Mine fit perfect off the printer, but my tolerances are about +-0.03 mm

 

Thanks. Will give that a try.

 

Thanks for the info. What length are your arms also? I was trying to use the openscad delta calculator from a link on this page https://sites.google.com/site/3dprinterlist/delta-3d-printers/delta-robot-calculator , but with the different size extrusions & vertex blocks, the numbers do not come out right. I will draw this up also, but having a good starting point always helps.

Thanks.

 

Arm length is something that gets debated a lot. That calculator was setup for very specific parts, basically you want enough to push the carriage to the edge of your build platform. So your arm length will be determined by effector and carriage offsets and distance to the furthest point to print at.

I'm going to use a bit longer arms, I'll lose some build height but there are some pros to this as well. I think my arms will be about 340 mm

 

Has anyone every thought about using the misumi t-slot for the extrusions on the base where linear motion is not needed? You can save a little bit of money buying these & have them cut to the size you want. http://us.misumi-ec.com/vona2/detail/110302684350/?Inch=0&CategorySpec=00000042730::c
For the options on this page, I picked Clear Anodize, L Dimension Configurable, 375mm Length, Type HFS5-2040 & qty of 12 & price is $48.60. 3x1500mm V-Slot would be $58.50 & you would have to cut them. I can not tell what the shipping cost would be or know what the difference in quality would be.

I am also looking into seeing if I can modify the openscad delta calculator script for the mini Kossel, to work with the 2040 extrusion & delta 6 plastic parts. https://github.com/Jaydmdigital/mk_visual_calc if anyone has any interest in this.

 

Kyo, Thanks for the info.

I made some progress on the Delta 6 calculator today & think I have it working well for calculating the Build plate & placing the 6 vertices & the aluminum extrusions. While I was at it, I made it so you could use 2020,2040,2060 or 2080 for the horizontal extrusions by changing some variables. There are no external stl files that are used with this script. It does not calculate the build height or rod arm lengths yet. I have attached a couple of screen shots & the rev1 version of the openscad file. If you try it, let me know how it works for you. I used Arondite's 375mm horizontal & 1000mm vertical lengths as the default.
Here are the dimensions I used.

 

It was a little strange using openscad at first having used AutoCAD & similar programs for years, but after using it for awhile I really like it. Of course I enjoy programming also, so it works well for me.

I worked on this some more last night & changed most of the variables to reflect the names from original openscad file to make it easier to use his formulas. It seems to be working quite well now except the rods are not quite the correct length & do not line up on the effector & carriages at the right points. The really nice change about this versus the original scad file is that since I am not importing any stl files, I am able to create a stl file from the design. I left out a lot of the detail in the parts because they are only used for visualizing. Openscad will complain that it might not be a 2-manifold, but will create it. It would be quite helpful to use some numbers others are using in this design & what the outcome numbers came out to on their build. I attached a screen capture, the latest openscad version rev6 of the calculator & a sample stl file output with 375mm horizontal & 1000mm vertical extrusions.

 

This is about as close as I can get this to work for now without knowing more about how some of the variables related to the rods work.

 

Looks good. Is there a place where we can enter the arm length we used and see how it effects it?

 

I forgot to take out some hardcoded variables on Rev7, This Rev8 fixes that & you can change the arm length.

You should be able to change the arm length by changing the DELTA_DIAGONAL_ROD variable
Uncomment this line on Line 79 & put in the number you want. This is just below the calculated Length.

//DELTA_DIAGONAL_ROD=500;

 

I have fixed all the problems that I know with this now. Seems to be calculating quite nicely & the numbers look good to me. I cleaned up the code quite a bit & put most of the user definable variables at the top. The length of the Diagonal Rod is the only variable listed further in the code currently at Line 87.

I found an image on https://sites.google.com/site/3dprinterlist/delta-3d-printers/delta-robot-calculator that gave me a better understanding on how to calculate the Diagonal Arm lengths & locations. I also used the 30.77 initial angle from this.

DELTA_DIAGONAL_ROD = Arm_Len; //Change Arm_Len to what the value you want.

You can also change the Angle of this arm instead of the Arm Length with the variable currently located on line 39:
Delta_Ang_1st = 30.77

Here is the section of the code near the top to change the variables you are probably interested in:
//These 3 variables are used to show the complete part in the View Area (Comment out if you do not want these)
$vpt=[-96.47,108.38,198.51];
$vpr =[55,0,25];
$vpd =3985.23;
//========================================================
Frame_Ext_Len = 375; //Length of Horizontal Aluminum extrusions
Frame_Ext_Ht = 1000; //length of extrusions for towers, need cut length
Frame_Ext_Wid = 20; //Width of extrusions
Delta_Ang_1st = 30.77; //Angle to use for Arms for Calculating Initial Arm Length
Rod_Rad = 6/2; //6mm carbon fiber rods? Just for show anyways
//Change these 4 variables below for different size Horizontal Extrusions
Frame_Motor_Ht = 41; //height of motor from vertex (Change to 60 for 2060)
Frame_Top_Ht = 41; //Height of Top Vertices (Change to 60 for 2060)
AlumExt = 40; //20,40,60 or 80
Alum_Hor_Z = 30; //use 10 for 2020, 30 for 2040, 50 for 2060 & 70 for 2080
Glass_Z_Offset = 4; //Offset of Glass above Frame
Glass_Thk = 3; //Glass Thickness
Eff_Ht = 11; //Thicknes of Effector
HotEnd_Ht = 45; //Height of Hotend
//========================================================
Here are a couple of screen shots showing difference using 2040 & 2080 for Horizontal extrusions.
To change the size of the verticals the Vertex plastic parts would have to be changed for visualization. Those parts would also have to be changed anyway if you used the larger extrusions on the Horizontal.

 

I might have gotten a little carried away with this design. I decided to make a calculator for 2060 & 2080 extrusions & made a vertex for vertex_20x40H_20x60V & a vertex_20x60H_20x60V(See next message for this part). One interesting thing I noticed when going to 2060 or 2080 vertical was the bed diameter increased. Thought I had an error in my program until I realized it made sense that the wider verticals would put the horizontals further apart. I see one problem that needs fixing with the vertex_20x60H_20x60V part. I circled the area on the attached screen capture of the part

I am not sure how well those vertex parts will work & they probably need to be taller. The carriage would also have to be made wider, but I did not work on that yet. I also made the holes smoother, but maybe I did not need to do that.

Here are the stats comparing the 2040,2060 & 2080. You do loose a little in the vertical using the same numbers, but you maybe able to make the Diagonal rods shorter.

2040 Aluminum
ECHO: "DELTA_DIAGONAL_ROD = ", 364.813
ECHO: "Delta_Ang = ", 30.77
ECHO: "6x Horizontal Extrusions = ", 375, "mm"
ECHO: "3x Vertical Extrusions = ", 1000, "mm"
ECHO: "***** BedDia = ", 276.506, "mm ", 10.886, "inches"
ECHO: "***** Max Build Height:", 522.683, "mm assuming a narrow tower or cone shaped build."
ECHO: "***** Max Build Height in Inches: ", 20.5781, "inches"

2060 Aluminum
ECHO: "DELTA_DIAGONAL_ROD = ", 376.166
ECHO: "6x Horizontal Extrusions = ", 375, "mm"
ECHO: "3x Vertical Extrusions = ", 1000, "mm"
ECHO: "Delta_Ang = ", 30.77
ECHO: "***** BedDia = ", 299.734, "mm ", 11.8005, "inches"
ECHO: "***** Max Build Height:", 453.928, "mm assuming a narrow tower or cone shaped build."
ECHO: "***** Max Build Height in Inches: ", 17.8712, "inches"

2080 Aluminum
ECHO: "DELTA_DIAGONAL_ROD = ", 387.451
ECHO: "6x Horizontal Extrusions = ", 375, "mm"
ECHO: "3x Vertical Extrusions = ", 1000, "mm"
ECHO: "Delta_Ang = ", 30.77
ECHO: "***** BedDia = ", 322.827, "mm ", 12.7097, "inches"
ECHO: "***** Max Build Height:", 384.231, "mm assuming a narrow tower or cone shaped build."
ECHO: "***** Max Build Height in Inches: ", 15.1272, "inches"
View attachment 11510

 

This morning I noticed an error in the 20x60H_20x60V.stl file in my previous post. I had the end screw holes for the horizontal extrusions located wrong. I deleted those files from the message as to not confuse anyone. Here is the updated stl file & the openscad file. I rewrote a portion of the openscad file to make it easier for me to read & updated. By changing 3 lines, I made a vertex to use 2x 2040 horizontal.

 

I have fine tuned this openscad script into 1 file to be able to create most any combination of this Delta printer using combinations of 2020, 2040, 2060 or 2080 extrusions. I double checked the output of build size dimensions of a 2020, 2040, 2060 & 2080 size by importing the stl output from this into a CAD program & the dimensions were the same graphically as they were echoed in openscad. This is Rev18. Be aware there might not be a vertex for the 2020, or 2080 verticals in this design, but I could probably create one if someone needs it. I am still on the fence as to whether to build this design, but have enjoyed rewriting this openscad file just the same. I am probably done tweaking this openscad file unless someone finds some errors in the output.

I have put these 5 combinations you might want & just uncomment the variables you want to use. 13 variables define each different combination.
There are a number of other combinations you can use as well, mainly when using 2x 2020 in different parts of the design.
1x 2020 Bottom on each side & 1x 2020 Top on each side & 1x 2020 Vertical
2x 2020 Bottom on each side & 2x 2020 Top on each side & 2x 2020 Vertical
1x 2040 Bottom on each side & 1x 2040 Top on each side & 1x 2040 Vertical
1x 2060 Bottom on each side & 1x 2060 Top on each side & 1x 2060 Vertical
1x 2080 Bottom on each side & 1x 2080 Top on each side & 1x 2080 Vertical
You will also want to change the 2 variables for the height of bottom & top vertex to reflect the sizes of the horizontal extrusions.
Frame_Motor_Ht
Frame_Top_Ht

I also added a variable for a thickness of a base that I saw someone have a part to, I included a stl & openscad file that I modified. I have not printed any of the stl files, so have not verified their accuracy other than checking the dimensions in a CAD program.
Alum_BaseThk = 0 //Set to 16 for 16mm height below the bottom extrusion. This will decrease your build height by this amount.

I attached the rev18 of the openscad delta6 calculator as well as the openscad & stl file of vertex for 2x 2020 Horizontal & 2040Vertical, a screen capture showing the extra base under the horizontal & also a version showing 2x 2020 at the bottom, top & Vertical. The 2x 2020 combination with 20mm gap between them seems like it might give more strength than the single 2040's.

 

I did a google search on the animated gif Nugz has on page 4 of this thread & found the github of the OpenSCAD-Delta-Simulator. This allows for different delta datasets, so I am going to see if I can create a Delta Six data set for it.

https://github.com/PRouzeau/OpenSCAD-Delta-Simulator

I have found a couple of values I had incorrect in my openscad calculator & will upload that again when I look it a little more.
I was just looking into making movement for this Delta in openscad & found this. This source should make it easier to understand the math involved for this motion.

 

See if this is an reasonable animation of what this design should look like in motion. I used 500mm horizontals & 1000mm verticals for the animation. I have also attached my latest Delta6 calculator script which has only a couple of bug corrections.

 

I have just done my first print ever, still waiting on it to finish but in the mean time, what sort of speeds are considered fast for a printer like this?
I have used Slic3r and my defaults speeds are 90 for perimeters, 180 for infill and 200 for travel.

Can I go a lot faster?

Video to follow..

 

That sounds a lot faster than my 10" Prusa i3v. I will be interested in seeing the video of that.

On another note, I contacted the author of the Delta Simulator & gave him the data set I created for the Delta-Six. He made a couple of corrections to my data set to make it work properly with his main simulator routine. He also added a file to explain how to personalize your own data file. Here is his github link to that simulator again.

https://github.com/PRouzeau/OpenSCAD-Delta-Simulator/

 

Link to video of my first ever print:

https://goo.gl/photos/aKAiTGZ8DyqepoCQ6

 

Here is what I printed,

I had a few problems, my extruder was not set correctly and I managed to bend the filament and had to cut it off an reload 10% into the print.
The base is a bit wide, I am guessing I need to start a little further off the plate?
After that things started to settle until I started to get little bubbles in the latter part of the print, It was a brand new sealed roll of PLA, did it get affected by moisture in 30 minutes? Where I live it is not that humid or cold.

Any suggestions?

 

Looks cool. Can you tell me the extrusion lengths you used for the Horizontal, Vertical & what the lengths of your Arms are? Do you have the glass just resting on the painters tape on the extrusions? What size build Dia. & Height do you get from your setup? I am still on the fence as to what size to make my delta. I have 6 - 2020x500mm's & a 1 2060x1000mm on hand, so I am thinking about using double horizontal 2020's on the bottom & one on the top, then modifying the vertex to handle 2060's in the vertical.

 

arm_length 410 # # this is the length of an arm from hinge to hinge
arm_radius 223 # # this is the horizontal distance from hinge to hinge
# when the effector is centered

horizontals are 445mm and the verticals are what is left out of a 1.5m length, 1050mm.

I have a print area of 260 x 260 square and height of 550mm.

Yes, just sitting on the tape, will do something nicer later but at the moment I need to get the printer running to make parts for the printer

I have a 60mm on the bottom and 40mm on the top, If I had my time again, I would put an 60mm on the top so I have room to stop the verticals at 40mm and then have the 20mm for tension adjust. At the moment I have put a spacer on top of the top bar.

 

I have done a bit more work on the printer, I checked my extrusion length and found I was 40% over, set it up correctly and now it seems to be printing well.
I tweaked the speeds, from memory I am now running 100, 120, 150 and I am very happy. The print below started at the default speeds of 100 and towards the end I increased upto 150%. So it looks like 100 is a good number to start with.

 

Do you have some more photos of your build? In one of the photos from the designer of this build, I see a couple of bolts thru the top with a fender washer on them. Is that used to keep the verticals more secure? If I use 2060's for the verticals, seems like I might not need those.

 

Do you mean the bolts running vertical? they are part of the tensioning for the belts.
I will do a close up next time I have the top off but in short, I am using a 3mm x 30mm screw in each hole with a piece of aluminium between so when I tighten the screws it pulls the top frame up (the top frame has the idler attached) which tightens the belts.

 

I changed the X tensioning belt on my i3v that works great for that setup & seems like I might be able to do something similar on the delta, but it would be attached opposite direction as it is on my i3v. It attaches to the V-slot with 2 t-nut connections, then I can tighten the belts by loosening the bolts, moving it, then retightening them. I might have to extend the vertical extrusions up a little to gain better access to tightening them.
X-idler Mount Plate with X-axis belt tensioning for Makerfarm i3v 10" by GeoDave

 

Came across this on Thingiverse today. Looked like it might be a good solution for tightening belts.