Definitely Replacing the Soundboard

Saturday, April 18, 2009.

This morning I got up, took a look at my soundboard, and sadly I now find that a crack has developed in the glue joint between the two halves of the soundboard. Here are some pictures.


I guess this cements the decision I already made to remake my soundboard. I imagine I will attempt to fix this, but I am going to see what the experts say at the Cafe first.

My wife and I had another yard sale today so I wasn’t able to work on the mando through the day, but once we were done with that I was able to spend a few quality moments with my now-possibly-defunct soundboard. One of the luthiers at the Cafe suggested that, assuming I am able to salvage the soundboard, I should remove the sides completely and re-glue. So, as a first attempt, I gave a relatively gentle tug between the two and - pop -they came further apart. Another gentle tug and - pop-pop - almost completely apart with only the head-block remaining attached to the soundboard. Add a bit of water, pull again, and - pop - completely free. Now, either my initial glue job was a lot worse than I thought, or my water soaking method had a much greater effect than I previously thought. As I think about it, I am guessing probably both.

In any case, there are three distinct lessons I have learned from this; one - take extra care to make sure that you are gluing pieces where they actually belong (the first time), two - if you can’t make yourself follow lesson one, DO NOT SOAK THE SOUNDBOARD in an attempt to salvage pieces that simply be replaced (and they can all be pretty well replaced), and three - take extra care to make your joints correct BEFORE gluing - otherwise you get to do it over.

Finally, after removing the sides, I thought I would hold the soundboard up to a bare light bulb to see just how good of a job I did carving the thing. Below is a picture of what I saw.

As you can see, I did not do a very good job (yet another reason I am becoming happier and happier to remake the thing). From what I have read, there ought be a nice, distinct light ring about an inch or so in from the edge where the wood is thinner than everywhere else. This ring should be of roughly uniform width and go pretty much all the way around. The rest of the surface ought to be pretty uniform getting thicker in the center. On mine, it is clearly quite a bit thicker at the base and inconsistent everywhere else. Obviously checking this against a light is not a step I took before. I certainly will on the next one, though.

Some Backwards Steps

Friday, April 17, 2009.

Earlier this week, I posted some pictures of my work out on the Mandolin Café website and luckily (or unluckily I suppose depending on how you look at it) one of the REAL luthiers who frequents the site pointed out that one of my tone bars was glued in the wrong location. Sure enough, when I went back and checked it against the plans, I had indeed missed the mark. A little embarrassing to not catch something so big myself, but I guess that’s what learning is all about.

So yesterday, I decided to try to “unglue” the bar by soaking it with water hoping to soften the glue enough to get it off. After letting it soak overnight, not only was I NOT able to get the bar off, but I DID manage to unglue the soundboard from the side wood. Oh yeah, I also managed to warp the soundboard. And I suspect I might have ruined the glue joint between the kerf lining and the side wood, too. Way to go, Steve!

Of coarse one of the other REAL luthiers suggested that I simply carve away the mis-located bar and replace it with a new one. I thought about doing it this way (and probably should have taken his advice) but not having extra wood from which to make a new bar, I opted to try my removal process instead. Sigh.



So, all this damage done, I have decided to start over on the soundboard-side wood bending-kerf lining. I can’t say that this is a really bad thing, though, since I must confess that I was a little disappointed in the overall results I had achieved on this first assembly – my side wood bending was a little “off” from the shape of the mold, my mahogany blocks were not quite the right shape to match the thickness of the side wood, and I think I can do a much better job of shaping and finishing the soundboard before installing the tone bars. In all, I think I will probably be a lot happier with my finished instrument this way. So, I placed another order today with my friends at Steward-MacDonald for new top and side wood and, once that arrives, I can begin again.

This evening, I got home from work and took another last look at the tone bar, having decided to go ahead and continue to let it soak all day - after all what more damage could I do? When I took the soak rag off, much to my surprise, the tone bar popped right off. No additional damage to the soundboard.

I’m still gonna go ahead and build the second soundboard though. After considering my ongoing disappointment with the first one, I think I will be a lot happier if I do.


In the mean time, I still have plenty to do with finishing the back and getting the neck ready to attach. Here is how the scroll on the back is coming.


In the long run, my hope is to “salvage” the old soundboard-side assembly enough to use for a “mandolin-build-in-process” display some day. I have often thought that this kind of display would be kind of cool to have set up at a festival or such. Maybe this can be my first piece for such a display.

Cutting F-Holes, Tone Bars, and Gluing on Soundboard

Sunday, April 5, 2009.

While it was a short week and I wasn’t able to spend as much time on my project as I would have liked, I still accomplished some of the more impressive steps to date and am very pleased with where I am.

Having been out of town on Saturday, Sunday and Monday of last week, I got back to work on Tuesday evening by first removing all my clamps from where I left them last Friday and then cutting out the F-holes in the soundboard.


The photos above show the results of my gluing on Friday.

As I started to cut the F-holes I was a little nervous about this step since the soundboard is so thin (as little as .01”) and I was planning to use a powered scroll saw to do the work. My fear, naturally, was that the scroll saw would cause cracks or some kind of chipping. My other option was to try to cut these manually using a hand-held scroll saw with one hand while trying to steady the soundboard with the other. While the manual method would likely help prevent chipping and cracking, I opted for better material handling control and went with the powered method. I figured the gauze I glued in earlier was my insurance against damage. As Mr. Siminoff suggested, I made sure to cut slowly.

I started by drilling four holes, one in the center of each scroll “balls” (best description I can think of on short notice) using a small 1/6” bit as a pilot followed with a larger 1/4” bit which is big enough to allow the scroll saw blade to be inserted. Then with these four holes drilled, I moved over to the scroll saw and began cutting.

In this picture, you can see where I have made my first cut from one hole to the other and am getting ready to cut around the ball.

As I did this next step, I learned something - cut out the ball first and then cut the long stretches. Here is why; as I made the cut around the ball, the “stretch” cut (my term) completed earlier, I found that the wood was vibrating badly and realized that the inside point where the stretch cut meets the ball was having to handle all that vibration with no support. Had I cut the ball first and then the stretch, starting the stretch cut at the point, the stress from the vibration would have been shared by a lot more wood rather than just the point. I hope that makes sense.

In any case, the points on this first F-hole made it through my abuse and I went successfully on with the second one much happier with my new-found insight. Here are some photos of the finished results after some sanding.



That pretty much did it for Tuesday night.

Wednesday I started with the tone bars. Back when I started this project, I made it a point to save pretty much all my scrap wood so, when it came time to make these, I was able to cut them out of the scrap pieces I saved from the blanks used to make the soundboard itself. Once I had cut out my tone bar blanks as per the dimensions in the book, I needed a way to transfer the shape of the inside curve of the soundboard to the bars so I could cut them to fit. To do this I went about as simple as it gets. I took a small piece of scrap side wood and taped on some lead from my mechanical pencil. Holding this in my right hand while holding the tone bar blanks in position with my left, I gently traced the curvature onto the blanks.

The razor blade (that’s saw dust on it) is there for reference.

Once I had the blanks marked I cut them roughly to shape.

Then I went through the time-consuming process of sanding and shaving them to actually match.

Then it was time to glue them in place. For this, I used three clamps one each tone bar with leather cauls on the back side of each clamp and a board on the from to help distribute the pleasure more evenly. As with most gluing procedures, I finished of the night with this step and left them to dry for 24 hours.


These photos only show the first of the two bars being glued. I glued the second one on right after taking the pictures.

Thursday I removed the clamps and did an initial crowing and cleaned the edges of the joints between the bars and the soundboard.


That is a pencil mark on the front edge I used as a reference while fitting the tone bar, not a crack. It actually fooled me once while I was working on it, too.



Come Friday, it was time to glue the soundboard to the sides. This process was one I had been looking forward to with both excitement and a little bit of trepidation. I figure this is where it starts coming together or you find out you have to take some big backward steps. While I frequently compared my soundboard to the side assembly at each stage of the process, I couldn’t help but worry. Fortunately, nothing seemed too out of whack to prevent me from gluing up, so I went for it.

For this, I used my 40 cello clamps, the side mold for reference, and hide glue. After preparing the clamps and getting the hide glue to temperature, I spread the glue and worked as fast as I could to clamp the soundboard in place. Here’s how it looked with all the clamps in place.






I was concerned with the time that it took to get all the clamps on since I have read that the hide glue dries quickly, but I figured that even though it took longer to complete than I wanted, I would wait and see how it looked once dried and, if necessary, take it apart and try again if it didn’t hold. With that, I was done for Friday.

Saturday was completely a wash as far as mandolin building goes since my wife and I decided to go to an auction that ended up lasting all day. We got some good stuff, but boy, what a day. Once we got home and got unloaded, it was about 9:00 pm, so about all I had time for was to unclamp and inspect my glue joints. It appears that contrary to my worries, my clamping speed was sufficient and the joints are tight. Here is how it came out.






I’m going to have even less time available next week so I am not sure if I will get enough done to really be worth posting, but I probably will anyway. Until then.

Back and Sides

Sunday, March 22, 2009.

This week was less productive than others, but as I get further and further into this, I guess that is to be expected occasionally. What I did get done was to bend all of my side wood, glue most of it together, and to get the maple for the back ready for shaping. Now that I look at it, maybe the week was more productive than I thought.

On Monday, I decided to dress the sides of my maple back wood and glue them together. I clamped each of the two halves into my truing board and squared them with my hand planer. Before I could glue, though, I had to be sure that the two edges were true and square. I did this by holding the two boards together up to the light and looking for a gap between them. The first time I did it, there was enough of a gap that I took one of the boards back to the truing board for rework. The second time, while there was still a really tiny gap, I decided to do a trial clamp-up, and the gap disappeared. Time to glue.

Here is the back glued and clamped together. Once again, I used hide glue for this and did my best to make the bottom side (opposite the clamps) as flat as possible for the sake for when I plane it later.

Tuesday, I began the process of attempting to bend my side wood. I had a lot of trouble with every bend I tried and would end up breaking the wood. After destroying one full 2-ft length, I gave up for the night and did a little research to try to find out what I was doing wrong. I went back “the source” (www.mandolincafe.com) and found an old post where several professional luthiers shared their techniques. What I found was that there were two things they do differently than I was doing. First, most of them work their unbent wood down to a thickness of 0.70” to 0.8” before bending. This, they explain, makes the bending process, especially when using the hot-pipe bending method as I am, much easier but still allows the finished sides to be strong. Second, most of them use a backing material held against the wood while bending. This backing material is typically a thin strip of metal.

So on Wednesday night, I was ready to try it again, only this time I decided to use my drum-sanding rig to thin my strips down first. This process, it turned out, took quite a while, so after I was done sanding down three strips of wood my night was pretty much used up and I was only able to bend one piece of wood, relatively successfully, before shutting down for the night.

Thursday, I was back at it again, bending and breaking my wood (dang it). The piece I was trying to bend this time was a small section that connects the head block to the first point. This piece, I had read, is thought to be one of the more difficult ones to bend and now I understand why. It is quite a short piece with opposing sharp curves at each end and, as I discovered way back as when I was attempting my first practice bends, bending the wood right at the edges is much more difficult than just about anywhere else along the strip. Consequently, I broke several of these pieces before finally getting one close. At the end of the evening, while attempting to finish bending the second to last piece I needed, I was getting pretty frustrated with my inability to get the sharper bends without cracking the edges. That was when it dawned on me that I needed to try using the backing material the professionals had recommended.

Now before you begin to question my intelligence too much, I had tried this before, but with no success. But that effort and experience was on wood that was too thick, too wide, and too early in my learning curve. Once I realized this was poor experience to base decisions on, it was time to try again.

As soon as I broke out the backing strip it was the difference between night and day. All of a sudden, no more edge cracking and quicker, sharper bends were mine for the asking. It was unreal. The piece I had been laboring over was quickly finished and, where I just knew it was going to crack, smooth. This is one lesson I will not quickly forget.

The last thing I did before closing down for the night was to glue the head block onto the scroll wood.


In these photos, it appears that I might also be gluing the upper point on, too. This is not the case. This is where I had clamped one of the side pieces onto the mold to allow it to dry to shape after bending.

Friday was movie and dinner night with my lovely wife, so no mandolin building.

Saturday I bought a surface planer. My back wood was about 1-1/8” thick at its thickest spot and about 7/8” at the thinnest. My goal, as per Mr. Siminoff’s book, is 5/8” (and flat). Not only did my wood vary in thickness, it was also quite cupped and warped. Fortunately my gluing efforts allowed it to be relatively flat on one side so putting it through the planer was no problem.


Its really hard see much from these pictures, but these are both sides before planing.

And here I am getting ready to make my first pass. I did nothing extra to attempt to hold the part flat or square.

I am really happy with the results of my planing. The board came out really nice.



Here it is as it came out of the planer and then as I marked it with my template. (It's really hard to see, but there is a penciled outline there!)

Finally, I made new mahogany point blocks and was able to glue most of the rest of my side pieces together. I chose to make new blocks when I found that the shapes that Mr. Siminoff supplies did not match well with what I needed to match the side wood and the mold. While I am comfortable that this is purely a result of inaccuracy on my part when I made my mold, making blocks that match it are now necessary. To do this, I traced the curves of my mold onto paper, offset the lines by the thickness of the side wood and then used those to make new templates.


In this picture (above), I clamped the wild end of the scroll piece back to mold simply for the photo. I intend to make a couple of minor bending adjustments to this piece before gluing it to the tail block.