Calibration square assistance, height is too short

So assembling my first multi piece build last night, I found that things were not going together cleanly.

I found some 25x25x25mm calibration squres and ran a test.

Length width height
MC: 25.3 x 25.3 x 22-23mm (don't recall exact number, but off quite a bit)
Simplify3d: 25.5 x 25.6 x 22-23mm (don't recall exact number, but off quite a bit)

Obviously I'm slight over on the sides and quite a bit too short. How do I go about fixing this as it appears to occur in both programs? I am also going to try a different type of PLA to see if it's possibly specific to the material (using an amazon purchased glow PLA).

 

Reproduced with the stock material, so definitely a setup issue. I will try to look through some of ziggy's threads to find info.

 

That Z is way off. Do you know what version of firmware is running? If you look in the configuration.h file this line (near line 450):

#define DEFAULT_AXIS_STEPS_PER_UNIT {80,80,2267.72,753} // default steps per unit for RoBo 3D R1
​

contains the X, Y, Z and Extruder calibration constants. The 2267.72 value for Z should correspond to the threaded rod pitch and stepper microsteps per rotation for your printer. It's really a very simple relationship and unless you for some reason have a different value there it is very unlikely to be the problem. That value should be the same for everyone's printer unless they swapped the threaded rod out for one with another pitch. BTW - my extruder value (the third number) is different than the default because I recalibrated it for my particular grooved hobbed bolt.

What slicer are you using and did all the top layers finish printing?

The X and Y differences could be due to some over extrusion and bulging of layers and probably also Z ribbing as described in Ziggy's excellent thread and document. It is hard to get a precise X & Y measurement because if this. You will probably find that the X and Y measurement retain that error even if you print something that takes the entire length of the bed.

 

Thank you for the feedback.

I'm running the latest firmware available (MC updated it right away when I started the printer for the first time). Ironically after it updated, it has never again displayed the firmware and always states "detecting target firmware". As a new guy, I've seen people referencing config files, which I assume are on the device itself, that I have no idea how to access. I have not yet looked into that, more reading to do

Slicer is slic3r and integrated simplify3d, so same issue, two sets of software/slicing. The top layers look slightly thin when not using fill.

 

If you are curious and wanted to look at the firmware grab the zip file in the first post here: http://forums.robo3dprinter.com/index.php?threads/marlin-firmware-update-3-14-14.1857/ This is the non-autolevel version but is very close to the one you are probably using. You can find your version elsewhere on the forum. Open the configuration.h file in a text editor.

Basically when you do a firmware update the configuration.h file and all the others get combined or compiled or built together and loaded into the flash memory in the Arduino based microprocessor in the printer.

 

Does anyone know if there is a way to read back the stepper calibration constants in Mattercontrol possibly by sending some G code?

I doubt this is the problem but worth checking.

 

I guess you could try a test to override the Z steps per unit with this G code in your startup code:

M92: Set axis_steps_per_unit
Example: M92 Z<newsteps> Sprinter and Marlin

Allows programming of steps per unit of axis till the electronics are reset for the specified axis. Very useful for calibration.​

from: http://reprap.org/wiki/G-code

 

ok when I started mine and calibrated it with paper it was dead on accurate to where I wanted it... but for the life of me I cant find where I can redo the paper thing and set it up like the original calibration I had (after doing the paper thing) I needed to move the wires from the left side and change the way the wiring fed into the extruder area... now I cant get it to sit as level as it was.... How can I go back and do the paper testing all over again??? mc doesn't allow me to use the manual controls on the three points....

 

now I am slightly off in the Z direction and need that to be consistently level from right to left and its slightly off.... how can I manually recalibrate it?

 

Well, flashed with v3 (6.10) firmware manually.

25.20 x 25.15 x 22.40

Off to try to figure out exactly what 2267.72 really means...

 

See the leadscrew calculator here http://prusaprinters.org/calculator/ And use 5/16 pitch.

200 (steps/rev) *16 (microsteps/step) /(1.41111 (mm/rev)) = 2267.72 (microsteps/mm)

Also see the optimal layer height calc. None of this explains that difference of 25-22.4 mm. The calculator says that if you choose 0.3mm layer height the error will be 0.28mm over 25 mm. You can try one of the suggested layer heights like 0.2046mm or 0.3034mm.

Probably something else is going on.

 

You may want to check the screws on both the z couplers to see if they are tight enough.

 

Hrrrm..

2267.72 / 2560 = .885
22.4 / 25 = .896

I saw in another thread that they were thinking of going to metric. Methinks they might have and the config is wrong

 

Yeah.... upgrades good... not telling the consumer base... badddddd

Printer received 8/11/2014

 

Nice catch, I didn't think of that. What thread was that mentioned in. Did someone from Robo3D actually say that or just a user? You should be able to make the thread pitch measurement. If they did make the change they really should have posted on the forum and announced a firmware or constant change. Very typical of their poor communications.

 

Found this thread.

Mike Kelly
http://forums.robo3dprinter.com/index.php?threads/z-axis-ribbing-issues.2492/page-3

 

Thanks, I just read Ziggy's comment.

 

Thanks to this thread I've solved my z height issue!!!!

Long story short: My R1 is 2 weeks old, shipped with 8MM rods and the "Steps Per mm Z"setting was wrong. While it took 2+ hours of research, the fix was this simple: Open Matter Control -> Configuration Tab -> EEProm Settings button -> Change Steps per mm for z to 2560.00

My z height is now 100% correct! A huge thank you from me to this community for it's excellent care of each other

Next step, solve many issues preventing me from printing this: http://www.thingiverse.com/thing:440493

 

Make sure you have a sufficient amount of print cooling enabled. All those small sharp angles will tend to curl up if the plastic does not cool right away. If you are using ABS you will want to disable cooling for the lower section and have a lower max fan speed than you would use for PLA. For PLA you probably would have the max fan at 100%.