Official Luthiers Forum!

does it tell you by itself which frequency is the top and qhich the back?

A spectrum chart by itself won't tell you that, although you can often infer what you're looking at from knowing how instruments work. There are various ways of finding out where the peaks and dips are coming from, depending on what tools you have and can use on the instrument in question. Chladni patterns are easy to do if you have a signal generator and amplifier, but they can introduce things like glitter into the box, which can be hard to remove. If it's an historic instrument in a museum they may not allow it. Simply generating the sound in the room and using a microphone to locate the 'hot spots' can give you much the same information, although it takes more time.

Each different system is, in some sense. a language that you have to learn to read. you can translate from one to another, and may find different ones more useful for different purposes. To get the most out of such a chart you need to know how the test was done, and how the measurement is calibrated. As usual, more information is better. Alain's chart shows a part of the spectrum of several guitars, or maybe of one guitar under several modifications. He was using it to compare a particular feature of the sound that is of interest to him, in the range around 450-500 Hz, but without knowing more about the instrument(s) or test procedures it's hard to say what he's looking at. It could be an 'air' resonance that is actually the product of a complex 'couple' of the air and top, or it could be a 'cross tripole' top resonance of (a) classical guitar(s), or a different mode in a steel string.

No it does not, you have to understand it yourself.
In short:
Taping the top at the bridge will produce the strongest peak at the monopole frequency
Blocking the sound hole will remove the "air mode" or Helmholz peak from the graph.
Taping the back will give you the back resonnance. B(1,1)
The T(1,1)3 back reflex will be very near the B(1,1)
Taping near the wings of the bridge will give the cross dipole, etc.

Refer to: https://goreguitars.com.au/main/page_about_design_modal_tuning.html

There is no magic way there is only a lot of reading and experimenting to do.

The graph is actually from 2 Tenor ukuleles, same wood and shape, same thicknesses, with different bracing patterns, and I was trying to find the differences the bracing made to the frequencies.

Well i am trying to buy Gores book but custom fees are 130 euros and plus the books of 165 around 300 euros....i really want to read them and was ready to order them till i learned about the duty fees...:'(

You really do sound like the right person for these books. Keep in mind Gore is PHD engineer and his books are basically like 4000 level college university text books for a 4 hour class. 3 hours of lecture and a 1 hour lab every week for a semester.

It really is kind of like that. But it sounds exactly what you are after. So my advise, save up and get the books and then go join the Ausie Luthier Forum and poke around.

Well o have spent thousands of hour tweaking intruments so as to see rhat happens if i do....i know need to know what i do i mean gores book sound like thwy have thw answere as far as knowing what to do and not getting just close. Even from the things you mentioned on thia forum i start to understand some things.I thnk to get to the next level i need to have more knowledge...Its a pitty that this book doesnt excist as an ebook.i just need the information not the paper and the taxes haha.

pikolo, where do you live?
I'm in UK, and there are no tax or duty on books.
Replied to your PM

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

I live in Greece. Here the local costums told me that i will have to pay 150(actually told me ' around 130') euros on taxes.Dealng with custom fees in Greece is a nightmare. I dont understand why they charge so much for books...i mean it is not an iphone!

Wow that is pretty hefty tax. I just bought a book from New Zealand that cost 35 US dollars. Then another 45 dollars just to ship it!

I wholeheartedly agree with all the experienced builders above on there not being a formula (though Trevor might argue with that point to a certain degree . That said, here's what I would do to give you the best chance at getting in the ballpark on your first try:

1) Determine the species and dimensions of the target instrument, especially thicknesses of everything that makes up the body. Measure the top, back and sides to .001", braces can be to 1/32"
2) Buy some plans that match the target guitar shape and execute them using the dimensions and species you took off the original.
3) Put the target guitar on a stand, run a polishing cloth under the strings around fret 14, put your microphone a couple inches from the sound hole pointing straight into the hole and then record into your DAW while tapping firmly on each bridge pin. Do a freeze framed spectrum analysis and look for the lowest three peaks in the response during the louder portion of each noise. They should be nearly identical, other than volume. Write those down. they will be around ~100, ~220 and ~400 with the loudest one usually being the middle frequency. (note these are approximate and depend a LOT on the guitar design and size). If you can point out the body style I'll give you my target ranges to help identify these)
4) Once your guitar is done enough to temporarily bolt on the neck and put strings on it (before bindings or finish), do the same procedure with the microphone and DAW. This is where you can do a small bit of comparison and tweaking... but not much. The lowest of the three peaks is mostly related to the volume of air inside the body. You can't do much about that one. The next one will be the top of the body, which you can lower by sanding some wood off the edges of the top around the lower bought and maybe even a little toward the bridge. The third peak is the back, and you can do the same there. If you can reach them, removing wood from the braces of the top or back will also lower those frequencies.

Please understand I just reduced hundreds of pages of reference material and thousands of hours of experience down to a paragraph, so take it with a grain of salt. There are more holes than a good block of Swiss cheese in what I wrote above.

I suspect that's what you were looking for... but again, this will just be a way to get you close if you are lucky enough to end up with similar wood stiffnesses and grain orientations, and, and, and.

Best of luck!

Good advise but...

all instruments should more or less obey the physics rules...so i guess they will apply to mandoline and bouzoukis too....

A less expensive option to the Gore books:
https://www.amazon.com/Fundamentals-Mus ... 048626484X
It is not guitar centric, but may give you the fundamentals of how some of these things work in a broader sense.

Available as an eBook also.


Colin North wrote:
You might have mentioned it was a Greek bouzouki................
I don't have clue, but I do like Retsina and Ouzo

Pine tar and licorice..... two of my favorite flavors in a beverage.....

Well it looks very interesting!

Looking back at your sample from the first page of posts....

What size guitar is this? And, is that spectrum analyzer doing any "smoothing" or is it a single slice of time with actual frequencies captured? I've never seen any frequency response that smooth. And 50Hz is too low for a body resonance, even on our SJ-200 models.

Typically for this type of analysis, you need something a little more accuracy-minded than the typical spectrum analyzer.

Assuming for the moment that data is accurate, I suspect the 50Hz is noise. The 100 is probably your Helmholtz (body) resonance, that peak around 165 would be the top and the back is likely that hump around 300... or maybe that small bump at 250. Those are realistic for a body a little smaller than a Dread.

When you get to finally dialing in your frequencies, it's critical to have an accurate set up to locate frequencies precisely, because if one of your frequencies is too close to a real note you will typically start introducing intonation and wolf note problems.

r.e the 50 Hz signal. What part of the world are you in? More specifically, what frequency do your electrical mains operate at? In the US it's pretty normal to pick up 60Hz signals from transformers, fluorescent lights and such. If you're on a 50Hz system then that could be causing that interference.

_________________
Steve Smith
"Music is what feelings sound like"

It is a greek bouzouki...not a guitar.Maybe thats why?

..

Can you do the same process with your "target" instrument and post that image? Might be more helpful for purposes of explanation.

The picture i post is from my target image.I taped the bridge ten times and this is the results i got.That i posted above .The instrument was strung and the strings muted.

I apologize, I have not read the entire thread, but the cost of the Gore&Gilet books seemed weird. From Greece I would expect the best source would be direct from Trevor
https://goreguitars.com.au/main/page_the_book_international_customers.html

The books are AUD200 (€127) and international shipping is 55.

There is no duty on books imported in Greece, and VAT is 24%, calculated on the AUD200 price at origin.
https://www.easyship.com/en-au/duties-and-taxes-calculator/greece?value=200&currency=AUD&category=books_collectibles



Sent from my iPad using Tapatalk

If this is the target instrument, then my statements above stand. The biggest issue in providing a good interpretation/reference from this graph is that I'm convinced it's smoothed a great deal... or it's combining multiple samples into one averaged out graph (like most real-time analyzers do). Alan's software shows a much more realistic and accurate example that would be a great reference. I've also used WavePad with good results, though it's overkill for this task it does an accurate job.

IIRC Gore's book is fairly specific to guitars. Not just because of the shape of the guitar but because of the general public's perceived idea of what is good tone in a guitar. That was one of the things I found interesting in the book is that it's not all physics, it actually has some psychology and biology in it. What they did is gather up a whole bunch of guitars that were generally considered 'good.' Then they proceeded to measure the heck out of them. From that they built a model on what the general consensus was of a 'good guitar.' That model essentially boils down to a guitar that is fundamentally well built, with good joinery and so on, but that has a top that is just stiff enough to have about 2 degree's of bridge rotation at string tension. This seems to work well across the guitar family of instruments regardless of what wood is used. So with this model you will build a good guitar if the criteria that was met to make this model is 'good' to your ears.

I'm not sure how this would translate to a bouzouki which is shaped very differently and has a bridge with a tail piece. It would be interesting to try though.

I'm still kicking myself for not reading all the previous posts closely enough to realize we weren't talking about a guitar this whole time

Live and learn...

I am really Sorry guya for not mentioninv more that it was for a greek bouzouki but it is ok.In the end we are speaking for guitars too... I have repaired a lot of them so i would like to have all the info you give me!I really appreciate your time and effort to help me out in every question i have.