1st box closed - spectrograph result

You can ask questions here about Trevor and Gerard's exciting new book on Luthiery.

Moderators: kiwigeo, Jeremy D

Post Reply
User avatar
Posts: 196
Joined: Thu Oct 11, 2007 7:55 pm
Location: Adelaide, Australia

1st box closed - spectrograph result

Post by joel » Thu Jan 04, 2018 7:03 pm

Happy new year everyone!

Well, I finally closed the box on my first build (Hooray for me!). It's beginning to look like a guitar! I began building years ago, well before I had the Design and Build books. But I thought I'd take a spectrograph of the freshly closed body and see how my extremely amateur 1st effort measures up.
I took this using a spectrum analyser app on my phone. The Red line is the line of interest. When I tap the top I get a good couple seconds of sound. Reassuring - it's resonating a bit.

Details: X braced OM size. Top is King Billy Pine currently at 2.8mm thick braced with Sitka. Sides and Back are Victorian Alpine Ash - the back is X braced too. A couple gratuitous shots are included.
Box Closed.jpg
Back test fit.jpg
Top braces low angle.jpg
Next up is making the neck and the excellent binding jig shown in the Build book.
- If God had intended us to drink beer, He would have given us stomachs. - David Daye.

- The mouth of a happy man is filled with beer. -

Posts: 377
Joined: Mon Jan 21, 2013 2:59 am
Location: Seattle

Re: 1st box closed - spectrograph result

Post by johnparchem » Fri Jan 05, 2018 2:52 am

I am still a novice looking at these but here are my thoughts. If I look at the red it looks like your top and back are pitched close together with that double peak. If the peak around 220 Hz is the top resonance, it will move away from the back when the bridge is glued on; a good thing. If the back resonance is the 220 Hz and the top is the second of the double peak they will be very close after you glue on the bridge and string it up. Looking at the bracing pictures and the spectrum graph your top might be a bit stiff.

User avatar
Trevor Gore
Posts: 1306
Joined: Mon Jun 20, 2011 8:11 pm

Re: 1st box closed - spectrograph result

Post by Trevor Gore » Sat Jan 06, 2018 3:39 pm

Years ago, when I was experimenting with the use of spectrum analysis, I stumbled across Visual Analyzer. The reason I started using it was that from the very first tap I did I was able to recognise the shape of a guitar frequency response curve from the charts that were published in some of the key guitar research papers dating back to the early 80's. I immediately knew I was on to something, even though the image of that first tap remained on the screen for only a fraction of a second. That image also looked very much like the ones in the book, with a high first peak at ~100Hz and a higher second peak at ~ 200Hz. That, to my mind, is what the frequency response of a guitar should look like, because it was very similar in its characteristics to those that the 80's researchers had spent years to produce and spent thousands of dollars on their early specialist equipment to produce them. Now we can do it all and more for the price of a laptop computer or a smart phone, but the frequency response of a guitar hasn't changed, so should still look much the same. What the frequency response looks like depends not only on the guitar, but the mic you use, how you position it, how you tap the instrument, what you use to tap it with and how you set up the spectrum analyser. Get those right (and there are many versions of "right") and you will see a frequency response curve (FRC) like the 80's researchers produced (and like the ones in the book). If the FRC's don't have those characteristic features, how do you know that they are anything like right?

So the first thing I would do is to encourage anyone using these techniques to refine their procedures until they can produce repeatable spectra that resemble the ones in the book (easier to find than the original scientific papers from the 80's!). Then people have comparable results where we can be fairly confident we are comparing apples to apples and a wider number of people can weigh in with their help.

Regarding the results posted, I would be fairly confident that your peak at ~102Hz is the main air resonance, but after that I can't really be confident about anything. So I would encourage you to vary your technique using the guide in the book and with the set-up parameters I suggest for sampling rate and sample size until you can get FRC's that look "right", which may or may not be possible with a phone app, and is one of the reasons I continue to use VA, which seems well suited for the task. When (if) you can do that, I and others will likely be of more assistance to you. If it doesn't work out with a smart phone, there is always VA!

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests