Mini Kossel, Final Calibration: Moving 90mm with 85mm input

I am still working on my Mini Kossel (MakerGeek 1/19/2015 with all the upgrades kit)

I finally got my zProbe working and now the auto level process is running great. I'm on the last calibration step before a test piece. Really hoping to close today.

Problem: When doing the final calibration (Blomker instructions) I am moving from 0,0,1 to 0,85,1. I am moving in the Y direction 90mm not 85.

What I have tried and observed:

1. Observation: When input G1 X0 Y0 Z175 my LCD output shows X0 Y0 Z185. Maybe that is a symptom.
2. Tried: I read the instructions stating it could be my link length. It said if I'm long, then increase the link length. I'm not seeing any changes when i do that. Always come out about 90mm. I'm set at 215 and went to 210 and 217 with little change.
Currently: #define DELTA_DIAGONAL_ROD 215 // mm (Was 214)
3. I also have started to look at the AXIS STEPS. This is what I have: #define DEFAULT_AXIS_STEPS_PER_UNIT {94.1, 94.1, 94.1, 92.3} // ( Was {100, 100, 100, 450} Makergeek said 94.1176 for x,y,z and 92.3 for e)
4. Question: What elements make the distance travel reduce.

NOTE: When I did the beginning calibration I thought I was pretty much right on in travel x,y and z. I was doing this with a micrometer in 10mm increments which is kind of tough to cover.

Further Insight:

When I move in the Z direction by 20mm by clicking the 10mm move in Pronterface twice, I am pretty damn close to exactly 20mm.

BUT....

If I do G1 Z210 and move there. Then do G1 Z180, it' move greater than 20mm.

This must be the source the the problem.

Thoughts?

Edited 1 time(s). Last edit at 03/08/2015 11:57AM by Dark Skunk.

210mm - 180mm = 30mm not 20mm.

If the Z axis is consistently moving the right amount, then your X Y Z steps/mm (which should all be the same) are correct.

What value did your delta radius come out at after calibration? On my Mini Kossel it is just under 105m. My diagonal rod length is also 215mm.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

First: Typo on the 210 to 180 (I was doing 20mm checks before).

I ended up reseting the board manually and re-uploading the software. Then I did this:

1. Home Key
2. G28
3. G29
4. G1 X0 Y0 Z1 (confirm center is over my previous dot, it was).
5. G1 Z210
6. Click 10mm down z three times on proterface: I got 29.62 with the micrometer (I'm probably plus or minus 1-2mm with that measuring process).
7. G1 Z210
8. G1 Z180 I got 29.56 so I'm thinking that is the same.

I don't recall the term delta radius. I have a zProbe attached to the print head (zero offset) so I went through that section skipping a lot that did not look relevant. My print diameter (the glass is 180mm). I just read through it against but I don't see that term in the Blomker instructions (unless I missed it).

Question: What is Delta Radius? Where can I find it in the sketch or how can I confirm that value (maybe there is a G code that tells me?).

In RepRapFirmware for the Duet electronics, if you send M665 with no parameters then it reports the delta radius and other delta parameters. I believe Marlin supports M665 for setting the delta parameters, but I don't know whether it reports them as well. It may be that when you run auto calibration, the firmware reports the parameters it found at the end of the process.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

M665 Said sending but gave no output. Is this what you are looking for?

-4.278 -4.278 -4.278 -3.345 -2.912 -2.912 -2.912
-5.413 -5.413 -4.051 -3.188 -2.580 -2.461 -2.461
-7.101 -5.807 -4.247 -3.027 -2.455 -2.298 -2.576
-8.108 -6.544 -4.863 -3.511 -2.723 -2.253 -2.548
-9.048 -7.424 -5.651 -4.176 -3.360 -2.980 -3.056
-8.334 -8.334 -6.939 -5.301 -4.310 -3.617 -3.617
-7.426 -7.426 -7.426 -6.302 -5.103 -5.103 -5.103

BTW: Is there a way to slow down how quickly the machine homes the end stops. It's a bit violet as it's setup now. I want the speed in half if I could.

Edited 1 time(s). Last edit at 03/08/2015 02:53PM by Dark Skunk.

Quote
Dark Skunk
M665 Said sending but gave no output. Is this what you are looking for?

-4.278 -4.278 -4.278 -3.345 -2.912 -2.912 -2.912
-5.413 -5.413 -4.051 -3.188 -2.580 -2.461 -2.461
-7.101 -5.807 -4.247 -3.027 -2.455 -2.298 -2.576
-8.108 -6.544 -4.863 -3.511 -2.723 -2.253 -2.548
-9.048 -7.424 -5.651 -4.176 -3.360 -2.980 -3.056
-8.334 -8.334 -6.939 -5.301 -4.310 -3.617 -3.617
-7.426 -7.426 -7.426 -6.302 -5.103 -5.103 -5.103

BTW: Is there a way to slow down how quickly the machine homes the end stops. It's a bit violet as it's setup now. I want the speed in half if I could.

Looks to me that it's not doing auto calibration, instead it's doing bed compensation. You need to either not use bed compensation at all (because it shouldn't normally be needed on a delta), or run it only after doing delta calibration. On RepRapFirmware (for Duet electronics) and Smoothieware, G32 runs auto calibration. I don't know about Marlin.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Yes, this is the calibration for leveling the plane in relation to the actual reality. I thought that is what I needed.


Do you have a link for delta calibration? I ran through all the instructions I have but the standard 'calibration' in relation to x and y have not come up to my knowledge. I will search again in the meanwhile.

Edited 1 time(s). Last edit at 03/08/2015 06:38PM by Dark Skunk.

I describe delta calibration in my blog entry here [miscsolutions.wordpress.com], in section "Calibrating the endstop positions, delta radius and homed height". The description of how to adjust the settings in firmware is geared to RepRapFirmware on Duet electronics, so it will be different in Marlin, however the general principle is the same.

Perhaps more directly useful to you is the description of how to calibrate the Think3DPrint3D Mini Kossel kit here [docs.google.com].

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

To close this one out, the standard process did work. I didn't do the math originally and should have. I was making a significant change and seeking a difference. I went to far and it crashed into the bed. Once I did the math it came out fine. If I recall it was [desired distance / actual distance] * [delta diagonal] which then gave the new diagonal it should have been.

So that part is done. Now to the printing issues.

Edited 1 time(s). Last edit at 03/14/2015 06:27PM by Dark Skunk.