Calibration print

K8200 3D Printer K8200 - Your printer, creations, and ideas for imp
1

I found this on thingiverse:
thingiverse.com/thing:53668/#comments

This is the explaination:

This works pretty well. Within 2 prints I had my printrbot jr fully calibrated to .1mm tolerances. Well enough to print an iPhone case that fits. I’ve included the STL and a sketchup file if you want to mess with it.
In case your new to calibration, I’m going to break it all down for you:

Usually in the custom Start g-code area in Slic3r you’ll have 4 calibration values for X,Y,Z and E for Extruder.
If you don’t have custom g-code in there, you’ll want to add it. Here is what mine looks like for a printrbot jr.

G28 ; home all axes
M92 X63.0242 ; calibrate X
M92 Y62.7669 ; calibrate Y
M92 Z2071.4959 ; calibrate Z
M92 E650 ; calibrate E

You are welcome to use these as starter values. Just paste them into the custom g-code section of slic3r. (note: direct drive extruder starter value should be M92 E100)

To calibrate you first print this puppy out. When it’s finished you’ll want to measure it with a caliper, digital preferred. How close to 50mm for X and Y are you? X is the side to side when facing your printer. Y is the back and forth. And Z is the up and down. Bust out your calculator and run this calculation:
(Don’t be scared, it’s easy)
First multiply the current value of X by the length in the STL, 50mm. Then divide that number by what you actually measured. The result is your new custom-gcode value for X.
So in my case, say I measured the X of my print to be 48.25mm, I would do

63.0242 X 50mm = 3151.21 (current X value times the value in the model)
THEN
3151.21 / 48.25 = 65.3101 (result divided by my measurement)
So in my custom G-Code I would replace 63.0242 with 65.3101
Do the same for Y, and it works for Z too!

Re-slice by exporting it to G-Code and reprint that bad boy. Check your measurements, and do it all over again if you are off by more than .5mm

I hope this has been helpful, I know it’s tough when you are first getting started.
For speed, I printed this object with 1 perimeter, 1 shell top and bottom, and infill of .1 (10%)

How does this work? You have to fill in values in the Gcode and replace them with calculated values later. But after slicing another object these values will be replaced no? I would like to try this but don’t really get the principle…

Greetings,

Kenny

2

Hi Kenny,

the RepRap wiki notes on M92

This is good for calibrating, true, but after you find the right values, you should store them in the EEPROM (or by reloading an adapted firmware), so they will be used by the printer independent of the Slic3r profile in use.

As for the calibration itself: there should be not that much to do. Both X and Y axis use timing belts, so as long as the pulley is fixed to the motor the ratio of steps per mm is fixed, too. Standard value is 64.25. With smaller pulleys, e.g. GT2x20 or T2.5x16, it’s 80 steps per mm. In order to check those values, you don’t need to print anything, just attach a ruler to the frame and test whether some random movement commands (e.g. +100 mm, -50 mm, etc.) are processed correctly. Z axis is similar, the pitch of the rod doesn’t change, so the value of steps per mm should be fixed just as well. Pitch 1.25 equals 2560 steps per mm.

Repetier Host has a built-in calculator (Tools menu) for belts and leadscrews, and there’s of course also Josef Prusa’s website.

As for the E steps per mm, there have already been a couple of posts. Calibrating is possible by extruding a set length of filament. To get an accurate result, you need to make sure that the hotend is at printing temperature and, more to the point, you should use an extrusion speed close to the setting during an actual print. If you set the speed too high during extrusion, you’ll have high pressure in the hotend, causing slip between hobbed bolt and filament.

Cheers,
kuraasu

3

Ok, thank you :wink:

4

Today I started to check the calibration of the extruder and yes, there is a slight difference. When extruding 100mm, only 97mm is extruded (very small difference but I want it perfect offcourse). So I wanted to load this into the EEPROM. But all the values from the EEPROM are 0. is this normal? And if so, how can I restore these values to the correct ones?

greetings,

Kenny

5

I have been testing allot of settings the last few days but never with allot of changes (But I’m already quite pleased with the results I get). But there are a few things that got my attention.

The first one:
I have been trying to print some things that have very shallow overhang. Like when you embross a Logo in a print like this:

When printing it I get no relief at all, its just flat, so the filament hangs down and by the flatness/glossiness of it, it even looks like it is printed straight onto the bed:

2whiadc.jpg1600×1200

You can see the logo because of the orientation of the lines, but you can’t feel the relief/height difference at all. At first it did not bother me that much because it looked nice after all. But with a second print I wanted this relief and got the same result:

3135f9j.jpg1600×1200

On this print the height difference was almost twice as high and still got this result.

Btw I can print normal overhang without much problems, sometimes 1 strain of filament fails the first time but then it’s ok. So it got me thinking: Is it a to high bed temperature that causes this? I’m printing PLA on a mirror with hairspray at 50 degrees. Or is it the fan that is not cooling enough. It seemed that the fan was even not turning at all when filling in the logo. (because of the height difference not being high enough to trigger the fan?)

The second one:
I printed out 2 10mm calibration cubes (0,25mm and 0,15mm layer height) and something else came up which I thought was weird, as you can see, I got very little Z-axis wobble with the standard rod (but with fixed motor). But I see allot of X and Y wobble… What can be causing this? On this print the bed was not vibrating when laying the outer layers or anything.

2587685.jpg1600×1200

Btw: sorry for the upside-down pictures, when uploading to tinypic this happens, don’t know why…

Greetings,

Kenny

6

nobody that can help on this last post? In particular the question about the shallow overhang

7

You’re trying to bridge between two points that are just one or two layer heights away from the surface. That’s bound to go wrong. The plastic will always sag a bit before the thermal contraction kicks in.

8

Ok, Too bad that this is not possible :slight_smile: I will have to alter the design then :slight_smile:

9

I’m currently trying to calibrate the extruder again since I have a new Hotend (E3D v5) and i’m getting a really weird behavior:

Standard steps = 600
First calibration run extruded 86mm instead of 100 so this becomes 697,7 steps

Trying again:
extruded 110mm instead of 100, new calibration steps = 634.3
then extruding 80mm instead of 100…

What is happening?

I pulled out the filament to check for skipping but it looks like the filament is gripped just fine by the hobbed bolt

10

Are you using the Manual Control in Repetier Host? What’s the Speed [mm/min] setting?

11

100 mm/min

12

Try 60 or 50 mm/min, for the small nozzles 100 may still be too fast. Extruding 100 mm then will take some time, but you can test with shorter moves first for a rough value and, after that’s ok and reproducable, go for more precision with a longer extrusion.

If you want to have a reference value to estimate slip, test what setting you get without the hotend (i.e. without load).

13

[quote=“kuraasu”]Try 60 or 50 mm/min, for the small nozzles 100 may still be too fast. Extruding 100 mm then will take some time, but you can test with shorter moves first for a rough value and, after that’s ok and reproducable, go for more precision with a longer extrusion.

If you want to have a reference value to estimate slip, test what setting you get without the hotend (i.e. without load).[/quote]
Before you answered I retried a few times and for some reason now it happened correct for a few times in a row. I will check it later with a slower speed to make sure it’s still the same. It came out on quite a big difference. First it was 600 steps, now it is 704 steps.

First thing I notice is that the first layer is spread out much more (probably normal since it extrudes more). First I had the problem with slicer that the first layer was not filled up completely. Now I have a problem that after laying the brim around the object the extruder does not retract and starts out with a kind of large blob on the first place were it starts on the print… (printing with Cura). I’ll see if this behavior continues with other objects (for now I’m printing the ultimaker robots to test some settings)

Edit: now I noticed it does retract between the brim and the first layer and it still lays down a large blob to begin with

14

Tried another object, exact same thing. the printer lays down the outer brim, retracts, spits out a big blob and then starts the print completely normal without any problems from then on. But one corner is ruined for the upcoming 4 or 5 layers

(Problem started after calibrating the extruder from 600 steps to 704 steps)