Delta sizing?

Hello!

I am looking at [reprap.org] but i seem to always come out with different answers.

I am scrach building a delta. (2040 upright profiles, DUetWifi, SmartEffector, Linear rails). Or rather, I am getting a new frame for my current setup as i have come to realize that 1515 even with bracing was not sturdy enough. My current machine has 750mm uprights, 360mm horisontals and 330mm arms


Carriage height should come out at 14mm with 12mm rails and a 2mm PCB adapter i guess?
But i am trying to figure out the arm length.
I think i'll go with 900mm uprights since it is a standard length, and the aluminum corners will take 120mm total, compared to 80 on my current machine.

Currently i have a 250mm diameter build plate. But i would like to make room for a 300mm.

It seems that 410mm horisontals should suffice..

But i keep getting lost on suitable arm length, and I a not making friends with the calculators referenced on the wiki.

Any insight?

Go for width of the bed plus 10%-20% depending on how important z printable height is to you. There's less to lose with slightly longer arms than with slightly short. I went 330 then changed to 360mm.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

If you are looking for 300mm print diameter, you may find the dimensions I give at [miscsolutions.wordpress.com] helpful.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

So as per the Delta Geometry Page the minimum recommended is 20 deg. I use trigonometry to figure out arm length. cos(θ) = Adjacent / Hypotenuse is the basic formula.

Here is how I go about it (Printable Radius + Offset of carriage to Print Bed - Offset of rod centre and hotend centre)/.9 (.9 is cos(25) I use a minimum angle of 25 as I use printed rod end).

You may want to adjust the minimum recommended angle depending on your BOM.

Hope this is helpful.

I recently read the 20° rule is based on the Traxxas rod ends. Not sure if it's right, but IF, then the angle-restriction is meant for sideways motion.
Nevertheless, it works as a rule of thumb. The steppers have to work very fast to keep up with accelerations when printing at the outer border.

Just a sidenote: with two sets of rods, you could print narrow/tall or wide/flat parts, whatever is convenient. Replacing Magball rods is easy. Only requires two config-files to switch.

In this thread Haydn claims he can swap longer to shorter rods and print without recalibrating.

[groups.google.com]

I've always seen the 20-60 degrees as guidelines. As using magnetic arms and linear rails the effector stays level even at extreme arm angles. I agree you don't want to print at the edges as it is inefficient due to needing high carriage speed, but the with good motors and 32bit controllers high carriage speed with accuracy is not that hard to achieve.

Edited 2 time(s). Last edit at 02/10/2018 06:15AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I just followed my own advice and exchanged the 360mm rods with 270mm rods.
I had to edit a few files: config.g, bed.g, ( config-override.g if exist ) and a macro file. The printable diameter shrinked [shrunk?] from 280mm to ~120mm and the height changed from 220mm to 344mm. Pretty impressive differences.
I made a test print and noticed the 3 steppers were moving much more than before. I see it as a pro-argument, since the accuracy is higher now. ( I doubt it will be visible, but anyway )
Another pro-argument is: shorter rods are generally stiffer, so I expect a better print quality from that fact. ( less ringing/ghosting)

Edited 2 time(s). Last edit at 02/10/2018 07:42AM by o_lampe.

It's a nice idea swapping on shorter rods, if you coupled that with a multi zone bed you could almost have a lower energy daily drive config and a higher energy max config.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

A 6mm aluminum bed will spread the heat, no matter what diameter the heater has. The losses will be almost the same, except you add an insulator sheet that only leaves the smaller print area open.

AFAIK there are two types of limitations that dictates the rods angle, and thus the rods length for a desired printing diameter:

1. The maximum angle admitted by the rods articulations
2. The maximum speed allowed by the stepper motors/pulley ensemble

Both needs to be taken into account according to your type of rod joints and motors, and determine the rod's angle. Then, knowing the angle and the desired printing diameter, rods' length can be calculated.

If the rods are too small, either the motors won't be fast enough and will loose steps (when a rod is almost horizontally), or the rods articulations won't be able to circle the full printing bed (when the circle diameters are big).
If the rods are long enough, the problems above vanishes, but you will lose some printing height (giving a fixed rail length for rods).

I have a set of Traxxas ball joints (very cheap ones, bought from Aliexpress and most probably counterfeits). The maximum measured angle is about 45 degrees, so +/-22 degrees around the middle position. I will go with +/-20 to be safe.
Some other metal ball joints allow only +/-13 degrees, like the 'ZOKOL bearing SAL04P/K Male Thread M4' ones here [www.aliexpress.com]
Magnetic ball joints or cardan shaft bearings might allow wider angles, I didn't looked into those yet.

Regarding the maximum belt speed, this is a little more complicated. It depends mainly by the maximum step per seconds of the motor and the pulley diameter.
Yet, the number of steps per revolution, the driving current, number of micro steps, motor detent, load weight, moment of inertia, etc. all these need to be taken into account to determine the maximum belt speed without loosing steps.

Edited 1 time(s). Last edit at 02/10/2018 07:28PM by RoGeorge.

DjDemon's recommendation of the bed diameter plus 10-20% is usually a good one.
I use 360mm arms on my machines with 300mm diameter beds, and that works out with a bit to spare.

However, I also noticed that you mentioned that your horizontal extrusions are 410mm.

Is your print bed going to fit completely inside of your base triangle?

On my printers I use 360mm horizontal extrusions, and the bed extends slightly outside of the horizontal extrusions (by about 15mm).

Because your horizontal extrusions are relatively long, in your case it may be necessary to go with plus 20% or even 25%.
All you give up is a little bit of maximum height, and most people rarely print objects close to their maximum height.

Thank you all.

I ended up with mostly copying DC42's all metal delta. So, I went with 390mm horisontals, and 1m uprights, and sticking with the current 360mm arms.

I'll keep you posted

Quote
Haydn
However, I also noticed that you mentioned that your horizontal extrusions are 410mm.

Sorry if i was unclear. I had not yet ordered the horisontals. I was thinking about 410mm, but decided to go 390.

The die is cast!

Lower vertex with bottom plate.

not a _perfect_ bottom plate. But when one has a piece of "close enough" 3mm alu plate.. one uses it if one feels a bit cheap



Edited 1 time(s). Last edit at 02/28/2018 03:57PM by JanJoh.