Best tutorial so far. After watching your video it took me 30mins to recalibrate my delta kossel after some upgrades (perfect diagonal rods, aluminium carriages etc.) Thanks! :-)
Thanks for making your video complete, yet edited down with information displayed. I haven't built a printer yet but your video was very useful for understanding how different calibrations manifest in the final print.
Do you just add the (M666 X Y Z) command at the beginning of the g-code? I'm working on calibrating my Deltaprintr and this video is great but I don't quite understand what we are doing with the new M6 values we calculate during the test prints. I'm printing a 100mm cube just like in the video Currently the "X edge" on my printer prints ok the "Y edge" of the cube disappears halfway through the line and the extruder drags on the surface Same with the "Z edge" of the cube; the extruder drags until it gets halfway and is closer to the "X side" The front edge of the cube seems to print fine Please Help! thanks
If extrusion does not occur immediately on the brim/skirt would the cause be due to a shortage of E step calibration? My steps are accurate in producing 100mm when asked to, but the skirt/brim does not print until about 1/2-1 pass
Alvin Thai Can be attributed to several things: 1. ooze 2. retract without reprime 3. too much space in Bowden tube means you have to build up pressure before it will extrude. 4. nozzle is blocked from prior retract and needs to build up pressure to push plug into melt zone I always manually purge right before a print and pull any extrusion away with tweezers before it reaches the build plate.
Jay, a couple of questions. First, what are those red connectors you use on the hot end for the heater and thermistor? Those look functional and not monstrosities like deans connectors. Secondly, it seems from this and the previous video that your Z tower shows a lot more variability than X and Y. Might that be the end stop switch or some other slop on that tower? Lastly, I notice that your kossel has the same odd vibration on homing that mine does. Travel down at higher speeds is nice and smooth but homing sounds rough and rattles everything. I might play with homing speed, perhaps there is a resonance that the default homing (is that what you are using) speed is causing. Cheers, Michael
LOL, the one for the hotend heater is a Deans Ultra and for the thermistor is a Deans Mini ... were you thinking of the XT-60 connectors? The way my endstop switch wires runs on the inside channel causes some interference with the slider...I need to reroute it. There is no slop on the slider, but there is some in the Traxxas connectors. I use a rubber band to try and remove it, but I need to loo for some way to tighten this joint yes. Good point. Homing speed is set to 90mm/s. After I sprayed some dry Teflon lube the "honking" went way.
Hey Jay, what do you use to hold the glass bed? It seems you are using some sort of pad? My delta printers is using paper clamp on the edge. It is plagued when calibrating, because the clamp makes the glass curved too much and the flatness is just way off. I want to improve my printer just like yours I am using the Delta from SeeMeCnC
I have an issue with my flsun qq, in the middle Z offset is perfect, but a few inches outward (all directions) its round about 0.1mm or more too high. Thats after auto-level of course. Do you have any idea why that is?
Delta movement converts a set of coordinates into the number of steps for each axis motor. This is some Pythagorean math (A^2 + B^2 = C^2) and is calculated for each tower, for each move, and moves are subdivided by segments per line so there is a lot of computations. If any of your values like delta_rod_length are off, then movement would be parabolic from the center outwards. This is the fine tuning video after you gave a pretty good calibration. Sounds like you do not. Try ruclips.net/video/tDLbqLve128/видео.html and then come back to this one if you think you need to.
I'm not sure what you are doing here. How does this method compare to the paper-test on the center and on the area around the three towers and adjusting the endstops (which is what I use)? It is similar, or does this method do something else, or additional?
This goes beyond using a paper feeler gauge and uses a caliper to measure the actual layer of plastic put down. If you are happy with the quality you get with the paper calibration, no need to do this. However, with this I print directly to glass with a thin layer of gluestik.
I watched it again and now I get it :) It's a neat way! but isn't it hard to measure the filament that precise? Isn't the readout too dependent on how hard you push on the caliper?
Use of calipers does require some practice and using the same amount of force. However even with the cheap 6" digital caliper from Harbor Freight, one can get 0.1mm accuracy easily. I calibrated once for what you see in the video and can print at will without worrying if the base will be flat or if the part will stick. I posted this video as a follow on to my manual calibration using the paper feel gauge video. It's not required at all.
Nice method :) I actually do use something similar I believe. I print a spiral object that covers most of the bed, and I start at VERY low z-position (slightly above the height where it doesn't extrude at all) so it's really squashed on the bed. Then I can visually see if the flatness of first perimeter is similar all over the whole bed, and if in the area of one of the towers it looks different than near the other towers then I adjust that endstop slightly and try again (I got screws on top of the carriages that i can adjust). That also results in a very good calibration. However you can also clearly see then that the bed is not 100% flat. There are always some random areas that seem to be slightly higher or lower than the rest and they are not aligned with a tower or anything else that I can see. I guess z-probing would be the final step then to correct for the non-flatness of the bed...
I've found with first layer of 0.2mm I don't have to worry about the "non-flatness" of the cake plate I use on my printer. It's "flat enough" for my needs.
Stupid question: if you can trust your steps/mm on your towers, why can't you just print a dummy carriage that you can start touching some reference (like the bed), then raise the carriage and count how many steps it took to hit the endstop? Then repeat for each tower...
+GoatZilla This question can be answered by moving you carriages down as you suggest manually to see the physical limit or four machine. Mine won't allow it because of how my carriages are made and work....yours may.
+UniK Fanta - This video is for delta printers with the towers labeled as X, Y, and Z or alternatively Alpha, Beta, and Gamma. There is no F tower to my knowledge though this video is a year old now and things may have changed.
I know that that isn't a tower, but you showed a picture with a circle, and in the middle of X and Y was F thats why I asked.. But thank you for the fast answer and you helped me a lot with your videos :D
+UniK Fanta - Ok, you are talking about the points of calibration on the build surface not the towers. F is opposite of Z tower between X and Y towers. I like to check the calibration of F(front), L(left), and R(right) after I have dialed in points x, y, and z in front of their respective towers. reprapandme.blogspot.com/2013/12/the-basics.html
+Leo Mckee-Reid Unless you are moving something physically on the printer then no. I have actually taken my printer to work on several occasions and it has no impact on calibration.
+Bass Envy You are saying that the filament that is laid down in front of the Y tower is either thicker or thinner than where it's laid down in the X and Z right? If so, then if is always thicker or thinner or does it depend of how close to the Y tower it is?
When it lays down the filament in front of the y tower the thickness is .23mm from y tower start point and at the end of it is like .14mm (these are not exact numbers).
How big, and if it's not flat then you will need to look into one of the auto-probing options. Of course printing bigger, has a whole other set of issues after calibration.
This video is meant for after you have done the initial calibration video and effector movement is parallel to the build plate. See my other video for that process. I am looking for the layer to be 0.2mm. Y and Z were at 0.33 and 0.35 meaning the effector is 0.13 and 0.15mm too high. Deltas home to Zmax so it means I needed to move the offset from the physical switch away from Zmax towards the build plate hence the negative.
Not sure why it won't let me reply to your question. I have some blog posts that my help you. reprapandme.blogspot.com/2014/02/irst-tests-and-calibration-repetier.html Read that and then follow up with anything that's still unclear.
I have some blog posts that my help you. reprapandme.blogspot.com/2014/02/irst-tests-and-calibration-repetier.html Read that and then follow up with anything that's still unclear.
You should be able to yes. The fine tuning gets into making =very small adjustments to make the printed layer a uniform thickness. It's been some years since I did this on my delta running Repetier (and almost as long as I've used that printer). I mainly use my big printer running Smoothieware, and any reprap gcode compliant firmware should work.
Наука
Best tutorial so far. After watching your video it took me 30mins to recalibrate my delta kossel after some upgrades (perfect diagonal rods, aluminium carriages etc.) Thanks! :-)
Great glad you found it useful.
Thanks for making your video complete, yet edited down with information displayed. I haven't built a printer yet but your video was very useful for understanding how different calibrations manifest in the final print.
i know it is pretty randomly asking but does anyone know of a good website to stream new series online?
I wanted to thank you for the videos. Very helpful tuning my Kossel Mini and my Kossel XL.
Great, glad they are still useful.
Excellent video, thank you. This will help me take my printing to a higher level of precision.
THNX!!!! for both video's Kossel mini printing on 0.15 all round.
Great, glad the information is still useful 2 years later. I love being able to turn my printer on and know it's going to print flawlessly.
Thank you!
tomy983 Glad you found it useful.
Do you just add the (M666 X Y Z) command at the beginning of the g-code? I'm working on calibrating my Deltaprintr and this video is great but I don't quite understand what we are doing with the new M6 values we calculate during the test prints.
I'm printing a 100mm cube just like in the video
Currently the "X edge" on my printer prints ok
the "Y edge" of the cube disappears halfway through the line and the extruder drags on the surface
Same with the "Z edge" of the cube; the extruder drags until it gets halfway and is closer to the "X side"
The front edge of the cube seems to print fine
Please Help! thanks
If extrusion does not occur immediately on the brim/skirt would the cause be due to a shortage of E step calibration? My steps are accurate in producing 100mm when asked to, but the skirt/brim does not print until about 1/2-1 pass
Alvin Thai Can be attributed to several things:
1. ooze
2. retract without reprime
3. too much space in Bowden tube means you have to build up pressure before it will extrude.
4. nozzle is blocked from prior retract and needs to build up pressure to push plug into melt zone
I always manually purge right before a print and pull any extrusion away with tweezers before it reaches the build plate.
Jay, a couple of questions. First, what are those red connectors you use on the hot end for the heater and thermistor? Those look functional and not monstrosities like deans connectors.
Secondly, it seems from this and the previous video that your Z tower shows a lot more variability than X and Y. Might that be the end stop switch or some other slop on that tower?
Lastly, I notice that your kossel has the same odd vibration on homing that mine does. Travel down at higher speeds is nice and smooth but homing sounds rough and rattles everything. I might play with homing speed, perhaps there is a resonance that the default homing (is that what you are using) speed is causing.
Cheers,
Michael
LOL, the one for the hotend heater is a Deans Ultra and for the thermistor is a Deans Mini ... were you thinking of the XT-60 connectors?
The way my endstop switch wires runs on the inside channel causes some interference with the slider...I need to reroute it. There is no slop on the slider, but there is some in the Traxxas connectors. I use a rubber band to try and remove it, but I need to loo for some way to tighten this joint yes.
Good point. Homing speed is set to 90mm/s. After I sprayed some dry Teflon lube the "honking" went way.
LOL, yes I was looking at XT-60. Used to use those on my RC helicopter. I'll check out the Ultra and minis, thanks!
Jay Couture Nice, RC connectors are perfect. Not a fan of Deans myself, but instantly recognized them!
Hey Jay, what do you use to hold the glass bed? It seems you are using some sort of pad?
My delta printers is using paper clamp on the edge. It is plagued when calibrating, because the clamp makes the glass curved too much and the flatness is just way off. I want to improve my printer just like yours
I am using the Delta from SeeMeCnC
I use 3M Scotch Heavy Duty Mounting Tape, Clear (4010).
I have an issue with my flsun qq, in the middle Z offset is perfect, but a few inches outward (all directions) its round about 0.1mm or more too high. Thats after auto-level of course. Do you have any idea why that is?
Delta movement converts a set of coordinates into the number of steps for each axis motor. This is some Pythagorean math (A^2 + B^2 = C^2) and is calculated for each tower, for each move, and moves are subdivided by segments per line so there is a lot of computations. If any of your values like delta_rod_length are off, then movement would be parabolic from the center outwards. This is the fine tuning video after you gave a pretty good calibration. Sounds like you do not. Try ruclips.net/video/tDLbqLve128/видео.html and then come back to this one if you think you need to.
github.com/MarlinFirmware/Marlin/issues/11124
I'm not sure what you are doing here. How does this method compare to the paper-test on the center and on the area around the three towers and adjusting the endstops (which is what I use)? It is similar, or does this method do something else, or additional?
This goes beyond using a paper feeler gauge and uses a caliper to measure the actual layer of plastic put down. If you are happy with the quality you get with the paper calibration, no need to do this. However, with this I print directly to glass with a thin layer of gluestik.
I watched it again and now I get it :) It's a neat way! but isn't it hard to measure the filament that precise? Isn't the readout too dependent on how hard you push on the caliper?
Use of calipers does require some practice and using the same amount of force. However even with the cheap 6" digital caliper from Harbor Freight, one can get 0.1mm accuracy easily. I calibrated once for what you see in the video and can print at will without worrying if the base will be flat or if the part will stick. I posted this video as a follow on to my manual calibration using the paper feel gauge video. It's not required at all.
Nice method :) I actually do use something similar I believe. I print a spiral object that covers most of the bed, and I start at VERY low z-position (slightly above the height where it doesn't extrude at all) so it's really squashed on the bed. Then I can visually see if the flatness of first perimeter is similar all over the whole bed, and if in the area of one of the towers it looks different than near the other towers then I adjust that endstop slightly and try again (I got screws on top of the carriages that i can adjust). That also results in a very good calibration. However you can also clearly see then that the bed is not 100% flat. There are always some random areas that seem to be slightly higher or lower than the rest and they are not aligned with a tower or anything else that I can see. I guess z-probing would be the final step then to correct for the non-flatness of the bed...
I've found with first layer of 0.2mm I don't have to worry about the "non-flatness" of the cake plate I use on my printer. It's "flat enough" for my needs.
Stupid question: if you can trust your steps/mm on your towers, why can't you just print a dummy carriage that you can start touching some reference (like the bed), then raise the carriage and count how many steps it took to hit the endstop? Then repeat for each tower...
+GoatZilla This question can be answered by moving you carriages down as you suggest manually to see the physical limit or four machine. Mine won't allow it because of how my carriages are made and work....yours may.
Can I recalibrate it using Repetier instead Marlin? with G-code?
Yes, Repetier accepts the same gcodes.
Hello, can I adjust the F "tower" with M666 or an other gcode ?
Thank you
+UniK Fanta - This video is for delta printers with the towers labeled as X, Y, and Z or alternatively Alpha, Beta, and Gamma. There is no F tower to my knowledge though this video is a year old now and things may have changed.
I know that that isn't a tower, but you showed a picture with a circle, and in the middle of X and Y was F thats why I asked.. But thank you for the fast answer and you helped me a lot with your videos :D
+UniK Fanta - Ok, you are talking about the points of calibration on the build surface not the towers. F is opposite of Z tower between X and Y towers. I like to check the calibration of F(front), L(left), and R(right) after I have dialed in points x, y, and z in front of their respective towers. reprapandme.blogspot.com/2013/12/the-basics.html
If I move my printer to some other location in my room, should I be re-calibrating anything?
+Leo Mckee-Reid Unless you are moving something physically on the printer then no. I have actually taken my printer to work on several occasions and it has no impact on calibration.
+Jay Couture Excellent! thanks for the quick response.
trying to level my bead and having one issue my y axis is thickness is inconsistent the x and z are fine
+Bass Envy You are saying that the filament that is laid down in front of the Y tower is either thicker or thinner than where it's laid down in the X and Z right? If so, then if is always thicker or thinner or does it depend of how close to the Y tower it is?
When it lays down the filament in front of the y tower the thickness is .23mm from y tower start point and at the end of it is like .14mm (these are not exact numbers).
what about bigger build plates not planar ?
How big, and if it's not flat then you will need to look into one of the auto-probing options. Of course printing bigger, has a whole other set of issues after calibration.
i'm confused as to how you calculated your new endstop values in comparison to the layer thickness
This video is meant for after you have done the initial calibration video and effector movement is parallel to the build plate. See my other video for that process. I am looking for the layer to be 0.2mm. Y and Z were at 0.33 and 0.35 meaning the effector is 0.13 and 0.15mm too high. Deltas home to Zmax so it means I needed to move the offset from the physical switch away from Zmax towards the build plate hence the negative.
@@JayCouture3D where can i find the intial calibration video?
@@fastneonman1977 ruclips.net/video/tDLbqLve128/видео.html
What Glue stick do you use?
Elmers extra strength
Jay Couture Thanks Jay, your videos helped me a lot!
Not sure why it won't let me reply to your question. I have some blog posts that my help you. reprapandme.blogspot.com/2014/02/irst-tests-and-calibration-repetier.html Read that and then follow up with anything that's still unclear.
Can I use this tutorial in my Delta with repetier?
I have some blog posts that my help you. reprapandme.blogspot.com/2014/02/irst-tests-and-calibration-repetier.html Read that and then follow up with anything that's still unclear.
Jay Couture thanks a lot for this information! =D
My delta is calibrated, now I want to do this fine calibration. Can I do this the same way in repetier? Thanks
Can I do this calibration like the video of repetier? Print, take meager, difference*sets and do it for x y z
You should be able to yes. The fine tuning gets into making =very small adjustments to make the printed layer a uniform thickness. It's been some years since I did this on my delta running Repetier (and almost as long as I've used that printer). I mainly use my big printer running Smoothieware, and any reprap gcode compliant firmware should work.
r u in usa
Yes, I am
need help calibrating mine
Is the video not clear or do you have a question that is not covered by video? You can email, me at jay.couture at gmail dot .com if you like
that be great thanks