@dc42 Problem solved, thank you!
I'd already checked for any slip or backlash by holding the belt while jogging small amounts, but was going to double-check the grub screws to be certain.
They need a torch to see them clearly.
They are OK - but I was not expecting the pulley to have changed size!
A small fragment of fibre reinforced tape from somewhere was stuck on it, so the belt was moving slightly further for each motor rotation.
@dc42 No, just generic fibre reinforced ones.
Hi,
My Kossel XL has been working fine for the last couple of years.
(Duet 3 MB6HC; firmware 3.4.0).
I've not used it for a couple of months (too busy with work).
I just tried to print something and found that one axis is moving around 10% further than the other two - when it went to probe the bed centre, it was a long way off to the side.
Homing then jogging Z- shows the same effect - the front left axis (as I have it positioned) moves proportionally too far as it travels down, and the same jogging back up - so all three level gradually level out again as it gets back to the top!
I've checked the config file is not corrupt; all three axis parameters are identical.
Also tried power off and let it restart without attempting to print, no difference.
The motor pulleys and belts are all fine, no slipping.
Any ideas?
Thanks,
Robert.
What happens if you jog Z over the full height range - does the nozzle move purely vertical or have a lateral offset with height??
It's not likely, but the only things I can think of are a stepper with a different physical step count, or a pulley with a different size?
@Th0mpy said in Anybody wants a stepper motor analyzer?:
You can update your code to point to these two repositories:
That's got it working, thank you!
@dc42
OK thanks, I was using 70'C, that may be the problem!
I'll try it at 80'C
My Kossel XL has been working very well until about a week ago, when a PETG print lifted a layer of glass out of the bed, despite a fresh coat of bonding/release spray before each print...
I had ordered some Printbrite+ during lockdown, which arrived a few month back but I did not do anything with it.
Glass bed turned over, printbrite applied, all cleaning & curing instructions followed carefully.
The first print I tried broke loose after a couple of layers.. I re-read the instructions & realised I'd done the pre-print clean with acetone, instead of soapy water. Try again, pre-clean with soapy water and it worked as advertised!
Three or four prints, all OK, just releasing once the bed started to cool.
Then I set up a large area 12 hour print, same clean - and it broke loose in the middle of the night & wasted the lot.
I'm not sure if I'm doing something wrong with it, or it's just an unreasonably critical surface??
Any ideas? I have things I want to print but don't want to waste masses of filament if its going to be hit & miss on which stay attached to the bed.
(I have a new 300mm diameter Borosilicate bed on order just in case).
@Th0mpy
Excellent, thank you
I just use large zip bags to store each open reel, with a fabric bag of silica gel sealed in with the reel.
I also use a Sunlu dryer as the filament feed while printing.
(one of these types -
https://www.amazon.co.uk/SUNLU-Filament-Printing-Dimensional-Accuracy/dp/B081ZLG6KR )
@zapta
Hi,
OK, thanks for that, hopefully Th0mpy is around!
I can't find anything relevant in his posts in this topic, unfortunately.
Hi all,
I've assembled another unit from the original parts I got, but I am having serious problems trying to build the program for it..
I've changed machines since I did the original so its a fresh install of VS Code & platformio etc.
I cannot get the Wizio stuff to work; it looks like some part of that project has been either removed or renamed?
I never normally use Visual Studio & the only experience I've had with platformio was in one day, for the last one of these I built.
Any clues or fixes to get it to build properly would be greatly appreciated!
Working on Win 10 64, with all updates etc.
(I normally program such as PICs using MPLab & CCS compiler, or industrial controls - never visual studio stuff).
I use this setup from Thingiverse, with the E3D V6 hotend heatsink from the kit, titanium heat break, plated copper block & 0.4mm Olsson ruby nozzle.
https://www.thingiverse.com/thing:3893311
That's with a Flex3drive system, so like this on top:
https://flex3drive.com/product/g5-flex-for-duet-smart-effector/
The end result is very light but as precise as a direct drive setup.
I've only done a few bits for experimentation.
I just used Pritt (white glue stick for paper) and that worked fine, temperatures 70 & 240.
I agree with Rushmere3d, you need to add some limits for C.
I have 0 and 500 on mine.
; Axis Limits NEED CHANGING FOR TC. CONSERVATIVE GUESS NOW for XY
M208 X-14 Y0 Z0 C0 S1 ; set axis minima
M208 X330 Y265 Z350 C500 S0
And this is my home C file:
; homec.g
; called to home the C axis (coupler)
;
;G91
M400
G92 C499
M400
M913 C60 ; MOTOR TO 60% CURRENT
G1 H1 C0 F2000
;
M400
G92 C0
;G90
M913 C100 ; MOTOR TO 100% CURRENT
G1 C0 F10000
;Open Coupler
M98 P"/macros/Coupler - Unlock"
I have the lock position as C33 and unlock as C125
@techbutterfly said in Tweaking configuration on MB6HC with SBC:
I am a little curious about the M558 line under Z-probe - what is that for?
See the GCode (and MCode) documentation here:
https://docs.duet3d.com/en/User_manual/Reference/Gcodes
The narrow white column is scrollable to find the specific code you are looking for.
How have you got the CR Touch wired?
If it's anything like their version of the BL Touch, there may be more than one cable type and you must work from the physical pin connections and not just rely on wire colours..
I only ever use the configurator to create an initial rough config.
I keep the files for each machine in their own machine directory on my PC, then if I need to tweak the config I edit the file with a text editor and re-upload to the machine via DWC (the Duet web interface).
That also means you have a full working config available if you ever need to rebuild the SD card - burn the latest image, start the machine with that in it & open DWC, then drop all the files back in and you are running again in a few minutes.
@jrcl
I'd agree with Percar;
It looks to me as if it may be overextruding for the nozzle height, so forcing each adjacent strip sideways and breaking it free from the bed.
Try using babystepping and increasing the nozzle height as soon as the print starts, to see if it smooths out?
I went through this with my converted TronXY when initially getting that going, similar effects due to Z too low on the first layer.
@toddm said in Stringing with PLA:
What would you change first?
Both at 195C on the extruder,
I always run at 210'C for PLA and get excellent results!
To me it looks like the stringing is as the plastic is not molten enough, so stretching rather than separating as a liquid.
I have a junction box on the back of the machine with the tool distribution board and D15 sockets for the tool cables. That allows me to remove tools and just use a dummy plug to link through the CAN bus at any unused connector.
I also have the cables to each fitted tool permanently plugged in!
Although you may get it to work with boards being connected and disconnected, I suspect there will be CAN errors and require continuous resets as boards appear and disappear from the bus, as seen by the controller.
I've just copied the E3D cable support setup, which uses long springy steel cable ties to keep the tool cables formed in to upright loops.
More photos here:
https://forum.duet3d.com/topic/25709/meet-the-frankentron
Video here:
https://www.youtube.com/watch?v=x-ut81sn-io
This is the basic circuit you need for a reversing motor with limit switches:
https://www.halloweenforum.com/attachments/relay-controlled-dc-motor-jpg.252253/
The relay flywheel diode should preferably be a schottky or fast recovery type, with a higher current rating than the relay coil and a rather higher voltage rating than the circuit operates on.
It would also be a good idea to add a kind flywheel circuit around the motor, to minimise relay contact arcing; the most effective is four diodes arranged as a bridge rectifier with the AC connections across the motor and DC out connected back to the power source (eg. the 24V supply) with bridge positive out connected to supply positive.