Bed leveling

[gabrielk]

Once and for all. I want to know how to make proper bed leveling. I did it following Geneb manual, but lately i read abiout different approaches. Since there is many information in many threads, not all of them are overlaping.Please correct if I'm wrong or I missed something.
Briefly: measuring one tower, adjusting second and third, adjusting center, measuring z lenght

I'm using Repetier firmware with Repetier host. I have Phebe I. So to fit I made a circle of 80mm radius. Just like on the picture below.
So i have 3 calibration points for 3 towers:
Z= x0 y80
X= x-69.3 y-40
Y= x69.3 y-40

I will use method with thin sheet of paper.

Calibration steps:
1. I put in firmware bigger values fo Z LENGHT for example 370.
2. Heat up nozzle and bed to working temperatures for me it is 225 nozzle and 70 bed. They can warp and change leght.
3. Home machine using G28

4.Z Tower calibration:
4.1 Put in repetier host: G1 X0 Y80 Z20 F1200
4.2 Manually lowering the nozzle until it starts to snag the paper. Here comes the first question, what about errors in repetier host, should i move it down using only .1 mm or also 1mm button. And what about M114 command, sholud I use it somehow, when and how.
4.3 Repetier host shows in snagging position 363,00mm
4.4 Home machine using G28

5. X Tower calibration:
5.1 Put in repetier host: G1 X-69.3 Y-40 Z20 F1200
5.2 Manually lowering the nozzle until it starts to snag the paper. For example Repetier host shows 362.7 in snagging position.
5.3 Adjustin the screw. In this case screwing it in about .3mm down.
5.4 Homing machine using G28
5.5 Put in repetier host: G1 X-69.3 Y-40 Z20 F1200
5.6 Manually lowering the nozzle until it starts to snag the paper.
5.7 Repetier host shows in snagging position 363,00mm, if not adjust screw more. Each time home machine and put corresponding coordinates after homing.
5.8 Home machine using G28

6. Y Tower calibration:
6.1 Put in repetier host: G1 X69.3 Y-40 Z20 F1200
6.2 Manually lowering the nozzle until it starts to snag the paper. For example Repetier host shows 363.2 in snagging position.
6.3 Adjustin the screw. In this case screwing it out about .2mm up.
6.4 Home machine using G28
6.5 Put in repetier host: G1 X-69.3 Y-40 Z20 F1200
6.6 Manually lowering the nozzle until it starts to snag the paper.
6.7 Repetier host shows in snagging position 363,00mm, if not adjust screw more. Each time home machine and put corresponding coordinates after homing.
6.8 Home machine using G28

7. Check if every tower have 363 in snagging position, by puting corresponding coordinates for each tower
8. Home machine using G28

9. Checking if heated bed warps in the center up or down
9.1 Put in repetier host: G1 X0 Y0 Z20 F1200.
9.2 Manually lowering the nozzle until it starts to snag the paper. For example Repetier host shows 363.2 in snagging position. Sholud be 363mm
9.3 Adjusting DELTA_RADIUS in firmware. Uploading changed firmware.
9.4 Home machine using G28
9.5 Put in repetier host: G1 X0 Y0 Z20 F1200.
9.6 Manually lowering the nozzle until it starts to snag the paper.
9.7 Repetier host shows in snagging position 363,00mm, if not Adjusting DELTA_RADIUS in firmware. Uploading changed firmware. Each time home machine and checking snug level.
9.8 Home machine using G28

10. Measuring Z Length
10.1 Put in repetier host: G1 X0 Y0 Z20 F1200.
10.2 Manually lowering the nozzle until it starts to snag the paper. For example Repetier host shows 363 in snagging position. That is your Z LENGTH.
10.3 Put new Z lenght in fimware for this example it is 363. Upload.
DONE !!!

Please correct me if I make mistake somewhere, or i missed something.

Still don't know what M114 can give me in calibration process, and what about using Reptier host buttonsm since I read they not work very well and by using them you gathering errors.

I hope this can help somebody.

[daftscience]

Nice write up. I think this will be helpful to a lot of people, especially the delta_radius part.

I used the coordinates from Mhackney's build.
http://forum.seemecnc.com/viewtopic.php ... t=70#p5124

G0 X-77.94 Y-45
G0 X77.94 Y-45
G0 X0 Y90

I also found that when moving the nozzle down to the paper I got better results by sending a G0 command instead of hitting the down arrows in repetier host.

The M114 command is the reason why I don't use those arrows. Basically, the Z= value in repetier host isn't accurate. After moving the nozzle down look at that number. When you send M114 it will update to the actual value. Without that it was tough to get z=0 to the right height.

[Polygonhell]

One correction when adjusting the DELTA RADIUS you are not correcting for the bed warping when it heats up.
You must do the correction with a flat bed (clamp a piece of glass on it), you are correcting for errors in the delta geometry that cause motions to be none linear.

Linear motion of the delta carriages is converted to circular (technically conical) motion of the actual hotend, this is compensated for in the firmware by breaking down a requested move into small linear segments that approximate a linear path. The math to do this is very sensitive to the actual size of the components, adjusting the DELTA RADIUS, is just matching the math to your printer.

[mhackney]

That's a very important point Polygonhell. The Onyx is so large and not that stiff and only held at its perimeters so unless you have a flat surface like glass or my aluminum plate on it that itself does not warp and establishes a true plane, the calibration likely won't be effective. At 80°C my Onyx is .5mm bulged in the center. And the bulge is not symmetric as measured with a DTI. There is a "hill" between the center of the plate and the electronics pad area. The aluminum plate I made (and will make available for sale soon) eliminates this bulge.

cheers,
Michael

[gabrielk]

with my changed end stop screws i manage to position 3 points on the circle and the center with accuracy of 0.04mm . I used LCD controls to move nozzle down.

[chrishamm]

Check your alum extrusion to see if it a perfect triangle. Mine was out .06" on the one side.
I could get all four points zeroed with in .001" but as soon as I went between the alum extrusion
it was out up to .04" This was tested with a dial indicater.
I think you really need to check 6 points of the outer circle and the center to truely level it.
I am guessing they had a bad batch of lazer cutting.

[gabrielk]

.001" is 0.0254mm so my 0.04mm is not so bad. I think it's enough for now. Thank you all for helping me.

[aehM_Key]

Got mine also calibrated to "0.02mm". Thank you.

I had to change the additional offset in DELTA_RADIUS from +0.63 to -0.15 .
I glued my heatbed to the supporting plate, and I think it helped a lot in keeping it even. (And probably caused, that I can not heat it to more then 70°C..)

(I also used http://forum.seemecnc.com/viewtopic.php?p=5403#p5403 )

[Av8r RC]

Where can I find this DELTA_RADIUS value?

This is the only thing I can find that has DELTA anything

// Make delta curves from many straight lines (linear interpolation).
// This is a trade-off between visible corners (not enough segments)
// and processor overload (too many expensive sqrt calls).
#define DELTA_SEGMENTS_PER_SECOND 200

// Center-to-center distance of the holes in the diagonal push rods.
#define DELTA_DIAGONAL_ROD 269.0 // mm

// Horizontal offset from middle of printer to smooth rod center.
#define DELTA_SMOOTH_ROD_OFFSET 198.25 // mm

// Horizontal offset of the universal joints on the end effector.
#define DELTA_EFFECTOR_OFFSET 33.0 // mm

// Horizontal offset of the universal joints on the carriages.
#define DELTA_CARRIAGE_OFFSET 35.0 // mm

// Effective horizontal distance bridged by diagonal push rods.
#define DELTA_RADIUS (DELTA_SMOOTH_ROD_OFFSET-DELTA_EFFECTOR_OFFSET-DELTA_CARRIAGE_OFFSET)

// Effective X/Y positions of the three vertical towers.
#define SIN_60 0.8660254037844386
#define COS_60 0.5
#define DELTA_TOWER1_X -SIN_60*DELTA_RADIUS // front left tower
#define DELTA_TOWER1_Y -COS_60*DELTA_RADIUS
#define DELTA_TOWER2_X SIN_60*DELTA_RADIUS // front right tower
#define DELTA_TOWER2_Y -COS_60*DELTA_RADIUS
#define DELTA_TOWER3_X 0.0 // back middle tower
#define DELTA_TOWER3_Y DELTA_RADIUS

[gabrielk]

line with #define DELTA_RADIUS

for example for me it is:
#define DELTA_RADIUS (DELTA_SMOOTH_ROD_OFFSET-DELTA_EFFECTOR_OFFSET-DELTA_CARRIAGE_OFFSET+0.53)

so i'm adding 0.53

[daftscience]

If you're using the repetier firmware it will be called PRINTER_RADIUS

[aehM_Key]

It does not matter, if you add the offset at DELTA_RADIUS or PRINTER_RADIUS, because DELTA_RADIUS is calculated out of PRINTER_RADIUS and PRINTER_RADIUS is not used anywhere else.

[September]

G'Day all,

First time poster here, I recently purchased myself a Rostock MAX (last few weeks), and have spent many hours trying to calibrate the hotend movement, I think I have finally achieved this, now to calibrate the filament settings anyway. I found this image today, it's based on the Cerberus, which by the looks of it is very similar to the Rostock design. It shows what the different variables represent and may help others calibrate their Rostock. Especially if you are going to use custom parts.

[img]https://lh5.googleusercontent.com/-wqbD ... 008+AM.jpg[/img]

Cheers

[kbob]

That diagram is worth 1,000 forum posts! Thank you!

[Polygonhell]

It's worth noting that only 2 dimensions are actually used directly in the computation, the DELTA_DIAGONAL_ROD and DELTA_RADIUS. The other values are just there to compute DELTA_RADIUS from more easily measured values.

[dgirard]

Having a problem here after following these directions and the ones from here http://minow.blogspot.com.au/

In preparation for working on an auto-leveling scheme (still waiting on my rare earth magents to arrive), I thought I'd check everything. I've been having success printing smaller objects in the center of the print bed, but problems working outwards...

I've got the towers all equal, and the center equal with the towers..all are within I would say .05 of a mm...close enough for now.

HOWEVER...when I try to draw a radius, I see that the print head lifts up (domes) as it moves *between* towers along the radius...it appears to be different between each tower a bit, and it's fairly significant near the edge of the build platform (it lifts up by 2-3 mm!)...enough to make it impossilbe to print in this space.

I'm at a bit of a loss for how to troubleshoot this one.

FYI: I've installed the tricklaser carbon fiber arms, and measured them to be 1mm longer than the stock ones (I updatd the firmware setting for delta_diagonal_rod).

I'm running the production (not dev) build of the repetier fimware from seemecnc, and have made the adjustments for my v1 rambo.

Any ideas where to start on this?

[dgirard]

OK...I think I've got it...

I found in another thread that apparently the # of teeth in the pully's has also changed since the indiegogo campaign (I don't remember seeing that before...)

I changed the #teeth setting in the firmware from 20 to 15 and recalibrated...now it's not lifting between towers...

It amazes me that I was able to get some pretty darn good prints with this setting wrong...

[626Pilot]

Usually when it lifts or dips between the edge and perimeter it would be a matter of changing PRINTER_RADIUS. I found that increasing this by .05 would raise the print head by about 1 thou at the perimeter, relative to the center. I like to use a $10 depth gauge (accurate to 1 thou) and the SeeMe pen holder, although you can get the same thing for free off Thingiverse as long as your printer works well enough to print it.

I got the Hall-Θ endstop board from here: http://mauk.cc/webshop/index.php?route= ... duct_id=50
Then there's the holder for the endstop: http://www.marcoantonini.eu/doku.php?id ... :bed_level

Once I have the metal rod, magnet, and screws, I'll do a write up. I think a couple other people are doing so as well.

Edit: I will need to redesign the Hall-Θ holder to work with the Rostock MAX effector, which has different screw spacing than the (Kossel?) platform that guy is using. I'm going to see if I can redesign it to work with nozzles from different vendors, which have different heights between the nozzle tip and the effector. One size fits all, hopefully. Also with Imperial measurement stuff (3/8" instead of 5mm magnet cubes, and metal rods in sizes you can get from McMaster-Carr.)

[Eaglezsoar]

Usually when it lifts or dips between the edge and perimeter it would be a matter of changing PRINTER_RADIUS. I found that increasing this by .05 would raise the print head by about 1 thou at the perimeter, relative to the center. I like to use a $10 depth gauge (accurate to 1 thou) and the SeeMe pen holder, although you can get the same thing for free off Thingiverse as long as your printer works well enough to print it.

I got the Hall-Θ endstop board from here: http://mauk.cc/webshop/index.php?route= ... duct_id=50
Then there's the holder for the endstop: http://www.marcoantonini.eu/doku.php?id ... :bed_level

Once I have the metal rod, magnet, and screws, I'll do a write up. I think a couple other people are doing so as well.

Edit: I will need to redesign the Hall-Θ holder to work with the Rostock MAX effector, which has different screw spacing than the (Kossel?) platform that guy is using. I'm going to see if I can redesign it to work with nozzles from different vendors, which have different heights between the nozzle tip and the effector. One size fits all, hopefully. Also with Imperial measurement stuff (3/8" instead of 5mm magnet cubes, and metal rods in sizes you can get from McMaster-Carr.)

It looks as though the automatic bed leveling requires Marlin firmware, is that correct?

[626Pilot]

It looks as though the automatic bed leveling requires Marlin firmware, is that correct?

I think so. Someone in the other thread about auto leveling said he was going to try to port it to Repetier.

[626Pilot]

...Then again, here is this: https://github.com/Crash1/Repetier-Firmware

It's a fork of Repetier, a few months old. It seems to have the delta code in it. I don't know whether or how it works.