Measuring Plate Frequencies

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

Moderators: kiwigeo, Jeremy D

Post Reply
klooker
Wandoo
Posts: 14
Joined: Fri Aug 23, 2013 3:18 am
Location: Worthington, Ohio USA

Measuring Plate Frequencies

Post by klooker » Sun Mar 11, 2018 5:47 am

I was getting inconsistent & what I thought were questionable results with tapping so I tried to generate some patterns. On this piece I got 81Hz for Fl when tap testing which seemed too high.

I then tried to measure with Chladni patterns. I started around 30Hz trying to see Ft. I noticed that things got calmer when I increased the frequency so started going down and wound up at 26Hz to get this
Adi3-FL-Web.JPG
Adi3-FL-Web.JPG (34.6 KiB) Viewed 606 times
Adi3-FL-Signal.jpg
Adi3-FL-Signal.jpg (27.23 KiB) Viewed 606 times
I had a multi-meter connected to my analog signal generator to measure the output frequency & I also measured it with VA.

The meter said 26 Hz. VA showed peaks at 26, 54, and 81 Hz. So is is safe to say Fl is actually 81Hz?

On this next piece when trying to measure Fc I kept turning up the frequency until the Fc pattern formed but the meter said that was at 184Hz. I tired to get the pattern to form around 90Hz but couldn't do it, possibly because I couldn't get enough Db out of my speaker & amp?
SitkaFc-Web.JPG
SitkaFc-Web.JPG (86.56 KiB) Viewed 606 times
I was also able to get this pretty pattern at 142Hz, don't know if it really means anything useful.
Diamond Pattern-Web.JPG
Diamond Pattern-Web.JPG (56.16 KiB) Viewed 606 times
Also, what type of driver do you use - I just blew 2.

Thanks

Kevin Looker

klooker
Wandoo
Posts: 14
Joined: Fri Aug 23, 2013 3:18 am
Location: Worthington, Ohio USA

Re: Measuring Plate Frequencies

Post by klooker » Sun Mar 11, 2018 7:07 am

I forgot to mention my plate thicknesses which i think would affect the frequencies.

The first plate is 4.85 mm.

The second is 4.21 mm.

Thanks

User avatar
kiwigeo
Admin
Posts: 9310
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Measuring Plate Frequencies

Post by kiwigeo » Sun Mar 11, 2018 6:36 pm

If you hold the plate at the antinode with your (pinched) fingers and tap it you should get a prominent peak for Flong, Fcross and F diag. From memory for the Sitka plates I've tested 81Hz would be where I would expect to see Flong. Here are tap test results for a Lutz Spruce Plate from my stash.

Flong (circa 79Hz)
Flong MG_1196.jpg
Fcross (circa 128Hz)
Fcross MG_1195.jpg
Fdiag (circa 75Hz)
Diag_1194.jpg
Martin

klooker
Wandoo
Posts: 14
Joined: Fri Aug 23, 2013 3:18 am
Location: Worthington, Ohio USA

Re: Measuring Plate Frequencies

Post by klooker » Mon Mar 12, 2018 1:33 am

Thanks Martin,

Is that VA? The plots look smoother like they're filtered which make it easier to read.

I think I was initially concerned because my F-long values were always higher than the examples listed in the book but then I realized my plates are all thicker than the ones in the book.

Tap testing the second piece gave me Fc of 122-128 Hz. The pattern shown formed when the signal generator was putting out 184 Hz. I neglected to read that into VA to see what was actually seen by the plate.

I should probably also state that my "driver" was actually a 2-way bookshelf speaker so possibly the 2 drivers were doing something weird but I'd think the crossover wouldn't allow much of anything to the tweeter at that low of a frequency.



Kevin Looker

User avatar
Trevor Gore
Blackwood
Posts: 1316
Joined: Mon Jun 20, 2011 8:11 pm

Re: Measuring Plate Frequencies

Post by Trevor Gore » Mon Mar 12, 2018 11:52 am

klooker wrote:
Sun Mar 11, 2018 5:47 am
The meter said 26 Hz. VA showed peaks at 26, 54, and 81 Hz. So is is safe to say Fl is actually 81Hz?
Kevin, the VA screen shows those three frequencies and they are all close to integer multiples of 26Hz. It seems that your speaker is generating harmonics of the fundamental, which is not unusual, even when a single driver is used. As you have a pretty high peak at 81Hz (or there abouts) I'd say that is doing the driving and that is what you are seeing on your panel. So the 81Hz tap result is on the money.

Fc is the most difficult one to isolate as it can be hard to get a good "ring" from the panel that isn't contaminated with other modes. Holding the panel about 30% down the long dimension and at 22% along the cross dimension seems to work fairly consistently for me. Tap on the centreline of the panel also at 30% down the long dimension. Explorer a bit until you get a good ring, then measure that.

Regarding drivers, check out this thread: viewtopic.php?f=33&t=7817

Here's a picture of the driver I'm using:

Image

Make sure you don't have any DC on the driver due to DC offset on the sig gen. The sine waves from the sig gen need to oscillate say, +5v and -5v about zero, not from say, 0v to 10 volts. You may have to put a capacitor between the sig gen and the amp.

User avatar
kiwigeo
Admin
Posts: 9310
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Measuring Plate Frequencies

Post by kiwigeo » Wed Mar 14, 2018 1:35 pm

klooker wrote:
Mon Mar 12, 2018 1:33 am
Thanks Martin,

Is that VA? The plots look smoother like they're filtered which make it easier to read.
Yes Kevin the plots are from VA. Im tapping the plates with a rubber eraser stuck on the end of a piece of dowel. The mic is a Shure SM47 and its running via a preamp into a PC laptop. The preamp allows me to boost the signal if necessary. Note Trevor's notes on where to hold the plate.....usually you can hear a distinctive increase in the ring/sustain of the tap when you get your fingers in the right spot. As Trevor also says picking Fcross can be the tricky one.....with the Lutz Ive got in my stash the plot is usually fairly clear.
Martin

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest