Official Luthiers Forum!

The top is handled in much the same way as the back. I didn't have to mess with the join, the plates were fine just as they were. The grain was a little wider in the middle but also better quartered (or vertical, depending on how you mean it) and I don't mind that so I left it. I sent it a couple of times over the actual jointer and then got the plane out and went to the joint:



I took me a couple of swipes but this time I was successful with the plane and had a nice tight joint. I use hide glue to join the top plates and this time I used the tape method which has been working well for me on the last few. Here is a shot with the top set up with the offset tape:



And then the glue is run down the center and the back is taped to hold it shut.



As with the back, my next stop is to the deflection jig and to take some measurements to add to the pool.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

I have been testing the plates using the Chladni method for 10 guitars or so and I have been slowly adding more and more tests to each instrument. For this one I think I scheduled about 7 or 8 separate tests. My setup is straight forward, I have a small amplifier and a signal generator powering a handheld speaker. I use small foam pyramids under the plates and aluminum glitter that I found on ebay as the medium. I try to look from about 30hz up to 500hz for each test and I have printed out forms for all of the data. My setup is hidden behind my pinup of Neko Case with her old Gibson:



It folds down and is leaves everything at a comfortable height to dial in the correct frequency and keep the speaker cord from touching the plate.



And a shot from the side showing the setup and other tools



The first tests I run are on the glued up top plate thicknessed to .14. I am very influenced by the work Mark Blanchard does and specifically by the talk he gave at Healdsburg a few years ago about his method of identifying tops to be used for specific sized instruments. In an effort to glean some knowledge out of the unprofiled plate I am testing it as a rectangle and then at each shape down to the one I will be using. In this case I am making my CL model, currently my smallest shape, so I could cut it down slowly and test the top at each of the other shapes as well. The idea being that the information you get from the rectangular plate will tell you which sized guitar that top will naturally excel at. The problem is, of course, that we don't all build the exact same shaped guitars and it takes a lot of trial and error to arrive at the data you will use to judge your own sizes. I think it is worth the effort, however, and I am dedicated to keeping the data and expect that at some point it will help me to make much more informed decisions and to maintain a high level of consistency in my instruments. As it is now, though, much of it is somewhat meaningless. Especially these early tests as I have only just begun keeping these records.

The first one is of the rectangular plate. I will show the data I get and explain a bit about each shape. You can see how the shapes remain fairly similar but the frequencies change. Here is a shot of one of the modes shown in the glitter:



And the results for the rectangular plate:



After this I drew all of my shapes over one another onto the plate. and then cut it down to may largest size, the model B. It is 20 3/4 long and 15 3/4 wide. Here is the data for this shape:



The next shape is my model CG. This is the only shape I make where the waist is not at the golden section of the body. This model is 15 1/2 in the lower bout and is 19 3/4 long, as are the next 2 as well. Here is its data:



Next is the model CLM which is very similar to the CG, as you can see from the data. It is a bit less wide in the lower bout and has the waist shifted back a bit. Here is its data:

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

And finally we get the shape we will be using, the CL shape. Here is its data:



I also test the back although I test it a bit thinner, at .1. here is the data for the back at that thickness and cut out to the CL shape as well:



Having gotten what I wanted documented I can put those into the file and move on with the construction for now.

My top plate is still at .14 and cut to the shape we will be using. Before I thin it down further I need to inlay the rosette. I used some of the milling cutoffs from cutting the back and sides for this. I have a little plexi form for making an 8 tiled rosette and I traced each tile onto the longer blackwood piece.



Each of these pieces is then sanded flush on the back side to the line that was originally drawn.



Then, moving over the the table saw the miter gauge is set to 22.5 degrees and the 2 outer angles are cut. The first is done by eye to the line and the second is done with a stop so all of the tiles are the exact same size.



Once these are all cut successfully and fit together well to form the closed shape I bring back out some of the veneer I thicknessed for the purflings. The inner line around the rosette will be this same veneer and I want to use it as the intersections between each tile. It is easily cut with a sharp razor blade into small section that will fit between each tile.



The padouk veneer pieces are CA'd between the tile by hand pressure. It is important to keep the upper edges tight with each other when holding them together.



I generally do them in sections and end up with 2 halves to join up. Once all of the section are completed and joined I double stick tape it down to a sacrificial board. The board should have 2 perpendicular lines drawn on it and they are used to line up the glue up and to locate what will be the center. Once it is taped down I run it through the sander and clean up the top face.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Out of the sander a hole is drilled at the center point and I use my circle jig to rout out the circular rosette from the glue up. Depending on which double stick tape was used it may be necessary to let a little solvent soak in under the rosette to cleanly release it.



On my soundboard I mark where I will want the rosette inlaid and mark what will be its center point and drill my pilot hole there for the circle cutting jig. Tracing the rosette shows me where I want to open up for the inlay.



I try to get a nice snug fit. If there is some slop it can still work but more care needs to be taken when routing the thin openings that the purflings will fit into.



And then it gets caveman clamped in place and allowed to dry fully.



If it was a bit proud it will need to be leveled and then the channels for the purflings will need to be cut. In this case I used the same purflings as I did on the backstrip and cut a 1/16 channel on the inside and outside of the rosette.



Before I added the purfling strips I sealed the cut channels with a healthy bit of shellac. I want to use the thin CA to lock in the purfling pieces but I have been bit before with the CA wicking into the end grain of the spruce and staining it yellow. I use a brush to coat the channels and use quite a bit, especially where there may be exposed end grain.

The purflings are then gently hammered in around the rosette. With something as thin as 1/16 they generally do not need to be pre-bent, thy conform very easily when hammered in.



When flooding with CA in this case i want it to cure as fast as possible to avoid any getting through the shellac barrier but i am wary of the accelerator still. After the flood I quickly scrape the protruding purfling pieces. The little shavings bond with the CA and will very quickly cure it all. I am careful before sending it through the sander to remove any scraping pieces from on top of the spruce, I worry that the sander could embed them into it if they were there. Once they are gone though it goes through the sander again and we magically move from a gross looking glue pile to a nice clean rosette. There is some nice curl and silking in the blackwood, I think this is going to look great. The padouk pieces around and between the tiles is a subtle touch I like as well.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Burton, I went through the 3 pages (as of now) of your build and thoroughly enjoyed it.
It was a pleasure to see you again in Woodstock, I hope your finger is healed by now!
One comment I would have is when you cut your purf strips on the table saw: please use a bigger sacrificial block over the purf sheet and clamp it to the fence. Just lower the blade, put the sheet you want to cut up, set the block on it without pressure, and clamp it with one of those Irwin clamps. Turn the table saw on, then raise the blade until it is buried under the block, but not sticking out. When you want to cut up a sheet of different thickness, repeat the procedure. Does that make sense? Holding this block with one hand is inviting kick back and possible disaster.

_________________
Laurent Brondel
West Paris, Maine - USA
http://www.laurentbrondel.com/

Hey Laurent! I had a great time seeing you at Woodstock, it was great to get to chat in person.

Your suggestion is a great one for the table saw. I don't know why I didn't think of that, after hearing it it seems like the obvious way to do it. I will definitely be changing how I cut those. My finger still looks a little Frankenstein-y but in all other ways is back to normal. Not to make light of an injury, but it was great on Halloween.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Taking the plates down to the thickness they will be braced at is my next step. Considering how much time I spend doing deflection readings and running Chladni tests even I would think that I have a fancy way of knowing when I am at the optimal thickness/taper to begin bracing but unfortunately I don't. All of that data will be helpful at some point but for the moment I am still going by feel for this step. It has been working well, I have been very happy with the guitars I have been making and hopefully at some point I will see trends in my logs that correspond to what I see and feel and that will be great. For now though I have to trust my instinct.

For the back plate I think of it as similar in function to the top plate. I consider the upper bout to be very structural, it has to resist distortion for the heel to stay put. I want the lower bout to be looser and more responsive. I am looking for the frequency of the joined top and back to be the same before I go to finishing and considering the back has a more pronounced arch and is somewhat stiffer to begin with I need to thin it a bit. I usually leave the upper bout down to the 3rd brace more or less the same thickness and thin in a semi-circle from there back. I will run the plate through the sander at angles (I love the 10-20 for this especially) and thin in the lower bout one or more times depending on how much thinner than the upper bout I want it. I am looking for the lower bout to lose its internal strength. I don;t know how to describe that really, it feels stiff and then at one point while you are slowly thinning it it almost feels like it gives up a bit. That is the feeling I am looking for in the lower bout. Different woods hit that point at very different thicknesses. This wood hit it around .075 or so which I feel is pretty normal for anything with a hard surface(that is not a dense rosewood). Here is a shot of the criss cross sanding before I clean it up:



My tools to clean it up are a tiny Lee Valley scraper, a nice carpet razor blade, some 80 grit sandpaper and my ROS. I will work the transitions first with the scraper and then blend them with the 80 grit(usually also with a small block). The razor blade is a great less aggressive scraper. I finish up with 80 and 120 grit on the ROS, on both sides of the back plate.



Once I am done with this I can never resist a nice wetted shot of the back and do a little daydreaming about what it will look like when it is all together:



Then, before I start bracing I run yet another Chaldni test on the ready-to-brace plate. This one yielded more modes than the other tests.



I also ran another deflection test and filled out my back log page.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

I work the top in a similar fashion to the back. After inlaying the rosette I noticed a little inclusion on the face, you can see it in the rosette shot which I will show again here. It is below the rosette.



I also had an isolated bearclaw on the back side of the top that I did not want to show through on the face if I could help it. I had to keep these both in mind when bringing the thickness down.



I felt the inclusion on the face was by far the worse of the 2 and began my thicknessing from the front. I knew the inclusion was shallow and it did sand out quickly and with the littlest bit left in terms of thickness. It turned out fine, the inclusion was removed and the bearclaw is not showing through on the face. This is always an issue, I didn't see either of these things in the rough plates and luckily they did not become an issue here.

I taper the top in the same fashion as the back. I do start my tapering lower though on the top plate. Since this guitar will have the buttresses I also tapered the top leaving it thickest from just above the soundhole to a bit behind the bridge. I felt it was at good strength at .102-.105 in that area. I tapered it down to just above .09 on the edges at the back and a bit thicker at the top. I blended everything with the same tools as the back and made sure to fill out my data form for the top plate:



My next step, before running my Chladni tests, is to deeply score the soundhole in an attempt to mimic a little bit what the top may look like with the soundhole cut out. I started doing it like this a while ago and am feeling now like it probably doesn't need to happen or maybe even is a bad idea but for now I stick with it. I don't fully cut it out because I want to glue a support on before cutting all the way through and I want to see the chladni data before committing to a bracing pattern which would affect the soundhole support. Hence the dilemma. I try to leave as little wood left as I can when scoring the soundhole and then I run the Chladni test:



After running the test I flip the top over and run them all again but this time draw the most powerful lines (the most crisp ones) onto the back side of the top. For a while I felt like this would more or less tell me where my bracing would go, I would just use lines I was seeing on the top. Now I see it a bit differently. It all depends on the premise that it is the top itself that is the driving factor in the sound and the braces are only aiding that along. Using that thinking (which I admit seems simplistic in this case) I try to find areas where many lines are crossing. I think of these as points on the top that are not vibrating most of the time (the lines show points of no vibration) and places that can be strengthened without much loss to the plates' efficiency. In other words, the peaks to the scallops. Again, this is still a premise and has no actual science behind it. It is based on many assumptions that I know people find hard to stomach. It keeps it interesting for me however.

Here is the backside of the top with the nodes identified:



And then a bracing pattern laid out with the nodes marked for later brace shaping:

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Once my bracing scheme is laid out I can finish up on the soundhole. I cut out my support piece to fit up against all the braces that will flank the soundhole and glue it on:



I then run the center hole through this new piece and cut out the soundhole opening from the back. It is important to leave the circle cutter setup with the exact same bit and position as was used to score the soundhole before the chladni testing.





This cut is usually very clean but I still give it a little sanding just to be sure. For this guitar I want to bind the soundhole with the same wood I used for the backstrip and will be using for the binding. I set up my bending iron and bend the rosewood into the correct shape to fit inside the soundhole.



A little fitting is next to make sure it fits in there nice and snug and then I can glue it in. I try to leave more overhang in the inside rather than the outside if I can help it. It is easy to scrape the top flush with a few strokes and then send it through the sander to flush the inside. Since the support is there I don't risk sanding off my bracing pattern.



And all cleaned up it is looking very nice! Gavin and I had discussed the guitar being much more classic than extravagant. No pearl,inlay, marquetry etc... I think this will be a clean, simple, and proper treatment for this guitar. A little figure and a perfect match for the back and sides.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Hey Burton,

Your work has a professionalism that I really appreciate. I am very excited that you are documenting here. I know it can be hard to deal with the photography and posting and I thank you for the time you are taking. Your work is a cut above the rest and I will study this intensely.

_________________
http://www.booneguitars.com
https://www.facebook.com/pages/Stephen-Boone-guitar-builder/488208541257210

I love the way you made your rosette. I make a hardwood rosette, but just have routed it out of a whole large piece. Do you have formulae for different rosette diameters? If so, where are they available? I'd never figure out the math on that! Thanks again, Burton. Your work style is similar to mine (though much more advanced). I love the detail work, so I am really appreciating this build...and will learn alot! Beth

Thanks guys, I appreciate the comments.

Beth, I am not sure what you mean by the diameters? Do you mean the angles to cut the tile ends at? I will admit to cheating as my incra miter gauge has them printed on it! I always draw everything out first though and I made my little plexiglass piece to trace the tiles from a drawing. I hope that helps.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Yes, Burton, I was referring to the tile angles. I'll have to look at an Incra miter guage. I think the angle would depend on the diameter of the rosette and the number of pieces used, so I was trying to visualize how to get there. Beth

Burton,
I'm not trying to pick at your words here, but just get a clear understanding. I usually think of a "dip" as something that falls and then rises back up - like a small valley or relief in the middle of the fretboard. But I think you mean "fall off" or "ramp" to use my words. You are sanding a fall off from fret 14 all the way to the sound hole end and it does not, obvioulsy, rise back up. The fretboard ramps down slightly over the guitar body. I have read where others do this. Correct?

Aslo, when you sand lastly with a flat bottomed level along the string lines, you will create flat spots on the fretboard along these lines. How do you handle this to get back to a true smooth radius?

Thanks much for your excellent effort. It is very helpful to me.

Ed

HI Beth,

Actually, I checked again and it was easier to remember than I thought. The angle to cut at is 180 divided by the number of sides. If there were only 4 sides you would cut 45 degree angles. Mine had 8 sides and 22.5 degrees etc.... Good old math. Mr. Dudack would be proud of me. (probably prouder if it had been on my brain right away!)

Ed,

You are correct, it is as you say more of a ramp than a dip. I don't have a set measurement I go for in terms of how much of a ramp. I should though.

In terms of sanding with the straight bar I am not trying to remove hardly any material. It is mostly just for checking to make sure that each string path has a straight and level ride to the saddle. I am also not super anal about having an exact true radius at every point on the board. The biggest advantage to having it perfect would be for the fret to be ultra perfect when pressing in with cauls. In reality, there probably are very small flat spots and not exact radius' along the plane. As long as each string path is clean I can live with it. If they are so flat that it is easily noticeable though, then I would consider that a problem and smooth it out.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

You are setting the bar very high, bravo!

_________________
Bobby Masten

Hello everyone! I have some more to post here. I have been reading back over this and thinking recently and I have been feeling some doubt as to whether all of the extra work I put into some of the processes is worth it. Is it worth all the documenting, all the elaborate parts, or even making my linings the way I do? There have been some persuasive arguments made that constructing the linings to be very rigid and structural is moot once you glue the top and back on. I felt at one time strongly that my method was what I wanted to be doing. And, truthfully, I still do. I am less certain though every time I play an amazing guitar of someones and it is built in the classic standard method with x braces, tone bars, kerfed linings, etc...that all of the fancy inside stuff is in any way more important than the person putting it all together. That may be totally obvious to many of you and in my head I have known that but recently, and seeing all of this laid out, it hits home. Why did I feel the need to re-configure some of these things? I am not sure. In some cases I feel that I am making things stronger and more stable but is there a cost to that? Have I thought it through completely enough to still have balance in my guitar? Sometimes it feels almost heretical to change things up that seem so set in stone. I know that those things are probably not as set in stone as we think, and certainly changing things up a bit is not any sort of heresy but I have been questioning a lot of these issues in my own work lately. It feels a bit strange to be in the middle of documenting all of this and having a period of doubt. Then again, it has been that way the whole time. Little steps and then little steps back. Little insights, little setbacks. If I am honest, I get this feeling at some point in every single guitar. I guess in some ways this document would not be complete without this. I wonder if some day it will be more of a smooth process where I know the steps and methodically and exactly follow them through. For now though, I still sweat each decision, even the ones I have made successfully before and should be totally comfortable with. I don't doubt that the guitar will sound good, I have had enough success lately that I am not worried about that. Rather, I worry about each little step and how much I can improve them. How many little parts could be optimized and then, conversely, how much I could be clouding the issue with overthinking. For something I claim to love sometimes it is a major PITA! Luckily it has worked itself out enough times that I don't sink into total doubt (I used to sometimes and it was a groaner). Sometimes a little break is good and sometimes putting it to the grindstone is too. This time is not so bad, writing it down makes me see it as just another step that gets repeated over and over and in that vein, what's the big deal? Get on with the guitar already!

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

The sides are glued up and sanded and my blocks are complete so it is about time to clamp those guys together. A little trick I have been using to locate the neck block is to glue 2 little strips onto each of the sides. I pre-cut my tenon in the neck block to 5/8" so each of the 2 little strips are 5/16 and get glued onto the sides like so:



These make sure that the neck block is perfectly centered as it gets glued onto the sides. I end up using a lot of clamps (thinking of it in money terms, almost $100 of clamps!) to get the bock tight to the sides. I struggled with this for a long time. Since the block is already radiused I would have clamps popping off and it would drive me crazy getting it all clamped and glued up. I don't sweat it too much now, I glued all of my cauls to a piece of fiber so once the center ones are clamped the outer ones can't go anywhere. That has made it a lot easier. Here is a shot of the neck block getting glued up. I make the top flush and the bottom (which has the taper already shaped in it) is inset.



The I do the same with the tail block. I leave it a bit proud on the top and bottom. I have been using the LMI white glue for these processes. I don't want to rush these enough to use hide glue and I have a funny feeling about the fish glue. I wish I didn't but I do. I could use Titebond but prefer the LMI white.



And then the rimset is starting to look like a real guitar!

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Now that the rims are glued up with the blocks I have to shape the tapers. Since I don't use the dishes to do this it is a little trickier. The back has a pronounced taper that starts just behind the waist and goes in a straight line to the neck block. My neck block is already shaped to have the correct angle so all i need to do is transfer that to the sides and cut them down. To do this I use the old half pencil:



Using the bottom of the neck block as my guide I trace the taper onto the inside of the rims.



Then I rough cut on the bandsaw, plane, and then sand to this line. When it is done I have a nice tight taper on my rimset and a flat plane for the upper bout.



The top is tapered as well but less than the back. This taper is not cut into the neck block. I used to do it that way but had trouble consistently gluing the neck block in to the correct offset on the rims. Now, I mark my offset and just shape the block and rims together to the correct taper. I use a flexible rule to rough mark the taper.



Then I rough away material on the belt sander with a low grit and finish up with some sandpaper on a flat board to be sure the plane is true. When it is done my upper bout looks like this:



Now that both upper bouts have their taper in i can begin to prep the linings and get them installed.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Once the tapers are in the rims I can install the linings. I am doing them exactly the same as I did when I published the double sides and linings tutorial so to save myself some time here I will refer back to that again which is here:

viewtopic.php?f=10117&t=21455

The only thing different for this guitar that was not detailed directly there is the extra piece for the bevel cutaway. When I am doing one of these I glue in a solid piece that butts right up to the neck block and extends to the closest vertical brace. You can see it glued in here:



The linings get glued in and any squeeze out gets cleaned up. I always leave them short at the ends (otherwise it is really hard to get them properly in place) so I have to add little shims against the tail block. I use offcuts from the linings themselves and make the little pieces so they fit in tightly.



Once these are in and the insides of the rims/linings are cleaned up I use a big sanding disc to clean up the top and bottom. I made a piece of thin MDF that locks into the miter slots and ends of the table tightly. Until I get my granite plate installed (:)) it works great. I use this to sand the linings (which were installed just proud of the rims) to flush with the rims.



I'll put it aside for now until the top and back are ready to join and the brace pockets cut. I will do a final clean up just before it is ready to join but for now it is going to sit like a little vulture on the shelf. It feels good, nice and solid but not too heavy.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Great work Burton.

Do you have any shots of your side bending?

Ed

Sorry Ed, I don't. I can't imagine it is too much different though than what you would expect. I shoot for fast and hot.

For the radiusing of the braces I have a bunch of forms that I made that correspond to roughly 1 through 5 sixteenths of offset over a 16" span. I figured it all out once using a circle radius program (http://www.1728.com/circsect.htm) and they were:

1=42
1.5=29
2=21
2.5=17
3=14
4=10.5

I made the 5 a long time ago but have not used it since, I didn't even write down what it corresponded to. My forms look like this:



I made them by flexing a piece of wood until it touched the offset in the middle and was held tight at the outer edges. I knew it wasn't exact but it has worked well. It forms a nice circle section in the center and flattens out a bit at the edges. I have since obtained some nice CNC'd forms from Tracy and plan at some point to make new templates from these although I have not done it yet. They are like this:



To radius the braces I also made the other matching piece to my homeade radius forms. I double stick tape the braces down to these forms and shape them on the router table. A little smoothing after and they are ready to glue. Those forms look like this:



I made all of the different forms because I end up using most of them at one point or another on each guitar. For the top I radius my x braces and the main cross brace to the 2.5, or 17 (most of the time at least, depending on what I am looking for I may use a different radius scheme). All of the other top braces (usually) are glued flat. For the back I leave the upper brace flat (for the flat upper bout), the second (waist) brace is shaped to 2 (or 21), the third to 4 (or 10.5) and the fourth brace to 3 (or 14). Since I am putting a pronounced bend in the back plate as it moves from the flat upper bout to the heavily radiused lower bout (and the flattened plane of the rims) it needs the compound radius to not feel strange. It took me a while to get at the proper curves for each brace but now that I have them the back looks pretty normal on the guitar. For the top, the upper bout is flat for the fingerboard extension and the lower bout I glue flat so when it pulls up slightly it pulls up into its normal looking radius. So far this has worked well also. This may have been a section I over-thought at one time but now that it is normal for me it seems, well, normal. I have not found a lack of responsiveness with the heavily radiused (compared to some at least) top plate and I have not had a problem tuning the back to where I want it as long as I take the bend into consideration for the back plate.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Hi Burton,

Thanks for sharing your build. Nice work.

Cheers - Luc

Burton,

I am a wanna be luthier...having not completed my first guitar. All I can say about your posts is WOW! I am very intrigued with your methods, particularly the rounded neck block and your jig to sand the shape. I know you haven't got there yet but how do you round the neck to fit the guitar? If this will be shown later, I am OK with waiting for that step in the process.

Tom

Hi Tom,

Yes, I will get to that eventually. I am way behind in keeping up here, I am almost to that point in the actual work. I hope to catch up soon.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.