I need some serious help

[LuckyImperial]

OK confirmed! From z0 to z100, I only measure it moving 80mm. So to fix this I afjust.....?

There's a fancy equation you can do to figure out what you need to set your Steps Per MM to, but you can always just increase it from 80 until you get what you need.

[Nickschaos81]

I think I got it I'll post pics of the print I'm running at this very moment. I have learned sooooo much from you guys, so grateful

[Nickschaos81]

https://www.dropbox.com/s/d89247ymvg...31746.jpg?dl=0
https://www.dropbox.com/s/9330fvc76l...25158.jpg?dl=0
https://www.dropbox.com/s/5qkh0149m6...25124.jpg?dl=0

I think my prints are improving. You guys agree? lol especially from where I started. So it seems after I adjusted my steps other calibrations were thrown off lol , so the little 2 piece sub there has some bowl shape too it , I checked after print and yes those calibrations have been thrown off. No big deal.
So here is my game plan and a couple more quick questions.
Im going to double check steps per mm are as close to dead on as possible
going to do a manual bed leveling and then adjust my diagnol and horizontal settings again to elimenate dishing, or bowling w/e its called
then double check my total z height
then triple check all points to make sure we are homed in, including e Steps which brings me to my first question
1. Increase or decrease esteps to reduce how much extrudes ( I think decrease but i could have sworn i read increase some where)
2. how do i check to make sure steps per mm is correct on x and y ( mark bed , move head and measure? and if so what do i adjust to fix it if its off?)
3. How do i post pics in this forum so i dont have to use dropbox anymore?
4. anything else you guys can think of ?
5. autolevel? should i use it ? lol and if so , how?

Again , thanks you guys your awesome! wouldnt be were i am with out yalls

[Alibert]

OK, you are making good progress here. Sorry about my delayed response but I am in a different timezone and usually (!) asleep when it is evening over there....

Repetier does use different naming conventions, thanks LuckyImperial for the corrections.

I think you have already figured it out by yourself, but for the other readers this explanation:

When you told the firmware to go from X0 Y0 Z0 to X0 Y0 Z100 with (according to the firmware settings) 80 steps per mm, the firmware moved all three towers by 80*100=800 steps. As the belts in reality only moved 80mm, the actual steps_per_mm = 800/80=100 steps per mm. If you change that in the firmware you should be allright (apart from the manual re-calibration).

As to your question:

1) You can calibrate the extruder too. Just take the Bowden tube off the extruder. Put a mark on the filament referencing to a fixed point, like where it comes out of the fitting. Extrude 100mm (G91 <enter> G1 E100 F300 <enter> G90 <enter>). Measure the extruded length with a digital caliper. Calculate the correct steps_per_mm = (old_steps_mm * 100mm) / measured_mm_extruded.

That will give you your actual steps_per_mm for the extruder. However, the amount the slicer will request for a move is calculated from the diameter of the filament (a setting in your slicer). It needs to lay down a certain volume of plastic for a track, and calculates the volume from the filament diameter, and gives the mm of extrusion (linear) to the firmware. If you filament is slightly off, this volume calculation has an error. Measuring your filament with (again) the digital calipers, you can enter the correct value in your slicer.

Depending on your nozzle, the type of polymer and the temperature you print with, the actual extruded volume differ from the calculated ideal. Most slicers will have a fudging parameter. In Simplyfy3D it is called 'Extrusion multiplier' and is found on the Extruder tab settings. For PLA a value of 0.95 is usually OK, for ABS I use 1.00 as multiplier.

2) For a delta this is different than for a cartesian bot. Your X-Y moves are recalculated to tower heights, and are non-linear as such. Basically the tip of your hotend is recalculated to the intersection of three spheres defined by the tower height as center and the diagonal rod length as radius. To get dimensional correct X-Y moves, you may need to have different rod lengths for all three arms as explained here: http://www.thingiverse.com/thing:745523

Print out that object and the pdf chart as explained, and measure. The adaptations as explained in the article are for Marlin firmware, but I have seen that Repertier has those corrections already implemented in the firmware. I am in the process of migrating from Marlin to Repertier myself, so will soon be able to help you further there.

3) just use this button:
picture.jpg

You will need to save your images as .png type first.

4) Loads and loads, where to begin?

5) If you have a heatbed that buckles when heating due to poor heat distribution/too thin material, then autocalibration might be useful. The consequence is that you propagate first layer errors upwards through your print and I would not use it myself. Better to have a good, absolutely flat (heat-)bed (did anyone say glass?) and do the hard labour of manual calibration. My experience is that when once properly calibrated, you don't need to mess with it again for a long, long, loooong time. All time spent with good manual calibration pays itself off in hundreds (thousands?) of good prints.

[Nickschaos81]

Haha, thanks. Wasn't saving them as .png

I am thinking of changing the effector plate for a solid pieace. The one I got is for auto level and the center (which holds the extruder) pivots so that it can trigger the autolevel switch like an endstop. A good idea I think but poorly implemented with 3d printed parts. So your answer on autolevel was what I was hoping to hear. If I got rid of the pivot and went solid effector , I think it would give better accuracy because there would be no unnecessary play in the head.

I'm going to get those calibrations redone tonight , man I'm so excited. No longer is there a light at the end of the tunnel, but now I can actually see outside

[Nickschaos81]

NEW PROBLEMS!!!! so not excited

I went thru and did all calibrations again , everything seemed fine except my diagnol rod length (for dishing) went from like 240 to 640 .... that normal? Horizontal length had to be increased too , its now at 180 from 140 ish (dont remember exactly)
I also did the extruder calibration as well. Esteps went from 147 to 165. My Max Z went from 350 to 302 , then down to 268 and seems to steadily be decreasing.... but this could be because the heating and cooling of the extruder , or the stupid auto level mickey mouse rig of an effector i have.

The real problem that i seem to have is it now tries to print everything in micro. When I load the calibration tool to print , I slice it ... even in the preview window it still shows that its going to print out close to my max bed size , but once printing starts it prints it 1/4 what it is supposed to. I loaded the calibration cubes and it tried to print those so small they would have fit on a nickel, with a 4 wrap skirt. So either I inadvertly built a shrink machine or I screwed up somewhere.

[Nickschaos81]

I sat down , and started to just go thru the eeprom list and find out what each thing was.... then i sat down and started to measure everything and comparing it to what was in the eeprom (that was based on manufactor instructions) .... had some descrepencys , and now it seems that I had things off and didnt understand what numbers where super important , and just important. Im printing some more calibration cubes right now. If they come out like i expect them too , and the calibration circle tool does good as well. Im going to then post a final post explaining what i have learned and hopefully it will help some else out there as well. Stupid math , should paid attention in school instead of checking out chicks.

[Nickschaos81]

https://www.dropbox.com/s/o0396u2ikl...02201.jpg?dl=0
https://www.dropbox.com/s/fk4pot8zpl...02209.jpg?dl=0
https://www.dropbox.com/s/850oy15fpu...02220.jpg?dl=0
https://www.dropbox.com/s/y6g216raat...02229.jpg?dl=0

Belive it or not , that first cube cal. was printed using almost the same slicer settings , and no scaling on either , guess thats the difference in proper calibration. woot , go me (and you guys for helping)
I just tried doing a dry run on the delta calb. tool and at the extreme edges between towers (not at the towers) the head is lifted probably 6 mm ish off build plate.... I have now adjusted the horizantal rod and diagnol rod to get the head to be the same height at towers as its at at the middle at z0 , what do i do when it lifts between towers (while making a circle) I can post a video if you guys need me to .


Edited : Im going to bed , talk tomorrw

[Alibert]

When using a limited number of points for calibrating radius and rod length, I have found that there are multiple solutions that will fit those limited points. I have done this using a few combinatorial optimization techniques (gradient descent, downhill simplex and genetic algorithms). This explains why you can find 'crazy' combinations that should work, but do not in the spots you didn't include in the optimization. The printed size of the object directly relates to the diagonal rod length as well. If it is not the actual (real) rod length, the size of the printed object will scale to larger/smaller.

I have found that the best approach is to follow these steps in this specific order:

1) measure the diagonal rods carefully (I have a 12" digital caliper for this purpose) and use this value in the firmware
2) adjust the endstops so that all three towers just touch (as close as possible to the tower) at a given max z height
3) adjust radius value only to get a as flat as possible z-plane (using opposite points of the towers as well. You should be within 0.3mm over all of the build plate at this point. If not, redo steps 1-2.
4) print out the calibration object according to the link I gave above. Measure the correct angles between x,y and z tower and enter these in the firmware.
5) redo step 3, you only need very small adaptions
6) print out the calibration object again, but now only compensate the diagonal rod sizes for each tower using the calculation explained in the link. Leve the tower angles from step 4.

With Marlin firmware and following the above I am within 0.1 mm z-height error over nearly my whole build plate (280 mm diameter) and less than 0.05mm in X and Y direction. I am in the process of converting to Repertier 0.92 and will be able to tell you where to plug in the adjustments soon.

[Nickschaos81]

Yea, I'm confused now albeit .... the only adjustments I have made have been too diagnol rod length, horizontal rod length, max z height, and steps per mm. Is there a video tutorial showing me what your talking about doing? I feel like there is a bunch of steps I'm missing. Because if I preform the exact same process of calibrations between the towers wouldn't I just be splitting the difference, because adjusting the diagnol rod length to get the tower points closer then adjusting it between the towers, wouldn't that just move the tower points into the bed. Yea I'm confused , defiantly lol you win... ��