Spreadsheet for calibrating printrbot reprap or probably any 3dprinter – Beta 2 with some scripting: https://docs.google.com/spreadsheet/ccc?key=0An3o-3QkLawpdHkxYUpsRFZTTkNkbHRWSW5vLVJYZmc
This topic has already been discussed endlessly - it is simply wrong to mess up your steps/mm to “fix” other inaccuracies.
Do not “calibrate” your printer via this method.
Regardless, many people do this and some vendors (e.g., Printrbot) suggest it and this is a tool to make it easier.
Well then, i guess the Printrbot guys don’t have a clue either.
The spreadsheet indeed makes setting up your printer with inaccurate values easier.
Have never understood how anything but the extruder should need this kind of calibration. Wade extruder use printed gears and a DIY hobbled bolt so there is some room for calibration …
Just a quick google shows that they aren’t the only one suggesting this. I understand the theory (which is why there is an ideal column) but I also understand the practice. But if you feel so strongly perhaps you should go pen corrections to the pages on the RepRap wiki? Or point folks to some other resource that would help the rest of us. In the end that’s all I’m trying to do is help others to the best I can.
I am actually surprised myself that the Z requires any calibration. On my machine with timing belts, I don’t find it very surprising. Of course, I’m just an electrical engineer so not really my specialty.
Not all of these are related but: https://www.google.com/search?q=calibrating+steps+per+mm+reprap&aq=f&oq=calibrating+steps+per+mm+reprap&aqs=chrome.0.57j62l3.5295&sourceid=chrome&ie=UTF-8#hl=en&tbo=d&spell=1&q=calibration+steps+per+mm+reprap&sa=X&ei=v__3UIaZKeqE2QWxpYAI&ved=0CDEQBSgA&bav=on.2,or.r_gc.r_pw.r_cp.r_qf.&bvm=bv.41018144,d.b2I&fp=df19f1b4a17a9572&biw=1424&bih=1001
Meanwhile if the spreadsheet offends you, I’d challenge you to point the rest of us to something specific and constructive. Otherwise, the idea that you don’t agree with the premise is noted.
X, Y and Z are always set to the calculated values. (Hint: combining imperial mechanics with metric firmware calls for trouble. Use metric only for best results.)
This is what you do to calibrate your extruder: http://richrap.blogspot.de/2012/01/slic3r-is-nicer-part-1-settings-and.html
For fine tuning, measure the width of the skirt (at least 15 layers for best results) slic3r generates and compare it to the “single wall thickness” it includes in every .gcode. Use the ratio you get to tune the filament multiplier in slic3r’s filament settings.
So far this procedure gave excellent results on every machine i’ve used it on.
Any remaining inaccuracies in the parts are simply a result of mechanical inaccuracies like backlash, inaccurate guides, an eccentric hobbing on the filament drive… These can not be calculated away but need to be resolved by carefully improving your printer mechanically.
Again this is one of those things that was in vogue if you will awhile ago but not so much anymore. Its out there including the wiki b/c thats the whole idea behind free and open information.
What Ive found if you start with calibrating steps/mm this way is that you will never get a 100% accurate sample reading. Usually your smooth rods are a little off parallel, a bearing isnt doing something right, or some other mechanical problem is causing your linear motion to not be consistent.
So if you start with a dead set steps/mm as calculated by the standard formula, steps = motor_steps_per_rev * driver_microstep / belt_pitch / pulley_number_of_teeth, then you can start to look for other problems in your linear motion if your prints are not accurate enough.
And for the record, printrbot has up until recently been using printed pulleys. Any printer with this setup and the hard as hell to tension T5 belts are going to have problems here so either get a better belt system or live with a little inaccuracy.