Official Luthiers Forum!

Started on checking neck set today on my first one, did this as the book suggested before starting the finish process. When I set it up, I put a straight edge along the frets, to see how far above the sound board I am at the saddle location. Now I am stuck again, not sure how to proceed. Here is the dilemma:

Comp. Book says the gap over the sound board should be 3/8, + 1/16 max. Now I get confused with mixed fractions, so I converted everything to decimal inches. Com. gives the height above the sound board to be .3906/.4375 inches. I put the heel solid on the end of the guitar, and measured 5/8, or .625". OK, so I'm off, need to shave the heel to set the neck. Then, I noticed a gap under the fret board at the rosette.

This first pic shows the heel sitting solid on the body.


This condition lets the end of the fret board stick up above the sound board at the rosette:


Next step, I rotated the neck so the gap was closed at the end of the fret board. When I remeasured the gap above the body, I got a measurement of .4062", which falls within the range listed in the book. Wonderful. But now, i have a huge gap at the heel, which can be shaved down to fit. Unfortunately, I also have a gap between the fret board and the body at the neck joint.



After checking, it seems the top is not flat in this area, so no matter what I will do, I will have a gap under the fret board. Is this normal? If not, then should the box have been made flat in this area? I so, why is it never mentioned, such as how to make one area flat while the rest is made in a radius dish?

Please help, I am confused and beyond frustrated at this point.

Thanks,

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Don't worry Phil, the right people will be along shortly to help.

I don't use molds and radius dishes, but I would think that a domed top that's truly a dome would need a LITTLE flattening in that area, consistent with what you're seeing. Once that's done the rest will fall into place it looks like. I would flatten the soundboard and see how it sits with the heel flush the way you have it. Looks to me that there will be minimal flossing at that point.

BTW that's one reason I rough radius my fretboard prior to gluing down, and then do the final radiusing and fretting AFTER the neck is finally attached, so that I can guarantee everything is true over the neck to body joint.

Let's see what folks say...

_________________
Old growth, shmold growth!

I'll try to answer this the best I can. I am working on #1 right now but have done a number of neck resets so I understand the geometry of getting the neck angle correct. I just recently started a thread about this exact issue. At the time I did not have my top on and had a 28' radius on the sides where the top connects. Having some knowledge of what needed to be done my question was how to deal with the 28' radius and the flat fretboard extension. My fear was exactly what you are experiencing now.

After doing a lot of searching and reading I was as confused as ever because guys do this in a few different ways. From completely flattening the upper bout to just leaving the 28' radius on the heel block and putting a 60' radius on the upper transverse brace to what I ended up doing which was to just flatten the top of the heel block and putting the 28' radius on my UTB and then making a flat spot about 4" in the center of the UTB to accommodate the flat fretboard extension. (that seemed the most logical to me)

In my case, I have the top on and am getting ready to put the neck on and everything looks correct. I am confident at this point that the angle is right and the heel fits tight against the body and the fretboard lies flat on the body as it should.

Unfortunately in yours case something is not correct as you pointed out. There are some unknown things you didn't mention that may help explain why you ended up with the gaps. What radius is your top? Did you flatten the area around the neck joint on the sides where the heel sits? What radius did you put on your UTB? All of these things have a direct relation to the geometry of your neck angle and whether your fretboard extension sits flat on the top.

Clearly something is not correct and unfortunately I am not the one to tell you how to fix it but I do have a good understanding of how to avoid this problem.

Hopefully someone with much more knowledge can tell you the correct way to fix it (which I know you can). I just would not want to guess the best way since I am still working on my #1 build. I think if you can answer the questions above it will help someone who knows the correct way to fix it.

I hope the info I posted about how to avoid this problem will help you in the future and maybe help you understand what went wrong.

Bob

The problem appears worse along the centerline of the guitar. When I put the neck in place, whether held in either position, the gap is minimal on either side of the fret board. By this, I mean I cannot slip a piece of paper under the edge on either side, which tells me the fit is within .004" across the width of the fret board. Along the length it is a different story.

i probably messed this up without realizing it when I profiled the sides before gluing on the top. If I had known this was supposed to be flat, I would have made it flat, not radiused.

I thought about sanding the top, but the top is so thin compared to the gaps, I will likely sand all the way through before I get the board to seat flat.

Thanks for the help, and keep the ideas coming...

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

I've read Cump. several times today, and guess what? It doesn't address the issue of having a gap under the fret board. (Or I should say if it does, I missed it.) It talks about neck angle, straight edges, locating the bridge, measuring the height of the straight edge over that area, adjusting the angle, etc. Went through that many times today. As I stated, Cump. does give dimensions for height above the saddle location, being in the range of .3906/.4375". With the heel sitting solid, my height is .625", meaning my angle is way off. So, put a clamp on the end of the fret board and hold it tight at the rosette, the height above saddle location is now .4062", clearly in the range listed in the book.

Putting a shim under the end of the board seems like the easiest solution, but my saddle will be quite high.

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Phil, there are two issues: 1) what you should have done and 2) what to do to get out from where you are.

First up, you're far from the first to suffer this problem, because, as you have found, most of the books do not cover it well. But at least one does! On another forum, I did a search on "neck angle" which turned up 90 odd pages of results, so you can see how much of a problem it is and how far short most of the literature falls.

The things that should be regarded as constants are the string height above the soundboard at the saddle position and the action. The neck angle is then set to make sure that those targets are achieved. Choose an action height that you like and the string height above the soundboard at the saddle position is usually chosen to be somewhere between 12 and 14mm. Personally, I go closer to 14mm, but it's up to you. What sets that height is the projection of the longitudinal tangent to the upper bout extended over the saddle position, because for a gapless fretboard fit with no wedges, the underside of the fretboard and the top of the neck have to be coplanar with the upper bout. So a straight edge set on the upper bout and projecting over the saddle position should show a gap at the saddle position on the centreline of 2.0 to 2.5mm depending on which end of the range you choose. With the underside of the fretboard co-planar with the upper bout, a fretboard ~6.5mm thick with frets of ~1.0mm height and action of ~2.5mm on the D string you should get a string height above the soundboard in the 12-14mm range.

So the trick, such as it is in this, is to set the curvature of the upper bout correctly, so that when the neck is coplanar, the projection over the bridge falls in that 2-2.5mm range. This can be achieved by planing a flat on the top of the UTB to effectively reduce its curvature to something of the order of a 60' radius. The angle between the bottom of the fretboard and the heel is then whatever it needs to be to make the gaps close between the heel and the sides.

Cumpiano, in his book, doesn't explain very well how he does it, but basically his process is to cut the heel at 90 degrees to the fret plane then buckle (for want of a better word) the top and sides until he gets the right projection over the saddle position. He does this by tilting the whole neck block, then locking it in place by gluing on the back. I'll let theguitarwhisperer explain the detail of how this works as he has direct experience with Cumpiano. So, in a nutshell, you have this gap problem because you have the wrong curvature on the UTB.

So what are your options? First, check how much material you would have to remove from the top panel in the fretboard extension area to get down to the correct gap when the tangent is projected over the saddle position. You need to make this area flat, anyway, as the fretboard has to be glued down, but it also has to have the right "tilt" to it. If you can't get the correct angle without removing more than ~0.75mm of material, satisfy yourself with making it flat, and you will then need to wedge the fretboard extension. If your last picture (gap under the 12th fret) is the right neck angle, you need to create a flat on the guitar top, as above, then remove ~ 1mm from the underside of the end of the fretboard, tapering to removing nothing at fret 12. The objective is for the fretboard to sit down flat on the guitar top with the fret plane straight, with the right spacing of the straight edge at the saddle position. You will then have a gap at the bottom of the heel, so will need to trim the shoulders of the heel to close that gap whilst still keeping the joint tight and the centrelines aligned. As you do this adjustment, you will be effectively shortening the neck, so make sure you can still get the bridge to fall in the correct zone, with the bridge pins holes still passing through the bridge plate.

So you're looking at a couple of minutes with a good block plane to flatten the upper bout and to create the wedge on the end of the fretboard and a couple of hours with a very sharp paring chisel to trim the heel and adjust the joint.

Patience is a virtue.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Thanks Trevor, that helps. But I am still confused on what the UTB has to do with this situation. If I understand correctly, UTB is upper transverse brace, meaning it is the one that runs across the guitar. This puts it 90 degrees to the longitudinal tangent line that sets the string height over the saddle.

To go from the sublime to the ridiculous, a tube is round but straight, a sphere is round but not straight. To me, it seems the UTB would function more like a tube, causing an arch across the body but having no effect along the length of the body.

I can see where increasing the radius of the UTB would have an effect across the fret board, but best I can tell, I don't have that problem. When I hold the neck in position, where the fret board extension contacts the upper bout I cannot slip a piece of paper under the edge on either side. This means there is less than .004" (0.1 mm) gap in the direction of the UTB.

Your explanation of the Cumpiano method is the best I have read so far. I did build to that book, and though I didn't really understand why, I followed it as closely as I could. It is nice to know I'm not the only one with this problem.

So if you or anyone else can explain that UTB connection a little better, it would be much appreciated!

Thanks again,

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Speaking for myself, that is the problem. I can not do anything without understanding why (because that is my nature).

I think I am able to explain the side to side radius on the UTB for you (at least how I understand). You want to be able to lay a straight edge on your top at the upper bout just where the fretboard will sit and have the correct gap where the bridge is for the set up you are shooting for (string hight).

The straight edge should lay completely flat on the top from the neck block to the front edge of the sound hole. Beyond that the straight edge will be above the soundboard giving you the angle you need for the setup you are trying to achieve. The side to side doesn't matter since that will be the radius of your top. You are only looking for that flat spot under the fretboard extension. The radius of the UTB is what will give you the correct geometry for your specific setup.

At least that is how I came to understand what I was trying to do.

SO... Experts, is my explanation correct? Please feel free to correct my understanding of this if I am wrong, I have yet to glue my neck on but I feel confident that my angles and geometry are correct and everything looks perfect on a dry fit.

I hope I explained it for you correctly and so you can understand it. It took me about a month to wrap my head around the "why" of it.

Cheers
Bob

I think Bob meant neck block...and this is what I think you want to achieve as well. And after it's flat you want a straight edge laid on that new flat surface to be 2 - 2.5mm above the sound board at the saddle location.

How you get there from where you are is the question. I've taken a block with 120 grit sandpaper and sanded that area flat but you need to constantly monitor the angle so you're seeing the 2 - 2.5mm over the saddle location that Trevor is recommending. Not that hard really.

I've also raised the lower bout of the guitar with a small block and sanded the area over sandpaper laid on a flat surface, very much like sanding in a radius dish only sanding just the UTB area over flat sandpaper. The small block is set at a height that will give that same 2 - 2.5mm over the saddle location.

After that area is flat and the angle is correct the neck needs to be set to accommodate.

Hope that helps.

Here's a really bad illustration if needed.

_________________
Thank You and Best To All

If you put an "angle finder" on the soundboard hanging over the side, how does that compare to the angle on the cheeks of your heel? The last picture, where you show a gap under the heel side of the fretboard end betrays a straight plane there.
By no means do I have enough experience in this arena to be throwing out suggestions, however, it would seem logical that if your tenon was allowed to slip further down into the mortise that would eliminate that gap. Assuming, of course, that the tenon is bottoming out. If it is, trim it back a smidge on the bottom, but don't touch the heel itself. If it isn't, then make sure the binding isn't holding the neck "up", so-to-speak. I think you're closer than you think you are, ftr.

**Edit** I'm not sure i was being clear, but basically what I'm asking is this; If you remove the neck, and lay a straight edge from the soundhole across the soundboard towards the neck, is that area flat? If it is, then see the advice above for the tenon adjustment. If it's not, then you've got a hump.

OOPS... Yes I did mean the neck block and not the heel...

I will correct it so there isn't any confusion.

Thanks,
Bob

I'd also add that all this 'flattening' should be done before the neck angle is cut as the angle at the neck block will very likely no longer be 90 degrees.

_________________
Thank You and Best To All

Sure!
I didn't really mention it it as it didn't seem relevant 'cuz the method being used wasn't Cumpiano's, but essentially, Cumpiano does the same thing domers do, but in a slightly different way that yields a flat spot under the fretboard and the correct neck angle, nothing is really buckled, unless you consider a domed top buckled, and the perimeter of the soundboard is flat to glue the sides to. Cumpiano just works WITH the dome in the rest of the assembly.

Here's Cumpiano's take on it from his blog, in the section dated July 24th 2011: http://dolcecano.blogspot.com/search?up ... date=false

Essentially, when the UTB is curved, the wood where the neck block is glued is tipped back. The middle 2.25 inches of the UTB are left flat so that the curve encompasses the rest of the brace.Instead of carving out a domed section in the solera, he simply uses a workboard shim that's screwed into place, with a hard section where the neck block is glued to the soundboard that is sanded to match the angle of tilt back induced by the curvature of the UTB. When the sides and back are glued on, everything is locked into place, and the result is a neck block orthogonal to the soundboard curvature at the right angle to yield the correct string height above the soundboard, with a flat spot to glue the neck to, and a straight fretboard from nut to fretboard end. The actual angle of the joint is 90 degrees to the neck, but the shoulders are tipped back slightly.

With outside mold construction, the block is glued to the sides in the mold, and the domed top is glued to the sides. Same with the back. As in the Cumpiano method, everything is locked into place at that point, and the result is squared sides, a domed top for rigidity, and a neck angle that yields the correct string height above the soundboard. Various methods can be employed to yield a straight fretboard from the nut to the end of the fretboard, but the actual neck angle will have to be measured and transferred to the neck heel, and the soundboard may need flattening under the fretboard.

_________________
Old growth, shmold growth!

Curve the UTB, hold it in place under the flat unbraced soundboard, and watch how the region where the fretboard glues down tips downward as you conform the soundboard to the brace...

This happens with any domed top unless a method is employed to avoid that, such as the soundboard facet described above by Larry, in which case the soundboard is bent down around faceted ribs anyway.

A domed top is geometrically a section of a sphere, because the totality of the bracing yields that result, despite what any individual brace is doing at any one section.

_________________
Old growth, shmold growth!

BTW, what method are you using? You mentioned radius dish so I assumed outside mold/radius dish?

_________________
Old growth, shmold growth!

OK, here is the method of assembly. First of all, Cumpiano mentions two braces, a wide flat one called Upper Transverse Graft, and another between it and the sound hole, called Upper Face Brace. I am assuming what is referred to here as the UTB is the same as the upper face brace.

I did my best to follow the book, but this being my first attempt, who knows how close I got. That being the case, i started with the Upper Transverse Graft, and it was flat. It was glued as shown in the picture below by clamping it with a stiff backer board, and a stiff board on top of it. This one had no radius whatsoever.



The next pic shows it after gluing:



The next one, according to the Cumpiano book as I understand it, is flat in the center, with a slight taper on the ends, as shown on page 167 for the UFB. In this photo, you can see this brace resting where it will be glued. Now maybe I put more radius on the ends than the book called for, but you can see that the center of the brace is flat at the sound hole.



Lastly, this pic shows how this brace was glued in place. As you can see, it was clamped to the sound board without any stiff backers, allowing the board to follow the profile of the brace.



So, for better or worse, that is how the braces were made and assembled. Hopes that help to answer some questions, because no, I did not use a radius dish.

Comments? Suggestions on what to do better next time?

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Spyder, I'm guessing when you profiled your sides you continued a radius on the top of your sides from the sound hole towards the neck. That would account for your gap.
Mike

The other guys have answered this and added drawings (thanks, Larry), but here's another way to look at it:

The X-bracing pulls the whole top into an approximation of a dome. To reach the peak of the dome you go "up hill" from all directions. When the UTB (upper transverse brace) is glued on, the curvature of the "dome" is locally modified, so that the approach to the top of the dome from the neck end could be more or less steep depending on the curvature in the UTB. Typically you want to make it less steep (and constant angle i.e. not curved) so that when the neck is coplanar to the upper bout the projection of that plane over the saddle position gives you the 2 - 2.5 mm clearance, and that is done by using a flatter UTB.

It doesn't matter which order you glue the braces on, the outcome is much the same.

Ideally, the top in its braced state but before gluing it to the rims should have the correct curvature on the upper bout to give the 2-2.5mm clearance, but it will only stay this way if the curvature of the rims matches the curvature in the top, so you don't change the geometry as you glue the top on. The corollary is that if the curvature of the top panel is not quite right, you can fix that by how you glue the top to the rims.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

I thought maybe he was mixing and matching techniques.

_________________
Old growth, shmold growth!

Here's how a 12 fretter should look if made according to Cumpiano's method.

The junction is 90 degrees square, the surface for the fretboard is flat to the soundhole, just the barest minimum of scraping assured flatness.

I'm thinking that if you followed the book pretty closely, you may have tightened the show down too much and introduced extra curvature to the soundboard.

The other possibility is that the humidity increased in your shop after the box was assembled and the wood rosette helped increase swelling at that spot.

Either way, I think Todd's original suggestion should clear things up for you.

This looks salvageable to me.

_________________
Old growth, shmold growth!

If your top is thick enough, you can use this technic to sand a flat area:
http://theamateurluthier.com/moreprojec ... ive20.html

The last 2 I built I used a flat top with a radiused UTB to achieve the angle and height at the bridge similar to what SeaGull guitars uses. I think this method has potential and will attempt to perfect the UTB radius to get that angle just right.

_________________
Thank You and Best To All

Thanks for the pic of the one "done right!" That sort of thing helps, it really does. So, for comparison, here is a pic of the same thing on mine. If you look close, you can see that the angle is indeed 90 degrees, or as close to dead on as I can see. but once you get past the slot for the truss rod clearance, the top begins to slope inward toward the sound hole. By carefully feeling the surface, it looks like the 90 deg. section is over the head block, and as soon as the sound board clears the head block it begins to slope downward.



What did I do wrong?

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Ok


Sent from my iPhone using Tapatalk 2

OK, may be on to something here. I have no idea what happened, but it appears that my top has sunk inward between the head block and the sound hole. After posting the last pic, I gave it a good look with a straight edge. I basically have two planes: one 90 Deg. to the heel, the other between the head block and the sound hole. Now if i put a straight edge along the sloping plane, guess what? The straight edge actually hits the far side of the sound hole. This indicates to me that this area has sunk inward. Why? I have no idea.



Something odd seems to be going on here. Humidity? Geometry? Clamping pressure?

If this is something not easily overcome, ie. inherent to the box on this one, then I will ask the next question. What happens if the saddle height is higher than ideal? I have seen it ad nauseum that the height needs to be low, but I can't recall why. If it works to be higher on this guitar, would it be so bad? Fixing this with a shim under the free end of the fret board would be the simplest solution.

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."