Official Luthiers Forum!

From what I've read, seen and heard, tap testing is something learned over time, a lot of time. I'd like to fine tune the braces but don't want to go too far. How do you know when to quit? This is where things are right now:

Tone bar


Top of X-brace


Any help is appreciated.

Thank you!

_________________
Julie Moriarty
http://JulimorCreations.com

I think it was one of the old lute tutors that said in relation to tuning the highest string to " tune it up until just before it breaks". That is how some people treat soundboard and brace construction. I like to leave a little more for the "ages" to work their magic on it.
From what I see in the pictures your bracing looks like it should work O.K.. It doesn't look too massive, and it doesn't look too weak.

First: have faith. The guitar designs we have are pretty robust. If you've built carefully and used decent materials you will more than likely end up with a better than average guitar.

Testing of one sort or another comes in when you want to do better than 'better than average', and want do that consistently. Wood varies a lot in the things that determine tone, so building to dimension won't get you a consistent product in that respect. This is a plus for manufacturers: with decent quality control they know that all of the their instruments will be at least 'pretty good', and the variation they get in tone ensures that they'll be able to satisfy the tastes of a broad clientele with varying needs.

I'm convinced that not all testing is conscious. After you've built up some experience you come to 'just know' when things are right. An educated thumb, or ear, are just as much testing apparatus as are dial gauges or signal generators. The problem with those 'organic' methods is that they can be awfully hard to teach or to learn without some hands-on time with a 'master'. That's why some makers have moved toward using more 'mechanical' methods, and quantifying things. It's not very romantic, but you can progress a bit faster, I think, and communicate what you find more easily with others.

There are several kinds of test you could do now, or later, but others that are already precluded by the fact that you've glued the top to the rim. The most basic 'up front' testing involves ascertaining the properties of the wood you're working with. Traditionally this is done by flexing and tapping, but there are more technical methods that can be used. Many makers use a static deflection test. You support the top (usually) on something like a couple of lengths of pipe some distance apart, put a weight on it, and measure the deflection. Some makers use this data to calculate the the correct thickness for the top, while others simply thin it out until they get a deflection they feel is right. A search on 'delfection testing' should yield some information.

One of the more useful things such testing can do is tell you when you've gone as far as you should. Making the top lighter and more flexible can help it to produce more sound, but it can also allow it to fold up too soon. One simple test that can be run on an assembled guitar is to measure how much the bridge rotates forward when you put the strings on it. Trevor Gore maintains if it pulls forward more than 2 degrees the top is likely to be too weak. You can reduce this either by beefing up the structure or by lowering the height of the strings off the top, or by using lighter strings. Do a search on Trevor Gore, and be ready for the deluge....

David Hurd wrote a book that concentrates on deflection testing of Classical guitars, entitled 'Left Brain Lutherie'. I don't know if anybody has applied his methods to steel strings: you could be the first!

Dana Bourgeois uses 'tap testing' both of the top before it is mounted on the rim, and of the assembled box, as ways of fine tuning the tone. This can be hard to get across except person, but he has made information on both methods available. A more technical version of these is testing the 'free' plates and the assembled box using Chladni patterns. Both methods yield similar information: Dana's takes longer to learn, but is quicker to do.

The bottom line is that you can go any number of ways. So long as you don't get too carried away carving braces you will probably be alright just winging it. Since nobody out here knows just what the properties of your top or brace woods are they won't be able to say for certain whether you've exceeded limits, from the information you've given: it looks as though you've got a reasonable safety margin, but then, maybe not. I'd have gotten information on the top properties before I even joined it, and followed along with other sorts of tests as I went before this stage. It's not too late to get some useful information, and even f it doesn't help to optimize this one, it might make your subsequent efforts better.

If you are going by a plan I would stick to the plan and then lighten everything up. Every plan I have ever seen seemed to be over built, which makes sense if they were traced off of factory guitars.

I started doing deflection testing, it takes a few iterations within each model guitar but it starts to make sense soon enough. For me the process starts from the very beginning, handling the book matched tops. You can tell which ones are stiff relative to others in any given thickness just by flexing them, in time it starts to make more sense. As much as I've read and or watched videos I never could get anything out of tap testing. I always do it though hoping that some day the light bulb might go off in the head.

But it is interesting to get a top and join it. Then tap it and hear this beautiful ring. Then you start to thin it out to your target thickness or deflection thickness, tap it again and it sounds like cardboard. Then you start bracing it and you bring it right back up to that nice bell like tone THen finally when it's all assembled into a box is where you get a true tap tone imho. That's the one that for me makes better sense, that and 'echo' testing the chamber too and finding out what resonant pitch the body resonates at.

i do constant tap testing and a basic hand flexing test. i have a pretty good idea of what i want out of a given top so i'll detune it down to a near flubby state. i do this carefully making sure i have as even as possible tap tone throughout the entire plate.

the hand flexing double checks me to make sure that i don't go too far. once assembled, i know that my hardwood bindings and finish will knock the pitch back up a bit to where i hope it will be. so far it's working.

Leave those upper X legs alone. Lower bout bracing can probably be carved down some.

I go by a combination of stiffness and tap tones. Mainly stiffness until the very end, because tap tones change too much with each step of the build. Trevor Gore's 2 degree bridge rotation rule that Alan mentioned is my ultimate measure of stiffness, but when thicknessing the soundboard you can flex it with your hands and see how boingy it feels, and at the open backed box stage you can press on the bridge area with your thumbs and see how plush the soundboard feels. Assuming you haven't glued the bridge on yet, you have to guess how much stiffness that will add.

Do tap and listen at all stages of the build as well. The more information you can get, the better. Record some taps as well, for future comparison.

If you spool clamp the back on, you can really get an idea how the box will sound, while still being able to open it back up and carve some more. The tone will be a lot lower pitched due to the mass of the clamps, and a bit shorter sustain due to the cork pads on them damping out the edges, but it's still a useful test. The top and back generally maintain the same semitone relation after gluing and removing the clamps, so you can use that to tune the back braces (I aim for 4 semitones above the soundboard, as recommended by Gore/Gilet). I record the spool tap in Audacity and do a spectrum plot to see the frequency peaks, but some people can do it by ear. Again, the bridge will change things once it's glued on. It can either raise or lower the soundboard pitch, depending on how stiff the bracing already is in the bridge area, and how heavy the bridge is.

I had the same nerves on my first guitar. I had a Antes OM plan, onto which an experienced luthier scribbled the correct starting dimensions (1/4" wide bracing, 1/2" high at the x). Then with that starting point I tried to match the brace profiles on a picture of a Santa Cruz OM. I decided to err on the side of under building rather than over building - figuring that if the guitar started to distort after a year or two of tension I could always salvage it with a bridge doctor. Fifteen years later that guitar is still has great geometry and sound.

Thank you for all the great information. This is such a great place to learn!

When I was tap testing the soundboard and doing some carving, prior to any glue ups, I did notice some changes that sounded better. Since the final sound is affected by all the parts that come together, I had to admit I really have no idea if what I'm doing will be good or bad or not matter at the end of the day. But that didn't stop me from carving a wee bit off one area that sounded a little dead.

I'll take what's been shared here and give it another look before moving on. Thanks again.

_________________
Julie Moriarty
http://JulimorCreations.com

learning to voice a guitar is an art. I use a few basic rules and as I am sure you maybe seeing is many of us do things a bit differently . A good rule is to keep a building log so you learn how and why.
Best advice I can give you is to build a sacrificial guitar. Poplar will actually make a decent sounding guitar. Start with unscalloped braces. Many have a misconception that you have to scallop braces to get good tone. Many of Martin guitar don't have scalloped braces.
Back to the poplar , make a guitar out of poplar and you can put a trap door on the back of the guitar , then start working the braces. I learned more doing this than 15 years of building.
Mr Carruth has a great deal of experience and knowledge and while I learned a lot from him at meetings like ASIA and such I learned a lot more in building the experiment.
There are 2 things I will tell you that I often read on the forums and that may me misinterpreted information
A Scallop the upper braces
don't do that. The structural part are there for a reason. Yes you may scallop or taper the ends but 3 inches in to the rim . If you scallop these you can expect top deformation and shortened life of guitar
B break angle
This one is a matter of physics . The fallacy that you have to 45 degrees of break angle is not true . I used a spectrum analyzer and oscilloscope and you won't see much breakdown until you get rather low at about 5 degrees the tonal decay was noticeable.
C Tapping on an open top may help you to a degree but don't trust your ears you can gets apps that show wave amplitude and what notes you areas of the top are moving to. Find what you like. I have learned that open top deflection and closed top deflection are 2 different things.
D Helmholtz frequency This can be a help but this is the frequency your back naturally is moving to
E things change Humidity can alter things so be wary of the RH control
F Don't force a joint if your joinery is not pristine you loose energy and efficiency of energy transfer.
G You won't learn it all over night but you will find some journeys of information are dead ends. You can learn something from everyone even if it is not to do it that way.

enjoy the journey


There is an art here so I have only build 250 instruments and I am still learning. A good rule of thumb to start with is when scalloping think about your loading of the top. Since material usually are more efficient taking a load under tension than compression you can use that to your advantage. I do not take my X scallops below 5/16 in. As a rule I start the scallop at about 3 inch in from the rim. As for the X joint I have seen too many guitars that were over scalloped and over time the top bellies and the bridge will rotate forward . So the marriage of design to structural load is desired. Too small bridge plate or overly scalloped braces may lead to failure.
Experience and knowledge will tell you what you what you need.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

Personally I use deflection testing. It gives consistent results and automatically compensates for differences in wood density, resin content etc. Here is a link to my process https://howardguitars.blogspot.com/2016/01/deflection-testing-of-acoustic-guitar.html

In regards to your main question about when to stop. While I do not share my target values in the article as they took me a while to figure out and represent my "tone" I will share a way to determine them. Start with one guitar built exactly to a known good plan and measure it during construction. Next build three variants, each varying in a different way. ie, one with a more flexible top but same stiffness braces,one with same stiffness top and lighter braces and one with both and compare. That was basically my process start. Alternately one could reverse engineer a starting point by removing the top form a decent sounding factory guitar and measuring it , then shaving off the braces and measuring the top and use those numbers as a maximum stiffness.

No matter how you decide to approach this it will require gaining experience from building a few guitars.

_________________
Brian

You never know what you are capable of until you actually try.

https://www.howardguitarsdelaware.com/

Thanks guys! I'm going to leave well enough alone and see how it goes. You've made me realize I can't tell if what I'm doing is right or wrong without having some reference. The back goes on today.

_________________
Julie Moriarty
http://JulimorCreations.com

I read Brian Howard's paper some years ago and adopted that method and it is beginning to work very well for me, in fact it really took the voodoo out of the tap testing and so on. So thank you Brian!

The best part about this method is that you can use any wood you have and get very similar results and consistency is very important in this business. Trevor Gores method is very detailed but you need to have very precise dimensions in your panels to get it right. IDK about you guys but I buy wood where I can get it and it's often cut around knot holes and not shaped to perfect rectangles. That doesn't mean it's not good wood, it is, it's just cut around the defects. So doing deflection testing on the top that is roughly cut to dimensions is more consistent in the long run.

Basically what you are doing, perhaps without even knowing it, is measuring the Youngs Modulus of the wood. Even with deflection testing if using perfect panels you can actually calculate YM but that's not really the point in deflection. Keep a log book, record everything, and when you start to get guitars sounding the way you like then do deflection on cedar, spruce, hemlock, pine, whatever... and you can expect to be around that target.

What everyone else said, and:
After the guitar has been played a while, compare it to the best guitar you can find that is made to the same basic pattern (especially the bracing) and wood types. If it sounds better than yours, start comparing the sound you get when you thump it with your thumb. Go all over the top and look for an area where the better sounding guitar gives a lower tone than yours. That is an area where you want to take extremely small amounts of wood off your braces. I usually use a small curved chisel or a razor blade. You only have to remove a miniscule amount of wood to make a noticeable change. Whatever you do, don't try to make yours lower than the model guitar. Besides making the guitar fold up in a few months, you can end up with a loud but nasty-sounding guitar. Don't get greedy, you can't make a ukelele sound like a bass viol. All you are trying to do is to take an already great well-made guitar and bring out the best in it by copying a proven successful design.

Anytime you get your hands on a great guitar, thump it all over from headstock to endblock, stare at the inside with a mirror, push gently but insistently on the bridge. And this goes double if you ever get to examine a great guitar with its top off.