This weekend I spent some more time on the Dreadnought Guitar project. In the last post, I had assembled the sides, created the inside mold for the guitar body and installed the kerfed linings. This weekend it was time to profile the kerfed linings to match the eventual contour of the guitar top and back.
As I mentioned previously, both the top and back of the guitar incorporate a radius. These components are radiused from neck to tail and side to side creating somewhat of a domed shape to each. These domes on the top and back plates of the guitar serve the purpose of making th components stronger and help them to resist changes in humidity more than corresponding flat components would.
The top has a more gentle radius angle of 1.5 degrees whereas the back is more severe at about 5 degrees. As a result, in order to adequately attach the top and back plates to the sides, the appropriate angles must be made to the edges of the guitar sides to create a good gluing surface. You’ll recall from my last post that I installed the kerfed linings 1/32″ proud of the edges of the sides so that they could be contoured appropriately in order to effect this assembly step.
Building the Sanding Stick
While there are many potential ways of profiling or contouring the kerfed linings and edges of the guitar sides (i.e. creating a set of radiused dishes with abrasive paper adhered to them to sand the edges), the method I used was simpler. I created a sanding stick to be used to sand the edges at the two different angles for the top and back. This jig was fairly straightforward to build – the only tedious part was getting the angles correct.
I started with a piece of straight stock about 24 ” long by 3 1/2″ wide and some scraps to create wedges of the appropriate angles. The two wedged sections are on opposite sides of the stock and are about 8″ long. One wedge is at 1.5 degrees and the other is at 5 degrees when measured to a point about 4″ from the other end of the stock. 60 Grit sandpaper is adhered to both faces of the stock at end opposite the wedges so that the midpoint of the paper is located at the 4″ mark that I designed the angles to intersect.
Using the Sanding Stick
To use the sanding stick, I mounted the guitar body into my bench vise using the waist clamp as a mounting mechanism. I took the time to mark the edge of the sides with a yellow grease pencil so that while sanding the kerfed linings, I would be able to tell when I had reached edges of the sides.
With the guitar back facing upward I placed the 1.5 degree wedge of the sanding stick at the tail block as a pivot point and sanded the neck area, moving the abrasive end of the stick across the upper bout of the guitar. I continued this process, moving the pivot point along the edge of the sides until I made my way around the perimeter of the guitar. Once I had sanded away the grease pencil lines I made sure that both the neck and tail blocks were sanded across their width to the same angle as the kerfed linings.
Once the sanding of the top edges was completed, I reversed the guitar body in the waist clamp and repeated the process on the back. This time with the 5 degree wedge starting at the tail block to match the more severe angle of the back. In the picture, you can see a closeup of the neck block area with the kerfed linings and edges of the sides sanded to the correct angles.
Installing the Rosette
Because the top and back will be radiused, the braces that will be eventually applied to them for support are also radiused. In the case of the top, there is a decorative rosette to apply around the sound hole. It was necessary to install the rosette now while the top was still flat.
The sound hole and rosette channels were pre-cut into the top. The materials for the rosette consist of three pre-bent, ringed segments. A herringbone center pattern and two narrow segments (black/white/black rings) for the inner and outer edges. As you can see in the photo, the rosette rings were cut and needed to be fit to the routed channels in the guitar top.
Because of the way the fingerboard will overlap the top, only the inner ring of the rosette will show so, the joint on it needs to look good. The other two rings will not be seen, but they afforded me the opportunity to practice the technique for fitting and installing them. I started with the outer ring by putting glue into the channel and spreading it with a toothpick. I then lightly pushed the rosette material into the channel following it along around with the rounded back of my chisel handle. There was not an opportunity to practice fitting and cutting on this ring because there was not enough material to make it around the entire sound hole.
However, on the center ring there was enough material to make it around. So, I began by stabbing an angled cut on one end of the material with my chisel. I then dry fit this along the channel and marked with a pencil at each edge of the cut on the trailing end of rosette material. I removed the material, cut the opposite end with a chisel and then installed the same way as the first ring. The result was close, but not perfect. On the last ring (which will show) I carefully followed the same procedure being more careful to track to the channel in the dry fit before cutting. The result was a much fit after gluing. The last step in the rosette installation was to apply some clamping pressure using two MDF cauls and some wax paper in between to keep things from sticking together.
Scraping the Rosette
After about 3 hours of drying time, I removed the clamps and cauls from the rosette. There was some glue squeeze-out as expected so, some judicious scraping of the rosette was in order. I sharpend the card scraper, put on a fresh burr and went to work.
The scraping was going along well until I heard the crack! I looked down and saw that the top had split along the previously joined glue line just above the sound-hole. Not good! I did not take a picture of this but, it was obvious that the book-matched top had split along the center glue line because I could see dried glue along the split edge – luckily there was no real wood tearout.
I was not sure if I could fix this or not – the top was already profiled in the shape of the guitar and the lower bout of the top was still glued together so, gluing and clamping this was very problematic. I first scraped the edge joint area as best I could with my card scraper to get down to bare wood. However, I did not want to remove any material that would leave a gap, just dried glue. Eventually, I settled on a method to clamp the pieces together by adhering pieces of scrap wood to the top of the guitar adjacent to each side of the glue-line with double stick tape. I then carefully clamped these together while also clamping the top to my workbench to keep it perfectly flat. I let the glued joint dry over night. When I took it out of the clamps and completed the scraping of the rosette area and glue-line, I was pleased to see that the repair had worked! In the picture, you can see the completed and scraped rosette area.
Because of the way this split looked, I think that it was the result of a poor original glue joint and not any mistreatment from me. However, in the process of repairing it I learned some things that will be useful to me in the future. We have to remember that wood is not a perfect material and sometimes making repairs or fixing mistakes that we make is part of the woodworking process. This is especially true when we venture into new areas but, working through these pitfalls is also one way that we grow as woodworkers.
The first step in building the Dreadnought Guitar was to start assembling the body. This involved the two sides of the guitar, the tail block, the neck block and the kerfed linings that run around both the top and bottom edges of the sides and serve as a gluing surface later.
Assembling the sides
The two Rosewood guitar sides are connected with to two Mahogany blocks. One block for the tail of the guitar and one for the neck location. The neck block has a mortise for a bolt on neck (some guitars may use a single large dovetail joint here). The tail block is just a raduised solid block to match the curvature of the sides. These two blocks must be glued to the the sides precisely at the center of the body and as square as possible to the edges of the sides. This is especially true of the neck block because this setup determines the eventual set of the neck on the guitar. Because the neck block must eventually be glued to the guitar back and top, it must match also the angles formed by back and top. So, it is machined with a 5 degree bevel to match the back and a 1.5 degree bevel to match the top.
The two guitar sides are machined flat on the top edge and have a radius on the back edge. To start the assembly, the centers of both the neck and tail blocks were marked with a pencil. The pattern of the top of neck block was outlined with a knife through just the top layer on a piece of cardboard. This relief was created so that the 1.5 degree bevel of the neck block could extend below the cardboard surface allowing the block to sit square to the workboard. The sides were placed flat side (top) down on the cardboard and the blocks were aligned to match up with the seams created by the sides. It took a few dry fits to align the seams as well as the top and bottom edges of the blocks with the sides. After these dry fits it was time to do the glue up. This started with weighting down the sides so that they would remain stable during the glue up.
The tail block was glued first (with the neck block just dry clamped to the sides). As you can see this took quite a few clamps in a small space – two of these needed extra deep capacity as well (it looks like some creative clamping solutions may be necessary before this project is completed). Gluing the neck block followed once the tail block assembly was dry. Again, a lot of clamps and extra attention to detail were required to get the neck block set as precisely as possible so that the eventual neck set on the guitar will go smoothly. We’ll have to wait until later in the project to see how well I did in that department! In the picture you can see the completed glue up of the neck block, the tail block and the sides. Also in the background are some parts of the cardboard form which I’ll talk about next.
Building the cardboard form and waist clamp
Often, guitars are built using a specialized workboard or custom wooden form. The kit that I started with has a very economical way of creating a workboard and form. This was done with a flat workboard, two cardboard cutouts that mirror the shape of the guitar body, a couple of wood blocks and a shop made waist clamp.
The two cardboard pieces were cut to match the shape of the guitar body. The first piece was inserted into the body over two scrap pieces of 3/4″ material. This was done to elevate the cardboard so that there would be room to attach the kerfed linings to the edge of the sides later. I made a couple of wooden blocks to stand off the second piece of cardboard from the first. One block in the upper bout of the guitar (the smaller section of the body) was 2″ high and the second block in the lower bout was 2 1/2″ high. This offset is because the body actually has a taper – the body is deeper at the bottom than at the top. The blocks were glued to the lower piece of cardboard and then the top piece of cardboard was inserted and glued to the blocks as well.
The second part of the form is a shop made clamp that slides over the waist area of the guitar body to keep things stable. I built this clamp from some scrap MDF. I simply cut the shape out on the band saw and then radiused the inner edges so that it would slide snugly over the waist without cracking the sides. With the inner cardboard form and the waist clamp installed, the assembly was fairly rigid. At least, rigid enough for the next step.
Attaching the kerfed linings
The top and bottom edges of the sides require some material to be applied to both support the guitar body as well as to add a gluing surface for both the top and back of the guitar later. Because of the curves in the sides of the guitar, these linings are kerfed with a narrow saw cut at even intervals so that they can bend with the shape of the guitar. The linings were not long enough to make it around an entire side of the guitar so they were installed in two pieces.
The kerfed linings needed to be attached about 1/32″ of an inch above the guitar sides. This was done so that they can be sanded flush to the top and bottom at the appropriate angles later. It does not require much pressure to clamp these linings to the sides. So, as you can see they were clamped using ordinary clothes pins – though, there were a lot of them! This was done by simply applying a sparing amount of glue to the backside of the lining and then applying the clamps one by one, carefully aligning the edges (1/32″ proud) and making sure that there were no gaps between the lining and the guitar sides. As I mentioned, two kerfed linings were required for each side (top and bottom) for a total of eight linings. In the photo you can see the the glued up body with the completed linings.
Just a note about glue. As you can see in the picture, I am using original Titebond I for this project. Traditionally a luthier would use Hot Hide glue for an instrument. This is done both for its strength as well as its reversibility (i.e. the ability to take a joint apart later if necessary). It is also important to use a glue that creates a very rigid glue line and does not exhibit any creep while making an instrument. Hot Hide glue has all of these properties. Liquid Hide glue is kept in a liquid state by adding Urea and as a result looses some of its strength making it useless for an instrument project. I do not typically work with Hot Hide Glue so I chose the next best choice, Titebond I. Titebond I is the best choice of the aliphatic resin glues for instrument making. It is not nearly as reversable as Hot Hide glue but, it can be reversed. Where Titebond I is better than other versions of Titebond is in the glue joint it produces – it is very rigid and does not exhibit creep. This will ultimately result in a stronger and better sounding instrument.
I get a lot of questions via e-mail on things I’ve posted about here. I usually just reply to the e-mail directly but, I thought that it might be instructional to post some of these questions and my responses because there may be others out there with similar questions who could benefit from the dialog.
I recently received a question from Jeff in North Carolina regarding the Mortising & Loose Tenon jig that I previously posted about. Actually, I’ve received a number of questions regarding this post so, I thought that this might benefit a number of you out there. Jeff and I actually had a couple of e-mail exchanges on this. Below is a transcription:
I’m pretty new to woodworking and enjoyed your article on loose tenon joinery. I’m going to give it a try and will soon purchase the DeWalt 518PK Router Combo kit.
Do you have dimensional plans for the router mortise jig? A rough idea will do since I can adapt it to scale. Material construction
appears to be maple and MDF. Correct? I’d appreciate any other tips that you can offer for constructing and using the jig (i.e., do’s and dont’s).I don’t have a router table so I envision some problems cutting and shaping the tenon stock (1/4″-3/8″ thick). I may try my hand using a low angle block plane on a test piece. Any other suggestions?
Many thanks for your help.
-Jeff
Here was my reply:
The jig I made was adapted from one found in Bill Hylton’s book Router Magic. It is fairly simple to build. Mine was made from scrap MDF and Poplar, but really any stable wood or sheet good would suffice. My jig is about 14″ wide x 8″ high x 5″ deep. It originally built mine from available scraps but, if I built it today I would make it higher than 8″. This is because after clamping the stock into the jig, sometimes there is not a lot of material left available to clamp into my bench vise. Other than that, the only thing I would double check is the depth based on whatever edge guide you are using. Depending on how much adjustment you have on the edge guide, you may need to adjust that dimension to give yourself enough extension to cut mortises.
When you use the jig, you want to be moving the router so that the rotation of the bit tends to draw the edge guide fence against the jig. Do this by standing in front of jig and moving the router from right to left. Also, when mortising, I usually take small bites, no more than 1/8″ to 1/4″ deep at a time, using a spiral upcut bit. It helps if you have a vacuum attachment for the router to clear the chips but if not, just stop after the mortise (or when it gets too filled with chips) and vacuum them out.
As far as the tenon stock, it is nice if you have rounded the corners of the tenon stock to match the mortises. However, there are other ways of dealing with it if you do not have a router table. One way would be to use square tenon stock of the correct thickness and just square the ends of your mortises with a chisel to match. Another option would be to just chamfer the corners of your tenon stock with a block plane until it fits. Most of the strength of the joint comes from the long grain glue surface provided on the sides of the tenon stock so, if it is a bit short in height because of the chamfers, that’s OK. Just be sure that you align things a bit more carefully when assembling the joint because you will not have the benefit of the tenon height to help there.
Jeff made some progress on the jig and wrote back with a couple of other questions:
Thanks for the advice. I’ve just about finished making the jig. I ended up using particleboard because my local Lowe’s doesn’t stock MDF. The 5″ depth works out fine for my DeWalt 318 plunge router and its accompanying edge guide. There is plenty of back to front adjustment to accommodate thicker stock if necessary; I’ll probably be in the 3/4″-2″ range most of the time.
I’ve just finished constructing my own version of the “New-Fangled Workbench” and I’m playing with the configuration to secure the jig in its clamping system to allow cutting a mortise in the vertical position (stile?). The vertically adjustable planing bench is a nice feature for supporting the bottom end of the workpiece. However, I’ve gotten pretty good at simply cutting a classic tenon on the table saw and using a shoulder plane for fine tuning so I may go that route.
You noted moving the router from right to left in order to draw the edge guide fence against the jig. This seems counter to router practice where the clockwise rotating bit is moved from left to right. Is your suggestion based on a peculiar characteristic of the spiral up-cutting bit?
Thanks for your help.
-Jeff
Here was my reply:
Glad to hear that you are progressing on the jig. I think the particle board will be fine as long as it is flat. You may want to coat it with a light coat of finish to keep it from picking up moisture and swelling over time.
With respect to the router feed direction, I am going by memory here without trying a piece in the jig, but I think that what I stated is correct. It is different from a typical edge router operation (like you described), because of the edge guide. You should try this for yourself, but I think you will find that if you feed right to left that will pull the edge guide in toward the jig. When you go left to right, the edge guide will pull away from the jig and you will have a typical climb-cut scenario (because there is no bearing on the router bit stopping this). One other thing that I sometimes do when using the jig is to plunge both ends of the mortise straight down to the final depth first and then route between them R to L until I’ve completed the mortise.. For deeper mortises, this ensures that the ends of the mortise remain vertical.
Doing tenons on the table saw is also something that I do after making mortises with this jig (look at the Queen Anne Side Table on the blog for an example). This is fine. You get the benefits of a clean, precise mortise with the router jig and then a tenon that you can fine tune on the table saw and/or with the shoulder plane.
Hopefully, this kind of question and answer is beneficial to others out there. Please let me know what you think. You can either leave a comment using the comments link at the end of the post or send me an email at thecraftsmanspath@gmail.com.
Tags: loose tenon, mortise, router mortising jig
Now that I am in the final stages of the Queen Anne Side Table project (more to come on that in a future post), it’s time to start thinking of what should be the next project to tackle in the shop. I have several furniture projects that I am considering but, I won’t start those until the current project is finished and out of the shop.
I have been thinking about something to bridge the gap between major furniture projects and also something that might be able to continue in the background during downtime. In the past, I would often do some turning during these periods because most turning projects can be accomplished fairly quickly. However, I’m always looking for a new challenge so, I have been considering some other things.
Being a guitar player from way back and of course, also a woodworker, I have always wanted to make a guitar. I have done a fair amount of research on lutherie but, I was always reluctant to take on a project. In the past I’ve evaluated both building from scratch and building from a kit of parts. A couple of years ago I went as far as purchasing a kit of parts for a dreadnought-style acoustic guitar from Stewart MacDonald. For whatever reason, I never started the project after receiving the parts. Though, I always had this nagging feeling that I should! As you may have already guessed, I’m going to use this opportunity to finally build a guitar. Starting now, between projects and continuing on in the background of my next furniture project.
Lutherie is a unique brand of woodworking. It involves some different skills, techniques and strict attention to detail. Though I will be starting with a kit of parts, don’t let that fool you. There is a lot of woodworking involved in producing a quality instrument even without milling the parts. Essentially, the basic head start that the kit gives you is as follows:
- All of the parts are preselected and available rather than having to acquire them separately
- The guitar sides are pre-shaped so that you are not required to bend them on a hot pipe
- The stock for the guitar top, back and sides is pre-thicknessed to a uniform dimension
- The guitar neck is rough shaped and the fingerboard is pre-slotted for the frets
So, you get the basic pieces but you still need to perform all of the joinery to create the guitar. The body needs to be built and all of the braces and blocking need to be joined and hand shaped. All of the binding channels must be routed and the bindings must be applied. The neck must be shaped and set with the truss rod, the fingerboard must be applied and fretted and the bridge must be set and the string action adjusted…and the list goes on.
I decided to go the kit route with the hopes of getting my feet wet, honing my skills and then transitioning to a scratch-built guitar in the future. The guitar I am building is a dreadnought styled after the highly desired vintage C.F. Martin guitars. The picture shows a representative view of what it will look like when completed. I chose the kit with the Rosewood back and sides, Mahogany neck and spruce top (shown on the left in the picture). I’m hopeful that I will be able to complete this project without the need to purchase a lot of specialized tools. We’ll see how I do in that department…
I’m looking forward to traveling this new path, acquiring some new skills and hopefully building a top-quality instrument. I’ll be posting about my efforts here along with whatever else is going on in the shop. So, please follow along and let me know what you think. Who knows, when this new guitar is complete I may even start playing regularly again!
With the bulk of the construction on the table complete, it was time to move on to the detailing tasks and finish work.
I spent a considerable amount of time scraping and sanding all of the parts of the table. There were areas on the legs where the knees transition into the posts that needed some cleanup. I carefully worked these first with a card scraper and then followed up with hand sanding. Of course, the rest of the table also required a bit of sanding as well. Because all of the parts had previously been smoothed with a plane and/or and scraped, my sanding schedule for the table started with 150 grit and finished with 180 grit. I tried going to 220 grit but, it seemed to be burnishing the Cherry so, I do not think I will continue to that fine a grit. When all of the sanding is complete, I will give the table a thorough vacuuming and then a final rub down with Naptha to remove all traces of sanding dust and to determine if any areas need further attention.
In the picture (click for larger view) you can see the Queen Anne Side Table in its current state. All that remains for the construction is to put a decorative edge on the top, attach the top to the base and add the hardware. Speaking of hardware, I recently ordered a set of drawer pulls and a matching escutcheon from Horton Brasses. The style is reminiscent of the hardware typically on antique Queen Anne furniture and should work nicely on this piece.
I have also started to consider how I will finish the table. Most Queen Anne furniture that you see is stained dark. Generally for Cherry, I like to use a natural finish allowing the wood to oxidize over time with exposure to sunlight, darkening naturally to a rich patina. However, for this table I am considering adding just a slight bit of coloring with Garnet shellac to even out the tone all of the components.
A quick primer on shellac:
Shellac is a universal binder and a very good finish in its own right. The substance is made from refining and drying the secretions of the Lac bug which is found in India and Asia. Essentially, any finish will adhere to shellac and shellac will adhere to almost anything. Additionally, shellac serves as a perfect sealer between different stages of the finishing process. For example if a dye or stain is used and then sealed with shellac, none of the dye or stain will be removed when the topcoat is applied. One important thing to realize is that only dewaxed shellac will support all topcoats. Most pre-mixed shellacs (aside from Zinsser Seal-Coat) contain wax and will cause problems with adhesion when another topcoat is applied. Shellac that is purchased as dry flakes is dewaxed and must be dissolved in denatured alcohol to produce the finish. Because alcohol is used as the solvent, shellac has the benefit of drying in minutes. Also, shellac has the properties of dissolving into previous coats effectively creating one single thicker coat when multiple coats are applied.
Shellac solutions can be mixed to different strengths or concentrations. This is referred to as the “cut” of the shellac – the premixed stuff is usually about a 3 lb. cut which would equate to 3lbs. of shellac flakes by weight mixed in 1 gallon of alcohol. I usually mix to around a 1 or 1 1/2 lb. cut. This produces a good workable consistency and fairly light coats. I only mix what I will use within about 6 months because shellac in the liquid form has a limited shelf life after which it will have a tendency to not dry or harden. For me, this means a 1 lb. cut is 2 oz. of shellac flakes (by weight) mixed with 16 oz. of alcohol (by volume).
There is nothing real scientific about mixing shellac – it will work no matter what strength you mix it to. It’s easier if the flakes are ground to a fine consistency before dissolving in the alcohol. Using an old coffee grinder would work well, or you can do what I do – put the flakes in a plastic bag, seal it and then hit it with a rubber mallet until you have very fine pieces left. Put the alcohol in a glass jar, add the shellac and seal the jar. Swirl the mixture around every half hour or so for about the first four hours and then let it sit over night. The next day it should be ready to use. The last step before use is to strain the shellac mixture through paint strainer or coffee filter to remove any impurities left after the refining process.
I took some time to create some finish sample boards from scrap Cherry following the same sanding schedule as the table. On the left half of the top board I used Watco Natural Danish Oil followed by two coats of Garnet shellac and then a top coat of General Finishes Arm-R-Seal. The other half of that board did not receive the Watco. On the lower board I used Boiled Linseed Oil followed by the same sequence of Shellac and Arm-R-Seal with the right half of the board not receiving the BLO. Sorry for the picture – it does not really show the variations too well. I am considering the using BLO/shellac sequence because it is not showing the blotching that the Watco seems to. Some of the Cherry on this table is a bit curly and may be prone to blotching – I do not want to obscure any of the curl in the boards.
It’s very easy at this stage of a project to speed through the final finish preparation steps and finish process in an attempt to get the thing out of the shop. I always find myself battling this. In the long run it’s always best to spend the necessary time to prepare the surfaces as well as possible and to test all of the finish steps to assure a desired end result. Of course, this takes more time and delays moving on to the next project but, in the end it yields the best results.
Tags: Arm-R-Seal, cherry, danish oil, Finishing, garnet shellac
