Calibration issues

G'day all,

I've recently (as in the last 2 days) got my very new and shiny printer (Ord Bot hadron) working and now experimenting to learn how to use and how to build these things.
However, I'm being stumped by 2 problems, that no amount of googling seems to be able to fix. (I think after the last test, I've got an idea about how to fix the first one but I do want everyone's advice).

Problem 1:
Things are printing very nicely, but they are the wrong size. I've printed somewhere in the order of 20 0.4mm and 0.5mm calibration squares in the last two days, and they are all turning out slightly the wrong size. In an attempt to localise the problem I built a 100mm calibration cube and printed it, also with similar results.
The 20mm calibration squares come in at 19.8mm square and the 100mm square comes in at 98.3mm, or 99% and 98.3% respectively. (A 8mm bar that I printed for a filament stand came in at 7.8mm or 97.5%, and a circle printed at the time part of the same object should have had a diameter of 40mm which came in at 38.9mm or 97.25, though take this with a grain of salt because I'm not confident that I got the exact diameter)

My guess is that this is the steps per unit in firmware, as I changed and calculated it this morning, based on a 20 tooth pulley and a 1.8degree stepper and 2mm pitch belt, coming up with 80. Up till that point I was using the default steps in marlin of 78.7402, and I measured the 0.5mm calibration cube this afternoon, and I believe it should be 22mm and it comes in at 22.25mm, so my gut feeling is that i need to alter the steps value to be somewhere in the middle of those two.

That brings me to my second problem, which I've got no idea where to go with.
All the walls are the wrong size.
The 0.4mm calibration cubes come in at 0.61mm, the 0.5mm cubes at 0.59 and the 100mm cube which should have been 1mm, at 0.7mm.
I've calibrated the e steps and measured and averaged my filament, but none of this seems to have fixed it and would welcome any assistance.


For information my setup is as follows:
Printer: Ord Bot hadron
Controller: Ramps 1.4 with RRD Full Graphic LCD
Slicer: Slic3r
3d Design program: FreeCAD or OpenSCAD depending on how I'm feeling.


(Also if anyone can suggest how to stop it oozing when its at temperature before printing that would be good, I've added a retraction value, but it doesn't seem to have helped)

Thanks in advance for your help
Knight

to set your steps use this formula
(expected size)/(actual size) * current setting = value for steps

so
20mm cube / 19.8 acual cube = 1.0101
1.0101 x 78.7402 (current step setting) = 79.5356 <- enter this value in your steps

print and remeasure and adjust again until you get the result you are expecting

Hi,
As Cat.Farmer said, (expected size)/(actual size) * current setting = value for steps
But, you don't have to print and measure you can just move the X, Y & Z from the printer interface to 100mm from home and measure the actual movement then apply the formula.
mostly the X & Y will be the same value but the Z is different.
for the E, Just extrude 100mm and measure it on your caliper, if less increase the nozzle heat or the printing speedor little of both.

I'm still new to the reprap, please guys correct me if i'm wrong.

Good luck

Yasser
Prusa i3 8"

Quote
yasrkasem
Hi,
As Cat.Farmer said, (expected size)/(actual size) * current setting = value for steps
But, you don't have to print and measure you can just move the X, Y & Z from the printer interface to 100mm from home and measure the actual movement then apply the formula.
mostly the X & Y will be the same value but the Z is different.
for the E, Just extrude 100mm and measure it on your caliper, if less increase the nozzle heat or the printing speedor little of both.

I'm still new to the reprap, please guys correct me if i'm wrong.

Good luck

Yasser
Prusa i3 8"

Not only do you not have to print, you shouldn't print. The correct way to calibrate is from the movement of the printer. Period. Since you are seeing that your walls are coming out thicker than intended, you can see how this is going to add a bias to the measurement of your printed parts.

Calibrate your X, Y, Z, and E. Then play around with you slicing settings you get the walls better calibrated. However, I wouldn't focus too much thin walls. I've found that Slic3r calculates extrusion with the assumption that the extruded plastic forms a rectangle (layer thickness X extrusion width). This doesn't account for rounding/bulging edges.

Okay, thanks for confirming that, X, Y, Z and E seem to now be calibrated to within a measurable tolerance.

The walls are still not correct. And PvtDBJackson, while I note what you said, I need to fix it (now if that means I can't use slicer, so be it, but it can't just be me with this problem, which leads me to think its a problem with my printer)
The reason I am concerned about the walls is it is affecting my prints. any parts that need to fit together have issues. For example I am currently (as in printer is buzzing away in the background) printing a fan mount, unfortunatly, the mount, which should have M3 screw holes, has screw holes measuring 2.3(ish, hard to measure while the printer is working - its doing the second part, hence the ability to check it). I don't mind drilling them out, but it does present me with a problem if this is going to effect every other print I want to fit together with something.
Its fine for my prints, I can build a tolerance into my designs, but for anything I get off thingiverse or another site, it becomes problematic.

For the record, I printed both calibration squares at the same time this morning, both the 0.4mm and 0.5mm thin walls came in at 0.65mm. Which causes me even more confusion... (though that being said for some random reason slicer printed one of the squares with rounded corners, don't mind, but might give someone else a hint as to where my problem lies)

Ok so don't print.. when i did mine i strapped a pencil to my x carraige and taped a peice of paper to the bed. Moved it, measured with my caliper, and did the math.

Quote
cat.farmer
Ok so don't print.. when i did mine i strapped a pencil to my x carraige and taped a peice of paper to the bed. Moved it, measured with my caliper, and did the math.

Makes logical sense, and I'll remember it next time I need to calibrate X/Y. But I don't understand how I can use this to fix the wall thickness issue. Would you be able to elaborate?

None of this will fix wall thickness. This is just to make sure that when you tell your bot to move 10mm, it moves 10mm. Wall thickness, and I'm sure I'll be corrected again, is your extrusion rate and layer height. You also need to calibrate your extruder.

Well it seems to be behaving now.

I checked the X and Y values, both correct.
I checked the extruded value, made a very minute change, now as close as I think I can possibily get it.

Extrusion multiplier set at 0.83

Wall thickness is now correct. The calibration cube comes in marginally smaller (X, Y and Z) than it should be, but when I printed a set of nautilius gears they all fitted together perfectly, so I'm going to call it at this point

Thankyou for you assistance, now time to build....something