New Builds
TO ADD FORUM POST USE THE DISCUSSIONS TAB OR ADD NEW TOPIC IN THE SIDE BAR
Organizer:
- Organized by
-
-
Spectrum analysis of Guitars
-
Spectrum analysis of Guitars
I figure this could be a fun thread.
For anyone who has done any of the spectrum analysis work based on the discussions with Giuliano Nicoletti on their new builds, or existing guitars, I figured we could post our results here.
-
23 Replies
-
1) An 0-14 fret Martin-style build with Adk Spruce and 35-style Indian Rosewood B&S
-
2) A 12-fret Size 5 martin-style build with Redwood and Koa
-
3) A Small Parlor(slightly smaller than example 1) of my own design with Adk top, enlarged soundhole, and Cherry
-
I just did my first test measurements and will upload later in the week after I make sure I have the system dialed in. There are four instruments: a Fairbanks F35, Bourgeois 00, Guild GAD30 and a StewMac OM top and sides that is mid build. This is going to be pretty wild. The information generated is going to take me some time to ‘grok’. And then of course, can the analysis lead to tweaks and correction?
-
Guild GAD 30 PCE
-
Bourgeois Country Boy 00
-
Dale Fairbanks F35
-
The same Dale Fairbanks F35 with smoothing applied to the graph
-
StewMac Rosewood OM kit – in construction – top and sides only – no back
-
Same StewMac Kit with a 7 gram weight on the bridge plate
-
The Fairbanks has Honduran Mahogany back and sides and a Red Spruce Top
The Bourgeois Mahogany back and sides and Adirondack Top
The Guild – Padauk Back and sides (with a soundport) Sitka Spruce Top
-
Hi Jonathan,
Sorry for the late reply, a very busy period. Sharing measurements is for sure one of the purposes of this space, I love to see that there’s already a lot of interest on the matter, let’s keep on rocking!
Cheers,
Giuliano
-
Hello everyone,
I see that already several of us are using the same software, and that’s a great facilitation for sharing data; I am just adding a quick reminder of guidelines for sharing graphs, to make more simple to read and compare curves;
– always use a log frq axis
– set limits to: 45-95 dB, 60-1500 Hz
– export the picture at 810 pixel, write details on the comment field
I know that at the beginning this procedure might seem boring and pretentious, but then again having beautifully formatted curves to compare will quickly pay back, trust me!
Here an example, and thanks for tolerating my obsessiveness (I am getting older…). 😇
Cheers,
Giuliano
-
What software is this?? In addition to these guidelines, do you have recommendations for software settings we should be using for frequency analysis?
-
Never mind, I just found the answer in Giuliano’s book!
-
-
-
Heres a 1972 D-35
-
Muted top monopole on this D28 copy is higher than unmuted. I’m wondering if there might be some strange coupling going on. I can’t identify a clear cross dipole and the back is not active, but I think it shows up (220hz) when muted. It’s a sitka top with Gore-style X-bracing (no scalloping).
-
If anyone is interested I posted a 3d printed speaker housing for the speaker Guiliano recommends in Loothprints.
-
I’m about to join the club- my cheap microphone I bought a year ago doesn’t work with my laptop so I just ordered the UMIK-1 Omni-Directional USB mic in Giuliano’s book. In the meantime, I’ve run REW with my PC microphone, so not posting anything until I get the UMIK.
It looks like my peaks are all lower in frequency that the lower limit in the charts: I’m interpreting that as “braces not stiff enough”/ too much scalloping. We’ll see if a good, calibrated, USB microphone changes the capture.
-
MAke sure your graph is set up properly to view the info as well.
-
-
All three of these guitars were made from the same Torrified ADK Spruce from Stew-mac, and sinker mahogany back and sides from the same slab
-
Nice plots. Have you discovered the overlay feature in REW? It’s a fantastic feature for comparisons like this.
-
-
The first repeatable FFT plots I’ve been able to get. Used REW and the MiniDSP Umik-1 from Giuliano’s book. This was quite the rabbit hole to get proper, repeatable FFT plots, but the pre-work is done. Figuring out the data implications will be another, like the double humps at 290hz.
Log in to reply.