CNC gun Stocks?

Been a while,, life gets in the way. Anyhow:

Hobbyist level, I have built 3D printers and refurbished an old Epilog Laser, Now I am looking at building a task specific 4 axis router for gun stocks.


I have seen what appears to be 2 distinct styles of carving a gun stock on a 4 axis router.


one has the 4th axis just "indexing" the stock and the tool carving Longitudinal as shown in this video: https://www.youtube.com/watch?v=P5_nT1hsQf8


The other is where the stock continually rotates on the 4th axis as it is carved "Radially" https://www.youtube.com/watch?v=Ik_UKwVTMv8


Questions:


Is one method better than the other?
Does it require different CAM packages to perform one or the other?
What would be a good "Hobbyist" CAM package to perform one, the other, or both?


It looks like they are using a Ball end bit for their carving, for the final cut it would seem the point of contact/cut would be infinitely small, requiring the passes to be in the millions for a smooth finish without ridges?


I have found books and articles explaining generic CNC router theories, but nothing in depth dealing with 4th axis, bit choice and carve paths for desired results,, Does anyone have a good source for reading material for these subjects,


Many Thanks,
Slack

To do this particular kind of work on a 4th axis, I'd likely opt for the indexing lathe because these setups are capable of asymmetric work. 4th axis work requires software that can "wrap" a model. One good example is Vectric's software. I you will use someone else's model, you can cut with VCarve Pro. If you want/need to do your own modeling, you'll need Aspire. There is certainly other software that will movel do the work, too.

Ball nose tooling is used for 3D cutting specifically because you can raster and use other path techniques to gradually get to your shape by managing stepover as you note. Do not expect to get a perfectly smooth surface off the machine regardless of what you work with...it's not going to happen. Even at the most minimal stepover measurements (which would result in enormously long cut times), the surface will still require normal attention just like you carved it by hand. The CNC is not a "finish" tool when it comes to this kind of milling. It does the drudgery work for you and makes it repeatable. The "finesse" woodworking is still on you. :)

One other thing...4th axis work is complex. You are going to want to get a solid handle on 2D, 2.5D and 3D work on the three axis side of things before you even consider moving on to the 4th axis. There's a learning curve.

My intuition says it's likely one would be preferable to the other depending on whether you're milling solid wood or a laminate. And I have no idea which is which. I've milled one glued up from Baltic Birch ply by indexing, and it went pretty well. There were a few places where it wanted to chip out, but BB is a poor substitute for a proper laminated blank to begin with.

There are more efficient tools than a ball endmill. A 'bullnose' specifically, a plain endmill with a radiused corner. But you'll likely have to finish up some details with a ball end anyway. With a big enough machine you can use a big long cutter and 'sidemill' with it (aka 'Swarf Cut"), which would be much faster and probably a better finish. CAM software will have to support that as you need the cutter off center of the 4th. In some ways, this is harder to generate than 5 axis toolpaths. And there are still the areas you can't reach ...

As Jim mentioned, there will be a fair amount of hand work no matter what. So even a 1/8 to 1/4" stepover with a 1/2 to 3/4" ball will get you about as good as you need. That comes off pretty quick with a good rasp or coarse sandpaper. And you're going to be leaving 'stock' for finish sanding below that anyway.

And, how are you going to generate a model to cut? These shapes are not easy. I use Solidworks, and have basically no experience with free form shapes. So it's a real bear to get anything other than the forearm and butt profiles pretty close to finished, as well as the inletting. But around the pistol grip was more or less just cutting the profile and then getting to work with a rasp. I've been watching video's about advanced modeling, but at 60+ there's a good chance I'll never get proficient enough at it.

The other option is to hand finish a pattern and digitize it. If I get around to building a machine I'll probably think hard about adding the hardware/software for that.

Or experience has been unless your ganged (4+ per setup more like 6-10) it's nearly impossible to make money one at a time.

I do not plan on going into "mass production", I am more dissatisfied with the products currently available and I am initially exploring "one off" examples for my own use.
"If" there ends up being an interest in the future, than that avenue will be explored in the future.

Quote:

---

Originally Posted by Mark Bolton

Or experience has been unless your ganged (4+ per setup more like 6-10) it's nearly impossible to make money one at a time.

---

Thanks for your reply.
Yes, the hand finishing is what will "make or break" the finished stock.

My strong area is fabrication, mechanical and electronics, My CAD is week, only from lack of time.
My 2D vector work is good,, I was going to grab a small "throw away" import 3D carver to get up to speed on 3D design/3 axis carving, while I am fabricating the larger machine to carve stocks.

I am thinking my approach to gun stocks would be 2 sided 3D carving, then the let in as a separate operation.

Thanks again for your insight.

Quote:

---

Originally Posted by Jim Becker

To do this particular kind of work on a 4th axis, I'd likely opt for the indexing lathe because these setups are capable of asymmetric work. 4th axis work requires software that can "wrap" a model. One good example is Vectric's software. I you will use someone else's model, you can cut with VCarve Pro. If you want/need to do your own modeling, you'll need Aspire. There is certainly other software that will movel do the work, too.

Ball nose tooling is used for 3D cutting specifically because you can raster and use other path techniques to gradually get to your shape by managing stepover as you note. Do not expect to get a perfectly smooth surface off the machine regardless of what you work with...it's not going to happen. Even at the most minimal stepover measurements (which would result in enormously long cut times), the surface will still require normal attention just like you carved it by hand. The CNC is not a "finish" tool when it comes to this kind of milling. It does the drudgery work for you and makes it repeatable. The "finesse" woodworking is still on you. :)

One other thing...4th axis work is complex. You are going to want to get a solid handle on 2D, 2.5D and 3D work on the three axis side of things before you even consider moving on to the 4th axis. There's a learning curve.

---

Wes,

Thanks for the info on the mill bits,, I am sure it will be a learning curve, but that gives me a starting point.

For me, building a gun stock specific machine should be basic (not quite simple), as I am good with fabrication, mechanics, electronics. I have built a 3D printer from scratch and rebuilt my obsolete Epilog machine.

The modeling software and setting up a CAM package to the parameters of a home built machine are going to be the tricky part.
Specifically the 3D modeling,, I am "almost" not looking forward to learning a modeling package.
I too have thought about rough cutting and hand shaping an example and having it scanned. But that would defeat the purpose of having a machine to make "one Off" stocks that come out only needing minor sanding/finishing,, not major shaping.

We shall see.

Quote:

---

Originally Posted by Wes Grass

My intuition says it's likely one would be preferable to the other depending on whether you're milling solid wood or a laminate. And I have no idea which is which. I've milled one glued up from Baltic Birch ply by indexing, and it went pretty well. There were a few places where it wanted to chip out, but BB is a poor substitute for a proper laminated blank to begin with.

There are more efficient tools than a ball endmill. A 'bullnose' specifically, a plain endmill with a radiused corner. But you'll likely have to finish up some details with a ball end anyway. With a big enough machine you can use a big long cutter and 'sidemill' with it (aka 'Swarf Cut"), which would be much faster and probably a better finish. CAM software will have to support that as you need the cutter off center of the 4th. In some ways, this is harder to generate than 5 axis toolpaths. And there are still the areas you can't reach ...

As Jim mentioned, there will be a fair amount of hand work no matter what. So even a 1/8 to 1/4" stepover with a 1/2 to 3/4" ball will get you about as good as you need. That comes off pretty quick with a good rasp or coarse sandpaper. And you're going to be leaving 'stock' for finish sanding below that anyway.

And, how are you going to generate a model to cut? These shapes are not easy. I use Solidworks, and have basically no experience with free form shapes. So it's a real bear to get anything other than the forearm and butt profiles pretty close to finished, as well as the inletting. But around the pistol grip was more or less just cutting the profile and then getting to work with a rasp. I've been watching video's about advanced modeling, but at 60+ there's a good chance I'll never get proficient enough at it.

The other option is to hand finish a pattern and digitize it. If I get around to building a machine I'll probably think hard about adding the hardware/software for that.

---

Got a little side tracked,, but actually getting back on this project with help from a VT electrical engineer and programmer :)
Getting ready to ask some CAD questions in the CAD forum

If you haven't yet touched base with him Gary Cambell is very knowledgeable on this subject. I believe this is his website https://www.centroidcnc.com/shopbot_gcnc.html

Interesting topic... my dad was a gunsmith from after WWII and eventually went into machine shops. We have always had a lot to do with guns and built a lot of custom rifles. Back in the early 80's I drove to Tucson to Harry Lawson's shop and hand picked a piece of walnut and had a semi-inletted Cochise Thumbhole stock made for a Ruger 77V in 25-06. While I was there, I got to see his machinery for making stocks. He had a Terrco stock duplicator, may have been modified, it held one master stock and could machine four blanks at a time. All the stocks were chained driven to each other as they rotated and all the actual machine work was done by hand with a stylus following the master.

In today's age, I would probably design the 3D model, then set it up to cut it from the four sides, top, bottom, right and left. Programming would be relatively straight forward, most 3D software can machine/design to a splitting point in the model. You could fourth axis the whole thing, honestly not sure how solid a 20" or longer piece of wood is going to keep from chattering and flexing, so fixturing it where you could put intermediate clamps might be better.

Any way you look at it, this is a big project and will definitely be a labor of love, not a profitable situation.

Quote:

---

Originally Posted by Brian Lamb

...You could fourth axis the whole thing, honestly not sure how solid a 20" or longer piece of wood is going to keep from chattering and flexing...

---

Ding ding ding...You hit the nail on the head.

Most who machine custom stocks and make money at it, just do them on 2 or 4 sides. I have several customers who send me stocks to scan on a regular basis and this is how they do it. You can't exactly blair through a piece of English walnut between centers and expect to put your name on it in the end.

Yes, there are sweatshops in the far east that use multiple ganged rotary duplicators, both manual and automated, to machine mass produced stocks. But what you don't see is the amount of sanding required to produce a sellable product. Even so, there's no money competing with these generic stocks, but there is in one off customs...and honestly, machining time isn't a huge deal as machining is a small part of the entire custom building process...so I wouldn't get too hung up on machining other than doing whatever results in the highest quality piece.

Quote:

---

Originally Posted by Brady Watson

Ding ding ding...You hit the nail on the head.

Most who machine custom stocks and make money at it, just do them on 2 or 4 sides. I have several customers who send me stocks to scan on a regular basis and this is how they do it. You can't exactly blair through a piece of English walnut between centers and expect to put your name on it in the end.

Yes, there are sweatshops in the far east that use multiple ganged rotary duplicators, both manual and automated, to machine mass produced stocks. But what you don't see is the amount of sanding required to produce a sellable product. Even so, there's no money competing with these generic stocks, but there is in one off customs...and honestly, machining time isn't a huge deal as machining is a small part of the entire custom building process...so I wouldn't get too hung up on machining other than doing whatever results in the highest quality piece.

---

This has always pretty much been my answer when people come in the shop either wanting me to machine a stock or telling me I should start making stocks. The mere knowledge, time investment, customer investment, in just FITTING a custom stock to an individual is insane. Many have multiple sample stocks on hand, fitting jigs and fixtures, etc.. the actual machining of the stock is a pittance in the overall process.

I have 2-4 sided a few stocks for friends just for fun and it comes back to when you add up your drawing time, machining time, tooling, forget about the finishing if you hand it off to them to sand, you will never sell the first one to the average joe because they simply will not pay the cost-of-entry to get to a even a rough stock off the CNC (rough meaning tight stepovers, super long machining times, etc). And a lot of super custom stocks you are not going to machine them close without 5 axis and will require substantial post processing if your coming off 3 axis operations.

The individuals willing to pay for a serious custom stock are not going to deal with some shop with a cnc machine. They will be looking for a true custom gunsmith.

Plus one to Brady and Mark's comments. I have built a dual rotary machine for an individual that makes custom shotgun stocks. His claim to fame is canting the butt stock left or right off the gun centerline in half degree intervals as part of his custom fitting process.

To avoid stock twist and vibration a second stock holding chuck was fitted in lieu of a tailstock and it was driven by a shaft and timing belts to match the headstock. This feature increased cut quality and accuracy by a long measure. And if you consider how much shorter than a rifle stock the 2 components of a shotgun are, this makes close tolerance machining of a rifle stock even more of a challenge on a traditional rotary axis.