Wolf notes.

[printer2]

...and a one, ...and a two, ...and a three, ...and a THUD!

So what have you done in the past, or do to tame ghost notes? I know they are out there waiting, just waiting to pounce.

[Bruce Sexauer]

To the best of my ability, I build them OUT of the guitar. All other solutions are band-aids . . . usually mass added to the nodes.

[printer2]

Hi Bruce, I was thinking about your guitars when I thought about this topic. Assumed you had learned how to avoid them from years of building. I realize now that I should have added what you and others do to avoid them. One of the reasons I thought of you is that your guitars are of a high Q value (Quality as in the quality of resonance circuits, how free it is to vibrate), guitars with a low Q value would have less of a problem with wolf notes, they are the equivalent of the family grocery getter as compared to a racing machine.

So what do you do to avoid them? You certainly do not shy away from using different woods or build to an exact shape. I would assume this would throw in a number of extra variables than for someone building to a specified design.

[jomaynor]

Originally Posted by printer2

...and a one, ...and a two, ...and a three, ...and a THUD!

So what have you done in the past, or do to tame ghost notes? I know they are out there waiting, just waiting to pounce.

I don't consider wolf notes and ghost notes to be the same thing, though.

A wolf note is what you're referring to here - the usually unwanted harmonic resonance that causes a certain frequency range to "take off."

A ghost note is a performance obtained hint of a note or a beat. Almost a rest, but not quite silent.

[printer2]

Originally Posted by jomaynor

I don't consider wolf notes and ghost notes to be the same thing, though.

A wolf note is what you're referring to here - the usually unwanted harmonic resonance that causes a certain frequency range to "take off."

A ghost note is a performance obtained hint of a note or a beat. Almost a rest, but not quite silent.

Oops, how did the ghosts slip in there? Something subconscious? Sorry about that. The ghosts I can manage. The wolfs are the ones I am wondering about.

[MC5C]

I think of wolf notes as dead notes that don't ring clearly. I get them when the body resonates freely at the same frequency as the note (or so I suspect). It is as if the resonance sucks the energy out of the string. I think there are many things that affect this, body volume, shape of the waist, location and size of openings. I make archtops and I theorize that the tap tone of the front vs the back has a big impact - my current build will have the tap tones as far apart as I can arrange, just to see what happens. I built a guitar with a side port - it has a strong resonance at D (fifth string fifth fret is the main offender. When I block the side port the resonance drops to A (sixth string fifth fret) and the effect is cut in half. Oddly the same notes played on open strings are affected far less than the fretted notes, which I attribute to the length of the vibrating string. I am currently making some different bridges to see if that changes anything, the bridge on it is non-adjustable and very light. I am making an adjustable bridge of ebony, and I have a (shudder) metal tune-o-matic that I will put on the same base, just to see what happens.

I would love to know if someone has a methodology for building guitars with no dead notes. I read about trying to tune the dead note to an infrequently played note, but I play all the notes...

[Paultergeist]

Originally Posted by MC5C

I think of wolf notes as dead notes that don't ring clearly....

That was my understanding as well.

[dekutree64]

Originally Posted by MC5C

I would love to know if someone has a methodology for building guitars with no dead notes. I read about trying to tune the dead note to an infrequently played note, but I play all the notes...

Unless you switch between A440 and A432 regularly, there should be some frequency space between notes where you can stash the wolf

But I would like to know other methods to "build it out" as well. Seems like usually the stronger the wolf note, the better everything else sounds. At least when using rosewood. Raising the back pitch or making the back more massive reduces it, but gives up some of the airy reverby quality of a good rosewood back.

I've considered trying a tornavoz type soundhole duct to restrict the airflow, but that seems like it would do more harm than good.

Multiple smaller soundholes are another option to try. Keep the total soundhole area around 12 square inches, but not all in one place so the large scale air movement is restricted. But if your side port did more harm than good, then that obviously isn't a surefire strategy. But unless you reduced the area of the main soundhole by the amount that the soundport adds, then maybe the problem in that case was too much hole area.

[Sam Guidry]

wolf notes happen when a strong resonance is too close to a scale tone. can you "build them out"? Of course! All you need to do is make sure none of the strong resonances land on those scale tones. The worst notes i have found are around f# to g# (92hz, 98, 103 respectively) as this is the range for the main air resonance. Another range is between 164 (E) 174 (f) and 185 (f#) as this is the normal range for the top fundamental resonance. Finally, 196 (g), 207 (g#) and 220 (a) are common areas for the back main fundamental resonance to land. Adjusting the top and back resonances is not too difficult and can be accomplished by adding small amounts of mass to the top or back around the center of the lower bout. However, the air resonance can be a troublesome one, as the only options for adjusting it without messing with the top or the back is to choke down or open up the sound hole.
I track these resonances while i build to ensure that they do not land on scale tones and i find that a guitar that is made this way sounds more even and balanced.
It is a good chance that most guitars have at least one woofy note and what i have noticed with most people is that if you have never noticed it, dont look for one because once you know its there it will drive you crazy!

[Bob Womack]

I always "go with the pros" and use


BearPee Wolf Repellant to keep journalistic notes away.

Bob

[Rodger Knox]

Originally Posted by MC5C

I think of wolf notes as dead notes that don't ring clearly. I get them when the body resonates freely at the same frequency as the note (or so I suspect). It is as if the resonance sucks the energy out of the string.

Originally Posted by Sam Guidry

wolf notes happen when a strong resonance is too close to a scale tone.

At the resonance frequencies of a guitar, there is little impedance to energy at that same frequency. The closer to the resonant frequency a note is, the quicker the string energy will transfer to the top. At a resonant frequency, all of the string energy is in the initial attack, and there's no sustain, so the note sounds dead. No new information, just a little explaination of the physics of the energy transfer. Maybe Al Carruth will drop in, he's an expert wolf hunter.

[MC5C]

Well, I did my little experiment today. Taught me a lot, but none of it particularly good... Background - my archtop has a one piece Jatoba bridge that weighs 21.8 grams on my little digital scale. I feel that this is exceptionally light for an archtop bridge, which was the goal was for that bridge, for no particular reason. Jatoba is somewhat lighter than the ebony I have, around as hard, and I thought it would make a nice bridge. Today I made an ebony base and put a Gibson TOM top on it, and the total weight was 89 grams, so a little over four times as heavy. The guitar just about died when I put it on, the sound was horrible and tinny, volume was gone, the shimmer to the sound was gone. What an extraordinary difference! No difference to the wolf tone, except that it sounded worse along with all the other notes.

Step two will be to make an ebony topper and see what it sounds like. I'll do that tomorrow.

[printer2]

Any other luthiers care to share how they learned to avoid the odd bad note?

[Bruce Sexauer]

I believe that nearly everyone misunderstands the primary cause of wolf notes. I do not call them wolf notes, I call them "local resonance nodes" , for that is what I think they are. I am not referring to the body cavity resonance, nor to the native frequency of the individual plates, but to smaller parts of the guitar (in this case) which resonate at specific frequencies which harmonize, or conflict, with the specific note intended by the player. I consider local resonance nodes to be conceptual design flaws, the result of inferior materials, or in some case the result of less than perfect joinery, and to be largely avoidable. Once the concept becomes familiar, a luthier can learn to spot these issues before they are incorporated and largely eliminate them. After learning how to build a guitar at all, and then coaxing superior performance from the result, it is the pursuit of subtleties such a minimizing "wolf notes" that are the road to mastery. Once the instrument is built, the fixes are band aids rather than solutions. IMO, of course.

[Alan Carruth]

I have a pretty broad definition of a 'wolf' note: it's a problem caused by one or more resonances of parts of the guitar. Thus the 'thuddy G', which is generally the outcome of the 'main air' (Helmholtz-type) resonance of the box being too close to the played pitch, so that it converts the energy of the string into sound more quickly and efficiently than the notes around it. You get a sound that's twice as loud for half as long. That's a simple one; only one resonance (well, two if you allow that the 'main air' and 'main top' modes are the two halves of the 'bass reflex couple', but let's stay away from that for now...).

There are two main objections to that definition of the wolf:
1) it's too broad, and
2) guitars don't really have 'wolf' notes like a cello does.

Actually, the usual cello wolf is caused by virtually the same resonance couple that gives the 'thuddy G' on the guitar. The main difference is that cello strings are bowed, while guitar strings are plucked. If you pluck a 'cello at it's wolf pitch you get pretty much the same outcome you get when plucking a guitar with a thuddy G; a loud note with little sustain. Bowing the string drives it into non-linear territory, and that's what ultimately gives rise to the growling or howling that the cello makes at that pitch. This is complicated stuff in some ways: it took a Nobel prize winning scientist (Raman) to figure it out (as a summer project before he went on to his 'real' work).

As for the definition I use being broad; I feel that's an advantage. It allows us to home in on what these problems have in common, which can help to suggest common types of solutions. Solutions tend to fall into two general classes:
1)alter the pitch of one of something, or
2) alter the impedance match between the string, which is driving things, and the resonance.

Altering the pitch of the resonance involves changing either the mass or the stiffness of the system, and changing either will also alter the impedance. To go back to the simple thuddy G problem, you can tune the strings up or down a bit so that you don't play that pitch, or you can alter the air resonance of the box by adding a port or changing the hole size, or you can add mass to the bridge drop the top pitch, or you can shave braces on the top to lower it's pitch. All of these things will make other changes, and will alter the tone of the guitar to some extent, and you need to figure out which is most likely to be acceptable. There can also be cost and warranty issues.

'Q' has an interesting relationship to wolf notes. A 'high Q' system has resonances that are tightly defined in terms of pitch (that's what 'high Q' means). A high Q resonance can be a harder target to hit, so it 's likely to be easier to position the resonance between notes so that it won't be a problem. However, a high Q resonance also tends to be much more active when you do hit it, so when instruments like that have wolf problems, they're big ones.

The usual 'wolf eliminators' that they make for 'cellos are tuned dampers. They have a mass and spring of some sort that allows you to set them for the problem resonant pitch, and some sort of damping in the system that absorbs energy when they're active at resonance. Since they really only add damping near that pitch they don't effect other notes much, but can kill a wolf. A really bad wolf can require a couple of these things to control.

Some of those cello wolf supressors also alter the impedance of the top of the bridge by vibrating out of phase with the bowing force, effectively 'nailing down' the bridge at that frequency. This can be a useful trick on a guitar as well, although it's normally most effective at higher frequencies. I can probably illustrate that best with an example.

One of my customers had a guitar that developed a buzz on the D note, third fret on the B string, after he'd had it for about three years. We spent some time looking at all of the usual suspects, with no luck. Finally I recorded the pluck on my computer and looked at the spectrum. The 3d, 6th, and 9th partials showed 'split peaks', the string was not vibrating at a single frequency at those pitches, but at two closely related frequencies. The buzz was the beat frequency between the two.

Driving the top at the 3d partial frequency (~880 Hz) showed a Chladni pattern that had a strong active area (antinode) where the B string crossed the saddle. The bridge, and thus the end of the string, was moving 'vertically' at that frequency, but not 'horizontally'. As far as the string was concerned, it had two different lengths depending on which way it was vibrating, and so it made two different notes. Some way had to be found to reduce the vertical motion at that pitch without messing up the tone otherwise.

We did some experiments with poster adhesive, to see where adding mass or reducing stiffness in the top would be the most help and do the least harm. In the end I removed a few shavings from the upper end of the upper tone bar, which shifted the node line for that resonant mode closer to the B string. This reduced the motion of the top at that point enough to get rid of the problem without otherwise messing the tone up to much.

So, to get to the original question: you make sure you don't build in problem resonances. High Q can help because a narrow resonance can fit in between notes better. It can also hurt if it manages to get out of the box.

Finding some other resonance to couple in can help. For example; tuning the 'neck' resonance to match the 'main air' pitch will allow the neck to work with the system. Instead of a single tall peak in the spectrum at that pitch you get a pair of peaks that are somewhat lower. The overall output in that range (the 'gain bandwidth product') is generally larger (more power overall), but the response is also more even. Without a single loud note there you don't have the 'wolf' issue. Some wolf issues around the open G string pitch can be similarly helped by tuning the back to work with the 'main top' mode at that frequency.

Simply adding mass to the bridge, say by putting in Depleted Uranium pins, helps 'keep the energy in the string' by nailing the bridge down: the less it moves the fewer wolf issues you'll get.

Generalizing from there: the way to totally avoid wolf notes is to always play cheap, heavy guitars that make no sound. And use light strings. Most guitars do show some of the tell tales, such as 'split peaks' in the output spectrum, of wolf notes, Usually they're not a problem, and that's why my definition includes the proviso that they're 'problems caused by...'. When it's not a problem this sort of behavior generally just makes the sound more 'interesting'. All good guitars have lots of these notes, which is one reason they're good.

Sorry for the long post....