Wow thank you @JohnJohn for this thorough explanation. I’ll unpack this and see what I find when I get home.
Thank you!
1 Like
Ha makes sense. Last night I checked every screw and retighten the belts and still am getting some variance on the right side of my etches. Will snap some pics when I get home of what’s going on.
I think it is supposed to be $RST=* instead.
While $RST=* resets all controller codes, $RST=$ only resets the machine settings dealing with motion - ‘the $$ settings’. The test as described will show if any of the $$ settings were changed from defaults and show whether the machine is intended to Auto-home or not.
The coordinate system settings in the controller are reset with $RST=# and all settings are reset with $RST=* . In this case, I don’t want to push a full reset yet.
I don’t want to risk blowing out a workspace offset for a CNC machine until I figure out the homing.
If you’re interested, the $RST variants are a little bit past halfway down this page:
3 Likes
Heres what my console is saying.
$$
$0=10
$1=25
$2=0
$3=1
$4=0
$5=1
$6=0
$10=1
$11=0.010
$12=0.002
$13=0
$20=0
$21=0
$22=1
$23=3
$24=200.000
$25=2000.000
$26=250.000
$27=3.000
$28=1000.000
$30=1000.000
$31=0.000
$32=1
$38=1
$39=1
$40=250
$41=40
$42=1300
$43=5
$44=1
$45=1
$50=1
$100=80.000
$101=80.000
$102=250.000
$103=100.000
$104=100.000
$105=100.000
$110=6000.000
$111=6000.000
$112=1000.000
$113=1000.000
$114=1000.000
$115=1000.000
$120=500.000
$121=500.000
$122=10.000
$123=200.000
$124=200.000
$125=200.000
$130=400.000
$131=400.000
$132=200.000
$133=300.000
$134=300.000
$135=300.000
ok
$RST=$
[MSG:Settings reset done]
[MSG:Position offsets reset done]
ok
$$
$0=10
$1=25
$2=0
$3=1
$4=0
$5=1
$6=0
$10=1
$11=0.010
$12=0.002
$13=0
$20=0
$21=1
$22=1
$23=3
$24=25.000
$25=500.000
$26=250.000
$27=1.000
$28=1000.000
$30=1000.000
$31=0.000
$32=1
$38=1
$39=1
$40=100
$41=40
$42=1700
$43=5
$44=1
$45=1
$50=1
$100=80.000
$101=80.000
$102=250.000
$103=100.000
$104=100.000
$105=100.000
$110=6000.000
$111=6000.000
$112=1000.000
$113=1000.000
$114=1000.000
$115=1000.000
$120=500.000
$121=500.000
$122=500.000
$123=200.000
$124=200.000
$125=200.000
$130=400.000
$131=400.000
$132=200.000
$133=300.000
$134=300.000
$135=300.000
ok
$i
[PRODUCER:MKS DLC32]
[V2.20(8M.H35.20230626)]
[VER:1.1h.2023062602:]
[OPT:VPH,63,512]
Target buffer size found
[MSG:Using machine:MKS DLC32]
[MSG:Mode=STA:SSID=:Status=Not connected:IP=0.0.0.0:MAC=FC-B4-67-87-97-14]
ok
I cant really say what I did to get the faults to clear as its working for the most part now as of a few days ago.
My current issue can be seen above in what Ed replied to about potential mechanical issues. I retightened every belt and checks for loose screws everywhere. I also tried starting prints from different locations last night and got the same results. The left side of the etch is always clear and crisp, the right side tends to get distorted and have issues. Also its probably worth noting im not running these etches fast at least I dont think. most of these are set to 1000mm/min.
Betcha didn’t take the laser head apart to see what’s rattling around inside … 
It’s less a matter of screws being tight as all the parts being secure. Check for stripped threads, overly large holes, bent brackets tight this way & loose that way, eccentric wheel adjustments with slop, stuff like that.
@parsec pointed to this earlier, but it’s worth review after you think you’ve checked everything:
So many parts, any of which can be loose / broken / bent / misadjusted!
The “Stagg” picture is the same file you showed earlier. Run that same job at the same speed on cardboard and show us what it looks like now.
1 Like
For arguments sake, turn your scan angle 90 degrees. Try again and compare to see if there is a difference in error placement on the design.
1 Like
Here’s my findings
After this crashed during Ed’s test print I power cycled he the system and software to do @JimNM print idea and now I can’t communicate with it again 
Console keeps spewing this and I can’t talk to it even after power cycling a second time lol this things an abortion.
The laser head is catastrophically loose: fix it before doing anything else. Why would you expect precise results from jank like that?
The next time you say this I’m gonna demand proof: “I checked every screw”.
I won’t take credit for it, but typically a crappy USB cable or damaged USB connector will do that, regardless of the machine’s manufacturer. Throw out the cable you got with the laser and install a known-good cable, ideally with ferrite cores on the ends to make it cost enough to ensure good QC, then verify wiggling the connectors while running a job doesn’t cause any adverse effects.
2 Likes
Oh man! There should be ZERO wobble anywhere in the laser machine. Belt movements should be smooth with no backlash. The laser head is not even attached, in the video. It is merely in the neighborhood.
1 Like
Nothing I can do about it there’s no slots to move the wheel closer. The screws on it can’t be tightened more. Thank for all your help guys but this thing is trash and I’m beyond over it. Longers stuff is garbage so I’m Returning it now. Probably going to get a xTool D1 Pro instead. Any good ones under $1K you can recommend ?
Think of this as a learning experience: take the head apart and find out where that slop lives. Once you find it, it’ll be easy to fix with ordinary stuff you have in the shop.
IMO, all of the low-end diode lasers are basically low-QC parts kits that may generally resemble a decent lasers after you reassemble them with considerable attention to detail. Do not expect any other brand to be better in any regard; they all ship too many duds to be coincidence.
IMO, based on what I read around here, XTool is the perfect master of incompatible firmware hackery. If you don’t want to fix hardware, you can’t fix firmware.
4 Likes
Okay something over $1000? Surely there’s something out there with a reputation for quality. If not @ednisley I have a business venture in mind and will need a SME . Grabbed a beer and put my mechanical engineer hat on. No worries some electrical tape and she’ll be right. Let see how some test prints go now.
For a while, anyhow. Works for me.
What’s with the two flat-head screws standing proud of the surface on that thing?
Are they supposed to be in a countersink that just didn’t happen to make it into the final design?
Are they supposed to expand a wedge against the sides of their slots?
Or are they just supposed to somehow not wobble?
Like I said: consider it a parts kit.
Yeah, I found this out earlier when I had to flush a G54 I did not set. Somebody should write a GRBL For Dummies book.
3 Likes
You’re getting there… this is great…
Here’s what I see, You can click to blow the picture up.
Before the Reset, the first Modified Setting $21=0. This turns off the hard limits. With Soft limits off $20=0 (before and after) and $22=1, which allows homing.
Homing without limits could make the engraver collide with the rails.
Homing Feed $24 Before reset 200 mm/min : $24 After reset 25 mm/min
Homing Seek $25 Before reset 2000 mm/min : $25 After reset 500 mm/min
Homing switch pull-off. $27 Before reset = 3 mm : $27 After reset = 1mm
This shows a great deal of confidence in the quality of the switch.
$27=3 is a safer setting.
Ok, now we have to depart from Regular GRBL so we need to know more about the controller. I’m glad I asked.
$i
[PRODUCER:MKS DLC32]
[V2.20(8M.H35.20230626)]
[VER:1.1h.2023062602:]
[OPT:VPH,63,512]
Target buffer size found
[MSG:Using machine:MKS DLC32]
[MSG:Mode=STA:SSID=:Status=Not connected:IP=0.0.0.0:MAC=FC-B4-67-87-97-14]
ok
This is an MKS-DLC32. H35 is the 3.5" screen. It’s a great controller. It’s newer hardware and has additional settings:
$40
$42
You can read more about the $$ settings here:
I’ll have to go dig through GRBL-stm to discover what $40 and $42 do.
When the controller gets confused it can do some pretty wild stuff (but not any engraving). Do you have the files or a little SD card that shipped with the controller?
Step 1 is Unplug all power / USB cables for about 5 minutes (With No SD card installed) then try again.
This could be custom position feedback or a crash report of some kind. If you can repeat this behavior scroll back in that window and have a look. You could have a strange error message like “Guru Meditation Error” - I wish I was joking.
LightBurn may also be having a fascinating conversation with a pair of headphones or a mouse… scrolling back should be informative.
Interesting… I’ll have to revisit this if I have similar issues with my xTools D1 Pro, after over a week of banging my head on the table I decided to return the Longer Ray5 while I still could. Ed going to hate me for buying this lol but I didnt want a new laser with the same form factor and this was under $1k with a rotary so I pulled the trigger. Hopefully this one doesnt have as many issues. In the event it does what brand do you like in that price range?
Not me!
Given that it’s an xTool, though, I’m gonna run the “Not My Problem” algorithm, because nearly everything I think I know about those is wrong. 
1 Like
You may have just gone from not bad to worse.
1 Like