Build 12.4 - approximating B(1,1)

[joel]

Morning all!

I've glued my back onto my sides. Build section 12.4 says that an approximation of the uncoupled back main resonance can be estimated from a spectrogram of the back+side assembly. Screenshots of the test attached.

What am I looking at? Which peak is the one?

Once B(1,1) is identified, what then? Section 11.1.1.5 only discusses top resonances. If I'm reading 12.4 and 11.1.1.5 correctly the B(1,1) is roughly equal to T(1,1)3. If it's higher than my design goal, I might be able to scallop out more of the back main brace to lower B(1,1) and therefore also lower T(1,1)3 in the completed instrument.

Also, how do you make VA spectrograph reduce/change the frequency window down to 0 to 1000Hz? Nothing I do seems to make a difference to the pics shown. I'm unsure these spectrograms can be read with enough accuracy for tuning my B(1,1).

[Dave M]

If you left click on the spectrum and hold it down you should get a box which will select that part of the spectrum that is inside the box.

Dave

[Dave M]

Again to accurately measure a peak you can repeat the process to get just the peak. Note the control at the bottom that says zoom out.

[Dave M]

Joel what I say above is true of the version I have been using. However having just downloaded a copy to a new PC I discover that you need to hold the control key and the left mouse button down to select a region of the spectrum.

Since the two copies seem to be the same version I'm not sure what is going on yet. Still it should get you going with your measurements.

[kiwigeo]

Should normally be a simple click on the window and then a drag out of the yellow rectangle to include the part of the spectrum you want to zoom in on.

[Dave M]

Yes quite Martin. That's what I have been doing for years on my Windows 8.1 machine. But when I downloaded the exact same version of VA - 9.0.6.60 - to my newish machine running Win10 this no longer works, you need to hold the CTRL key down as well.

I haven't done much to this machine so can only guess that it is something to do with the interaction of VA with the new OS.

Is there anyone else out there that has experienced this?

[joel]

Thanks guys. I knew it would be something simple!

OK, here are the zoomed in versions.

Can someone double check my thoughts regarding the interpretation of these spectrographs?

I have a strong peak at 230Hz (the first of the double peaks). I believe this is the B(1,1)?

If I'm reading the books correctly, I want my back resonance at around 226Hz for a small steel string guitar (Build Table 22-1).

If so, I need to scallop the main back brace down slightly to achieve the target 226Hz?

[kiwigeo]

LOL...my Windows laptop running Windows 8 is in my workshop and its never connected to the internet for fear of being forced to upgrade to a more recent version of Windows. Im a Mac head and the windows machine is only there to run VA.

[AKEric]

You might want to check out Trevor's videos he made with Robbie O'Brien. The latest one explains a lot about the construction process and fine tuning frequencies at different stages, including B(1,1). Most excellent.

[joel]

Mr Eric,

Do you happen to have a link to those videos?

If that 230Hz peak really is the B(1,1), then I think I'll leave the scallopping for now. It's close to target, so it might be wise to see how the body as a whole measures before tuning.

[WJ Guitars]

I have the videos and they are very helpful instructional videos. They can be purchased on line from the O'Brien Luthier Academy. To access the videos once you purchase them you need to have setup your user email address and password with the Academy when every you want to view them.

https://lutherieacademy.com/

The videos in my opinion are worth spending the money in tandem with the Gore / Gilet books. These videos cleared up many aspects that I had not fully understood regarding building a falcate braced guitar.

Wayne

[GregHolmberg]

As described in the Design volume, the natural resonance of the top, the back, and the air are coupled together. It's easy to calculate the uncoupled frequency of each. For example, the air resonance depends on the volume of the cavity and the area of the soundhole. But calculating where they end up when coupled is more complicated. This is what the four-degree-of-freedom model does. You can see in the book that the back resonance will be pushed higher when coupled to the top. And the top will be pushed lower. They "repel" each other.

I've used the 4DOF model to calculate the coupled frequencies for several sized guitars. I see the back resonance pushed anywhere from 18 to 25 hz above its uncoupled frequency.

So if you want the coupled back frequency to end up at 226, then uncoupled might be 201 to 208. It depends on the frequency of the top and the air. It's a complex system and playing with my spreadsheet, I see that it's very sensitive to small changes. Getting it to hit all three of the targets is not easy.

Greg

[joel]

Thanks all.

Rather than perform complex calculations, I'll continue with the build and adjust once done.