Connection problem or what?

Hi guys.

I installed marlin 2 on a 3d printer where I put a 5w laser module,The board is an arduino mega with ramps 1.4.

I connect to lightburn without any apparent problems… The problems come when I have to start a job or when I have to start a frame.

So,the axes all move smoothly and the response to commands is immediate,but when I press the “frame” button the console reports the following:

Starting stream
Stream completed in 0:00

and nothing happens, so I press ‘frame’ again and the console reports:

Starting stream
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
echo:busy: processing
Stream completed in 1:11

And yes,after more than a minute manages to make the frame.

But the big problem is that the same thing happens when I start the job, it takes a very long time to process every single line of gcode,performs a movement and then nothing seems to happen for 40-50 seconds, then he performs another movement and so on… One very curious thing is that if I save the gcode to an SDcard and start the job directly from the 3D printer SDcard reader, nothing changes, it seems to generate a badly made gcode, but I open the gcode with an editor and it looks like a normal gcode (in this particular case it is the focus test offered by lightburn itself)

No problem with LaserWeb, everything works perfectly




Is 250,000 baud rate indeed the best baud rate to connect to the board? That’s not atypical but worth validating.

Many Marlin implementations seem to work better with “synchronous” Transfer mode vs buffered. You may want to try that.

Other than that you may want to consider dropping Marlin and putting GRBL on the board, especially if can dedicate the machine to lasering.

Thx.
Yes I tried changing the mode to ‘synchronous’ but nothing changes, and when I compiled the firmware the baud rate was set to 250000.
I used marlin because it’s something I already know,I will switch to GRBL if it works,but I don’t know it well and I need to find a tutorial on arduino and ramps for GRBL

In any case the performance should be better than what you’re seeing.

What is the speed setting for jogging in Move window? That should be the framing speed. Is it possible that the setting is just slow?

6000 mm/min seems excessively fast but at least shouldn’t be causing what you’re seeing.

If you issue single g-code commands to the controller do they complete without fuss?

Commands such as G1 Zxx or G1 Yxx work fine, also commands to set offsets such as M206 X -4 and M500 work fine

I’m confused then, a square frame operation should only be ~4 gcode commands. At 6000 mm/min that should go quickly.

Where exactly is it going slow?

Try enabling “Show all” in Console and then run a frame. That will show you the code.

as I wrote the first time I press ‘frame’ nothing happens, the second … see screenshot and use zoom :slight_smile: The machine is not slow, it is the commands that are slow to arrive, when the command arrives the machine executes it without problems.In this case, after more than a minute, the command has still not arrived (it has now been 4 minutes and it is still there)

when i press for one time “frame”


after 1 min he executed it
Interestingly, if the frame succeeds (even if after 2 min) the subsequent ‘frames are almost instantaneous, as if it had unlocked’.

Not sure what’s going on. But I’d focus on Marlin buffer or other interface configurations.

However, if you can manage it, GRBL Is a far better platform currently for laser operations.

Thx,yes I am already working to compile GRBL,but GRBL don’t know if it supports a display with an SD reader.my machine is far away from PCs

There are several GRBL implementations that will work on the Mega. You may want to explore different ones. I’m not specifically aware of any that support offline loading from SD Card.

Unfortunately, I have found that the problem is not the board or anything else, but simply lightburn.
The problem is solved by reverting to version 1.2.04, I don’t know what differences there are between this old version and the new one, but I certainly regret having to use outdated software.
When I have the resources I will buy an offline controller for GRBL, in the meantime I will make do with this version

Can you take a screenshot of Edit->Device Settings in 1.2.04?


go changed the baud rate to 115200 because I recompiled the firmware with the value 115200. Now the frame works but there are still other problems, the focus test runs at awkward speeds, in jerks. On laserweb everything is OK.

https://streamable.com/wdttij

If framing is now working that’s a good sign. At least the communication issues seem to be resolved.

Can you specify the issues?

One thing to try is to enable “Fast whitespace scan” in Edit->Device Settings and set a reasonable speed. In Marlin, this value is used for traversal moves if I’m remembering correctly.


This is a short video of speed work https://streamable.com/wdttij

this is the result, I do not know why the lines are not parallel as in the preview

Yes my friend,this work.now everything moves more smoothly,thank you very much

This could be a mechanical issue or a situation where the machine is losing steps.

Can you push “Save gcode” in the focus test, save with .txt extension, and then upload here?