Chladni Patterns

[turnertj]

Hey all...

My first build came out okay, but honestly it sounds like a good guitar with a cold. So I started to look around at ways I could be a bit more scientific about the next build. After seeing Darryl's setup for bringing out the Chladni patterns I thought I'd try something similar. I built a box that clamps in the soundboard (or back), and then put a 5" speaker underneath. I'm driving it with my computer and a freeware frequency generator. The video linked in below shows a quick frequency scan of the unbraced soundboard with no soundhole. It's a start. I think the hope is that I could find what patterns are good at this stage, and watch them change as I brace. Then maybe have a better way to tune the bracing, plate thickness, etc.... later. I may put my taylor in the machine and make it vibrate to see what a good guitar has after bracing and the box closed.

http://www.youtube.com/watch?v=Ujq81nsi ... e=youtu.be

[tippie53]

Anything that can give you a number is more scientific that an ear. When you can apply science and you learn to control the process , that will take your building to the next level

[Herman]

There are several items on the net about the Chladni-patterns of a guitartop. But there is no one that I know of, that can tells us what to do when a patterns does not show what would be expected. I made some of these patterns a time ago, just for fun. But IMO it takes a lot of work, braces and experiments to bring you an idea what to do with it.
I don't want to be negative, but for amateurs like us, the guitar is a very complex thing that cannot pinned down with some measurements. In my view we better trust ourt ears and make tight joints. That can take us to the next level too. Let me put it this way: If the Chladnipatterns were the way to the best guitar, everyone was using them.

Apart from that: Have fun in your builds and measurements. Let us know
Herman

[Ken Hundley]

I bet there are a ton of guitars out there that have chaldni patterns that do not fit what people "like to see", yet the instrument its self sounds great. I agree....use sound building principles, good craftsmanship, a little luck when needed, and you'll have a great instrument.

[turnertj]

Hey guys...

I agree with what you all said. Certainly it might take years to build up a database/experience to be able to use these patterns, but it really costs nothing to do, just an extra step...so why not. Maybe it will just be a way to gain some insight and tune my intuition. I'm a scientist and engineer in real life, so I'm trained to look for these cause and effect relationships and quantify them. But part of why I love this sport is the fact that there is as much art involved as science. I'll keep posting what I find as the build progresses and maybe it might help me or others. If not, I may just be making abstract art with a partially built guitar and some sawdust! Nothing wrong with that either!

Tj

PS I think Al Carruth has a video using Chladni patterns. Anyone watch it?

[Herman]

That is the spirit TJ! If you like to make those patterns, go for it. Don't let me hold you back.

The problem with the scientific approach is that for comparing things in a rightful way, you need to make all objects the same before alter one or two things. And the woods we use show more or less different qualities. Often two similar guitars sound different.

So when you get confronted with 2 different patterns, The question comes: where does the difference come from? As I see it, when you want to find patterns in patterns (Yo!) it will take lots and lots of samples.
Who wants to make that effort?

TJ, maybe you're the man. Maybe one day you are not famous for your guitars, but for your research on the patterns. I for me won't, I'm too lazy.

[Darryl Young]

On your setup, it's better to be able to move the speaker around to drive specific locations on the soundboard to develop the patterns. You want the speaker over the area that moves when it is vibrating. The node lines that form are the areas that are NOT moving so the speaker is located in-between these lines. Also regarding finished guitars, once the soundboard is glued to the rim, the pattern shapes are "forced".....and all guitars with similar body shapes will look alike wether it's a crap guitar or an amazing guitar. However, you can learn a lot from the frequency of the various vibration modes.......just not the shape.

Per Gore (and many other folks) the sound of the guitar is determined by the frequency and volume of these vibration modes. So if you want to match the sound of an existing guitar, try to match the frequency of the main, lower vibration modes. It's impossible to match all vibration modes as the ones that form at higher frequencies are extremelly difficult (if not impossible) to control. Even so, if you matched the modes where the bridge bounces up and down (monopole modes), where it rocks side to side (cross dipole), and where it rocks fore and aft (long dipole) and maybe throw in the cross tripole (2 node lines running along the grain).......then those guitars would sound similar.

A point, why does mearsuring and collcting data (no matter how you do it) require you to throw out your senses (ears, touch, etc.) as useful tools? Surely no one believes that as soon as you try to collect data that you must completely ignore other feedback. Why not add it to the feedback you receive from your senses. If the perimeter feels too stiff in the lower bout around the bottom of the tone bars/bass X leg.......I sure wouldn't ignore that. Same song, different verse.......why would collecting data make one stop thinking tight joints are important.......and proceed with reckless abandon? these aren't mutually exclusive approaches.

With the above said, I agree with a lot of what has been said. Creating and measuring Chaldni patterns won't make your guitar one iota better........unless you use the data to make better decisions. Unfortunately, there is no 1,2, 3 step method that says "if you see this, here is what you need to do". It does give you feedback on the distribution of stiffness of your top.......but here again, it may not be intuitive how to read that information.

If doing Chladni patterns, shape is only one piece of information you learn. The frequencies of the vibration modes are just as, if not more important. The frequency of the vibration modes is an indirect measurement of the ratio of stiffness to mass. So if you select reasonable material to build with (quarter sawn, well dried wood, etc.) it gives you good feedback about the stiffness of your top. This can help out a new builder.....or at least I feel it helped me out. Also, if you produce the Chladni pattern (as opposed to just tapping and measureing the frequencies) you know for sure which patterns are being formed at what frequencies and this wasn't apparent to me.

I think I've learned from this.......but I think you learn a lot from building your first guitar no matter how you do it so I can't say for sure how much this helped me since you only get to build your first one time. I think probably many of us here can build a better than average guitar even on our first build. I think focusing on getting the perimeter of the guitar loose for good bass and power from the monopole mode helped make this guitar good. Watching the frequency of the monopole mode and keeping it in the ball park of what other good builders have done probably helped me obtain good power without getting the top too weak. I consider that one good thing that came directly from my measurements.

Again, I think experienced folks can build great (not just good) guitars without doing any of this.......so it's not required. Maybe anyone can get lucky and build a very nice guitar, not sure. I personally think I built a better buitar because I used additional info/feedback from the Chladni patterns.......but I can't prove that......and wouldn't even try to. If it appeals to you go for it, and if it doesn't, I'll bet you still have fun and still have a good chance of building a nice instrument.