I have been googling for an hour with no solution so hear I am again, smilieface Now my problem is the laser cutter does not cut or engrave curves well. It goes increadably slowly and stutters seemingly at every segment of the curve. I am using inkscape to generate the graphic, laserweb to make g code and openbuilds control. I have also used lightburn with the same result. I have been googling for an hour so hear I am again. Thanks for any help
Paste your Grbl Settings please (Click Backup Grbl Settings button on the Grbl Settings tab, open file, paste contents)
Thanks for replies I have increase acceleration from 150mm/sec to 1000mm/sec, in several steps, makes no difference. grbl text file below $0=10 ; Step pulse time, microseconds $1=255 ; Step idle delay, milliseconds $2=0 ; Step pulse invert, mask $3=0 ; Step direction invert, mask $4=1 ; Invert step enable pin, boolean $5=0 ; Invert limit pins, boolean $6=0 ; Invert probe pin, boolean $10=1 ; Status report options, mask $11=0.020 ; Junction deviation, millimeters $12=0.002 ; Arc tolerance, millimeters $13=0 ; Report in inches, boolean $20=0 ; Soft limits enable, boolean $21=1 ; Hard limits enable, boolean $22=1 ; Homing cycle enable, boolean $23=1 ; Homing direction invert, mask $24=100.000 ; Homing locate feed rate, mm/min $25=1000.000 ; Homing search seek rate, mm/min $26=250 ; Homing switch debounce delay, milliseconds $27=5.000 ; Homing switch pull-off distance, millimeters $30=1000 ; Maximum spindle speed, RPM $31=0 ; Minimum spindle speed, RPM $32=1 ; Laser-mode enable, boolean $100=72.000 ; X-axis steps per millimeter $101=72.000 ; Y-axis steps per millimeter $102=199.100 ; Z-axis steps per millimeter $110=3000.000 ; X-axis maximum rate, mm/min $111=3000.000 ; Y-axis maximum rate, mm/min $112=2500.000 ; Z-axis maximum rate, mm/min $120=150.000 ; X-axis acceleration, mm/sec^2 $121=1000.000 ; Y-axis acceleration, mm/sec^2 $122=1000.000 ; Z-axis acceleration, mm/sec^2 $130=1100.000 ; X-axis maximum travel, millimeters $131=610.000 ; Y-axis maximum travel, millimeters $132=0.000 ; Z-axis maximum travel, millimeters Thanks for reading
Because you effectively only increased it on one axis. Z doesn't really need much. Perception of "slowness" is almost always acceleration settings. If you're running belts, you should easily be able to set X and Y accelerations to 5000-10,000mm/s^2.
Did you try generating gcode in Lightburn, or just sending the LaserWeb gcode with Lightburn? I used to be the lead dev on the Laserweb Project before I joined OpenBuilds, and fair enough there are some inefficiences - lightburn's gcode is a little more lean
Eh, I also use LaserWeb- especially for rasters- and have had zero issues with it through grbl and bCNC. Whatever inefficiencies there are, I think are imperceptible in a properly set up machine.
True, but still interested in the answer to the question. A second CAM may give another data point (or point out a mistake in his LW config) Also, what is the specs of the PC driving the controller?
Oh Silly me Acceleration is now set at 10000mm/sec on X and Y (Z @ 150mm/sec) But it has made no difference PC Spec: Memory 3.8 GiB Processer intel pentium(R) CPU 4415U @ 2.3 GHz x 4 Graphics intel HD Graphics 610 Kaby Lake GT1 Os type 64bit Ubuntu 16.04 I don't have time now to test Lightburn again. Tomorrow. Thanks
Control PC for me is i5-4670K (4x3.40GHz), 16GB RAM, Win 10 Home 64 bit. Correction, my mistake- acceleration should be in the 500-1500mm/s^2 range. For some reason my brain ran off with max rate numbers. I doubt your machine will like accelerations that high unless it's running some beefy servo drives! Mine is currently at 800. 0.01mm or 0.02mm Junction Deviation shouldn't make any difference here- if anything, 0.020mm would calculate the cornering speed to be higher. Might need a snippet of G-code, because it seems like the control end of things is set up right.
Hi everybody I have changed x and y acceleration to 800mm/sec and arc tolerance to 0.01 and it has made little difference when running the job thru laserweb. (still very slow round the arcs) I tried running the same job with Lightburn and it worked loverly I changed arc tolerance to 0.008 and it made little difference to the job run in laserweb I would prefer to run an open source g-code software if possible so and explanation would be welcome Thanks for all your help Aaron
Oh, and I tried the Openbuilds Cam program, but I couldn't get it to start from the top left of my machine. I would be interested to know is the Blackbox is open hardware? Thanks
LaserWeb is not as optimised as Lightburn (slower serial sender) - "you get what you pay for" in reality. Also, most of the laserweb developers have moved on. Lightburn is worth the price (consider that this reply comes from "the" guy that started LaserWeb...) OpenBuilds CONTROL is just a sender. It doesnt dictate where the job starts, your gcode does.
The g-code attached above has zero G2/3s so it's not a feed rate issue, it's either a streaming issue inside the sender, a buffering/planning issue inside the controller or the CAM is just outputting way too many lines. If using CONTROL/bCNC/etc which are optimized senders, that discounts the first option. They're very fast. The second two are two sides of the same problem. Perhaps I haven't run into the "too many lines" issue because I only use LaserWeb for raster etching, and Fusion 360 for everything else. LaserWeb only produces lines, period, it seems like, which given that grbl internally splits curves up into lines for [Junction Deviation] acceleration planning seems like it's just adding way more work than necessary: (Though you could try changing this number in LaserWeb settings to say, 0.3mm, and see if that helps) Whereas Fusion 360 allows for very easy etching of curved toolpaths and has a grbl laser post processor, allowing direct cutting of sketch lines: If you want to do both raster and vector etching/cutting in the same file, it gets tricky though. That's where Lightburn comes in. At only $40 for the g-code version and at-cost upgrade to DSP, seems like a no-brainer to me. I'll probably pick it up at some point too when I'm actually running machines instead of building them constantly. There's also LaserGRBL that's free but I've never tried it so I don't know whether it's worthwhile or not.
Hey everybody. Stand down and breath a sigh of relief, I think I have finally got a usable machine! So thank you all, especially Peter. It's been a bit of a battle but worth it in the end, although I am still missing my old Lasersaur Best wishes Aaron
Went the Lightburn route? as its so dedicated to lasers I bet in no time you won't even mis Lasersaur anymore
Hope you found the solution you were looking for. What kind of work are you doing with this laser cutter of yours?
I use lightburn now which works well. I've given up with laserweb which is a shame cos I'm really in favour of open source software. I do the odd job for the general public and some teaching/training/makerspace stuff. www.cnccraft.co.uk/macerspace. Cheers
just been messing with this code. Average line length is about 0.03mm, which is barely a few steps for each motor. Wrote a quick script to filter it and only output moves greater than 1mm (filtered.gcode) and 0.3mm (filtered.3.gcode). Maybe you'd like to give them a try on some scrap and see how it goes and let us know, for science (-:
