Can’t access the video…
Whoops, forgot to set the permission. Try it now.
And it looks like I could even see the firing of a speckle/dot. Doesn’t tell us anything but is cool to watch. Most everything is cool in slow-mo.
It is pretty cool. I was just going over the optics again and wanted to disprove that theory. it fires when the nozzle reaches the very end of the overscan, so not reflection. I can’t wait to forget this problem, these things bother me more than they should.
2 Likes
You might find this interesting: https://www.youtube.com/watch?v=PHU7V8tCS_c
Many, including myself, quickly replaced the M2Nano board with a SmoothieWare compatible board and use PWM into the LPS-L input while leaving LPS-IN controlled by the POT. And in the 5 years the K40 machines are made by far more than the couple of manufacturers originally making and selling to resellers so now other controllers are often used and I’ve even seen Ruida controllers on K40 machines. Russ Sadler did a whole video series on upgrading a K40 into what he calls a K40 Xtreeem laser cutter including 2 videos on installing Ruida and setting it up.
Soooo I’m would love to try the LPS adjustment but no idea how? 
@Bklynghost @jkwilborn a power supply guru had a look at the video and while I’ve not confirmed it but he said he stepped through the video and the laser head moves specifically to the spot that the dots are fired at, fires and then continues. This type of thing looks like firmware as opposed to LPS since the LPS would not effect motion.
It could be, but this is a problem with pretty much the second machine I’ve got.
Before this one, I had a K40 with a Cohesion board that worked well.
I don’t remember if I started getting this problem before, or after a tube change, but here is a rough timeline of my “journey”
- K40 with C3d board running well ~2.5 years
- Dots start appearing
- Tube change
- LPS change (generic eBay, but better than stock)
- Burned C3d board after messing with settings too much to fix the problem
- Replaced burned C3d with RDC6442S-B (EC)
- Gave a shot to another LPS from Light Object (40-50w model)
- Ran this setup for about a year with the dots
- Bought a new 50w machine
- Switched the stock knockoff controller to the 6442
And now we’re here.
Now this timeline would make me think that maybe it could be the controller, but a few days ago I did connect the original controller that came with the machine and there was no change.
My current FW version is V8.01.65
I did find a FW version .67 on Ruida website and I’d be willing to upgrade to see if it makes a difference, but I want to make sure that it is the right one for my controller since mine is a “-B (EC)” and there is nothing on the website about that particular model.
Maybe it is something in the settings after all, or this is some sort of a laser covid that spreads to other machines.
If yours is a 6442, I don’t think the fw applies…
The other thing to check (you too @Bklynghost & @BaronPork ) which I’m not sure has been discussed here is that it may help to include or upgrade your optical isolation between the controller and the LPS - eg. here is the optocoupler circuit which sits between my RDC6442G and HY-T60, it’s easily to replace or remove:
1 Like
Oh man, that’s a great idea. You wouldn’t happen to know if Thunderlaser sells those? I did a quick search and didn’t get any results for ready circuits, or much information on the design. I might just try to copy yours.
Not sure, they probably wont be interested in selling you one if you don’t have one of their machines, but no harm in asking - you can see a little more info about them here
You might be able to find a generic breakout board for the digital 6N137 OC or one of these - like this or this or from here but get the high speed 5V model.
I wonder if the low voltage for the PWM signal is being lifted by LPS ground noise - this may cause false highs and trigger the dots, the optocoupler would solve that.
1 Like
Mine definitely doesn’t have opto-isolation. As basic as it comes - power filter, lps, 24v psu, 2 drivers, 2 motors, dumb Z motor (AC mains), controller (6442s EC(B?)), CDWG 80W tube and nothing else.
I did get a spark to check for proper machine grounding at the socket though.
You could enlighten me on how this is even beneficial. You are, at minimum, complicating something that was designed as, and is a simple ‘wire’ interface. It’s working fine on a ‘ton’ of these machines.
Is this for an emi/rfi type fix?
At this point I’m trying every single fix before shelling out over $300+ on a new tube and LPS that may not even fix the problem.
I will grab the 4 channel high speed AL-ZARD board.
I am a bit confused about the “cable length” voltage option. Should I get the 1.8V-5V or 5V-5V?
There’s been a question as to if this occurs with vectors too so I wonder if one of you who have this problem can generate a bunch of short vectors such that it’s like the vertical lines have been spread out vertically so there’s a 1mm gap between vertical vector lines?
I’m not quite sure what about what you mean. Aren’t shapes created in LB technically vectors? I’ve generated the line grid that way with the square tool. I was actually going to rasterize it and see if there is a difference, but I think I’ve done that before.
when you create a box shape or circle shape in LightBurn they are not vectors that I know of since you can’t edit their points. You can set LightBurn to either Line or Fill and it will do the operation. But to make them Vectors you’d need to select the box object, convert it to a Path and then it’s a pathed vector…
Maybe try looking at this file and see how it’s small lines, grouped so that it will run one horizontal line of vectors after the next.
dot-test-vectors.lbrn2 (24.3 KB)