Tutorial on feeds and speeds


(Phil Thien) #46

Cancel my comment, I didn’t realize I could click on a material and the chart would highlight my selection.

Very neat! Thanks for doing all this work!


(William Adams) #47

My pleasure!

Feedback on which elements to provide to be clickable would be welcome, and their hierarchy / ordering.

Converting the balance of this post into a new topic — please see:


Interfaces and data and so forth
(Julien Heyman) #48

The order most useful to me would be:

  1. machine (clears out the field and one less reminder about the opportunity to spend money on a beautiful Nomad :slight_smile:
  2. material (input constraint on 99% of my projects)
  3. endmill size & #flutes (input constraint when you have a limited set and/or the features to carve demand specific sizes)
  4. min/max RPM (I personnally really dislike going to the higher RPMs range (noise!), and prefer to stick to e.g. 12.000 and adjust feedrate accordingly, when this is possible/compatible with SO max feedrate)

(William Adams) #49

Come to the Nomad side, we have fixtures.

(and peace and quiet, and an enclosure, and a tool length sensor, and ER-collets, and can use an edge finder…)


(Evan Day) #50

You can’t use an edge finder on a SO3?


(William Adams) #51

Not a mechanical one with a trim router — you can use it if you have a spindle or VFD which spins at low enough speeds — that said, I think my Nomad is jealous of the probe on my SO3 XL.


(William Adams) #52

Okay, I’ve added some more values, and adjusted things a bit further — getting a bit cluttered, but hopefully useable still.


(Julien Heyman) #53

Cool. Is it possible to make diameter a filter (like machine & material), instead of a legend/key ?


(William Adams) #54

Not that I can tell — numeric ranges seem to have a limited number of options — I suppose if I hard-code them to have the unit that would then allow that, but it would then disable using them as numbers to calculate the size of the datapoint (which aspect I like)


(Jude Marleau) #55

Your chart is easy to read, and use. I love it. All information on one screen, no scrolling, sorted amazingly. Terrific job Will, many thanks to your determined efforts. Jude


(William Adams) #57

Thanks!
Here’s a version with the current Carbide 3D feeds and speeds:

https://public.tableau.com/profile/willadams#!/vizhome/Carbide3DCNCFeedsandSpeeds/Sheet1?publish=yes

I’ll start in on filling in the first version from the wiki, then we’ll fill in all the gaps.


(barthelemy bach) #58

The nomad side… Shapeoko for ever !!!


(Temujin Kuechle) #59

This is a relatively complicated project that I am interested in contributing too but don’t foresee me doing much for at least 2 weeks. I like where it is going. I have been using CNG-Wizard up until the end of summer, then I realized that I needed an enclosure and much better dust collection (which is all still in the works).


(Dave Richard) #60

Thanks Will for all your hard work. This should be extremely useful.


(William Adams) #61

Okay, back at this.

Here’s the thing which I’m trying to wrap my mind around — Chip Load Per Tooth — using as an example 0.004, for a 2-flute endmill this value appears at:

36" @ 4500 RPM
40" @ 5000 RPM
50" @ 6250 RPM
60" @ 7500 RPM
72" @ 9000 RPM

(and at various interpolated numbers — I suppose a graph might help)

and for a 3 flute endmill at:

45" @ 3750 RPM
60" @ 5000 RPM
75" @ 6250 RPM
90" @7500 RPM

So in theory, each of those cuts ought to take the same size / shape chip (accounting for variations in the flute shape as determined by diameter) and the reason to pick one over the other would be determined by the spindle power needed to spin at a given RPM, the desired Surface Feet Per Minute for cutting a given material w/ a given endmill, and the power needed for the machine to move at that rate while making the cut and not losing position.


(William Adams) #62

Some reference PDFs:


(William Adams) #63

Going to try again w/ Jupyter Notebook:

which has an on-line server provided by Google: https://colab.research.google.com/notebooks/welcome.ipynb

and if that doesn’t work, guess it’s processing.org


(William Adams) #64

Okay, working on this in Jupyter Notebook but wanting to update progress here so:

Pine — need feeds and speeds for 1/8" endmill, so start w/ the Nomad figures from: http://docs.carbide3d.com/support/#tooling-support

0.03" depth per pass, 4500 RPM, 72" feed rate, 32" plunge rate

which per: https://www.shapeoko.com/wiki/index.php/Materials#Nomad_883 yields a chipload of 0.008″

which can be achieved by the feed rate and RPM pairs of:

5.76" @ 3600 RPM
6 @ 3750 RPM
6.4 @ 4000 RPM
7.2 @ 4500 RPM
8 @ 5000 RPM
9.44 @ 5900 RPM
9.7776 @ 6111
10 @ 6250 RPM
10.08 @ 6300 RPM
10.4 @ 6500 RPM
11.52 @ 7200 RPM
11.68 @ 7300 RPM
11.84 @ 7400 RPM
12 @ 7500 RPM
12.48 @ 7800 RPM
12.8 @ 8000 RPM
13.44 @ 8400 RPM
14.08 @ 8800 RPM
14.4 @ 9000 RPM
14.56 @ 9100 RPM
14.72 @ 9200 RPM
15.2 @ 9500 RPM
16 @ 10000 RPM
17.4 @ 10875 RPM
18.4 @ 11500 RPM
20 @ 12500 RPM
21.6 @ 13500 RPM
23.2 @ 14500 RPM
24.8 @ 15500 RPM
25.6 @ 16000 RPM
26.4 @ 16500 RPM
26.8 @ 16750 RPM
27.2 @ 17000 RPM
28 @ 17500 RPM
28.8 @ 18000 RPM
29.6 @ 18500 RPM
30.4 @ 19000 RPM
31.2 @ 19500 RPM
32 @ 20000 RPM
32.96 @ 20600 RPM
33.92 @ 21200 RPM
35.2 @ 22000 RPM
36.4 @ 22750 RPM
37.6 @ 23500 RPM
38.8 @ 24250 RPM
40 @ 25000 RPM
41.12 @ 25700 RPM
42.4 @ 26500 RPM
43.2 @ 27000 RPM
44.4 @ 27750 RPM
45.6 @ 28500 RPM
46.8 @ 29250 RPM
48 @ 30000 RPM

For a 1/4" endmill, pine has an SFM speed of 1317.5325 — a 1/8" endmill spun at the above speeds has an SFM of 117.9 – 982.5 — unsurprisingly a 1/4" endmill ranges from 235.8 – 1965 — twice those numbers

The Shapeoko wiki https://www.shapeoko.com/wiki/index.php/Materials#Pine has feed rates ranging from: 31" (extended reach), 800mm (~31.5") to 35" with chiploads ranging from

  • 0.014937759336099585 — 31.5" @ 16870 RPM

(William Adams) #65

Okay, Google making Colaboratory available makes this a lot more workable — still hacking at a sample workbook, but it seems promising:

https://colab.research.google.com/drive/1uqRD28E_T_3h4ZR2fU-DM54-T2g3wkG8

Just wish there was a natural way to graph a Google Spreadsheet or represent it as a table.


(Wheeler) #66

This is very helpful, thank you for all your work on this!