29 Replies Latest reply on Aug 18, 2011 8:48 AM by Adam Meyer

# Variable thickness, on an archtop Guitar, impossible?

Hi All,

I was wondering if any of the experts here can shed some light to this problem...might be easy to do, but for the life of me, I do not seem to know of a way to do it in SW2011...

I have attached a pic of the model I am building:  It is that of an archtop guitar, which as you can see, I have modeled the top surface.  I did this using cycloid curves which I calculated, and then building the mesh, which allowed me to create the surface.  So far I am pretty happy with the way it looks, although it needs minor tweeks....now here is where it gets interesting:

I want to create a solid for maching BOTH the top and the bottom, but the thickness of the top varies down it's length...for example, the thichkness at the center is about 1/4 inch, then it goes to 3/16 and then down to about 1/8 as you approach the edges...

How can I thicknen this surface, or what tricks can I use, to achieve this?  Obviously I want the part to follow the surface from the inside....any ideas on tools or a method?  I am thinking something with a sweep, but it beats me how to achieve this....

Thank you all for any suggestions or ideas!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Welcome to SW forums John.

Just throwing some balls your ways:

Create the bottom surface at required thickness and then loft both the faces.

OR

Thicken the surface and then cut the extra material.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Well, creating a bottom surface, that would be a replica of the top, and again, I have the same problem...the thickness varies from the center to the edge...so I do not know how to vary that thickness, while still maintaining the surface variances constant...

Trying to thickness the whole thing uniformly and then cutting what I do not need, poses the same problem...creating some sort of geometry to do the cutting, while maintaining the curvuture of the surface, but varying the thickness...see where I am going with this?

I am sorry if I am not making any sense..just trying to figure this thing out.....

:-)

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Well can you thicken the surface and do a revolve cut with a taper kind of thing. Sketch the line tapered to surface such that it one end point is in the the center and distance increases or decreases as it goes outwards. Sorry don't have SW with to explain with a pic or model.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

John,

I can't claim to know proper guitar terminology but I think you and Deepak are referring to different "bottoms'.  If I'm not mistaken you have modeled the surface that the strings will be against, I'm assuming this is a hollow body acoustic guitar.  What I think Deepak means when he refers to the bottom is what you would consider to be the inside surface of the top.  Try Lenny's suggestion for making a surface as Deepak suggests.

Erik,

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

offset 3 surfaces, one at .25, one at .188 and the last at .125. then like a topo map offset the outer edge on a plane the distance you want the transition between those surfaces then project the geom onto those surfaces which become your crosssection for the new loft of the bottom.

this is a simple explination of how i might aproach this.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

This is a GREAT surfacing puzzle.  My take on it:

--Think of the top surface as a 3D mold surface or freeform datum.  You may want to change the definition of the surface so that it has a range broader than the "trimmed" surface pictured.  i.e. One might make the surface definition as a large rectangular or oval patch, then trim to shape.  Using tight-cornered border edges to define surface geometry can lead to problems later.

--Don't be afraid to make surfaces simply to use as tools, e.g. making a surface that is 3/16" offset, then trimming to create an edge that will be used as a loft curve.

--Remember you can loft to a single point (such as a single point in a sketch).

--Think of how to use guide curves and tangency/curvature controls rather than simply throwing in more loft sections.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Thank you for those ideas.

-Yes, I am actually using a much larger surface, which I then trim down.  There is simply no way that extruding any surface around the "horn" of the guitar, will yield good results.  surface needs to go in all sorts of directions, and this throws the program for a loop....making it oval, then triming back, seemed to do the trick, for what i have so far.

-I see what you and Lenny are  both implying...basically using the existing surface I have as a means of creating 3D countours in space, which then can be used to extract the final surface, at given depths.  I will give this a try.  I am afraid however, that I might loose some definition or resolution to the top carve, since it was created with quite a few cycloid curves.  i guess I will try and see what happens....

i will report back my results, and keep those ideas coming!  Thank you very much for your time...

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

I don't get it.

You want the back to be flat or curvy? If you want it to be curvy, what kind of curvy? You say variable offset, but I don't get exactly what you mean. Will it be convex or concave? Do you basically want to copy the existing surface and move it offset a given distance?

You don't want to use thicken. You want to use ruled surface to the edges, and set that normal to a plane. This will create straight extruded edges. Then, you want to fill in the new edges of that extrude somehow. But I can't provide any help on that without more information on what you want the back to look like.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Well I guess a picture is worth a thousand words....

Here is the cross section, at the mid-point of the top (looking through the back..).  The top, I want to look like I have posted on my original post, and the bottom, to be following that, EXCEPT, that it varries in thickness, as you can see from the image above.  It is about .25 at the peak, tapering to about 1/8 at the edges....

Hence my question:  If I simply thicknen the top surface, I will get the bottom to follow, but it will have a uniform thickness, as a result of the thickness I use in the thickness tool.  I want it to vary, so I guess I have to thicken it a certain amount, and then figure out a way to cut it, to achieve the results above...

Does this make more sense?  again, my challenge is that I do not want the top surface to change, since I have used curtate cycloids to create it, and looks pretty good...I just want the bottom to follow the top surface, but vary in thickness from the peak to the edge, as per the above curves...

thanks again!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Thanks for that image, that does make it quite a bit clearer. You will not get this to work with thicken. This is very similar to the work that Robert Stupplebeen does, so I will make sure he sees this.

One idea that I wouldn't discount though, that I recently thought of to do something like this:

Use move/copy body, and make a copy of that face, but moved over the proper distance for the edges.

Then use the freeform tool to tweak the center areas to the thickness you require.

I've never actually used freeform successfully before, but this is the type of thing where it should theoretically work very well.

If you were able to mathematically calculate the top surface, can you mathematically calculate the bottom surface?

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

aahhh...I tried using freeform in the past, and somehow, I got lost!    One of the reasons I LOVE solidworks, is that it allows me to think like an engineer, and be VERY precise on my measurements, mid-points etc....I get lost in modelling applications that do not have this, like for example the polygonal modelers...(3D max, Modo, etc...)

Tugging a little here, pushing a bit there, is a way that I just do not like modelling....hence my quest for trying to find out if there is some sort of a way to achieve this...more precisely... :-)

Thank you for your time however, and maybe your friend has more ideas on how to achieve this....it is eating at me!!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

John Withers wrote:

Does this make more sense?  again, my challenge is that I do not want the top surface to change, since I have used curtate cycloids to create it, and looks pretty good...I just want the bottom to follow the top surface, but vary in thickness from the peak to the edge, as per the above curves...

thanks again!

John we are not tell you to change the top surface, if you offset new surfaces they change only when the top surface changes not the other way around, we do this all the time for ID models, you have to build something to make something else. we offset, trim, fill, make new lofts that are tied to the top surface. hope this makes sense....................

again put up the top surface and let us have a go to show you and we'll send to file back to you so you can see how we think and go about it.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

John Withers wrote:

I just want the bottom to follow the top surface, but vary in thickness from the peak to the edge, as per the above curves...

Deepak has it right.  You are going to have to model the bottom surface somehow and then loft the two together.  The bottom cannot "follow the top surface" if the thickness varies.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

darn...I hope that is not the case....it just way too much work to model all those cycloids for another surface, while varying the height....

I am trying to think along the lines ofhow I can use the surface as I have as a "tool" to create the final result....but my head keeps spinning!  LOL!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Here is my attempt.  You will have to model the bottom surface before you do the overall trim to get the shape because the cutout around the neck will give you problems.  In my example I had to trim away the cutout and mirror the top surface to get it to work.  Just ignore the features in the folder.  You can play with the dimensions on the spline to adjust the curvature of the constraint curves.  If I had the extended surface I would have done things differently.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

YOU NAILED IT!!

The result looks awesome!  I am trying to understand exactly what you did, but the cut looks absolutely what I was after!

Adam, you are a surfacing GENIOUS!

Thank you very much for taking the time to do this.  I just looked at the results, and they look exactly like I had envisioned.  May I ask you a couple of questions if I have them after studying this?

again, thank you very much sir, for your work and GUIDANCE!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Sure, my email is in my profile information.

John Withers wrote:

YOU NAILED IT!!

The result looks awesome!  I am trying to understand exactly what you did, but the cut looks absolutely what I was after!

Adam, you are a surfacing GENIOUS!

Thank you very much for taking the time to do this.  I just looked at the results, and they look exactly like I had envisioned.  May I ask you a couple of questions if I have them after studying this?

again, thank you very much sir, for your work and GUIDANCE!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Put up the file and we'll take a crack at it to show you how on your file!

we can type all day and you still won't see untill you have something in your hands, it's more for an aproach, you have to make it your way, though

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Thanks for that offer!

I would love to see how folks might approach this problem...

I hope the attached file works?

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

I'm getting a bad knit between the bottom two segments. Can you re-export that at a higher tolerance? maybe try .STP, or if possible, .x_t? The edges just aren't clean between the two, and it's making my method not work properly.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

OK..I drove the knitting tollerances down all the way, and exported as a parasolid....maybe this will work now?

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

John,

Basically my best ideas have already been presented.  1. Define a 3D function of the outside surface and then mathematically offset the surface for the inside.  Matlab or similar package would be needed here.  2. Create construction geometry with known offsets and use this as the guides for the internal surface.

If you can post what you have as inputs to the design that will aide in helping you with more options.  Do you have multiple thickness sketches and profile sketches that you are working to?  Is your concern with the internal surfaces that the acoustics will be affected?  How accurate do you need them?  The internal ones may acoustically be more critical than the external 'show' surfaces.  Keep in mind your manufacturing tolerances when considering what is good enough.

You could also look into reverse engineering an existing product with a 3D scanner or CMM.

In my process I have a known surface and a desired thickness for some regions.  I then create less controlled geometry to mate up to these 'engineering' surfaces.  For my processes I am trying to hold near 1 micron accuracy in the engineering surfaces and allow for 10-20 in the other surfaces.

I hope this helps

Rob Stupplebeen

PS: Thanks Charles for pointing me to this thread!

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

Thanks Rob!

I think that is what I need to do....I will use every cycloid I created for the top surface, and then create an equivalent with the given offset for th bottom surface.

I was hoping not to have to do this, as it is a long and tedious job, but sometimes, ones has to do what needs to be done.

Yes, exactly, the variations on the thickness, is what gives it the proper "tone" and hence the reason why there is a variance.  This is more important for the back of the guitar, since it basically needs to vibrate, to amplify the sound.  Somewhat less important for the top, since the top will have "f-holes" cut into it, which make it more flexible, and hence more vibration prone if you will...

Just thinking that machines can do this much more accurately than humans, but the difference is in the details, and ultimately the accuracy of the model....

thank you again.  I will give it a whirl, and see where it goes...

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

John,

Attached is a technique that I think should work for you (*** credit to Lenny for his suggestion ****).

I essentially defined split lines to break your surface into three faces.  I then offset the different faces (like a topographic map) and created filled surfaces that joined their edges together to create a smooth surface.

Given that carved top wooden instruments are a combination of art, science and luck, I think this approach is probably as good as any.

Best regards,

John

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

I don't have time to make you an example, but if you made the top solid with a flat "bottom" you could shell the item with a multi thickness shell. You may have to add some sketches to define the thickness areas, but I think this may be a solution to your problem.

Take it one step further, create a solid body of the entire sound box, then shell it the same way. Cut sections out of the result and save each section as a part.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

john go look at T-splines for rhino and then T-splines for SolidWorks......this might be the thing you need.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

HI Lenny,

So i did the trial for t-spines about 4 months ago....the product has potential, but what drove me nuts, is that I needed to model something in Rhino first, then bring it into SW, then work from there...

Do not take this comment wrong, but I can not stand the modelling workflow of rhino....just drives me crazy...no design tree....it is a VERY capable surfacer from what I have seen, but the interface and workflow, needs a major update in my humble opinion...working in the SW interface, feels like "home" to me...

I had provided that feedback to the T-splines folks, and they said they have gotten this from several folks, and were apparently working on a newer version, which will be SW native, and no need to exit SW....I do not know if they are there yet, as I have not visited their site lately...but it would then become interesting, as it would make SW a quite better "organic" modeller....

Hear that SW?  Maybe you should "buy them???" LOL!

Best regards.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

all I would do is make my model in SW the open in rhino tur to tspline and reopen in SW. rhino is just the filter..... isn't it.

• ###### Re: Variable thickness, on an archtop Guitar, impossible?

OK..that works.....but it was not told to me that way....I will take another look I guess....

regardless, thanks for all your ideas sir!