So we have a modified K40 with a beta release of the Cohesion 3d mini board and glcd at RCL. I tried doing a test cut with various power and speed settings but I ran into a problem that I couldn’t get the laser to fire with anything lower than 20%. It’s not that it just wasn’t cutting or etching, but the controller wasn’t sending any current to the laser until I set it to 30% power. I checked the gcode, and it was sending the correct s values to the laser, so I am thinking that its a setting in the config.txt or through the glcd.
Does anyone know what setting might be the cause? From the top of my head in the config.txt we have the min power set to 0.0 the max power set to 0.8 and the default power set to 0.2. On the glcd the laser pwer scale was set to 100% and the s value was at 20%.
Also for future reference, what are those two settings on the glcd? I couldn’t find any documentation on them and how they effect any other settings. I know the s value is a power setting, but does that affect the gcode, or just the test fire through the lcd, or is that the default power? Do any of these settings effect the 0.8 or 80% maximum power I have set in smoothie? @raykholo
@Ryan_Branch if you pull the center pin of the pot up to 5v the laser will be able to fire up to full power when you “Laser Switch” and issue PWM. Although that may be good for a quick test that is not a good idea for longer term.
I would just put the pot back in, you will need it. Is your PWM wired to the “IN” or the “L” pin?
After putting the pot back in push “Laser Switch” at various positions of the pot to insure that the laser is working.
Then ground the L pin and verify it fires.
Then use the “L” wiring and setup the config properly.
@Joe_Spanier Yeah, I have the new Cohesion board myself. I was having engraving problems. Last night I changed from the 3 wire PMW setup to the L wire running to Pin 2.5 (bed) on the PCB. It fixed my engraving issues. I also discovered my laser doesn’t fire on fine details if engraving at 20% power in Laserweb. It was suggested that I change the config file’s default Laser Min power to something just above 0. Which would help the laser to fire at lower values. Something like 0.05 or 0.08. Just note I haven’t tested this yet (will tonight)
Im 95% sure its connected to L from what I can see in my build pictures. Im not near the laser now.
What is gained/missed from having or not having the pot. I liked having it for on the fly fine tuning but at the same time, this is in a makerspace so its nice to have the software limiting out the tube output.
@Joe_Spanier here is the whole novel on the subject. The short story is that the laser is a consumable and wears out. So unless you like constantly adjusting software settings and config files you need a 'intensity" control.
Also there are many other variable factors from the software’s pwm to the actual power that need tweaking and if you connect to L all that tweaking can be done with that pot.
@donkjr so how is all that handled on DSP based machines and machines that never had a pot to begin with. Ive always just adjusted power levels based on results. I know enough to know its never going to be a solid number.
@Joe_Spanier Its hard to answer that question without knowing the specific design of the machine, below here are my suspicions.
What pot-less and DSP machines are you thinking of.
…There are controls similar to the pot in a machines laser control (not teh DSP) software. Maybe an adjustment pot somewhere used during alignment but not on the panel.
…Ignore the problem with laser wear-out. I have found that many design challenges, in cheaply designed machines, are just ignored and we do a lot of tweaking. Machine dies over time
…Some cheap machines do not print grey scale they just dither the image so the intensity is not as critical. The standard K40 does this.
…They assume you will keep changing the parameters of the job to account for machine changes.
… The software uses an image power DF to image its grey-scale range ignoring actual power levels (today we tweak max power). It then adds to that value an intensity bias before it sends it to the LPS as an adjusted PWM. In this case the software would have a pot object :).
We have hacked the K40, it was not really designed to print grey shade, we are mod-ing the power control to do that.
If I did this from scratch I would have digitally controlled the LPS rather than PWM controlling the LPS PWM control, which is what we are doing. The software would have a PWM setting and an intensity pot :).
There’s a bunch of people with whom keeping the pot and pulsing L seems to work better.
Hook it up to bed mosfet pin 2.5, change the laser pin in config from 2.4! to 2.5, I’ve covered this in 10 separate threads in the group.
And yeah you’ve got an early board in there so don’t take any board specific advice from anyone besides Carl or myself It should be fairly transferable though, I didn’t change the board that much between revisions.
So I tried hooking the pot back up and then ran L to the bed like you said. I changed the config file and reset the board and made sure it was saved as well. Unfortunately I wasn’t able to get the laser to fire at all after that. No matter what I had the potentiometer set at. I put it back the way it was and it still only fires at 30%. Another thing to note is that I ran some gcode that had different power levels throughout. 21%, then 22%, 23%, 24% etc. And it is exactly at 30% that the laser will fire (current meter doesn’t even budge until then). That’s why I was thinking that it was a firmware issue. I may try changing the minimum power in the config to see if that does anything.
Good news! Changing the “laser minimum power” to 0.1 from 0.0 in the config seems to have solved it! I have it hooked back up the way it was originally as well without the pot. Not 100% sure that I won’t run into another problem, but I will keep a closer eye on the cohesion 3d g+ group. Thanks!
It should be fairly transferable though, I didn’t change the board that much between revisions.