m Hi team. Hereās my description of a problem and my suspicion that the Superlongboard is the problem.
Iām open for any guidance
Cheers
David
Iām trying to make hangboards for a local climbing sports shopā¦
This is serious for me as I canāt use the router to finish any hangboards that are usable.
My guess is that is memory handling in the SLB.
Somewhere the gcode is being misinterpreted
Examples
About 1 hour into a job the router suddenly digs in more than the gcode says. Once down at that level, it thinks that is the new level and continues the job with this error. This job was started 5 times during the roughing toolpath because of this problem, and, of cour
The router fails to retract as directed by the gcode, thereby creating a groove through the material.
The router retracts too much, and spends the rest of the time waving around in the air!
It never happens until near the end of the first hour of a job
To me (and this is probably from my 30 years running a software company) it seems like a memory handling error somewhere in the SLB.
Its not Gsender. Iāve been able to replicate it with Open Build
Thinking that it might be interference from the VFD cable, I wrapped the cable in an aluminum shield. The fault still occurs after this.
Connection with SLB is via network
Hereās a small amount of Gcode which is annotated to show the activity in the couple of seconds before the error. (at 57 minutes)
5:55:25] [ Y32.240 ] ok Turning At the end of the row
[15:55:25] [ X669.473Z29.500 ] ok Surface0
A gap of proceeding along path
[15:55:44] [ X180.104Z29.500 ] ok back to the top
[15:55:44] [ X188.795 ] ok
[15:55:44] [ X189.322Z29.400 ] ok
[15:55:44] [ X189.585Z29.166 ] ok
[15:55:44] [ X189.849Z28.802 ] ok
[15:55:44] [ X190.112Z28.157 ] ok
[15:55:44] [ X190.639Z20.440 ] ok
[15:55:44] [ X190.902Z20.000 ] ok bottom of hangboard pocket
[15:55:44] [ X223.299 ] ok
[15:55:44] [ X223.562Z20.139 ] ok
[15:55:44] [ X223.826Z22.885 ] ok
[15:55:44] [ X224.089Z25.857 ] ok my guess thi is the command misinterpreted
[15:55:44] [ X224.352Z27.971 ] ok
[15:55:45] [ X224.616Z28.747 ] ok
[15:55:45] [ X224.879Z29.166 ] ok
[15:55:45] [ X225.142Z29.400 ] ok
[15:55:45] [ X225.669Z29.500 ] ok Z axis all at the new wrong level
[15:55:45] [ X314.957 ] ok
[15:55:45] [ X315.484Z29.400 ] ok
[15:55:45] [ X315.747Z29.166 ] ok
[15:55:45] [ X316.010Z28.784 ] ok
[15:55:47] [ X316.274Z27.992 ] ok
[15:55:47] [ X316.801Z18.006 ] ok
[15:55:47] [ X317.064Z17.500 ] ok
[15:55:47] [ X412.673 ] ok
[15:55:47] [ X412.936Z18.005 ] ok
[15:55:47] [ X413.463Z27.992 ] ok
[15:55:47] [ X413.726Z28.784 ] ok
[15:55:47] [ X413.990Z29.162 ] ok
[15:55:48] [ ALARM ] ALARM: 10 - EStop asserted. Clear and reset [ X414.253Z29.378 ]
[15:55:48] ALARM: 10 - EStop asserted. Clear and reset [ X414.253Z29.378 ]
[15:55:48] [ X414.253Z29.378 ] ALARM:10
[15:55:48] [ X414.253Z29.378 ] [MSG:Emergency stop - clear, then reset to continue]
[15:56:56] [ [clear alarm] ] Operator clicked Clear Alarm: Cleared Lockout and Emptied Queue
[15:56:56] GrblHAL 1.1f [ā$ā or ā$HELPā for help]
[15:56:56] [MSG:ā$Hā|ā$Xā to unlock]
[15:56:56] [MSG:Caution: Unlocked]
[15:56:56] [ JOB COMPLETE ] Job completed in 00h57m
[15:56:56] ok
[15:57:04] [ $J=G91 G21 Z1000 F2280 ] o
[15:57:09] [ $J=G91 G21 X-1000 F2280 ] ok
[15:57:21] [ $J=G91 G21 X-1000 F2280 ] ok
[16:13:42] [ $J=G91 G21 Y-1000 F2280 ] ok
[16:13:52] [ disconnect ] PORT INFO: Port closed
Also attached is a photo of a situation where the router suddenly went down into the material further than the gcode said.[
@david_kelly When you say that the connection between the SLB and your CNC machine is āvia networkā, do you mean that you are connected to your network via router or are you connected directly via a dedicated NIC?
Thanks for the prompt responses. I appreciate your care.
The machine is a Version 1 Longmill 4x4 ft
No. I havenāt tried an air cut. Without astock, I doubt Iād be able to notice when the error. This is because the machine doesnāt stop or anything. It just cheerily carries on following the Gcode, but at the wrong level.
Thinking about it again. Its happening only on the Z axis. Unlikely to be SLB memory handling?
Hereās the relevant Gcode, annotated., including hitting the stop button.
5:55:25] [ Y32.240 ] ok Turning At the end of the row
[15:55:25] [ X669.473Z29.500 ] ok Surface0
A gap of irrelevant code proceeding along the pass
[15:55:44] [ X180.104Z29.500 ] ok back to the top
[15:55:44] [ X188.795 ] ok
[15:55:44] [ X189.322Z29.400 ] ok
[15:55:44] [ X189.585Z29.166 ] ok
[15:55:44] [ X189.849Z28.802 ] ok
[15:55:44] [ X190.112Z28.157 ] ok
[15:55:44] [ X190.639Z20.440 ] ok
[15:55:44] [ X190.902Z20.000 ] ok bottom of hangboard pocket
[15:55:44] [ X223.299 ] ok
[15:55:44] [ X223.562Z20.139 ] ok
[15:55:44] [ X223.826Z22.885 ] ok
[15:55:44] [ X224.089Z25.857 ] ok my guess thi is the command misinterpreted
[15:55:44] [ X224.352Z27.971 ] ok
[15:55:45] [ X224.616Z28.747 ] ok
[15:55:45] [ X224.879Z29.166 ] ok
[15:55:45] [ X225.142Z29.400 ] ok
[15:55:45] [ X225.669Z29.500 ] ok Z axis all at the new wrong level
[15:55:45] [ X314.957 ] ok
[15:55:45] [ X315.484Z29.400 ] ok
[15:55:45] [ X315.747Z29.166 ] ok
[15:55:45] [ X316.010Z28.784 ] ok
[15:55:47] [ X316.274Z27.992 ] ok
[15:55:47] [ X316.801Z18.006 ] ok
[15:55:47] [ X317.064Z17.500 ] ok
[15:55:47] [ X412.673 ] ok
[15:55:47] [ X412.936Z18.005 ] ok
[15:55:47] [ X413.463Z27.992 ] ok
[15:55:47] [ X413.726Z28.784 ] ok
[15:55:47] [ X413.990Z29.162 ] ok
[15:55:48] [ ALARM ] ALARM: 10 - EStop asserted. Clear and reset [ X414.253Z29.378 ]
[15:55:48] ALARM: 10 - EStop asserted. Clear and reset [ X414.253Z29.378 ]
[15:55:48] [ X414.253Z29.378 ] ALARM:10
[15:55:48] [ X414.253Z29.378 ] [MSG:Emergency stop - clear, then reset to continue]
[15:56:56] [ [clear alarm] ] Operator clicked Clear Alarm: Cleared Lockout and Emptied Queue
[15:56:56] GrblHAL 1.1f [ā$ā or ā$HELPā for help]
[15:56:56] [MSG:ā$Hā|ā$Xā to unlock]
[15:56:56] [MSG:Caution: Unlocked]
[15:56:56] [ JOB COMPLETE ] Job completed in 00h57m
[15:56:56] ok
[15:57:04] [ $J=G91 G21 Z1000 F2280 ] o
[15:57:09] [ $J=G91 G21 X-1000 F2280 ] ok
[15:57:21] [ $J=G91 G21 X-1000 F2280 ] ok
[16:13:42] [ $J=G91 G21 Y-1000 F2280 ] ok
[16:13:52] [ disconnect ] PORT INFO: Port closed
The PC and the SLB are connected via a 3 metre
CAT5 cable using IP address 192.168.5.1 from my memory
Iāve no idea why the command is ignored. Yes there is a gash created after the error.
Yes, Iāve checked the belt. Its not slipping. I havenāt checked the pulleys. But if either of the pulleys was slipping, the bit would not be driven into the material. Thanks for your careful analysis and response.
David
What goes wrong at that point in the code is that the router misinterpreted, in this case , failing to obey the Gcode command and fails to retract to the correct Z position. Other times, it can go further into the material than the Gcode command stipulates .
Belts donāt slip unless there is significant damage.
What endmill are you using? Based on the fact that you are cutting several pockets, I might guess that you are using an upcut endmill. If anything is loose (pulley, antibacklash nut, spindle, endmill, etc) the upcut endmill can pull itself into the stock. If it is a loose pulley set screw or issue with antibacklash nut, it can cause inconsistent motion. It could fail to retract the correct amount which will cause a deeper than expected cut on the next plunge.
If you jog your Z axis up and down several times through most of its range, does it end up where it should?
I donāt want to say thereās no way it could be an issue with the controller, but the SLB is based on well-established firmware running on a well established MCU. Iāve been around hobby CNC communities since well before Sienci Labs was around, and the issue youāre seeming to have is usually a hardware one.
H Neil,
Thank you so much, I really appreciate you taking the time to respond to this,
All the conditions you describe here exist. I had the Z axis carrier off, and it was stiff getting it back on to the antibacklash nut, and Im usinng a 1/4 inch upcut. Im also drilling deep pcket with quite a bit of travel each time, havenāt seen this before on shallower projects.
Cheers
David
It could be that itās losing a tiny bit per retract and at that point a threshold is reached that causes bigger issues.
When you get a chance, try the jogging up and down and evaluate if you have an issue without the cutting forces involved.
Sometimes the machines just need a little tune up and tightening.
@david_kelly FWIW, I would try @NeilFerreri 's advice of trying an air cut.
1 Put a piece of material on your spoilboard.
2 With an end mill in the router, set Z0 to the surface of that material.
3 Remove the material.
4 Run a toolpath.
5 At the end of the toolpath, return to Z0.
6 Replace the material.
Thank you all so much. Everything that has been said is extremely relevant to the situation.
First, my apologies to Siencsi Labs for impugning the SLB.
Thanks Neil, for your initial investigation and suggestions. You were right about the Z axis not being smooth. It was getting stuck in the linear rails. Some graphite grease helped here.
Thank you Grant for your suggestion, and your explanation when I clearly didnāt understand what yo were describing
Thanks Jay (on Facebook) for your suggestion about keeping the motors active. I didnāt know that this stuff at all. ( incidentally, the code i $1=255) Also above the Z axis bearings itās a second hand Longmill. Despite th grasig, thereās still a slolight grinding sound on the way down. This could be the bearings. Ill order some. Iām always a bit reluctant to order small parts because its $50 minimum freight out here in New Zealand.
Thank you Elnore and Steven (faceu book) about static and grounding the SLB. Iām not sure how to do this. But Iāll ask SIENCI.
Thanks also to Joshua for your input too. I am usin a LAN cable.
@david_kelly Iām glad that you seem to have tracked down the root of the problem.
One thing though. $1=255 should not be used with the SLB. Jay on Facebook may not have known you were running the SLB. His information is good for the LB. Look here for the technical doc information
Thanks everyone. I set the $37 command etc. as you said Grant, made sure everything Z -wise was moving freely., and the job was running perfectly at the right height for about 45 minutes I guess. Then the error occured spontaneously. relevant Gcode below, including the last 50 lines from Gsender.
Y31.640 Turning around at the end
A section of code thatās irrelevant
X538.044
X538.307Z23.154
X538.571Z27.746
X538.834Z28.700
X539.098Z29.166 Bottom of previous pocket.
X539.361 Across the bottom
X539.624Z29.310 Beginning of up the other side ALL BEING MILLED AT THE CORRECT LEVEL
X539.888Z29.403
X540.414Z29.500
X550.423Z29.494
X550.687Z29.457
X550.950Z29.400
X551.213Z29.310
X551.477Z29.166
X551.740Z29.155
X552.003Z28.578
X552.267Z27.110
X552.530Z22.386 IGNORED COMMAND
X553.057Z17.500 End of āup the other sideā Meant be āAt the top nowā : Meant to be back up to this height but didnāt come up far enough by about 5mm
X564.646 It now thinks itāsgoing across the top, but itās cutting a small trench between the pockets
X564.909Z19.872 Down the next pocket:also 5mm too deep- now digging a 5mm unwanted trench in the bottom of the pocket.
X565.173Z22.889
X565.436Z27.475
X565.699Z28.679
X565.963Z29.166
X566.226
X566.490Z29.310
X566.753Z29.400
X567.280Z29.500fileā¦ see the last 50 lines from the Gsender Console below
Gsender console output: I was a bit slow getting to the STOP Button, So I canāt see where the error occurred.
ok
X564.909Z19.872
ok
X565.173Z22.889
ok
X565.436Z27.475
ok
X565.699Z28.679
ok
ok
X565.963Z29.166
X566.226
ok
X566.490Z29.310
ok
X566.753Z29.400
ok
X567.280Z29.500
ALARM:10 (EStop asserted. Clear and reset)
[MSG:Emergency stop - clear, then reset to continue]
\x18
GrblHAL 1.1f [ā$ā or ā$HELPā for help]
[MSG:ā$Hā|ā$Xā to unlock]
ok
$X
ok
ok
ok
0 - SLB_SPINDLE, enabled as spindle 0, DIV, current
1 - Huanyang v1
2 - Huanyang P2A
3 - Durapulse GS20
4 - Yalang YS620
5 - MODVFD
6 - H-100
7 - SLB_LASER
ok
[MSG:Caution: Unlocked]
ok
G90 G0 Z0
error:1 (Expected command letter)
$J=G21G91 Z10000 F640.000
ok
\x85
$J=G21G91 Z10000 F640.000
ok
\x85
$J=G21G91 Y10000 F800
ok
\x85
@david_kelly Your Z moves seem reversed? Can you share your entire gcode (you can PM me if you donāt want it public)?
What CAM program are you using?
Where are you setting your Z-zero?
I guess the Z axis movements arenāt reversed, but I may be reading them in reverse!
Itās 7pm (actually, now 8) on Friday evening here . It will need to be tomorrow before I can get the code to you. Iāve got photographic evidence as well, and Iāll put that up too.
It was humming along nicely from starting the job from the start yet again!
So the bit was retracing its previous toolpath, without touching the material. The first time it had touched the material was when it failed to retract on coming out of the previous pocket. So the bit had not pulled itself into the material
Personally, Iām leaningtowards static, largely out of frustration.
I donāt mind people seeing the code
Iām using VcarvePro as the generator of the toolpaths and Gcode.
I have a student designer doing the model. Iām not sure what sheās using to produce the stl file Iām importing.
Anyway, thanks for your continued interest. I appreciate that.
David
@david_kelly Neil may already know this from the code that you have posted, but please humour me. What post processor are you using in VCarvePro? thanks.