Anycubic kossel calibration help, endstops, concave print area.

Hello,
I am new to kossel type printers coming from a makerbot cupcake! I have spent the best part of 2 months trying to get this thing calibrated without success.
I am using Repitier host on W7 with marlin firmware on the printer.
Typing M119 for the status of the endsops and when the 'xyz' carriages 'engage' the endstops they all come back as triggered.
When I move the carriages down only the z min is open, all others stay triggered, is this normal behaviour?
Also when I try M28 (home) I expect the 'spider/effector' and the carriages to rise up and hit the endstops but all it does is go down 1cm. Is this also the expected behaviour? If not what might the issue be?
Does the wiring of the endstops matter? ie on the ramps is there a +ve /-ve pin order or does this not matter for the endstops?
I would like to get my endstops working correctly before sorting out the concave print area.
Any advice on which way to change the delta_radius is also welcome (which variable to change and in which direction (+ve or -ve)).
thanks for any help, it is appreciated.
Ki

I have discovered M666 which adjusts the endstop delta. Anyone know how to use this gcode?

Quote
kibodwin
Typing M119 for the status of the endsops and when the 'xyz' carriages 'engage' the endstops they all come back as triggered.

That is wrong. they should be open.

Quote
kibodwin
When I move the carriages down only the z min is open, all others stay triggered, is this normal behaviour?

No.

Quote
kibodwin
Also when I try M28 (home) I expect the 'spider/effector' and the carriages to rise up and hit the endstops but all it does is go down 1cm. Is this also the expected behaviour?

No.

Quote
kibodwin
If not what might the issue be?

You have logic of your endstops setup in firmware just the opposite way it should be.

Quote
kibodwin
Does the wiring of the endstops matter? ie on the ramps is there a +ve /-ve pin order or does this not matter for the endstops?

For simple microswitch endstops the two wires can be swapped. For endstops requiring power (e.g. optical it cannot be swapped). This also depends on your electronics board. I do not know what +ve -ve means on your board. You must find this in the documentation for your electronics.

Quote
kibodwin
I would like to get my endstops working correctly before sorting out the concave print area.

Very good idea because without working endstops you cannot properly home and therefore also level the bed.

M666 info: [reprap.org]

Hi hercek,
Thank you very much for your help, it is greatly appreciated as I was beginning to think no one would reply.
I will take a look at the setup I have following what you say and go from there.
Again, thank you very much for your answers.
Ki.

Did you solve your endstop issues? Not sure which firmware you are using- the AnyCubic Marlin firmware? To adjust the Delta radius you should increase or decrease DELTA_SMOOTH_ROD_OFFSET depending on whether you want to lower or raise the center of the plane.

Hi siddharta,
Sorry for the delay, life got in the way...
I use the anycubic marlin firmware that came with the printer. I need to see if the endstops meet to be triggered when activated or not and make sure the config.h is correct. Can you tell me what your endstops say when triggered and not when sending a gcodetools M114.
Once the endstops configured correctly i will centre the nozzle and then make the smooth_rod changes to get my delta_radus sorted.
Thanks for helping, it is appreciated.
Have you got your delta up and running correctly?
Ki.

Hi, sorry for the delay in getting back to you, I neglected to follow the topic and haven't checked in for a while. Are you still having the endstop issue or did you manage to resolve it?
If not, what are you getting when running M119 when the endstops are triggered and when not? Also, check that you have them connected using the outer two of the three terminals on the microswitches. You can use M666 to calibrate the printer but you need to have the endstops working and the effector homing well before you can move on with that.

I bought a second Kossel kit from Anycubic, the larger version with linear rails again and have that one up and running quite well. I installed the Duet electronics as well as the IR sensor of dc42 and it's working like a charm. I designed a little mount for the sensor, if you're considering to use the differential IR sensor I can send it to you. I'm waiting on a set of new arms for the Kossel Mini and will rebuild it from the ground up when those arrive to get it as straight as possible.

Hi Siddharta,
I too apologise for not getting back to you, family got in the way but I will have time soon to test the endstops and make sure they are working ok and work on my delta_radius. I will get back to you when I have had time to play with my config.h.
So you now have 2 deltas?? Trying to make us all jealous??!! =)
Thanks
Ki

Hi Siddharta,
Would you be willing to share your config.h as I am wondering if I do not have an issue with me steps per unit config. I say this as I changed the smooth rod value and this had no effect on the curvature of the print area. I will also be able to see the other values you have for the same printer.
Do you know if I change the Steps per Unit setting in the config or is the change to the motor driver pot.?

Hi, I mailed you the firmware for two AnyCubic models (regular Kossel which they call "Linear Plus" and the Kossel Mini ("Pulley" or "Linear"). Note that for any other printer model things may be set up differently.

The steps/mm value is derived from your hardware, i.e. the stepper motor (steps/revolution), microstepping (should be 1/16th if you're using the A4988 Allegro drivers), belt pitch (2mm for GT2) and pulley (16 to 20 teeth are the most common). Have a look here Prusa steps/mm calculator. This is set in configuration, if using Marlin look for DEFAULT_AXIS_STEPS_PER_UNIT.

To my knowledge curvature (concave/convex) is solely determined by the Delta radius. To calibrate this properly you would first need to ensure that you have the correct length for the six Delta arms (measured eye to eye) and that you have this set in your configuration DELTA_DIAGONAL_ROD and that your max endstops are well configured (i.e. you can home properly).

Move the effector to X0 Y0 Z50 (should be roughly 50mm over the centre of the bed), then lower the nozzle little by little until it is at the correct height over the bed (it should lightly grip, but not hold in place, a paper sheet). Use M114 to get the Z height at that point and adjust the value in MANUAL_Z_HOME_POS accordingly (old MANUAL_Z_HOME_POS - Z height from M114).

Then move the effector as close as possible to one of the towers and adjust the endstop position on that tower (either physically adjusting them or using M666 to adjust them), so that the nozzle is again set at the correct height at that location. Repeat for each of the three towers.

At this point, unless your Delta radius is already correct, when you move the nozzle to the centre X0 Y0 Z0 position it will be either too low (requiring you to increase the Delta radius) or too high (then you need to decrease the Delta radius). You can increase or decrease DELTA_SMOOTH_ROD_OFFSET to modify the Delta radius. When you have the nozzle positioned well again in the centre positions, move the effector to the tower locations and check nozzle height at Z0. If the nozzle is not well positioned again adjust endstops and Delta radius as in the last two steps. Iterate until the nozzle is well positioned at the tower and centre locations.

dc42 has an excellent and more detailed and rigorous writeup of calibrating a Delta at delta calibration. Hope that helps.

Hi Siddharta,
Thank you for the reply, help and advice, it is appreciated. Apologies for not getting back to
you sooner, I have since completely taken the kossel appart and rebuilt it. I found that my rods were of different lengths
so anyone who is having issues should check this even if the rods come 'pre-made'.
I will definitively follow the instructions you gave.
Happy new year,
Ki