Here is a draft of the LW4 Laser Raster operation documentation as best as I understand it at this point. We can add more detailed explanations to some of the other settings as we put them to use and understand them a bit better.
Please ignore the bad page breaks and table outlines as this is destined for a webpage and is only in doc form while I flush out the information we need to include.
https://www.dropbox.com/s/7bkidv8miwpxsog/LW4-Laser-Raster.pdf?dl=0
Variable speed need to be implemented, no alternative way for now…
Reviewing the above:
I don’t understand the reason for a laser power setting. Shouldn’t the grey scale mapping always be across 100% of the available range? Perhaps I do not know what is meant by “machine settings”.
I think that there are to many places to adjust the laser power levels that conflict and override each other. I think there should only be two places. That pot on the machine (adjusts overall intensity) and the PWM from the smoothie. These should be the only two power controls IMO.
My rational is embedded in this vary complex read (sorry but its still a draft) that is a spill off of the LPS and PWM work I have done. I believe that we need to treat these controls more simply to get better consistency and simplicity while accounting for the changing power output of a gas laser.
My opinion is that much of this will be cleared up with the materials database. Many of us don’t have a POT and the laser power in the CAM operation is the equivalent of setting the boundaries for min and max power that the gradients are allowed to live in.
So this is a bit of a carry over conversation that Don and I were having. I do agree that having a wider range is a good thing and can result in better quality gray scale images, but in the end many are working with machines that are not up to the same quality as some of the mainstream production stuff. For example, my tube doesn’t even mark if my low end setting is below 13% while if I go over 80% it just creates a sooty mess. So I’ve capped my smoothie config at .8 and typically will let the machine ride at 0-100 on the CAM side and do my best to accommodate different materials with higher speed settings but again the machine has it’s limits so my only option is to dial down the max power. Without a POT that means I have to do it via software.
+Peter van der Walt Thanks I will read the reference tonight.
However, I don’t think that my perspective on power control changes because of the use of an algorithm for linear-ization nor has any impact by the type of laser used.
The basic point is that you want the software to always have the full range of control no matter what the actual power of the laser is. The lasers power diminishes over time so if the dynamic range is adjusted the same engraving will change as the laser wears.
If a material reacts differently I believe that the overall intensity (the power pot) should be adjustable but not the range of control.
Having power ranges set in the smoothie configuration and LaserWeb and the machine certainly makes the setup tweak-able but I don’ think is it doing for the imaging what folks think it is.
Then again… I seem to be on the periphery of nuts on this power control caper…
@Carl_Fisher that why I am saying that we should keep the pot. The laser is constantly changing (decreasing) and let the software operate fully within its range? Otherwise the software setup for a job has to constantly change :)?
Hrmm. I suppose it does make sense. My question is would adjusting the max in the config file accomplish the same thing or are you saying that the relative power needs to be increased. i.e. what used to be 20% maybe now needs to be 22%? If that’s the case, how does the POT accommodate this if the gcode still says 20% power?
This brings up a question I had. I do prefer to be able to go 0%-100% when rastering, but when I have a job that has both a cut and a raster I end up setting my pot for the cut and reducing the maximum on the raster so that it doesn’t burn too dark. Is it possible to add an optional pause after the raster so that the pot could be adjusted for the cut?
Sounds like a tool change command. I wonder if you manually added it to the gcode what a laser would do.
I need to explain this better, and simpler. I will think about this more and read the materials recommended looking for a better and illustrative explanation of my thinking.
Kim’s problem is a perfect example of why the software control should assume a range of operation and let the pot set the max power for any given lasers capability.
Consider that the laser max power needs to be kept at or below 18ma no matter the power output at that current. Also we have to consider that the power of the laser changes (drop) with usage. Finally, a new or different laser would have a different max power. It is desirable not to have to tweak job setting dependent on what laser is doing the cutting and engraving. The adjustment can be done one time at the device.
The pot is set at its 18ma point (max) and when engraving the range that the software uses is 0-100% of that power employing whatever step granularity is necessary for the desired engraving quality. I can see step granularity being a programmable function?
As the laser power decreases the pot is turned up to get back to the same relative darkness but the engraving settings need not change. The software still has the same range and # of steps at the new pot setting.
This way the relative intensity between steps stays the same irrespective of the max laser current setting. The pot in effect adjusts the absolute max intensity but not the relative intensity between grey shade steps.
Sooner or later the laser wears out to point the absolute intensity at full pot setting will fall below a level such that the image intensity is not acceptable but that cannot be corrected by changing the grey shade granularity. At this point the laser is worn out.
In the case of cutting; the Gcode sets an S value that represents the needed cutting power % of max.
Indeed the S value for differing materials and thicknesses should be programmable from the software as a % of max power. In the cutting case the pots position still limits the power to 18ma. Software should not use the same S value for cutting and engraving in a composite job? If the max pot setting that works for engraving is to high/low for cutting the the S value and speed of cutting should be changed in the Gcode?
The problem is that we do not have closed loop control of the laser power so something outside the software needs to be adjustable, creating a reference center-line that works without tweaking the job parameters for changing laser power conditions.
“It is desirable not to have to tweak job setting dependent on what laser is doing the cutting and engraving.”
This sounds like you’re trying to make gcode which runs on multiple machines. CNC gcode and 3D printer can’t do that; laser is no different. Even trying to make gcode that works on a single CNC machine over a long period of time only works on expensive machines. The linear motion on hobby machines wears out over time, reducing max speed and acceleration.
I agree with +Peter van der Walt and @Todd_Fleming . This pot case is almost limited to the K40’s. I understand @donkjr point but most of laser machines doesn’t have it therefore I see no need to implement something special to deal with this “external” setting.
Btw good job with the documentation. The webpages will be filled quite quickly when available.
Darn, I am once again on an island nagged by data and models of the way things might be.
I will once again venture into the darkness and see if I can satiate my curiosities until science meets reality.
When I return from the abyss to reality I will let you know what wisdom it is hiding… if any.
There is one important thing we have to consider. If we limit the laser power in software, we loose resolution (gray levels). I would always prefer an analog pot to limit the max power, if there is one, because then you have the full resolution in the software an could adjust the max power with the pot.
It depends on the machine and material if this is visible in the result, but theoretically limiting power in software to 50% for example, would reduce the number of power levels from 256 to 128.
@cprezzi heh! … that is the key point that I was doing a terrible job trying to make. Along with that consideration is the realization that the max power is constantly falling…
I don’t know how many descrete stepps the smoothie pwm has.
Probably more than 256.