Delta Calibration
Posted by igyros
|
Delta Calibration September 26, 2020 04:55AM |
Hello everyone,
I have registered here 1. because I am an enthusiastic tinkerer and 2. because I despair of my new Delta printer (Tevo Little Monster).
I bought a used Tevo Little Monster. Unfortunately it had some problems, badly assembled (screws twisted and crooked) the hotend is permanently clogged etc.
Most of the problems I have solved now and actually it would be ready for printing. But unfortunately I can't calibrate the printing bed.
As you can see on the pictures, some of the spots on the printing bed are much too close and some are much too far away.
The printer runs with smoothieware.
The following things I have tried so far:
-Printer frame completely disassembled, reassembled with a right angle and measured with a digital spirit level (if I can trust the spirit level, the frame is assembled to an accuracy of 0.05°)
-I set the small pins for the optical endstops to an accuracy of 0.05 mm.
-I tried to do the delta calibration with G32 first and then the mesh calibration with G31. But I have the feeling that it doesn't make a difference if I do the mesh calibration or not, I have exactly the same result.
-I have done the "Least squares (David Crocker variant)" with a different firmware.
So far nothing has worked and the result is always the same. Unfortunately I could not measure the arms yet, because there are several screws broken and I only want to drill them out when I have the right spare screws.
I suspect the firmware or the arms, but I am not sure and ask here for advice not to waste more money.
Do any of you have an idea where this error could trigger the image?
Before printing the image I did a least squares delta clearing and a mesh calibration. But the result is exactly the same with a normal delta calibration and without mesh calibration.
The following steps I plan to take to correct the error, but I am afraid to be on the wrong track.
-I want to exchange the mainboard for a Duet Wifi to exclude the firmware.
-I want to exchange the Effector with a Duet Smart Effector to use MagBallArms and have the sensor directly on the nozzle.
-I want to make the bed manually adjustable with springs. But I have little hope for success.
-I would like to install MagBallArm from Haydn Huntley.
If someone has an idea or knows this problem I would be very grateful.
I have registered here 1. because I am an enthusiastic tinkerer and 2. because I despair of my new Delta printer (Tevo Little Monster).
I bought a used Tevo Little Monster. Unfortunately it had some problems, badly assembled (screws twisted and crooked) the hotend is permanently clogged etc.
Most of the problems I have solved now and actually it would be ready for printing. But unfortunately I can't calibrate the printing bed.
As you can see on the pictures, some of the spots on the printing bed are much too close and some are much too far away.
The printer runs with smoothieware.
The following things I have tried so far:
-Printer frame completely disassembled, reassembled with a right angle and measured with a digital spirit level (if I can trust the spirit level, the frame is assembled to an accuracy of 0.05°)
-I set the small pins for the optical endstops to an accuracy of 0.05 mm.
-I tried to do the delta calibration with G32 first and then the mesh calibration with G31. But I have the feeling that it doesn't make a difference if I do the mesh calibration or not, I have exactly the same result.
-I have done the "Least squares (David Crocker variant)" with a different firmware.
So far nothing has worked and the result is always the same. Unfortunately I could not measure the arms yet, because there are several screws broken and I only want to drill them out when I have the right spare screws.
I suspect the firmware or the arms, but I am not sure and ask here for advice not to waste more money.
Do any of you have an idea where this error could trigger the image?
Before printing the image I did a least squares delta clearing and a mesh calibration. But the result is exactly the same with a normal delta calibration and without mesh calibration.
The following steps I plan to take to correct the error, but I am afraid to be on the wrong track.
-I want to exchange the mainboard for a Duet Wifi to exclude the firmware.
-I want to exchange the Effector with a Duet Smart Effector to use MagBallArms and have the sensor directly on the nozzle.
-I want to make the bed manually adjustable with springs. But I have little hope for success.
-I would like to install MagBallArm from Haydn Huntley.
If someone has an idea or knows this problem I would be very grateful.
|
Re: Delta Calibration October 29, 2020 03:30PM |
If you already know your arms are not correct, forget about troubleshooting the print. The arms *must* be equal lengths for a Delta to produce correct results. If they are not, the effector tilts, and the nozzle deviates badly from the expected flattend curve it should follow.
Your plan for "upgrades" looks good. I use magballarms (from "blue eagle labs", which I think is Haydn's retail site?) and have nothing but good things to say about them. Built very precise, work beautifully. They've got thousands of hours on them and no visible signs of wear. My only complaint is that magnets are strong enough they like to grab m screwdriver when I'm trying to adjust the cooling fan mounts on the effector.
I almost went with the smart effector, opting instead to use a Precision Piezo Orion, along with an E3D v6 hotend mounted in Haydn's magball effector (available off thingiverse and others). Not a good option if you don't have a working printer already. Plus I'm far too cheap to go with the Duet controller.
The only thing I would point out is that you do not want springs on the bed if you are using a bed probe. It will flex, it will add error, and it will drift out of alignment. All bad. You want a fixed bed, and then correct any misalignment with the frame by either shimming or using an extended calibration method that accounts for skew. I don't know if the Duet firmware supports it, but there are online tools to complete it if necessary. Alternatively you can use something like bed tilt/mesh if your firmware supports it. ( Personally, I use Klipper, it has extensive delta calibration built in.)
With a delta, although you can probe the bed every print, you shouldn't ever have to. Calibrate it once and then print forever.
Edited 1 time(s). Last edit at 10/29/2020 03:31PM by Gecko23.
Your plan for "upgrades" looks good. I use magballarms (from "blue eagle labs", which I think is Haydn's retail site?) and have nothing but good things to say about them. Built very precise, work beautifully. They've got thousands of hours on them and no visible signs of wear. My only complaint is that magnets are strong enough they like to grab m screwdriver when I'm trying to adjust the cooling fan mounts on the effector.
I almost went with the smart effector, opting instead to use a Precision Piezo Orion, along with an E3D v6 hotend mounted in Haydn's magball effector (available off thingiverse and others). Not a good option if you don't have a working printer already.
Plus I'm far too cheap to go with the Duet controller.The only thing I would point out is that you do not want springs on the bed if you are using a bed probe. It will flex, it will add error, and it will drift out of alignment. All bad. You want a fixed bed, and then correct any misalignment with the frame by either shimming or using an extended calibration method that accounts for skew. I don't know if the Duet firmware supports it, but there are online tools to complete it if necessary. Alternatively you can use something like bed tilt/mesh if your firmware supports it. ( Personally, I use Klipper, it has extensive delta calibration built in.)
With a delta, although you can probe the bed every print, you shouldn't ever have to. Calibrate it once and then print forever.
Edited 1 time(s). Last edit at 10/29/2020 03:31PM by Gecko23.
Sorry, only registered users may post in this forum.