CNC router build

[difalkner]

7 minutes ago, micks said:

Nice! Fun to see the machine cutting something! There's definitely a learning curve to using a CNC router. Probably the single biggest obstacle for the new users I trained (and quite a few that have been doing it for a while) is determining the proper feeds & speeds for the tooling and material. You will break several more bits, but you'll learn something each and every time.

Yep, already broke another one.  When Fusion 360 generates the program it leaves the router at the last point, pocket in this case.  I added Y0 X0 to the end of the code so it would move back to home, or close.  But I added it to the last command on the line with Z raising to 0.5" and I thought it would raise Z and then do Y and then X... wrong!  It started Y and X while Z was raising and the bit just caught the edge of the pocket and snapped.  So I modified it again, not that I need the program, but I wanted to see if I did it correctly and put Y0 X0 on the next line.  Now it raises Z to the specified height and THEN moves Y and X; learning, learning, learning!

[Mick S]

Yep, I quickly developed the habit of running the file in the air after making those changes to make sure it does what I was expecting. Those bits start adding up!

[difalkner]

I was setting up to cut a 24"x24" board when it occurred to me that I had not checked the table for coplanar accuracy.  This is the initial test and it's out enough that I need to correct it before going much further.

Here's the setup I used -

 

[difalkner]

This isn't really part of the build except that it's a natural progression to at least show the CNC machine working, so I thought a little project was in order.  The build isn't 100% complete, though, but close enough to begin using the machine.  I still need to get the VFD talking to the software (Modbus, connected via USB adapter - right now I have to open the electrical enclosure and use the keypad to turn the spindle on/off and change speeds), I need to make an auto-zero touch plate for the Z axis, need to figure out how to mount the Z- axis proximity sensor, and I truly need a dust shoe - really, really need a dust shoe!!

A little background on the project - I'm building acoustic guitars which means right now I build more templates, fixtures, forms, and jigs than guitars.  One of the tools frequently used is a radius dish or a set of radius dishes for different guitars.  People frequently call them 'flattop guitars' but most aren't even close to flat. The top and back have radii ranging from 12' to 65' or more.  The radius helps keep the guitar stable, makes it stronger, and helps with the sound (very basic explanation). What I'm using is a 28' radius for the top and 15' for the back so last August Sandy and I got out in the driveway and made some MDF dust, lots of MDF dust.  I know the dishes aren't cut perfectly so this looked like a good project to start the CNC on its path to production.

The video is a little longer than most I've posted but still tolerable at 8 1/2 minutes and starts with cutting the first radius dish by hand and progressing to cleaning them up on the CNC.  It's interesting to note that the DC is much louder than the CNC and also how the mic picks up sound - the radio isn't as loud in the room as it sounds on the recording and the fans for the enclosure aren't that noticeable but they are in the recording.  The hearing protection I wear is for the DC, though, because it is annoying and just loud enough that I won't run it without ear plugs.

 

[Tom Cancelleri]

Awesome stuff! I hated making radius dishes by hand. I need to figure out how to make a 20' radius dish for my Go-Bar deck in Fusion 360, it'll be on my list of things to do.

[difalkner]

5 minutes ago, Tom Cancelleri said:

Awesome stuff! I hated making radius dishes by hand. I need to figure out how to make a 20' radius dish for my Go-Bar deck in Fusion 360, it'll be on my list of things to do.

Revolve is your friend in Fusion 360!  Once I figured that out it took less than 15 minutes start to finish to draw the disc and create the tool path.  Holler if I can assist - David

[-MattK-]

hooray for finally getting to make stuff!!  it's been great following along with the build, David.

and yeah, you definitely need to figure out a dust shoe!  Chasing that router with a hose looks tedious as hell!  Is there something standard for that router or will you have to figure it out yourself?

[Tom Cancelleri]

3 minutes ago, -MattK- said:

hooray for finally getting to make stuff!!  it's been great following along with the build, David.

and yeah, you definitely need to figure out a dust shoe!  Chasing that router with a hose looks tedious as hell!  Is there something standard for that router or will you have to figure it out yourself?

I see a lot of people make their own using their CNC. There are some retail options available, and are really nothing more than what most people are making themselves out of 1/4" acrylic and aluminum. 

[Mick S]

Great video! Glad to see it cutting a project. I downloaded F360 but haven't tried messing with it yet.

[difalkner]

Thanks for following along, Matt! 

I have only just started looking for commercially available dust shoes and have several I'm considering but I haven't checked on size yet.  This spindle is 100mm diameter but most seem to be 80mm.  I'm still thinking about building my own, though.

11 minutes ago, -MattK- said:

hooray for finally getting to make stuff!!  it's been great following along with the build, David.

and yeah, you definitely need to figure out a dust shoe!  Chasing that router with a hose looks tedious as hell!  Is there something standard for that router or will you have to figure it out yourself?

I have a few ideas I want to try so I might go through a trial and error period of 'thrown together' dust shoes until I settle on a final design.

7 minutes ago, micks said:

Great video! Glad to see it cutting a project. I downloaded F360 but haven't tried messing with it yet.

Thanks, Micks!  I'm getting better at Fusion 360 for basic things but certainly have a long, long way to go.

[Elroy Skimms]

Does the increase in accuracy from your ingenious spinning MDF solution to the CNC make a big difference in the finished product? It strikes me that the CNC may be more accurate, but at 90 minutes per plate, there is a lot of time lost in the process. You might be able to improve the accuracy of your old method by using the CNC to cut the arc path of your jig, and then going back to spinning the MDF. With Fusion360, you could accurately model the base of the router and make sure that the radius of the arc is very precise and adjusting for the depth of the router bit. Something something something tangent of the arc.

You know, this might actually be the first time in my life that I can come up with a use for calculus, outside of the classroom. It's a shame that I have forgotten pretty much all of it. You'd think that after taking the class 3 times I might remember it...

-E

[Mick S]

1 minute ago, Elroy Skimms said:

Does the increase in accuracy from your ingenious spinning MDF solution to the CNC make a big difference in the finished product? It strikes me that the CNC may be more accurate, but at 90 minutes per plate, there is a lot of time lost in the process. You might be able to improve the accuracy of your old method by using the CNC to cut the arc path of your jig, and then going back to spinning the MDF. With Fusion360, you could accurately model the base of the router and make sure that the radius of the arc is very precise and adjusting for the depth of the router bit. Something something something tangent of the arc.

You know, this might actually be the first time in my life that I can come up with a use for calculus, outside of the classroom. It's a shame that I have forgotten pretty much all of it. You'd think that after taking the class 3 times I might remember it...

-E

Don't want to speak for David because there may be another reason, but I suspect he's running that slow because he's still getting used to the machine. He should be able to up the feed rate by a factor of 3 - 4 for that process. 

[difalkner]

31 minutes ago, Elroy Skimms said:

Does the increase in accuracy from your ingenious spinning MDF solution to the CNC make a big difference in the finished product? It strikes me that the CNC may be more accurate, but at 90 minutes per plate, there is a lot of time lost in the process. You might be able to improve the accuracy of your old method by using the CNC to cut the arc path of your jig, and then going back to spinning the MDF. With Fusion360, you could accurately model the base of the router and make sure that the radius of the arc is very precise and adjusting for the depth of the router bit. Something something something tangent of the arc.

You know, this might actually be the first time in my life that I can come up with a use for calculus, outside of the classroom. It's a shame that I have forgotten pretty much all of it. You'd think that after taking the class 3 times I might remember it...

-E

I don't think it makes that much difference at all although if I said that on one of the Luthier forums where the radius of the back and top are discussed and debated daily I would probably get two dozen negative comments in the first day! 

This was more about me learning than achieving a true difference in the finished product.  I did realize the radius was slightly off and my OCD nature wants that to be correct so I took the opportunity to both correct something I knew was slightly out and to learn in the process.  And to be fair, it is more like 3 hours per plate because I had to get the back side level first and I took my time doing that, as well - LOL!  But I view it as time invested and not time lost.  Now, if I was in a production setting that would be different.

My previous method was fairly accurate but I think there was a bit of flex in the fixture plus a bit of runout spinning a 24" square centered on a 1" wooden dowel pivot point and four low friction plastic pads on the periphery.  I don't have any desire to recreate that and make it better.  Oh, wait, I did - I built a CNC machine!!

[difalkner]

43 minutes ago, micks said:

Don't want to speak for David because there may be another reason, but I suspect he's running that slow because he's still getting used to the machine. He should be able to up the feed rate by a factor of 3 - 4 for that process. 

Yep, see my replay to E above.  And I'm not certain how much faster I could have run that.  As I said in the video I got to a max of 175 ipm and it seemed to be moving along at a pretty good rate although 250 ipm doesn't seem to be out of the question.  I don't know how many times it went around but I do know it is just under 75k lines of code and that takes a while to run even though it's just going in circles.

I have the stepover at 0.0625" but I could change that to 1/8" and speed things up although I might end up with ridges.  Seems like Fusion 360 put it at something less than 1/16" but I don't remember.  The reason I slowed it down for the start of the cut and until it got to a few inches out is that at 100 ipm and cutting in a tiny circle of less than an inch the machine, while not walking around, was doing some serious movement!  So I ramped it up from 25 ipm and that smoothed things out.

[FineLineAutomation]

The machine rated to 350ipm cutting feedrates. You should be able to speed it up dramatically.

Also, one thing to remember is bits don't like to run too slow. If your feed is too slow you introduce rubbing and that will seriously impact tool life.

I highly recommend you get a tool feeds calculator like HSM Asvisor or G-Wiz that can give you an accurate feed and speed for a bit and material combination. I have HSM Advisor and can run the numbers for you if you need them.

[Elroy Skimms]

Have you looked at http://www.themakersguide.com/home/products/triple-edge-finder-2 as an edge finder? I've heard good things about it, but I don't know anyone who has used one with Mach 4. The Mach3 settings and plugins are easy to find. I have no affiliation with this, but I do have a vested interest in finding someone to be the guinea pig for these things and a Mach 4 setup

-E

[difalkner]

2 hours ago, Elroy Skimms said:

Have you looked at http://www.themakersguide.com/home/products/triple-edge-finder-2 as an edge finder? I've heard good things about it, but I don't know anyone who has used one with Mach 4. The Mach3 settings and plugins are easy to find. I have no affiliation with this, but I do have a vested interest in finding someone to be the guinea pig for these things and a Mach 4 setup

-E

Yes, I've looked at them but haven't gotten one.  I'm sure that Bill Griggs can help set it up, though, if you get one.

[difalkner]

I guess technically the build is complete but there are still a few things I associate with 'the build', like dust control.  Tonight I ran a 7' piece of 4" PVC for dust collection and connected it to my partially complete dust shoe.  A friend is suppose to be getting me some clear vinyl so I'm holding off ordering anything to complete the dust shoe.

The hose reaches the extents of the X, Y, and Z travel but I need to work on securing it so it doesn't flop around.  At this point I'm not certain what I'll come up with for that, though - just started thinking about it. 

I'm still not accustomed to coming up with a design/idea, being able to draw it in Fusion 360, and then be cutting it out a short while later.  It's a given that we all do that with the other tools in our shops but this is just a bit different for me and it's still pretty cool!

I had some scrap 5/8" MDF and 2x4 so that's what I made these brackets out of, then painted them Ferrari Red, Yellow, and Black (that's important, btw - LOL!).

Anyway, here's what I have to this point -

[Chestnut]

Good to see this guy complete, doesn't look like it takes up too much space .

So what's the first elaborate project? I feel like this CNC is itching to cut something fancy for a project.

[difalkner]

2 hours ago, Chestnut said:

Good to see this guy complete, doesn't look like it takes up too much space .

So what's the first elaborate project? I feel like this CNC is itching to cut something fancy for a project.

I'm so far behind on other projects that this may actually take a back seat for a little while.  There are several Christmas gifts I want to make, though, so I need to get (and stay!) busy for the foreseeable future. 

[difalkner]

After running the PVC pipe and connecting the hose to the dust shoe, yet to be completed, I needed a way to manage the hose movement.  I worked through several mockups before deciding on making this mount out of 1/2" Baltic Birch.  I'll try it for a while and see how it does but the first thing I need to do is make a collar of some sort so the hose doesn’t slip down.  It won’t actually be that big of a deal because I imagine most of my work is going to be in the first 24” of the table and the hose doesn’t move in that area.

 

[Mick S]

7 hours ago, difalkner said:

After running the PVC pipe and connecting the hose to the dust shoe, yet to be completed, I needed a way to manage the hose movement.  I worked through several mockups before deciding on making this mount out of 1/2" Baltic Birch.  I'll try it for a while and see how it does but the first thing I need to do is make a collar of some sort so the hose doesn’t slip down.  It won’t actually be that big of a deal because I imagine most of my work is going to be in the first 24” of the table and the hose doesn’t move in that area.

 

Consider running the hose up to the bracket you made and cutting it there. Fx a collar to both sides of the bracket, then attach the hose to each side. The hose can then be quickly and easily disconnected on either side.

[Chud]

On 10/20/2016 at 9:31 PM, difalkner said:

I drained the water out of the system and am preparing for distilled water and the RV antifreeze but thought I'd mark the lines so the pump will know which way the water flows when I turn it back on...

You may have covered this but might I suggest you fit something that will shut down the spindle or sound an alarm if water isn't flowing? One way to do this would be to put a float sensor in the reservoir (if sourcing one is problematic try an aquarist/marine aquarist supplier).

Awesome build - wish I had the space to do something like this!

[difalkner]

4 hours ago, Chud said:

You may have covered this but might I suggest you fit something that will shut down the spindle or sound an alarm if water isn't flowing? One way to do this would be to put a float sensor in the reservoir (if sourcing one is problematic try an aquarist/marine aquarist supplier).

Awesome build - wish I had the space to do something like this!

Thanks, Chud!  I've thought about something like that but for now I'm just watching it closely.

****************************************

Quick update - I said I needed a way to keep the hose from falling down toward the spindle so what I came up with is a hose retaining nut, left hand threads, of course... Yeah, they didn't know what one was at Lowe's, either

It's possible I could add a second one below the mount and maybe hold the hose in place better but I don't think the little bit of movement will be a problem.  I could also put a retainer clip on the 'nut' and hold it in place but again, this probably won't be an issue.

 

[difalkner]

This isn't really part of the build except that I was testing not only the machine's ability to engrave small letters, 0.25" high, but my ability to do it in Fusion 360 which is definitely NOT friendly to this sort of thing.  I saw this small piece of Honduras Mahogany and thought it would be a good test and it turns out the machine had no issues at all with this.  I may try smaller just for grins - larger won't be an issue.  But one of the things I plan on using this CNC for is inlay work on guitars so I need to know it will perform well on small, intricate, delicate pieces of wood and Abalone (or Mother of Pearl).

Here's a short video of my test -