable laser power

[ArtF]

Seems to indicate a nonlinearity in power..
I would imagine that varies from system to system and power supply to power supply..
It will be interesting to see similar tests over various systems. I could be all wrong as to
how they react generally, I have only 2 that Ive cut with.. but they seem more linear that
that. 6ma still seems low though, what does it do at 15ma? do things get more linear as
power goes up?

Thx for the examples, interesting results.

Art

[ArtF]

Makes me want to add an ma meter to mine, I select only 0 - 100% and who knows what that means to ma..
it could be ma isnt linear to pwm command.. but then, I typically use 99% and vary feedrate.. still, interesting..

Art

[Richard Cullin]

I read in one of arts  posts that the  beam pwm output is not continuous when lasing  to allow the laser tube to recover  , i'm thinking this is not a good thing for a laser diode in that it limits the power unnecessarily  .  when viewed on the oscilloscope the beam pwm output  in indeed in bursts  at 50/50 mark space
is a  diode pwm mode possible ?
am I only getting 50% laser power ?

[ArtF]

Hi Richard:

  Well, that's dependent on a few things. There are two settings in the config that help control that,
Max Laser and MinLaser. They set the maximum and minimum laser power for any given situation.
So lets say you set them to 0 and 100. Your PWM now has the capability of 0 - 100 PWM, so it means you
can use Maximum 100% power. Auggie decides that based on Feedrate. Lets say you command a feedrate of
2000 mm/min , when you get to 1000, your PWM will be 50%, as you approach 2000 your PWM will
approach 100%.
  The normal trouble one has is you command 1000 inches per minute on a job where your machine
can ( due to accel or jerk limits on that job) only do 500 most of the time, the PWM will not exceed
50% in that case. Lowering the commanded feedrate will allow you to hit 100% of the feedrate,
and therefore hit 100% of the power.
  By doing this Auggie keeps you from burning so easily in corners or shallow curves where it must slow down.
Its all more about power/distance rather than just raw power if one wants a good engraving. It is possible
to turn off the power to feed-rate correlation and simply have the power you selected, 0-100%, but unless
its a cutting operation,. I dont recommend it. The DistCorrection mode button  can be turned off and power
will always be the selected power if one wishes. You may not have the laser control panel with a distance correction
button on it, Ill check I put it on the released screen.

Art

[Richard Cullin]

whats confused the situation is the test fire  button timing .  if I set  time to 100mS  then that's what I get
if I set the time to 200ms&nbsp; I get&nbsp; 139ms if I set it to 500ms I get < 250ms

[Richard Cullin]

screen shot of salea logic analyser measuring &nbsp;fire pulse &nbsp;
bottom trace is zoome in a bit to show duty cycle better

fire time set to 200mS power 72%&nbsp; &nbsp; &nbsp; result 139mS&nbsp; of pwm @72%

[ArtF]

Richard:

Hmm, that one is new. I just used testfireing this week to retest power, it did a good 38 second run at 38000 in the timer.
Im trying to think what could make that variable.. its hard to imagine as its coded to the tick timer..

&nbsp;I did have trouble last week with a system that hadnt been booted in quite a while so the tick timer went negative, I wasnt
checking for the upper bit to toggle. Can you reboot and see if the fire test works properly. When the time runs out, does the led go off as well or does the laser just stop firing? ( Like if you do a 10 second test, does the led stay on for 10 seconds on the firing panel?)

Art

[ArtF]

Richard:

Was that a 2 second test fire at 72% power?

Art

[Richard Cullin]

rebooted and tried again&nbsp; ,set&nbsp; time&nbsp; 1000mS&nbsp; power 72.7%&nbsp;
result&nbsp; 141.981 mS&nbsp; of 72%&nbsp; pwm

[Richard Cullin]

tried another pc , same result.
did notice&nbsp; the first test after a boot did for a 10 seconds test did fire led for&nbsp; 10 sec&nbsp; , all subsequent tests fail time wise