Laser stops and Lightburn gives integer error

When I try to engrave a simple word, the job aborts and I get an error in the console about a line in the GRBL code needing to be an integer:
image
I tried a new USB cable but it still happens.

Here is my gcode:
Trinitite a.txt (25.9 KB)

Try an online G-code validator to identify errors, or try using a different G-code sender.

And additionally, I recommend saving G-code files in plain text (.txt) format.

There’s nothing wrong with the G-Code, so this is almost certainly a USB glitch that’s very hard to track down.

Changing to a known-good / high-quality USB cable sometimes helps; ones with ferrite slugs on each end may have a higher likelihood of improving the situation.

If there’s a USB hub involved, eliminate it and run the cable directly from PC to laser controller.

If there’s a USB extension cable involved, it must be an active cable, not a simple extension. If at all possible, use a single good cable.

USB 3 ports seem to work better, except when they don’t work at all.

Fiddling with settings / drivers / configurations / whatever will not change the situation and will cause more problems by confusing the symptoms.

I ran it a couple of times, no errors, pix attached.

I’m using the original cable supplied with the equipment. I switched to another cable and it failed. I tried another USB port on the laptop and it failed too. I switched to another laptop. That failed too but got further then worked twice after that. It’s a single word so I don’t have a lot of confidence it will last but I agree it seems to be a communication issue.

Which is guaranteed to be the cheapest piece of ahem available that doesn’t disintegrate when touched. :grin:

Swapping the equipment helps rule out a bunch of possibilities, but it all boils down to using USB for an industrial process.

Try running the laptop from battery power (and without an external display) to eliminate weird glitches that can happen when two devices have a big loop through the AC wall outlet. It’s a low probability thing, but …

I got a new USB cable with gold connectors and a ferrite plug on each end but that didn’t work. The problem has gotten worse since I’m now having trouble even connecting and when I do I get errors. I contacted FoxAlien and after showing there were errors when using Candle too, they said it sounded like a bad control board and are shipping me a new one. Hopefully that will fix everyint and won’t be too much trouble to install. I’ll let you know how it works out.

1 Like

On the upside, you now have a spare: whichever one doesn’t work with the new controller. :grin:

Unfortunately, finding the cause of intermittent problems is hard, even in a lab with test equipment. Sometimes the job devolves to replacing bits & pieces until you hit the right one, which is unsatisfactory at best and expensive at worst.

Let us know how the new controller works out …

wait… was it suposed to not desintegrate on touch? :imp:

If USB glitches are notoriously difficult to track down, how can you be so certain that the issue lies with the USB connection and not with the laser controller or the G-Code itself?

The number of times folks have fixed intermittent communication problems (which is what your evidence implies) by replacing the OEM cable with something a bit higher on the food chain, plus the relatively low cost of even a brand-name cable, makes “replace the cable” a good first guess.

Getting rid of USB hubs and unpowered extensions is just good debugging practice.

While there have been occasional dud controllers (even dud replacement controllers), that’s unlikely. Replacing the controller requires more willingness to dig into the guts of the machine than most folks are willing to deploy on a random stranger’s recommendation, plus it’s relatively expensive, so “replace the controller” is not to be suggested lightly.

Aaaand, having been reading (and, worse, writing) G-Code for a couple of decades, it’s easy enough to decide a particular program is syntactically correct. :grin:

Call it an educated guess and you won’t be too far wrong …

Yeah, $10 for better/spare cable is no problem to narrow down the problems. I’m happy FoxAlien is sending the controller free of charge. I have had similar issues with my astrophotography hobby. I also replaced the controller on my 3D printer so hopefully this won’t be any more difficult.

1 Like

One can also import GCode into LightBurn for validation.
It won’t rebuild the original file and it will lose the speed/power settings but the Import will represent the graphics correctly and should reveal missing data.

There’s also a Plugin for Notepad++ that I’ve heard good things about.

Generally the key identifiers of the source of the issue are how the issue presents and repeats.

Presentation can mean there’s an error message, sometimes the message is absent. Sometimes the issue is strictly mechanical - which engravers rarely detect and manage.

Error messages inform about 90% of the issues, the error messages are most often the result of unfortunately chosen settings - most often defaults. These usually share a comment about geometry or limits. A nice clean error message is more daunting than it is unfriendly or unhelpful. People get spooked by them, but they’re a great ally and a great tool. It’s precisely the reason the Error Message exists.

Sometimes communications are lost and the controller just seems to also get lost, seems to begin ignoring the GCode or becomes unresponsive. This often happens after some time, but the elapsed time is rarely identical from attempt to attempt. These are usually communications issues.

Errors in generated GCode fail identically. These errors can be brought on by differences in controllers.

Inadvertently selecting iLaser as a device profile will cause most conventional GRBL devices to throw an error message at Line 0.

There is a Device setting called “Enable GCode Clustering”. If a GRBL controller doesn’t support GCode Clustering, the controller will consistently throw an error message most often in Line 4 or Line 5. The controller will stop and demand a fix.

Code that’s damaged on it’s way to the chip in the controller generates error messages at different times and at different line numbers. Generally the error messages are related to the grammar or syntax of the GCode and not related to geometry.

There’s a Troubleshooting 101 post in this… I’m feeling it.

Thanks for asking!

I received and installed the new control board. It was a little different so I couldn’t attach the fan the same way to the heat sink and the dip switches were set differently but I left them alone and it works fine. It’s easier to connect and works more reliably in addition to finishing the jobs now. Thanks for your help. FoxAlien took good care of me too.

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.