Tester body, removable top, fish glue

hoji

Well-known member
Joined
Apr 15, 2016
Messages
278
Reaction score
137
Here's an idea - build a ukulele as usual except attach the top to the body using fish glue. Fish glue, as you may know, dries very hard like HHG, except can be easily reactivated with water. Great for temporary joints.

This will allow to top to be removed and replaced with different experiments, to observe the effects of various thickness, stiffness, bracing, etc. It should be of a design that would not have too much, if any, overhang of the fretboard on the top.

I may try this. Its sinking in that I need some sort of semi scientific methodology to learning why these things sound they way they do.
 
I used to think along similar lines but thought a removable back was the way to go. In my case it was just bracing designs and not soundboard thickness that interested me. My solution was first to use a complete cheap Chinese uke and later to use an Ammoon kit simply replacing the laminate soundboard with something else.

My impression is that many on the forum build to a plan or maybe make small modifications to a proven design. I think you are in the minority in wanting to experiment.

However I would encourage you to have a go if that appeals to you. I would be interested to see/hear what you produce.
 
Last edited:
Hello, friends. I am back after another brutally hot and humid summer in the subtropics, and my shop is once again hospitable and the weather suitable for instrument making. Finally got this project resumed and the first experiment assembled.

The top is spruce, and the rest is poplar. Top and back are attached with fish glue. A little steam and they pop right off.

First go-around, the top is what would be considered VERY thick and heavy - .098" thick and weighs 41.5 grams (including brace and bridge patch). Brace and bridge patch rough and not shaped or sanded. Hardly any deflection. (I am in the process trying to put together a Kawika-style deflection testing rig, so I can measure as I go).

As expected, it does not sound good. But that's what I wanted! Now I know what a too stiff and massive top sounds like. :) It can only get better from here (I hope).

For the next experiment I will sand the top down a bit, measure weight/thickness/deflection, and re-attach and evaluate sound. Keep repeating until it starts sounding good. At some point I will start shaving the braces and sanding the patch to see what that does. The END GOAL is to get it to sounding good and then know what a good sounding top "feels" like (unattached) in terms of deflection, weight, etc. I may even keep it unattached as a "reference" for future builds.

(As is, it doesn't actually sound terrible! It is somewhat thin and boxy, not much volume or projection, but does have decent sustain. Sounds "toyish". Also interesting is that the treble response is quite poor and lacks definition.)

KIMG0908.jpgKIMG0982.jpg
 
I like the whole idea of this experiment. People talk about doing it, as have I, but never actually do it... What about end points? Theoretically you start with a too thick top and then take it all the way down to a too thin top with intermediate stages. Of course this would ruin the top but you would have found that too thin end point. Then go back and do the "right thickness" top. A lot of work.

Keep us informed please.
 
Interesting experiment! One thing you might want to do is record it at each stage so that you can compare the sound after you are through. Play the same chords/scales/songs for each recording.

I would think that you are going to have to go through many tops and braces. Otherwise, how would you know if a thick top with thin braces sounds better than a thin top with thick braces?
 
Eventually going to do the same with a guitar. Thought of giving the inside surface of the top a coat of hide glue and replace and adjust the bracing as desired. Thought with a coat of glue it would even the score between changes.
 
Yeah I agree with sequoia, it would be great to see the experiment deliberately carried "too far" to the point that the top is so thin it's no longer acceptable. Then you know for sure where your endpoint is because you went right by it.

So many variables with this. It would be interesting to be able to compare different bracing patterns at each thickness, but then how do you control for the top itself? Switching braces at each thickness sounds like a lot of work, using two different tops means there could be variation in the top itself.
 
Yeah I agree with sequoia, it would be great to see the experiment deliberately carried "too far" to the point that the top is so thin it's no longer acceptable. Then you know for sure where your endpoint is because you went right by it.

So many variables with this. It would be interesting to be able to compare different bracing patterns at each thickness, but then how do you control for the top itself? Switching braces at each thickness sounds like a lot of work, using two different tops means there could be variation in the top itself.

If you went "too far" with thinning the top, would you know right away, or a few months later? Wondering as I'm planning to build something....
 
UPDATE on this project. (change of plans :) )

It turns out that my disassembly/reassembly idea is not so simple. A couple things caused me to scrap this idea of progressive thinning, and jump right to the "end". For one thing, the fish glue is not as cooperative as I'd hoped - it takes a lot more heat and moisture to get it to separate, and more prying that I'm comfortable doing. However, contrary to this, the back popped free on it's own twice apparently due to ambient humidity, which was annoying. (I'm in subtropics.)

Second, there was a pretty saddening disaster during the first disassembly. The two transverse braces on the top came completely off. I was planning on these staying put (they were HHG'd) and to serve as accurate repositioning guides, indexing into their side notches. With these off and needing to be reattached, getting the top lined up again was a major pain.

So, for a final assembly, I thinned the top way down - .065 inches at most (and this is spruce), shaved the braces way down, thinned the bridge patch to about 1/16 as well. The top, with bridge patch only, weighed in at a final 24.5 g. The first version was 41.5 g including braces.

So I got it all back together, all HHG this last time. I also tried a finishing experiment which I think turned out really cool - new post to come on that.

First sound test.. real "poppy" and no sustain to speak of. Obviously too thin. And now I know what that sounds like. BUT overnight, after the strings settled and got it tune... it's not half bad! It's actually one of the better sounding ones I've made. That's not saying too much, as all my previous ukes have been not great sounding. But this one is close to what I would say is good. So, maybe I've just been making them too thick! I believe I learned something useful after all.
 
I applaud your stick-to-it-ness Hoji. And it might have been a pain in the ass, but you learned a lot which is invaluable regardless of how the uke sounds.

When it comes to erring on ukulele sound boards, too thin is the lesser of two evils but the both present their own unique evils. It seems that almost all first time builders over build their ukes and make the braces and top too thick which results in a dead, muffled sounding instrument. Boat paddles. Going too thin presents its own evils though which can be structural and that is no good either. Keep an eye on that bridge for rotation and watch for bellying.

Sounds like what you have is a very "responsive" top in that the energy in comes out very quickly and thus the "pop" you are hearing. This is not necessarily a bad thing depending on how the uke is played. Good for delicate finger picking in a quiet environment but not so good for heavy strumming in a noisy environment.

So now we get to the Goldilocks syndrome - finding the top that is just right. Not too thin and not too thick and just right. And then you throw in the variability of wood and... everything goes out the window. That is why luthiery is an art and not a science.
 
Every piece of top wood is different, no matter what the species, so you will have to know what to do with each new top plate, each and every time. Finding the best top woods, which are dry, stiff, etc., is the most important part of quality sound. Once you get to a point where your intuition tells you this is a excellent top plate, understanding comes easier. It takes lots of instruments to get there. Enjoy the journey!
 
Top Bottom