Carving tall, thin walls in a curved shape

Machines (Original/PRO) How To
1

I’d love to get some advice from the brain trust here regarding a machining strategy for tall, thin walls. Instead of bending the 3/32" thick sides of a guitar with heat/steam and clamping them into a mold (as one typically does for an archtop or acoustic guitar), I’d like to cut the sides out of a solid block of wood. I realize that this seems a little crazy but I have some unique design and construction characteristics that make this very desirable as well as strategies for reducing wood waste.

This would mean that the finished sides would be up to 3" tall or even a bit more but also very thin. The most obvious toolpath would be to use an upcut end mill with an extra long flute length that would just cut around and around until it was deep enough on the inside and outside of the wall. However, I worry about this creating a very deep slot that the machine wouldn’t like too much and that the first .25" of bit (depending on depth of cut) would dull so much faster than the rest of the flute length. I’m also concerned that the thin walls would start to flex during the cuts if they were pushed too hard during the latter runs as the walls thinned out. Finally, workpiece holding issues may come into play.

Any ideas/suggestions?? Thanks, as always, to all the smart folks here who take the time to share their experience and creativity.

2

I have built a few guitars and always bent the sides, I think milling them out of 3" thick stock would be very risky even with a 1/2" endmill, I think a 1/4 in will not work very good.
You would end up with end grain on both ends of the carve resulting in very weak structure even with the bracing and triangle corner pieces, I would recommend bending for the strength alone.
My 2 cents worth only.
Thanks, Pat

2 Likes
3

Hey Pat. You have very good points. These sides wouldn’t be wrapping all the way around as they would only be the most outer sides so the end grain wouldn’t be taking much traditional load. And, in addition to the sides being thicker than typical, there would also be additional bracing mechanisms in place as well. (A little hard to describe without seeing the design but it’s not a typical acoustic guitar structure at all.)

I have spindle with 1/2" collet if I wanted to go that direction.

4

I would agree that any areas of end grain that thin would likely not be strong enough to survive the machining operation. You would likely need to step down in small increments evenly on both sides of the part to keep the cutting forces from deflecting or breaking the part. You could also consider creating the blank from layers of wood with alternating grain direction to provide better strength.

1 Like
5

You can also cut one side of the profile inside or outside then hot glue some blocks on the finished side to support the finished part, have to be careful removing the blocks but you wouldn’t need much glue in my opinion, like to see the finished part when you are done.
Pat

6

Hey Jim,

I would use your CNC router to make a Fox bender 1 / Fox bender 2 (Build instructions) and fox bender patterns as required.

Further Reading

7

Thanks all. So, a little more background: Using my 1F I was able to build an awesome “side-bending machine” like the ones @Aiph5u linked to. Because my instruments are uniquely shaped, I had to make my own version instead of buying a standard template and it went great. Lots of plans and design ideas on-line but, of course, I did some things differently. However, it’s mostly traditional in regards to side-making. I also bought a stand-alone “bending iron” for some particular side curves and shaping the binding, etc. And I built a steam box as well, although I found that technique to be less useful for my particular needs. Anyway, no stranger to traditional bending methods.

My desire to carve tall thin walls for the sides is more about creating a different sort of animal than traditional guitar designs. Some adventurous guitar designers (Kasha, Ribbecke, Gore, Brosnac, Symogyi, Parker, etc.) have developed instruments that push the boundaries and that’s what I’m into. Lots of guitars use body elements that are carved from a solid piece of wood. Some are mostly solid (like a Strat) or topped two-pieces (like a Les Paul) or even carved from behind and have a back added to cover the chamber (like a Rickenbacker). Sure, all of these guitars have exposed end grain but lack of structural integrity to offset the string tension isn’t an issue with these designs. As the sides get taller and thinner, it starts to be something different completely.

@PezWoodworks That’s not a bad idea Pat. I had thought about bracing the inside but not the hot glue. (I have tried hot glue for workpiece holding and hate it for that application.).

All, what’s the proper toolpath if I want to the end mill to go round and round but always carve a little wider than is necessary by making multiple passes so that the bit isn’t in a tight channel while it’s cutting? If I go with a “Final Finishing Pass” will that help smooth the sides so that I have a smooth side and not ‘layers’ of ridges at the cutting depth? Would 1/2" shank end mills be better or actually cause more potential for failure because they are more aggressive?

Any other ideas would be more than welcome.

Thanks!

1 Like
8

I would suggest milling the outside straight, full depth, but leave some material for a finish cut.
This keeps the outside from bending of course because of all the material on the inside of the body is still there to support it.
Then start milling out the middle but leave some material against the walls uncut to give the walls strength. How much to leave will be trial and error depending on the type of wood. I would think stepping the inside would be good. (1/2" at the top to 5/8" at the bottom?) The idea is to allow the walls to release stress and move where they want to.
Then take a finish pass on the outside, straight down. Trial and error will determine if you can do it in one pass or step down.
Then work away at the inside of the walls but work only one step at a time (1/4" down? may be less?) This way the material down below the step still supports the wall where you are cutting. Finish off each step complete before moving on to the next step - you will not be able to come back and cut it later, it will be too weak.
On each step down step out a little ( 1 degree? maybe 2 degree?). This allows the wall above to move a little but still not hit the cutter.
In this way you may if lucky get 3" down with the outside straight and the inside stepped.
You may have to customize the wall thickness on different areas ( thicker on endgrain thinner elsewhere). Start out with a scrap of the same wood only 1" high see how thin you can get it in each direction. Then move on to 2" high and perfect your method as you go.

1 Like
9

One other thing, make sure your spindle is trammed as close as you can get it, I regularly cut 2" thick hardwood without a last full depth pass and there are no lines at the step down depths that can’t be sanded in a few seconds, my spindle in very near perfect trammed.
Thanks Pat

1 Like
10

Hey Jim, hey all,

it would add to precision if you were able to mill the half of the 3" depth from one side and the other after flipping the workpiece around. 3" long bits tend to bend. But I don’t know if you want to have the opening on one side only.

Do you know this video? They mill a rather deep groove into solid wood with an 8 mm (0.325") two-flute upcut bit with 24,000 rpm and 5000 mm/min (= approx. 200 IPM) feedrate and 3 mm plunge (approx. 1/8") (PDF project description), but only half the depth, then they flip the workpiece over by using registration dowels in the wasteboard. The final workpiece still stays attached through small wooden bridges that are removed after milling with a small hand saw.

Bits they used:

By the way, as motor, they use…

1 Like
11

Decided to play a little:
Piece of 2x4 pine. Cutting 1.9" deep (my longest cutter - a bit dull)
First side with straight wall - .23" deep per pass, .100 step over, 13000 rpm, 100ipm
IMG_0021
IMG_0022

Second side with 2 degree taper - leaving .340" thick wall - .23" deep per pass, .100 step over, 13000 rpm, 100ipm
IMG_0028
IMG_0029

I will leave it over night to stress relieve. It is also too late to make that much noise.
Tomorrow I will finish the straight wall then finish the tapered wall - trying to get 3/32" thick walls.

1 Like
12

Here is the result. .100" at the top and .160" at the bottom. 1.90" deep.
IMG_0031
IMG_0036
IMG_0034

At no point did I think the wall was going to break. The wall was solid while machining - even the end grain wall. ( Though it is incredibly weak now when I push on it )
Obviously having the end grain come out to the side was unwise. It was inevitable I was going to loose chunks there.
Finishing the taper side proved to be the most important part of the process.
For it I used 13000rpm, 50ipm, .125" step down, .20" step in first pass then .05" second pass then .03" final pass. Each .125" level was done complete before moving down to the next level.

I feel it could have gone thinner - it sounded so good milling. Perhaps .06" thick at top.
But it may be something that sounds perfectly fine until it breaks.

Hope this helps. It was fun.
Please let us see the guitar that results from this.

1 Like
14

To answer some of your questions directly:

All, what’s the proper toolpath if I want to the end mill to go round and round but always carve a little wider than is necessary by making multiple passes so that the bit isn’t in a tight channel while it’s cutting?

In steel I definilty would rough a slot a little wider than the cutter by cutting in one direction, then moving over 10% to 20% of the cutter width and cutting back in the other direction. Cutter deflection and chatter is a killer in slots that are size for size with the cutter.

With wood however I find it not so important. With 1/4" endmill I have no problem stepping down into deep slots that are size for size in pine and walnut. In hard maple I need to widen the slot. So it depends on the hardness of the wood and the amount the cutter deflects. With 1/2" endmill I think it would be stronger and less of a problem.

But chip evacuation is the problem. Stepping over and cutting back in the other direction will help a lot with getting the chips out.

If I go with a “Final Finishing Pass” will that help smooth the sides so that I have a smooth side and not ‘layers’ of ridges at the cutting depth?

I would suggest using a final finish pass cutting .005 - .010" more perhaps.

Would 1/2" shank end mills be better or actually cause more potential for failure because they are more aggressive?

I can’t see why 1/2" would be worse than 1/4". It should be stronger and more stable.
Aggressiveness is controlled by your feed per tooth. Just make sure the rpm does not make the cutter chatter. ( hitting sort of the resonant frequency of the cutter. )

1 Like
15

Hey Chris, hey Jim, hey all,

interesting experiment. I would not have thinked the wall would break either, even without the socle that you left on the bottom. You have a clamping problem this way :slight_smile: I would have given preference to the method in the video I linked above with the little wooden bridges – it would have made that socle unnecessary.

But if Jim hadn’t said he was building an electric guitar, of course I would ask, how can that sound? Or do you intend to realize some acoustic properties on this partly-hollow guitar Jim @jfleser?

This reminds me what I once saw on youtube, a crazy guy who builds a violin with sides made this way – not by bending the sides, but by using a solid hardwood piece and mill the sides with his cnc router, just like you show here. And additionally, he poured the holes of his workpiece with epoxy previously. When I saw this, I could only shake my head, how can someone think how a tone and stability get into a violin body? :slight_smile:

On the other hand, nowadays you can give children the chance of their life by allowing them to get a real violin into their hands they possibly would never be able to afford with a 3D-printed violin project ( [1], [2] ) – and it works too! There was even a violin printed with 3D-Laser sinter printer technology now. So maybe the sound and stability of wood fibers do not seem to be only thing nowadays… :slight_smile:

16

Hello Aiph5u,

I did not catch that Jim was building an electric.

I had thought Jim was intending to mill the sides and back (or front) of the guitar all from one piece and then glue on the missing face to this “pocket”. So it would be like what I did - the thin walls attached to a solid base - no bridges or tabs needed.
However I could have interpreted his intensions wrong.

Cutting out the sides only with no base and have it look like what you get from bending a guitar side would be more difficult. More care would be needed when it comes to the final pass at the bottom on the inside. Yes bridges would be needed.

As for sound - following tried and true gives great results, as we all know.
But it’s the areas that we don’t know that appeal to me most. We need people willing to go there. So go for it Jim.

2 Likes
17

Hey @Aiph5u @ChrisM et al,

Wow! What a great community. This is perfect. You guys are asking all the right questions as well and I’m pleased to address your curiosity. Hopefully I won’t bore you (too much).

Regarding the strength of the sides: One of the innovative luthiers I mentioned (Trevor Gore) has done some interesting research into the function of sides in guitar building; specifically acoustic guitars where wood vibration plays a greater role in defining the ‘sound’ of the instrument. This is opposed to solid body or semi-hollow body instruments where the wood vibration plays a lesser (but still important) role in the creation of the actual sound you actually hear. If you play these guitars without an amplifier, there really isn’t any real ‘tone’ that you’d consider putting a microphone in front of.

Archtop guitars exist in a world between the flat top acoustic and the typical electric guitar. I’m playing with the idea of carving portions of the sides to coexist with a dramatically different overall guitar architecture. It’s something that’s never been done before. (To the best of my knowledge and research anyway; there are tons of really smart, creative people in this world.). Although the ideas are patent pending now, I still don’t want to disclose too much publicly just yet. But I promise to post more here later.

Here’s a good summary of Gore’s research on another builder’s website. LINK. As you can see, the idea is that sides with more mass can actually help define a clearer more vibrant tone. Not ‘stiffness’ per say, (which bending is really good at providing), but ‘mass.’ So, if I carve the sides so that the integration of the sides with my other architectural innovations is cleaner and facilitates easier manufacturing, I can freely leave more mass because the sides don’t need to be extra thin yet also very stiff. The mass works in my favor.

However, the irony is that when it comes to fine “Archtop” guitars the ultimate expression is considered to be the ‘custom hand-built instrument.’ And, of course, our CNC machines are despised. But what if there was an incredibly innovative design that required such attention to detail and precision that it would damn near impossible for it to be carved by human hands? There are a couple of pieces of my instruments that would take so much time for a human to carve that it would make any guitar too expensive to truly be feasible. But these damn CNC demons… They can crank out these intricate carves in a manner of hours (sometimes six or more) without complaining and with great consistency.

OK, maybe my two Makita routers “complained” by simply bursting into flames. Buy my 220V, water-cooled spindle is cooking along fine and never seems to break a sweat. So far, anyway. No sense tempting fate.

Yes, I had seen the Hammer video as well as olive wood violin guy. There is also a team of folks who are led by Tom Nania (link) who has used CNCs to crank out carved archtop top plates for detailed testing. I met with him a few weeks ago at the Rocky Mountain Archtop Festival in Arvada, CO. We had a great talk and hoping to have him do some testing on my instruments for comparison.

Yes, flipping the piece over midday is something that I’ve considered as well as adding the back plate to the sides first and sort of carving them together. This type of suggestion and this back-and-forth in general is exactly what I was hoping for. Thanks again to all of you.

This was the primary reason for buying my 1F. It’s taken more than three years now to get where I am and though the CNC added to my already tough luthier learning curve and R&D, it’s been a joy to be so absorbed. My (very patient) wife calls it my “guitar PhD” years. The instrument looks amazing in Fusion360 and seeing it become a reality is just glorious; but hard as hell.

4 Likes
18

Hey all,

Here’s an interesting example of a guitar with carved (instead of bent) sides. This concept actually leverages an entire chamber hollowed out with the back carved thin but left in place. So, my design is very different (being much deeper like a hollow body and with sides much thinner as well) but you can see how it might be applied. The spruce top is dropped into place via an carved inset and I bet the industrial CNC carved the channels for the inset binding (with is usually on the corners) as well. This top is flat instead of arched as well but it’s an interesting design.

I played one for about an hour as a store during a visit to Vancouver this week and it’s an interesting take on the “acoustic sounds and electric sounds on one instrument” concept. However, the acoustic sounds were much better than the electric output and it played much more like a thin acoustic than a real electric guitar.

Screenshot 2023-11-29 at 9.31.07 AM
Screenshot 2023-11-29 at 9.31.07 AM790×1169 59.8 KB

1 Like
19

Hey Jim,

you did not provide a source for your image. What is the manufacturer?

EDIT: Okay, found it, it seems it is the recent Fender ACOUSTASONIC® Series.

It reminded me the Squier Telecoustic that I had seen in second-hand shop a few years ago, but it has a plastic body :frowning:

20

@Aiph5u Yep, it’s a major manufacturer. I’m not sure how big of success it’s been. Here the patent link. It seems what they’ve done is made a pretty good sounding acoustic guitar in a smaller size. So, the “problem” they’ve solved is that players don’t like playing an acoustic guitar that is full-sized. Maybe that’s not something a lot of folks are complaining about. The “electric-ness” of the guitar is left wanting. Other folks, like Godin, have done what I think are better attempts; here and here.

21

Hey Jim,

…and historically, there was also the ES-330 that was hollow and had no sustain block like the ES-335, and its cousin, the Epiphone Casino. It is known from John Lennon who played it and is recognizable by its neck meeting the body at the 16th fret (19th fret on ES-335)