Official Luthiers Forum!

Starting on 3 & 4 now, and 1 & 2 are in the books. Both built pretty close to the edge, with #1, a 00 having a pretty prominant wolf note at F# ( think that was it...) initially, that has almost entirely gone away over time. #2, a Parlor has a much less noticable wolf note at G#. I'm guessing it will fade over time as well.

I guess my question is this; Are lightly built guitars more susceptible to wolf notes? Are they avoidable, or are they something a builder reacts to after the fact, should they be present?

The answers to your questions are yes, sometimes, and yes. I realize those answers don't really help much, but I don't have enough understanding of resonant frequencies to offer any advice or explaination. There are those here who do...

_________________
A man hears what he wants to hear, and disreguards the rest. Paul Simon

I'd agree with Rodger's answers.

From what I've seen in steel strings, most wolf notes come from the back. If it's not too bad, time will often take care of it as you have seen. If it needs fixing, you can often find the hot spot by playing the note and feeling around. Once I find a suspect area, I use double stick tape and experiment with adding mass or stiffness on the outside of the box. When you find the best solution, add a similar brace inside.

_________________
http://www.chassonguitars.com

there is so much to learn . I make sure that my top and back are a good semi tone apart. Then there is the box frequency so it gets very specific in physics.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

I would recommend Trevor Gore's books for an insight/explanation.
It answered all the questions I had - and several I didn't know I should have to ask.

http://www.goreguitars.com.au/main/page_the_book_overview.html

Basically lighter built guitars may mean more responsive guitars, and wolf notes (dissonance) are then more likely.

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

Thank you Colin!


Yes. The reason is because the modes of vibration (the things that are responsible for radiating sound from a guitar, see my post in this thread) have greater amplitude. Simplifying things a bit, when a mode of vibration has greater amplitude, it also has greater admittance to the energy that drives that mode at its resonant frequency. So if a mode has a resonance frequency at, say, 110Hz, which is right on the note A, when you play an A note, the vibration energy from the string flows unimpeded into the guitar top, which exhausts the energy in the string rather rapidly, so you get a clunky sounding "wolf" note. What you actually want is for the energy from the vibrating sting to be transferred in a rather more orderly manner, which gives a note the sustaining characteristic we're familiar with. So "lightly built" guitars have to take into account where the modal resonance are, and they need to be placed so as to avoid scale tones and so you much reduce the likelihood of offensive "wolf" notes...which answers the second part of your question:

And to the third part of your question:

You really need to address the issue throughout the building process, so that you can be sure that you don't land the modal resonant frequencies right on a scale tone. In fact, you need to land the modal resonant frequencies exactly between scale tones to minimise the effect.

This is something the factories can't/don't do and so they build guitars that are not particularly responsive so that they avoid this problem (and a bunch of warranty claims from people not used to looking after responsive guitars). Custom/private builders can consequently do a lot better.

The details on how to do this are explained in that marvellous publication that Colin (above) mentioned. There's 800 pages of it, not easily abridged.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Any thoughts on why wolf notes would go away over time? If wolf notes are caused by a highly responsive guitar with a low frequency mode lying very close to a note of the tempered scale, then I would suspect one of those characteristics is changing over time to reduce the wolf effect. I don't expect that the guitar is getting less responsive over time (this would work counter to the idea of a guitar "opening up" over time). But I also don't see what would change to affect the frequency of the mode. The geometry certainly isn't changing and I don't think the material properties are either, unless the stiffness or mass is changing due to drying?

I wonder if perhaps the internal damping is what changes. If the damping goes down, the modal peaks get narrower in the frequency domain. So your mode at the note A (let's call it 109.25 Hz instead), is no longer excited as well by a 110 Hz input. This explanation would not work against the idea of a guitar "opening up", might actually coincide with it.

As the guitar ages, the resonant frequencies change (drop).

_________________
Formerly known as Adaboy.......

Interesting. Any idea why? Or how much? Also, that would imply that a guitar could develop a wolf note over time. Has anyone see that happen?

Great discussion. I like this. Not to ask anyone to give away treasured secrets, but the easy way to look at top/back frequencies is before the box is closed. Is there a rule of thumb for how much top and back frequencies are likely to change as we close the box? I suppose the bridge changes things too...

I like rules of thumb. This is not to diss the analytical types, whom I respect immensely. It's just that I pretend to be an engineer at work, hence I don't want to overanalyze stuff on my free time - I prefer to take something like an artist's perspective on things I do for fun. Balance...

FWIW, two weeks in and the wolf on my redwood/hog parlor is fading. I don't think I'm going to need to buy more double stick tape.

The vanishing wolf can be one outcome of 'playing in', and we know a lot less about that than we'd like.

Basically, the low-end wolf at F# or G on the low E string is usually a function of strong coupling, as Trevor says, between the string and the 'main air' part of the bass reflex couple. The top moves a lot, and extracts the energy from the string in a hurry.

If the 'main back' tap tone is very close in pitch to the 'main top' tap tone, the two will couple strongly and enhance the bass reflex action. In my guitars I've noticed that the 'main top' pitch tends to drop about 1/2 semitone in a month of hard playing, so the couple can start out strong enough to cause a wolf problem that fades over time. I've had this happen on a few guitars: they tend to have a little 'bark' on the attack, which goes away. It's also possible for the change to go the other way if the 'top' tap tone starts out higher than the 'back'. It can happen, too, that the resonant pitches will change some with variations in humidity, and this can cause a wolf to come and go. I had that problem last summer, when I took a guitar to the GAL convention to talk about wolves, and they all absconded.

The whole subject of guitar wolfs is pretty interesting; there are a lot of ways you can get them, and every time I've looked for them on a really good guitar I've seen one or two (or more). Often they're hidden by the simple fact that the guitar is producing so much sound, and in some cases they can add some 'sparkle' to the timbre. I think of them as a normal part of the tonal ecosystem, which only get to be problems when they get out of hand.

So if I want a guitar with some low end growl...

Do the freqs change that much?
I've been tracking the same bunch of guitars for some years now and the greatest drop I've seen is a bit less the 3hz, the rest averaging to about .5. It's only 4 guitars over 2, 2-1/2 years mind you...

Things change so much when you close the box (edge condition of the plate, coupling between components, addition of the bridge, influence and coupling of the air cavity etc. etc.) that it is very difficult to predict accurately where you're going to end up. It's the resonances of the closed box that matter, as that is what you hear.

Regarding how resonant frequencies change over time, on my guitars, generally they don't. I say generally because I've had one case where they actually rose, significantly. I think (one can never be certain) this was because I was using "young" wood, dry but not seasoned, that I shellacked on the inside. I think the shellac reacted with the wood resins to crystallise them, with the wood getting stiffer and stiffer as I was building the guitar and this continued on for about 6 months until it stabilised. Other than that one example, my guitars stay pretty constant.

I measured a bunch of top drawer classicals from Spanish masters when doing the book, and their top resonances were very close to precisely between scale tones. I got the impression that this was not a coincidence and that they had also stayed stable over 30-40 years. So it's definitely possible to have long term stability, even though there are numerous reports of the frequencies dropping, which is one explanation of why wolf tones "move" through the aging process. But as Al says, there's likely more to it than just that, with playing in effects also having some sort of impact.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

[quote="Trevor Gore"]Thank you Colin!

[quote="tjp"] So if a mode has a resonance frequency at, say, 110Hz, which is right on the note A, when you play an A note, the vibration energy from the string flows unimpeded into the guitar top, which exhausts the energy in the string rather rapidly, so you get a clunky sounding "wolf" note. What you actually want is for the energy from the vibrating sting to be transferred in a rather more orderly manner, which gives a note the sustaining characteristic we're familiar with. :[quote="tjp"]quote]
I have a guitar from a well known luthier (I don't think he is on this forum) that has a wolf note very close to 196Hz G. In discussing this with the luthier, one of the possible contributors to it showing up (I don't remember it being there when I first got the guitar) is my lowering the saddle height too much in order to lower the action where I prefer it. Does the change in saddle height (and therefore break angle) contribute to the transferring of energy too rapidly as explained above? or is there likely some other mechanism between saddle height and wolf notes?
BTW, I solved this issue by playing this guitar in a tuning that doesn't use G so often and never as an open bass string.
Dave

Saddle height can effect quite a few things, but a change in saddle height per se is not that likely to precipitate a wolf note. However, removing material from the saddle will lighten the saddle and even that small amount of mass change can make a difference. If that seems to be the case, the fix is to find some slightly heavier bridge pins so as to get your total bridge/saddle/pins mass back to where you were. Or alternatively, if your bridge pins are already on the heavy side, go to a lighter set. The plastic ones are lightest.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

In a series of experiments on saddle height and break angle I found that changing the height of the strings above the top can alter the resonant pitches of the top modes a little. So can changing the mass of the saddle. The mass change is likely to be the more important one, but, as Trevor says, you can't discount anything, since even a small change can alter the sound.

The demo guitar I had at the GAL last summer had a strong 'wolf' on the open G string pitch, but most people didn't hear it unless it was pointed out. In that case the large amplitude top motion caused the string to produce two pitches at the fundamental frequency, slightly above and below the actual G pitch, which resulted in a fundamental that had a strong beat frequency. It showed up clearly in a spectrogram, and an FFT of the note had a clear pair of peaks near 196, but the effect was masked by the upper partials, which were unaffected. This is what I mean by a wolf adding 'interest' to the sound. You can usually find these 'split peaks' in some partials of played notes on good guitars.

Once in a while this gets to be a problem, though. One customer of mine had an issue with the D note on the third fret of the B string buzzing. The problem tracked the pitch if you re-tuned the string, so it was not fret buzz (which we ruled out by other means anyway), but some sort of wolf. A recording of the played note showed that the 3d, 6th, and 9th partials had split peaks, so a resonance near 880 Hz was the culprit. A Chladni pattern found the resonance, and shaving one of the tone bars a little moved the node crossing the bridge closer to the B string location, reducing the wolf without ruining the tone of the guitar. BTW, he'd had the guitar for several years at that point, and has it still, about fifteen years later, and plays it a lot. Sometimes they 'play in' in funny ways...