Removable Back

[TallDad71]

I am new to this forum and it looks like it is full of builders with fantastic ideas, I would like to hear your opinions on my next project.

My questions are:
Does anyone have any experience of trying to build a guitar with a removable back and can they save me a week's work with some pointers?
Even if my idea works will I learn anything about braces that wasn't startling obvious to begin with, or have is there a better way to set up the experiment?

Experiment
I have build around a dozen steel string guitars some with very strong aesthetics and some with very pleasing tonal response. I have even built one or two that are both which is just beautiful and rewarding

The problem I want to investigate is that of bracing, specifically X Bracing, how it works, why it works, I can read it in Somogyi or Siminoff books but I want to experiment for myself.

To start the experiments I need to have a guitar where I can continually remove the back. By doing so I will have free and easy access to the bracing system and be able to add and remove material at will.

• The experiment will start off by loading up a bridge with no bracing underneath, can it get to pitch, what level of plate distortion takes place?
• I will then go on to add an X Brace. What difference has this made? It that brace enough on its own to resist the pull of the strings?
• Do I have to scallop the braces and leave a node 70mm out from the edge or can I taper them and still keep structural integrity.
• What effect to finger braces have, why use tone bars,
• how big should a bridge plate be,
• what effect do they each have on tone, resonance. etc etc etc

As you can see I have no choice but to have a back that can be removed and re attached without using glue and waiting 24 hours between each experiment

I have already made the decision to leave the back heavily braced, flat and unresponsive as having a plate that is sympathetic to the soundboard will complicate matters too much. I will run the back plate parallel to the soundboard to make adhesions better. What I need now is a system which will keep the back board in place whilst reflecting as much energy from the soundboard as it can.

My preferred method at the moment is to run kerfing around the outside of the sides of the guitar but to perforate it with strong 2mm thick magnets. I can then glue metal plates onto the outside edge of the backplate.. This would do an excellent job of keeping the back plate on but....... it would do nothing to prevent deformation of the sides from the pull of the neck and its need to rotate the neck block upwards when brought to full tension! So my next though was to build an internal rim and external which glues onto the backplate and fits snuggly either side of the guitar's perimeter, essentially locking the shape of the sides into place couple with the magnets.

Once I have finessed a soundboard for this setup I intend to use the same equipment to finesse the back, to keep experimenting to find a back plate that harmonises well with the soundboard.

My questions are:
Does anyone have any experience of trying to build a guitar with a removable back and can they save me a week's work with some pointers?
Even if my idea works will I learn anything about braces that wasn't startling obvious to begin with, or have is there a better way to set up the experiment?

[Craig Bumgarner]

I've played around with this idea for some time and the closest I have gotten to it is to tape the back on. I do this like taping bindings in place, but use enough pieces to completely close off the joint so that hopefully it is air tight. This does not seem to fully duplicate the glued up back, but comes close.

I'd be wary of adding much in the way of parts you would not otherwise use in normal construction. Steel plates, magnets, etc. have mass that will probably affect tone and may skew your results.

If the guitar is to be experimental, you might consider a number of very small screws into the linings and double sided tape to help secure air leaks.

Another idea would be to make the back a little over size, say an 3-4mm and then glue the back with a reversible glue like HHG or fish glue. You can then pop the back off fairly easily like is done on a violin. Heat, hot spatula knife and a little moisture. No bindings of course. Addition of strips of paper along the joint should make it easier to remove as will thinning the glue to weaken it.

[TallDad71]

Thanks for the reply.

I wondered about the effects of adding extra mass by the use of magnets and can't quite get it clear in my head.

My aim is to examine the effects of the constituent parts of the braces, and then go on to make the best soundboard I can. Is the best soundboard dependent upon the weight of the back, given that it's been made immobile. Or is the optimum soundboard confirguration independent of the back?

[johnparchem]

I would worry about the space between the magnets vibrating. Basically there will not be good coupling between the back and sides. I like the idea of a slightly oversized back and hot hide glue.

[Dave M]

I'm with Craig and John on this. A temporary glue joint of some kind is going to give the best simulation of the real thing and not skew your results.

And by the way we would all I think be very interested in your results. Do document them.

[Allen]

Weak solution of HHG to attach the back. Easy to remove and replace.

I've even heard of adding paper between the rims and back to help facilitate the removal, but don't know if it makes it easier or not.

[TallDad71]

HHG definitely seems to be recommendation. I just can't square that with the need to add or remove material after every measurement.

I decided the best way forward is to make the back and sides out of 4mm ply. The back will have two tracks glued to it in the shape of the sides, one mapping the inside of the guitar and one mapping the outside. The back will then just slip over the sides like a lid on a biscuit tin. 2mm dowel pegs will be inserted into the tracks and through the sides.

For good measure a 4mm wide 1mm high rubber band will sit between the tracks to make sure that the backs sits on the sides without rattle.

I am assuming that if I completely over engineer the back sides and neck it won't matter for this experiment. I am not trying to make the best guitar ever, I am trying to examine the effects of the bracing components on the structural and tonal integrity. I am also hoping that I will have built the best soundboard for a guitar of this shape and size.

Am I right in my thinking that a soundboard has an optimum configuration regardless of the build quality of the rest of the instrument?

[johnparchem]

I think that your experiments will provide valuable insight to how the bracing works. Using the overbuilt jig I am less sure that you will have built the best soundboard for a guitar of this shape and size. as the back and sides coupled with the top and affect the top's acoustic performance. Interesting project.

[jeffhigh]

Brian Burns, a very experienced USA classical and flamenco builder uses the taped on back on all his builds in the process of tuning.
Some form of tie between neckblock and tailblock may be needed on a steelstring.

[blackalex1952]

How about, on your experimental model, using a cylindrical light carbon fibre rod to connect the tail and headblocks together? Well epoxied into holes, of course.A tip I have- use household gelatine as a weak form of hide glue for the removable back. Works for temporary bridges too...( a bridge is a brace in one way, and how it connects to the internal bracing is also something to experiment with, along with the shape weight and timber of the bridge plate- I prefer rosewood for bridges and bridge plates, maple I see as a tone sucker. I would even prefer dense walnut to maple)
I once had, professionally in the 1970's, to record a talk given at a university by Peter Biffin and Greg Smallman when they were making guitars together. This inspired me so much that it is the only recording that I did that was a complete disaster due my not watching the record levels. Totally distorted....the boss was cross, but as a fellow guitarist, he understood. I still have the tape somewhere, don't know if it's salvageable.
Anyway, one of the things Greg did was with a steel string guitar that he had built with a rim around the sides at the back with captive bolts. The back was held on with wing nuts. Peter played it to demonstrate the sound. Greg then removed the back and with a resounding loud CRAK! he ripped out a brace from the soundboard. Needless to say- that got everybody's attention! I have no idea what my record levels did at that point and from then on...... Replacing the back, Peter played the guitar and Greg talked about the changes in sound.
Later,in the student union building, they exhibited the guitars and chatted, which is when I had some good interraction. Greg and I talked mainly about Indian classical music, which we were both into. Greg during this time, brace shaved a punters guitar through the soundhole and tapped and flexed, much to my fascination as I hadn't ever experienced something like that. The only other luthier I had met was Bill May, who told me about some of his ideas when I bought an early run Maton Messiah hand built by him from him. The punter in question noticed a remarkable improvement in the sound of his guitar. Later on I got to know Ian Noyce, another one of my inspirers to eventually try my hand at luthierie. When I mentioned Greg and his attitude, he told me of a luthiers convention back in the day, where Greg turned up and just stood outside selling his guitars! Cheers! Ross

[lamanoditrento]

I am assuming that if I completely over engineer the back sides and neck it won't matter for this experiment.

No it is the opposite. With heavy/over engineered back and sides you change the impedance/admittance between the soundboard the the sides and it will move the node lines and frequency response. Any tweaking you then do will immediately change when you do not have the heavy back and sides and you will be back at square 1.

As mentioned in another thread, this is covered in the Gore/Gilet books which will help you immensely in your investigations