Vertical Feed rates in partially-vertical cuts

Here's the place for discussion related to CAM/CNC and the development of the Path module.
berka
Posts: 42
Joined: Sat Aug 22, 2015 9:08 pm

Vertical Feed rates in partially-vertical cuts

Postby berka » Fri Nov 24, 2017 9:29 pm

This behavior surprised me a little, so I thought it may be up for discussion.
I feel like feed rate should consider both vertical and horizontal rates when making a move that's a ramp. Currently I see that the G-code generated picks one or the other feed rate rigidly.

A simplified example situation:
Horizontal feed rate: 20mm/s
Vertical feed rate: 5mm/s (plunge rate)
Contour cut around a box with a helix ramp entry dress-up.

My expectation:
Horizontal feed at (or near) 20mm/s. This is mostly a horizontal cut with a gradual ramp into the material.

Actual behavior:
5mm/s feed rate even though it's mostly horizontal motion.

Snippet from the end of the G-code inspector:

Code: Select all

G1 F5.000000 X-10.000000 Y11.587500 Z-10.415546
G1 F5.000000 X10.000000 Y11.587500 Z-10.965702
G1 F5.000000 X10.248344 Y11.567954 Z-10.972389
G1 F5.000000 X10.490567 Y11.509801 Z-10.979335
G1 F5.000000 X10.720708 Y11.414474 Z-10.986366
G1 F5.000000 X10.933105 Y11.284317 Z-10.993311
G1 F5.000000 X11.122532 Y11.122532 Z-10.999999
G2 F20.000000 I-1.122532 J-1.122532 K0.000000 X11.587500 Y10.000000 Z-10.999999
G1 F20.000000 X11.587500 Y-10.000000 Z-10.999999
G2 F20.000000 I-1.587500 J0.000000 K0.000000 X10.000000 Y-11.587500 Z-10.999999
G1 F20.000000 X-10.000000 Y-11.587500 Z-10.999999
G2 F20.000000 I0.000000 J1.587500 K0.000000 X-11.587500 Y-10.000000 Z-10.999999
G1 F20.000000 X-11.587500 Y10.000000 Z-10.999999
G2 F20.000000 I1.587500 J0.000000 K0.000000 X-10.000000 Y11.587500 Z-10.999999
G1 F20.000000 X10.000000 Y11.587500 Z-10.999999
G2 F20.000000 I0.000000 J-1.587500 K0.000000 X11.122532 Y11.122532 Z-10.999999
G0 F0.000000 X11.122532 Y11.122532 Z6.000000
Note how the feed rate is 5mm/s as long as the Z is changing and suddenly speeds up to 20mm/s when it reaches the bottom of the cut.

The behavior is the same for many other moves that change Z; it's not confined to ramp dress-up.
chrisb
Posts: 18991
Joined: Tue Mar 17, 2015 9:14 am

Re: Vertical Feed rates in partially-vertical cuts

Postby chrisb » Fri Nov 24, 2017 10:50 pm

Since a ramp has both components, vertical and horizontal it should use the lower of both - which is almost always the vertical feed.
berka
Posts: 42
Joined: Sat Aug 22, 2015 9:08 pm

Re: Vertical Feed rates in partially-vertical cuts

Postby berka » Sat Nov 25, 2017 4:31 pm

chrisb wrote:
Fri Nov 24, 2017 10:50 pm
it should use the lower of both
I disagree, but I'm not a machinist; I'd love to hear from an ME and a machinist. What this means to me is that I have to "fake" my plunge rates to get reasonable feed rates on this operation.

Look at this snippet from earlier:

Code: Select all

G1 F5.000000 X-10.000000 Y11.587500 Z-10.415546
G1 F5.000000 X10.000000 Y11.587500 Z-10.965702
That's a cut 20mm horizontally and 0.55mm vertically. I don't expect this to go at 5mm/s. Note that the Z component of that motion is actually closer to 0.14mm/s.

(Editing to include a file.)
Attachments
TestBottomPocket.FCStd
(51.06 KiB) Downloaded 16 times
User avatar
JoshM
Posts: 419
Joined: Thu Oct 05, 2017 5:34 pm
Location: New Hampshire

Re: Vertical Feed rates in partially-vertical cuts

Postby JoshM » Sat Nov 25, 2017 5:23 pm

Ramp entry is done where direct plunge would degrade Tool life, so err on the side of gentle seems consistent with the dress up, no? In 2d CAM I previously used, Ramp angle and Feed rate were specified independent of the OP Feed rate. In absence of that, lower Feed is safer.

Regards,
Josh
chrisb
Posts: 18991
Joined: Tue Mar 17, 2015 9:14 am

Re: Vertical Feed rates in partially-vertical cuts

Postby chrisb » Sat Nov 25, 2017 5:43 pm

berka wrote:
Sat Nov 25, 2017 4:31 pm
I disagree
  • I thought about it and perhaps you are right. I'm not a machinist either, but here are my thoughts:
    I keep the statement that we use the lower of both directions if we have a horizontal and a vertical component and do not investigate both components any further.
  • If we investigate it in detail the speed could be the vector sum of both components. That would be sqrt(Vx^2 + Vz^2). But this works only if you move at exactly the corresponding angle. Is there some standard technique how to adapt these speeds for different angles? If not I will think about it.
herbk
Posts: 1544
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Vertical Feed rates in partially-vertical cuts

Postby herbk » Sat Nov 25, 2017 9:37 pm

Hi,
a "ramp entry path" (with a reasonable set angle) don't need a reduced feedrate.
Using ramp entry always (mostly, - there are not many reasons why not) moves the "cutting side" to the cutting edge off the tool and you can work with full feedrate (thats the reason why i suggested to use ramp entry as standard in a other thread).
If the machine dont't allow the same feedrate for both, the machine settings (on machine controller) usual take care of that and reduces the feedrate to the allowed/posible.
Is there some standard technique how to adapt these speeds for different angles?
There is no real need for. A endmill has always (maybe there there are e few without i don't know...) a vertikal and a horizontal cutting edge. As long as you calculate the ramp angle flat anough, that the not cutting area of the tool is not touching the material, it's no problem work with "full speed".
If i use a tool which don't allow this, it's a "special situation" and i (the user) have to take care of it and switch the settings for that OP.
Gruß Herbert
GeneFC
Posts: 1025
Joined: Sat Mar 19, 2016 3:36 pm
Location: Punta Gorda, FL

Re: Vertical Feed rates in partially-vertical cuts

Postby GeneFC » Sat Nov 25, 2017 10:03 pm

OK herbk, you chastised me for not considering woodworking. Now I will return the favor by telling you that cutting in stainless steel is different. :D

I would not want the normal horizontal feed rate for ramped cuts. It would be hard on the cutters.

Gene
User avatar
JoshM
Posts: 419
Joined: Thu Oct 05, 2017 5:34 pm
Location: New Hampshire

Re: Vertical Feed rates in partially-vertical cuts

Postby JoshM » Sat Nov 25, 2017 10:28 pm

GeneFC wrote:
Sat Nov 25, 2017 10:03 pm
OK herbk, you chastised me for not considering woodworking. Now I will return the favor by telling you that cutting in stainless steel is different. :D

I would not want the normal horizontal feed rate for ramped cuts. It would be hard on the cutters.

Gene
Oh boy--it's on now :lol:
herbk
Posts: 1544
Joined: Mon Nov 03, 2014 3:45 pm
Location: Windsbach, Bavarya (Germany)

Re: Vertical Feed rates in partially-vertical cuts

Postby herbk » Sat Nov 25, 2017 10:35 pm

GeneFC wrote:
Sat Nov 25, 2017 10:03 pm
OK herbk, you chastised me for not considering woodworking. Now I will return the favor by telling you that cutting in stainless steel is different. :D

I would not want the normal horizontal feed rate for ramped cuts. It would be hard on the cutters.

Gene
It's a "question" of the ramp angle... the angle slows down the vertikal speed by itself. E.g.: If you have a ramp with 10° the tool makes 10mm horizontal and only 1,7mm vertikal.
Of course you are right, there can be some materials or other reasons where that is still to much, but that's what the user has to determine an adjust the angle.

In my mind each user should set (and it should be possible to do) the "base settings" to the parameters he usaual works with. For "unusual work" i (and i'm sure you to) im always more observing as on "daily work".
Gruß Herbert
chrisb
Posts: 18991
Joined: Tue Mar 17, 2015 9:14 am

Re: Vertical Feed rates in partially-vertical cuts

Postby chrisb » Sat Nov 25, 2017 11:31 pm

In fact the linear speed along the ramp path can be higher than full horizontal or full vertical speed. So if you have to earn money with it it might have some importance.

It turned out to be a bit of math, mostly good old Pythagoras, see the attached PDF.

If the postprocessor can read the feedrates he can take care of the optimization.

Edit: Corrected some Y to Z (see following posts), extended document to helix ramping.

If time is money then how late is it in my wallet?
Attachments
rampSpeed.pdf
(99.09 KiB) Downloaded 51 times
Last edited by chrisb on Mon Nov 27, 2017 12:54 am, edited 2 times in total.