He got the same behavior from two different controllers… QC issue?
1 Like
I have not upgraded the firmware, because I can’t guarantee that the machine will stay on long enough. I guess it will brick itself if the power drops out halfway through.
And just to be clear in case I haven’t explained it properly, the fault occurs whether the machine is running a job or just sitting idle. The video was taken when it was idle, not even connected to the laptop.
My (UK) workshop is around 12ºc at the moment, we have fairly high humidity so I don’t think static is a problem. It won’t hurt to run a ground to the metal frame, and I’ll try that later.
One thing that might be temperature-related is that the machine sometimes runs ok for a while (10-15 minutes) when I first switch it on after, say, overnight and after that the crashes become more frequent. At least, that’s the impression I have and I haven’t any data to back it up. To me, that indicates something like a dry solder joint or a bad connection somewhere on the board or maybe somewhere in the wiring loom.
Again, thanks for the comments, really appreciate it.
I have seen instances of the 3 things you mean.
Cold solders should be inspectable.
As per temperature actions on the machine and specially the power adapter those are variables you could test for easily. As i said bring it indoors, let power adapter accommodate to the ambient temperature and test. IF the behavior changes then you are on to something. If not, then for sure is motherboard level.
@berainlb Ah, missed that when I read through.
@Newlands Can you setup a camera so it can see the psu light and control board light and see which turns off? If the PSU light turns off along with the controller, it’s a PSU problem, if it stays lit, it might be the controller.
Also, have you tried a ghost run? Unhook the motor, laser, and limit switch cables, turn on the machine, and see if it dies. Maybe try running a project to see if the controller goes through the motions. Could be a break/short in a wire or other component.
Also, you haven’t said which laser it is, I know the 20/40W and the IR require a more powerful PSU (my IR came with a replacement brick because of this) and it might be cutting out for overload protection?
EDIT: nvm, see you’ve tried this too. I’m a bit slow this early. 
Only other thing I can think of, besides the above climate discussion, is are you controlling it via USB? Might be a fault in the USB cable, or the PC port you’re using causing a short. Try running it locally via the SD card. You can either click save gcode, rename the file tmp.gcode, and place an M17 at the beginning (edit with any text editor, or place it in the gcode header in lightburn) or use my wifi method to send it to the machine, keeping it isolated with no usb connection.
Brought the machine indoors as @gilaraujo suggests and sat down for some extended tests and recording results (in the warm!). Up until now I’ve been a bit random, chasing my latest theory, and decided I needed to be more methodical, so here goes:
-
Connected up the bare board to power, not in machine and usb not connected. Ran for 40 minutes and didn’t notice any shutdowns. Both LEDs on as expected. Measured voltages on the board as outlined in The Laser Power Drops Unexpectedly - xTool Support Center section 5, got 24v at the switch as expected, but didn’t get 24v on the laser ribbon connector header as they suggest I should (maybe these instructions are for an earlier version board, maybe I would only get that voltage if a job was running, I don’t know).
-
Powered on the complete machine, with the controller installed, laser head and stepper ribbon cables connected, USB cable not connected. On for 45 minutes, no power dropouts noted.
-
Power on to complete machine, Mac laptop connected via USB and running Lightburn. Ran my 1m 26s test job, on 0% power, four times successfully:
ok /comment - console output from job 4 of 4
MPos:0.000000,0.000000,0.000000,0.000000
Starting stream
Layer C02
M8 N1
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
M8 N4
[MSG:Pgm End]
Stream completed in 1:26 /comment - job completed, success
- A few minutes later, with no job running, the intermittent power outages started, connection to the laptop was lost:
/comment - no job running, the following is when the machine was on:
protect_flash_ret:0x105, flash id:c84017
�!���
N��!!�
����!�Xok
ok
ok
[XTool D1Pro:ver 40.31.001.01 B8]
ok
Cycled the power switch several times with no result, then removed the USB lead and had to cycle the power switch 3-4 times more before it powered up.
- Ran the test job again, failed about 50 seconds in with nothing in the console
ok /comment - machine restarted, same job started:
MPos:0.000000,0.000000,0.000000,0.000000
Starting stream
Layer C02
M8 N1
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4
Layer Outline2 Pass 3 of 4
Layer Outline2 Pass 4 of 4
Layer Outline2 Pass 1 of 4
Layer Outline2 Pass 2 of 4 - /comment - job failed at this point with no error msg
Job halted /comment - halted output from Lightburn
Stream completed in 2:43 /comment - restarted machine and reconnected to laptop serial port
protect_flash_ret:0x105, flash id:c84017
!�9��z1؝���
e���B�Xok
- Looking at the console output, and reading the post from @Skreelink made me suspect the factory supplied USB cable might be at fault, so found another cable - fortunately I had a couple of ferrites to fit at each end. I ran the test job again and repeated it 10 times successfully. That was about 30 minutes ago, and the machine has been on at idle since then with no power outages.
Have I solved the problem? I have no idea, but it seems promising. I have more tests lined up based on the suggestions made by all of you, but with this bit of possible success I’m going to do a bit of real work with it and resume tests if or when it plays up again. Thanks again, and fingers crossed…
Awesome approach and results. Hopefully that has indeed resolved it. You should test the cable for any shorts. That’s curious.
I’m any case, USB cable issues are very common.
1 Like
Hi.
Not really IMO/IME.
I didn’t reply before because I automatically assumed that @Newlands had tried with a different USB cable because I’m not the only one who has found the XTool USB cables to be most likely crap.
I unfortunately didn’t test the cable that gave me unnecessary grief right at the beginning of my laser/XTool journey about a year ago, I was just happy that everything worked after the cable swap.
I’ll see if I can still find it and test it throughly, but I may have cut it in half like I’ve done for decades with crackling audio cables.
With the exception of course that I do repair the audio cables when the gig is done, but I definitely won’t repair the USB cables.
I suspect that if tested/examined, we’ll find excessive stray capacitance or inductance rather than a plain short in those cables.
Regards,
Sam
Have you seen a cable that would cause a board to shutdown when all other downstream components were unplugged?
I’ve not seen this.
Hi.
Not shut down, but the behaviour with the supplied cable was so erratic that I wouldn’t rule that from happening out either.
As unlikely it seems, for some reason the USB cable XTool supplies with the machines is less than stellar.
While thinking about the subject (I could well be in the same situation as the OP at some point), one possibility at least in theory would be that the ground (or + & - voltage) potentials of the D1 and either (or both) of the laptops USB ports he’s using aren’t the same.
That’s fortunately very easy to diagnose by using the laptops with battery power.
I’ve ran across several SMPS powered USB audio interfaces over the decades that have been somewhat finicky about that.
None of those haven’t shut down either though, but there’s been all sorts of interference and weird behaviour in the audio signal.
Regards,
Sam
It was specifically the shutting down scenario to be curious. The other erratic behaviors would be more typical symptoms.
I think this is a potential general problem with USB but shouldn’t be cable specific. I’d expect a properly grounded system to mitigate this, however.
We’re just calling it dirt ground from now on. 
Hi.
Over the decades I’ve noticed that here’s no telling how a particular microcontroller may behave when it detects an unusual signal in its I/O port(s) or -connections.
I do agree that shutting down -or “shutting down” since without deeper diagnostics there’s no telling if the leds are just out- is somewhat drastic behaviour, but I’d suspect that it may have something to do with the HW and FW not being completely compatible with eachother.
I haven’t run into any severe problems -YET- with mine, but it does seem that xTool FW code does have a lot of room for improvement .
Just as their XCS does.
True.
It may well be a combination of issues though.
Also true, but even in the home recording studio circles, people tend to forget how important proper grounding really is.
Everyone just assumes that the system is properly grounded via cables.
SMPS powered equipment also sometimes make it very hard to do it correctly, at least without swapping the SMPS wall-/cord warts for something better.
Grounding may or may not have an effect to the 0 and 5 V data voltage potentials -or differences between them- though, and those can sometimes be real headscratchers.
Regards,
Sam
I ran the laser again for around three hours yesterday in its proper workshop setting with no problems at all.
I should have done, and I should have done some much more methodical testing from the outset, instead of making a wild stab in the dark at every wild idea I had about the cause!
I haven’t tested the old cable yet (it’s physically difficult to reach the connections inside the plug). And I have no way of checking RFI or EMI interference. I have a new, shielded usb cable ordered, along with a couple of ferrite cores.
Re the original cable - the two ferrites are not fitted right at the ends, there is about 80mm of cable between each ferrite and respective plug. And the power and usb sockets on the control board are right next to each other. I don’t know what the inside of the 24v power supply looks like (it’s sealed), but I suspect it’s not the best quality, so could be a source of noise.
Thanks for all the help folks, greatly appreciated.
I’ve now run the D1 pro for a few hours over the last few days and I’m pretty confident that it was the poor quality of the supplied usb cable at fault. I’ve marked the post from @Skreelink as the solution, but everyone who helped provided valuable information to help me - many thanks.
1 Like
Glad to hear the problem seems to be fixed and it seemed relatively harmless. Hopefully this thread will serve to help others who face similar issues. 
1 Like
This topic was automatically closed after 30 days. New replies are no longer allowed.