@robertbu Thanks for the advice. I am currently cutting at 2400mm/min; that is based on other advice from someone who has cut XPS. I was thinking of maybe bumping it up a bit; 3000 maybe.
I will input your settings and see what it does in Fusion.
for the finishing pass I’m actually using a scallop toolpath. If follows the contour of the wing edge nicer.
I guess I should rephrase my statement… I only do a single pass in foam. I guess technically I’m skipping the roughing pass and only doing a finishing pass.
In my early cutting I found the estimated time so wildly off that I stopped looking at it. I believe there are ways to tune this number to be more accurate, but I never bothered. Maybe something to revisit.
Your 1mm step over is probably overkill for this application, but it cannot hurt (other than more time). I found this foam incredibly easy to sand…too easy…so for large smooth areas and simple geometry, cleanup is quick. I’ve wondered about the best method to protect the foam when I do a wing. When doing cosplay props, I use monster mud. Monster mud is a combination of latex paint and some form of spackle or drywall mud. It forms a nice hard surface, but it is probably too heavy for an RC wing.
I am currently cutting at 2400mm/min; that is based on other advice from someone who has cut XPS. I was thinking of maybe bumping it up a bit; 3000 maybe.
One of the cosplay pieces I did was a blunderbuss stock. It was about 28" and double sided with long straight cutting and gentile curves like a wing might have. I just looked at the at the CAM for that project, and found I used a feedrate of 5000mm/min! I remember purposely pushing things to see the limits when doing this project, but it was a successful cut even at that speed.
I’m finding that fusion is pretty close. the wood parts I cut for the plane have been right on for timing.
It is tight, going for as little hand finishing as I can. I did a test cut on a micro plane wing about 18" long) I used .5mm step over. that resulted in no finishing, but I don’t think I need to go that tight.
Weight is very important. need it as light as possible. A lot of times people will use brown paper bag type of paper with diluted wood glue. it is really strong and in then end very light weight.
I’ve have also done other projects with diluted wood glue and talcum (baby powder). The talcum allows it to be more sand-able. Smells nice as you sand it too . you can mix it in different thicknesses for different applications. its tricky getting the right combination of glue, water and talcum though.
I might not go 5000 but knowing you went that high gives me confidence of 4000.
After you made this comment, I started noting the estimated vs. actual machine times again. The next two jobs I did, the estimated time was right on. But today the two are way out of wack. Estimate time of 17:25 turned into 1:14:00…an almost five times different. Worse a 4:16 estimate turned into 54:07 actual run time. Note the jobs that matched the estimated time were primarily contour cuts. Today’s jobs were heavily weighted to adaptive clearing.
Poking around I found the following by the person who apparently wrote the estimation code:
Furthermore, I don’t consider the acceleration and block processing speed of the CNC so that would also give some difference from the actual machining time.
I expect that at least acceleration and possibly processing speed may be way slower for my MPCNC than the stock Fusion 360 values. And the difference would have little impact on long contour cuts but hit adaptive clearing heavily. And I don’t see a way of modifying the values for machining time so that I could get more accurate estimates.
This is following up on a different thread that ended up not being related. So i moved it back to this thread.
@robertbu - following up on the z axis issue from the other post.
So it makes sense to start from the spoil board, move it up and re-zero at the correct height.
what i can’t figure out is how to set the X,Y.
I’m trying to shave off some of the material to get down to the the required thickness. if set my z axis to the desired height then I would not be able to set my X and Y for the facing mill to be centered at the corner of the stock.
It would intersect with the material I want to remove.
There are a number of ways of going about this problem, but the simplest for me is to just make sure that the home point I use in my authoring software is at least as high or higher than the highest point on my physical stock. So say I have a board that is around 20mm thick and I want to face it down so that I have a 17mm board. I might set the stock starting height in the software to 21mm and use a 17mm model. After I have set my home using the spoil board and the 21mm authored setting, I can move to my X and Y position without intersecting the stock. Your CAD calculation will then drive the router on passes to cut the stock down to the height of your model.
In practice I often do it the other way around. I find and zero out the X and Y for home. Then I move the bit off the material and set my Z, then I send the router back to (0,0).
Ok, so just to make sure I understand what you are saying…
Let’s say I want a part that is a precise 1" and say my non-precise stock varies from 1"-1.25 (for example only).
1 - in the software (fusion in this case) have my final part sized to 1"
2 - have a stock size that is 1.26 to clear the highest dimension of my non-precise stock. this allows the mill to clear it for X and Y.
3 - on the machine zero out at the spoil board
4 - move up to 1.26 (same as the stock part in fusion) and zero again
5 - set X and Y.
And one last question - how do you zero out the Z axis using the LCD screen? I know how when plugged into a laptop using repetier host.
how do you zero out the Z axis using the LCD screen?
That’s a good question. I don’t see a way to do it from the menu, so just put the GCode to zero these out (“G92 X0 Y0”) in a .gcode file on the SD card and run it. Or you can position your bit at your home position and power cycle your board. There is a small risk with the second method since your are de-powering the steppers.
Here’s what I would do to handle your example assuming you’re using Fusion 360
Set the material size 2" longer and 2" wider than your board and set the material thickness to roughly the thickest portion of your board (probably a little thicker).
set the offsets on the material so that your part is in the bottom left corner of the material.
Set the 0,0 point of the bit to the top right corner of the material, at the spoil board height
This allows your Z0 point to always be the spoil board, and it allows Fusion360 to know what the height of your part/material is. I chose the top right corner of the material to allow for any overhang of the Z axis on the machine to make sure the tubes don’t hit the material when going down to Zero.
I think Fusion 360 will probably complain about a collision, but the gcode generates anyways.
Alternatively, and I’ve done this before…
Take a scrap cut off the same height as your board. Use double-sided tape to stick it to your work surface. Put the corner of your board a fixed distance from the scrap both in the X and Y (say 1"). Add that distance to your material length and width. Then set the 0,0 point to be the bottom left corner of the material, which should be in the middle of the piece of scrap. You now have a fixed Z height for setting Z on new bits during a swap. This requires accurately measuring the height of the scrap so that you know your true height off the spoilboard.
OK, so I have been making successful cuts with correct size stock.
the next issue I’m running into is foam melting on the mill; the tip get clogged.
I’m running the mill at 3500mm/min. The spindle speed is tough to know. It is the DW660 which claims to be 30,000 rpm, but I’m using an AC speed controller set at the lowest speed; I’m not sure what that is.
So the question is do I increase feed rate or spindle speed, or both?
My guess is I need to increase my feed rate. Maybe 4000mm/min.
Any chance you could rig an air blower to help cool the bit and move the foam dust away from the cut?
I know it’s tough to tell with foam, but are you getting “dust” or “chips”? Chips are good, dust means more heat is getting generated. Moving faster on the same RPM should get you into chip mode, as I understand things?
So I ran the mill at 4000mm/mm and did a lot better. the chipping was significantly larger and minimal melting after about an hour of cutting.
What speed controller are you using?
and now I ran into yet another issue. My Y axis has chatter in it when milling. When I move it in Y using the movement function on the LCD it does not appear to chatter; only when milling. I thought it was just having a harder time where I have glue laminating the two sheets of foam, but it does it where it is cutting through only one of the sheets.
I have seen posts that mention checking to make sure connections are secure. The only one I can get at is at the board and it feels tight. Is there anything else I can look at before I start taking my wiring apart?
and here are some screen shots of the wing I’ve been cutting.
top and bottom half of the right wing and two sided cut on each. between the two halves is about 1mm out of alingnment.
I would say that is pretty good for my first parts cut.
That is what I do. I have a separate top and bottom half. Both are 1" thick at the thickest part and since the foam is not perfectly even I need to laminate a 1/2" piece to a 1" and then face mill it on both sides to the proper size.
In the main gantry part no, but I did notice that two bolts for the DW660 mounts are loose (lower mount), though I’m assuming that is a result of the chatter in the Y axis.
I’ll tighten those up, maybe add some locktight, and run a test.
Update - I double checked the gantry. There is some play when I wiggle it in the Y direction, but none in the X direction.
So I probably need to tighten some bolts. just need to figure out which ones.
So I finally finished cutting the main wing for my CNC RC plane.
Overall went very well. learned some stuff, particularly that I should have used loctite when I first assembled the MPCNC. had a few screws fall out. 5 to be exact.
The wing was cut in four panels, left and right, top and bottom and each panel was cut on both sides. for some reason the last panel I cut was a few mm off and is leaving a gap in the middle. fortunately foam is pretty forgiving and I think I can adjust it by sanding the inside areas where the wood structure is set; allowing that panel to shift over.
. you can mix it in different thicknesses for different applications. its tricky getting the right combination of glue, water and talcum though.
