Tap testing the panels - help, what am I seeing?

[joel]

Hi all.

Yeah, yeah, tap testing panels... Bet you all thought you were over that didn't you? Well, it's my first time to this party. I've read some threads and gleaned some good tips to help me record decent VA spectrographs. But I'm getting some weirdness in the spectrographs for cross grain vibration modes for both my Adirondack top and EIR back. They both have double peaks, and the frequency is higher than expected.

Vital Stats:
Top - Adirondack. 4.6 x 558 x 199mm. 232g. It's pretty dense compared to other spruce values in the books, but I haven't seen other Adirondack values listed anywhere so it could be typical? Dunno. Currently getting 3.07mm thickness. Seems a little on the thick side.
Back - EIR. 3.6 x 530 x 199mm. 293g. Target thickness = 2.2mm

First off, lets look at the long grain vibration mode. I think these are pretty clear.

Good strong peak at 68Hz for the top

And a good strong peak at 51Hz for the back

Here are my problem children, the cross grain vibration modes.

As you can see there is a strong double peak. The peak with the greatest magnitude is 191Hz, the second peak is 233Hz

And again, a strong double peak. The peak with greatest magnitude is 196Hz, and the second peak is 167Hz

The twisting vibration modes look pretty good.

Strong peak at 59Hz.

Strong peak at 49Hz.

Have I done something wrong with my tap testing and recording for the cross grain vibration mode? I must admit this was the most difficult to do. With the long grain and twisting modes I easily found good spots to hold and tap. I did my best with the cross grain to find an antinode to hold and a node to tap. At least these spots closely resembled the book diagrams and sounded/felt best. I recorded multiple sessions across the day on Sunday and got very similar results each time, so at least I was getting repeatable results.

Gear: Focusrite USB interface with matching condenser microphone. No peaks in the noise floor that I noticed in VA.

[Trevor Gore]

The cross grain mode is definitely the most difficult to get. However, the thickness result does not depend very heavily on it, which I guess is the good news!

I hold the panel about a third of the way down the long edge and tap at the same level down with the mic immediately behind the tap position. The right peak is usually the lowest frequency high peak, so in your cases the 191Hz and 167Hz peaks. Try holding on different edges and different ends and see if you can repeat the result. Then try moving your holding position from 1/3rd the way down toward half way down in small increments and see if you can get the same results repeatably.

The typical reason that the cross grain mode is difficult to get is because the node lines are curved, see Sections 1.6.2 and 4.3.3.3 in Design.

[joel]

Hey thanks for the reply Trevor! I really appreciate it.

I tried adjusting holding position and flipping the panel under test around to try all the sides. Moving the EIR back panel a little further away from the microphone did make the 167Hz peak stronger compared to the 197Hz. But I always got those two frequency peaks for my back panel. My top panel results were unchanged, regardless of edge or distance changes.

It's surprising how insensitive the results are to changes in where you hold the panel. So long as you hold it relatively close to the spots shown in your book.

Target top thickness = 3.07mm
Target back thickness = 2.33mm (up a bit with the lower cross grain vibration mode frequency)

Thanks again.