Top plate, skirt, neck, leg, nut, bridge, material and configuration, pickups and location, all in turn get the blame/credit for the particular instruments sound. Lets take the player/amp etc. out of the picture for the moment.

If we start with a simple top plate, and suspend it, then hit it (excite it)it will vibrate in several ways (modes). The vibrations will have a number of frequencies involved (a tone), and these frequencies will have a rate of decay (opposite of sustain). Different materials = different results (sounds).

If we now change the dimensions of the simple top plate, the resulting modal frequencies and their centers/locations will change.

If we now change the material of the simple top plate (say from wood to aluminum to carbon fiber), and repeat the experiment, we will get different responses re Hz, location, decay, etc.

So we now have a catalogue of various dings and clunks for a group on dimensions and materials in the form of a simple top plate.

Let us now take the simple top plate and add the skirts (dims and materials are variable), suspend the assembly, excite it, and record the Hz(s), and decay times; even the dings become clunkier.

Add the neck piece and repeat.

Add the end plates and repeat.

Add the legs and repeat.

Onto this rather substantial clunk, we can now add strings, and the mechanisms that hold the strings = Nut/tuners/changer et al.

All of the above materials, dimensions, configurations, attachment methods, coatings/coverings, and their summation of the various resulting dings and clunks are now directly and/or indirectly attached to the strings...they will have an effect upon the vibration (Hz/magnitude/duration) of the strings when the strings are excited. The question is, "how much will each contribute to the overall sound?".

This conglomeration of assembled parts is a "system". Systems have an efficiency of less than 1. Without excitation there will be no vibration. The energy that excites the PSG system is the vibration of the strings.

The system will absorb, or reflect this energy in varying degrees, depending upon the factors described above.

to the degree that the energy is absorbed, the strings go clunk, to the degree that the energy is reflected, the strings go ding.

If the wave(s) travelling along the excited string(s) meet with a soft material at the bridge/nut they will tend to clunk...with a hard material in a sudden way, they will tend to go ding. Materials/configurations farther removed from the excitation source will have much less of an effect upon the instrument's sound, these being "solid bodied" instruments as opposed to "resonant cavity" instruments.

Magnetic pickups have their own set of mechanical resonances (microphonics) and electrical resonances and "Q"s.

OK, so Ding and Clunk are not very technical terms! Just to carry it a bit further, a ding becomes a clunk when the "damping factor" becomes high. The damping factor is defined by the "Logariththmic Decrement". The LD is the amount of amplitude decay from the crest of one wave of vibration to the next...beginning to sound like "sustain" or lack therof?

Different materials have different damping factors... Sitka Spruce = 0.035, Mild Steel = 0.0049, Aluminum = 0.0034. Damping factors for the wood seems to be much larger than for the metals. I wonder what it is for the carbon fiber bodies, and I wonder if I should care.

Ok, there is food for thought re the sound...how about the stability of the sound?

Sound instability, or pitch shift is a function of body/mechanism flexing from change activations, or of the strings/body adjusting to thermal variation. The mass and material of the strings and body are different, so any ambient thermal change will affect the strings first/sooner. Now we are into the various thermal coefficients of the various materials and masses of the mechanisms and body. These values exist for the various materials involved. zero, or no, or none as descriptors is neither real nor desireable for any given material in this "system"...zero differential would be nice.

This rant is the beginning of a series that will again have to be taken off line (for the most part) because it won't "fit" well on the Forum.

The "series" will deal with the modeled (via Finite Element Analysis = FEA), and measured (via Frequency Spectrum Analysis = FSA) for the issues raised above. This is work that was applied to the design of the PSTL C69 BEAST, a 14 string monster with a 30" scale.

We will now carry it further into some general cases for materials, dimensions, and configurations.

There will be lots of charts, graphs, and instrumentation printouts. Those that wish to be included on the Email distribution list please email me direct and I will start the new list.

Come see the BEAST and the BABY BEAST at DALLAS. They will be at the Sierra Booth/Table.

The point of the 30" scale C69 with integrated changer/tuner on the players left, with rails and drop in/on neckplate, and simple terminator block with changeable sonic impedance nut is not that it is "better", the point is to show that it is possible without violating machanical/tonal sensibilities. The Baby Beast is the common S10 E9 24.250" scale using the basic mods of the Beast; so the comparison can be made between the "too wide, too long, built backwards" BEAST, and the much more standard BABY BEAST...not in Specs, but in sound.

You are right in that specifications will not mean much to some pickers, but then who plays what brand does not seem too impressive when they have an economic reason to do so. Most brands would pass muster for them if the price was right.

Of course you can also absorb vibrational energy and therefore sustain from the strings without creating volume. Imagine the bridge and nut attached to a mushy material. Therefore, there are two ways to absorb energy from the strings, one trades for volume by moving air, the other just absorbs it and doesn't move air in any useful way (I guess the energy dissipates as heat, but I'll leave that to the physicists)

But this is all acoustic sound. What happens with magnetic pickups? Unless they have become microphonic, which is usually not considered a good thing, magnetic pickups can only be activated by vibrating metal. The vibrations in the wood body of an electric guitar cannot be picked up by the magnetic pickup. However, any vibrations (or mushiness) in the body absorb sound from the strings and decrease sustain. Therefore, a solid-body guitar has the best sustain. A semi-solid body has medium sustain, and a hollow-body electric guitar has the least sustain. The solid-body also has a richer and brighter set of overtones. Hollow-body electrics sound darker, because the high overtones hold the least amount of energy, and so are preferentially absorbed compared to the lower ones. Notice that the change in tone and sustain is all subtractive. The solid-body adds nothing that is not already there in the vibrating string, but the hollow-body subtracts high overtones and overall sustain (that is, sustain of the fundamental). While that energy may be passed to the vibrating top of the guitar, this acoustic sound is usually so much quieter than the electrically amplified sound that it is irrelevant - through the pickup and amp we hear a darker tone.

Steel guitars have the most and brightest sustain of all, because the neck and body are one piece. There is no joint to absorb any vibartions (what we call the neck is really just a thick fret-board placed on top of the single-piece neck/body). Lap steels have the brightest and most sustain, because of the solid attachement of the nut and bridge to the body. Pedal steels have difficulty approaching the sustain and overtone richness of lap steels, because the moving parts of the rollers and fingers absorb some vibrational energy.

Therefore, a lap steel is the best test model for the effect of the body on the tone (which is simply the spectrum of sustained overtones). Softer woods absorb more vibrational energy, preferentially from the higher overtones, and give a darker tone with less overall sustain. Harder woods absorb less vibrational energy and give the brightest tone and most sustain. But as explained above, this is all subtractive. The resonance of the body can only take energy away, it cannot create more sustain. But of course it is true that if you can feel the string vibrations in the body, and even in the legs of the instrument, it is probably a good sounding guitar. That is because this indicates a lack of mushiness. The string vibrations are being transmitted solidly through the nut and bridge to the body and legs. That is a good sign that little energy is being absorbed by mushy body material or loose joints. But those body and leg vibrations do not actually contribute to the tone and sustain, except by subtracting higher overtones. The magnetic pickup cannot pickup the vibrations of the body.

One reader in a previous thread objected to this and said that if you knock on your instrument while deadening the strings, you can hear the knock transmitted through the pickup. I tried this, and yes you can. Then I realized that to hear the knock I had to turn up my volume pedal or amp to a very loud volume. If you pick a string at that volume setting, it is many time louder than the knock. At any playing volume the knock cannot be heard over the amplified string vibrations coming through the magnetic pickup and amp. Therefore, it is difficult for me to imagine that any body resonance can contribute to an electric guitar sound in any way other than the subtractive way described above. And therefore, I am very skeptical of all the theorizing about the resonance characteristics of body materials "enhancing" the tone and sustain of an electric guitar - make it darker with less sustain by subtracting vibrational energy, yes - make it brighter with more sustain than the natural strings by adding "resonance", I doubt it. I would imagine the maximum sustain and overtone richness would occur with a diamond body with nut and bridge carved into it. It would probably be so harsh, we would hate it.

Now having said all that, there are still some mysteries to me. I am not clear on the relationship of body thickness to tone. I would be very interested in taking various thickness bodies of the same material, and trying them on a simple lap steel, and measuring the sustain and overtones. I would imagine a too thin body would drain off too much vibrational energy. There is probably some optimum thickness where the least energy is absorbed by the body. I am not sure what the effect of a body thicker than the optimum would have. There may be no effect, so the extra thickness is just wasted material. Or maybe the extra mass prevents the whole guitar from moving with the strings and absorbing energy. This would be a very interesting experiment. Notice that any limberness or resonance of the body between the stationary endplates is more likely to subtract energy and sustain than to add to it. Therefore, I am very skeptical of testing various materials for "resonance" (ding or clunk) outside the system of endplates, nut, bridge, and magnetic pickup. Maybe clunk would indicate the mushiness type of absorption, but ding could indicate a too thin, limber top that is also not optimum for a solid-body instrument.

Finally, there seems to be some truth to the idea that screwing the neck too tightly to an Emmons push/pull dampens sustain. I don't know if this has to do with changing the mushiness factor, or the limberness of the body between the endplates (which has something to do with resonance), or what. This probably also has to do with the notion that solid wooden necks create a darker sound than hollow aluminum necls. But Bobbe Seymour once said that when he was working on the Sho-Bud Super Pro, the new sound of that guitar was not affected by what kind of neck they used. So there are some mysteries here that need to be examined experimentally.

A guy could get an old SHo~Bud...

Just funnin ya Ed.

I'm fascinated by the mechanics from the pix you sent.

I've long thought that a longer scale would yield higher dynamics.

I have nothing but the HIGHEST regard for "young" Tom Baker.

It's about time he got his hands on a helm somewhere.

My Best to you and him.

quote:

---

One reader in a previous thread objected to this and said that if you knock on your instrument while deadening the strings, you can hear the knock transmitted through the pickup. I tried this, and yes you can. Then I realized that to hear the knock I had to turn up my volume pedal or amp to a very loud volume. If you pick a string at that volume setting, it is many time louder than the knock. At any playing volume the knock cannot be heard over the amplified string vibrations coming through the magnetic pickup and amp. Therefore, it is difficult for me to imagine that any body resonance can contribute to an electric guitar sound in any way other than the subtractive way described above.

---

David ... I remember a similar discussion ... I'm not sure I am the guy you are referring to here ... but here's some "food for thought".

When dealing with "Standing Waves" ... in physics experiments ... the string is "fixed" at both ends.

In guitars (solid body, acoustic, etc) ... the ends are not really fixed ... since the body is vibrating.

Some may see this as insignificant ... but when dealing with wave mechanics ... nothing can be ignored

The body ... since it has a different mass, shape, composition, etc ... than the string ... vibrates in a different wave function.

This in-turn ... alters the strings wave function ... since its "fixed points" are moving.

The superposition of the standing wave functions of the strings and guitar parts (body, neck, nut, pickup, etc) ... is what gives the guitar its characteristic sound.

In other words ... the supposed "fixed points" are actually moving ... therefore the string does not behave as an "ideal string" would in a standing wave experiment.

Different body materials, masses, presence of resonating chambers, etc ... all will change the wave function of the body ... therefore altering the wavefuncion of the string.

------------------

Aiello's House of Gauss

My wife and I don't think alike. She donates money to the homeless and I donate money to the topless! ... R. Dangerfield

The source of the energy in the system is the vibration of the strings. It can only be absorbed or reflected. That which is reflected does not cause the vibratory decay as much as that which is absorbed.

If a body/mechanism has a resonance at a particular frequency it will require less energy to excite it than where there is no resonance, hence its acoustic impedance at the resonant frequencies will be greater, and drain (subtract) less energy from the string vibrations.

The highs do indeed fall off sooner than the mids/lows on the instruments that I have measured.

This brings me to a wise A$$ response to Eric...I do have a Sho-Bud Professional, it is one of the instruments for which I have measured the sustain; It is somewhat better than a Banjo.

I have no problem with the "traditionalists" that like to see keyheads, pretty woods, taxi stripes...I happen to prefer arch top jazz boxes like Stromberg, DiAngelico, etc. but that does not mean that the Les Paul's and Strat's don't have a function also...even the Steinberger.

( grasps chest and falls to floor...)..

I'd also like to be on the "send" list. The first pix were great.

I've been away on my Mission to Marrs for a week, but I seem to remember a conversion for old Sierras to this new system. That would be really great. There's a lot of them out there.

My banjo sounded really good when I put a pickup on it.

It sounded better when I put a station wagon on it.

EJL

[This message was edited by Eric West on 19 February 2005 at 09:55 PM.]

David, I think you effectively summarize the lumped-system view of an stringed instrument's interaction with the vibrating strings. Qualitatively, I think your intuition is spot on. But Rick's point about the 'fixed' endpoints moving is on target also. This is one of the areas where the vibrating string model studied in elementary physics and engineering ceases to be 'ideal'. For example, guitar players who want more sustain compensate by installing beefy (e.g., brass) nuts/bridges that have more inertia, and hence resistance (impedance) to vibration. {Note: As Ed implicitly conceptualizes, one can study lumped-mechanical systems using the same impedance concepts as electrical circuits.} In principle, this isn't the only way to control vibrations. In vibration control, an active vibration controller senses the vibrations and applies negative feedback control to reduce them. This would be reasonable on a space station or a long bridge, but I've never seen it done on a musical instrument (nor would I want to mess with a digital controller to do this.)

Another 'nonideality' in the vibrating string model is string rigidity, which works against free string vibration. Other things being equal, this tends to be less of a factor when the ratio of stringLength-to-stringThickness is large, obviously one of the motivating factors in Ed's beast. Unless there is an active source of energy (such as the vibration control I mentioned earlier), the only possible energy interactions are exchange (from one part of the instrument to another) and dissipation. String energy dissipation tends to be smaller when string nonidealities are smaller. Then one needs to consider energy exchange-to/dissipation-of components attached to the string. Again, using lumped modeling, one deals with the changer and nut separately, but this may or may not be justified if one really wants to accurately assess the effects of small design changes. Then there's the issue of exactly how the changer/nut are attached. Again, it's not always obvious what happens. Finally, energy is transferred to other parts, the body, legs, pull rods, linkages, pedals, levers, etc., and thereupon dissipated or exchanged in turn.

Another thought. Those of us who are 6-string guitarists probably remember the Travis Bean guitar, made out of aircraft aluminum. The wood on the body was mostly for looks and ergonomics. Now some people love these, and they sustain all day. But it's the overall frequency response that counts. To wit, I believe that a Strat (with a great deal of phase-cancellation when using pickups 1+2 or 2+3) sounds better to most ears. Now I know that guitar is different than steel, but pure, sustained sine waves are not necessarily the entire goal, at least for me.

Of course, the ultimate question "Why worry, be happy with what you have, they sound nice now, why mess with them?" always comes up. My answer: because we can! With that, however, the caveat is that there is a science to it, but also an art. When I was first studying system simulation (a very long time ago), a very wise man told me, "The purpose of simulation is not numbers, but insight." I have often found it hard to translate numbers into insight. People tend to like simple 'rules of thumb'. Great, if it works, but it isn't always simple. IMO.

ps, Ed, just saw your last response. Obviously, I agree, although I took a bit longer to get there. Also, I'm getting my Sierra U-14 going here, I'm very interested in any copedent suggestions you have, you mentioned a paper sometime back before I got it.

It reminds me of my favorite uncle, a very brilliant man, retired from Research and Developement at Techtronix and a steel player and builder since the '50's. He has in a way already "been there and done that", but mostly applied to non-pedals, which were his main interest, and their electronics. He never had any interest in patents and commercial uses for his designs...pity.

Sometimes all the scientific data and good theories don't add up the way you expect in real life...

Once he was very excited about building a steel cabinet from a special wood he had imported from Africa, it was so resonant it rang like a bell compared to normal wood like mahogany or maple.

He said it was the worst-sounding steel cabinet he'd ever made.... the resonant-wood theory was wrong on that one. I didn't get all the details, but maybe it was too resonant at one specific frequency and caused too many frequency cancellations, resulting in dead spots? Just my guess.

On the other hand, he built a pedal steel that even when pulling several strings at once with one pedal, had pedal-pressure so light you could push the pedal with one finger, and a new type of pickup with super-wide frequency response... and other great innovations. I kinda wish he'd gotten some of them to the commercial market. He's getting up there in years and has quit playing and working with steels now.

Matter of fact, he is sending me all his parts, pickups and experimental steels, I can't wait to get them.

Sounds to me like with guys like Ed and others working on the PSG, we can look forward to some great developments in the near future.

[This message was edited by Jim Phelps on 19 February 2005 at 10:30 PM.]

DM; Re 14 string setups... Julian Tharpe, Bill Stafford, and Al Vescovo have all provided such setups. Bill S and Al V are Forum members, so you can contact them directly to get info. Fred Shannon can probably supply info on Julian's setups. Re mine, it comes in two flavors = 25" scale E69 9X7+lk, and 29.730" scale C69 10X7+lk. The later was used as the basis for the CHORD LOCATION thread and mailings.

Re the "vibration" subject we now address: It would be nice to have the use of B&K, Endevco, HP, and other equipments for these "experiments", but that is not in the cards. Probably just as well, because then it would tend to get a bit "whiggy" for some. It will be muddy enough anyway. Mostly, we will use the output of the pickups as the signal source, and the strings as the stimulus for the FSA data.

Much of the Modal vibration and associated data will be from an FEA program.

When we get to the pickup magnetic analysis part, another FEA program will be used.

The last two programs will be the work of a long time friend (Physicist from France) who makes a living selling and consulting re such software packages. His name is Jean-Marc Gery, and his company, located in CA, is HY-MARC. Look him up on a search engine for previews of coming attractions.

Not sure if I should start the Emailings before the Dallas show or not...TBD.

The object of this activity is to show some methods that may be employed to analyze the sound of musical instruments, particularly those that are vibrating string based as is the PSG. As "good" sound is subjective, it will be up to the individual to associate the visual symbols (graphs, charts, and other data presentations) with the terms good, bad, and anything in between.

JP; I can dig your fathers era, having been there. My approach before computer modeling and the FSA/Endevco/B&K equipment was to use phono cartridges and oscilloscopes for body/mechanism vibration data, ..yeah, I am that old.

Steven Errede's Notes

Heres a very look at Modal Analysis

The Homepage for that study.

------------------

Aiello's House of Gauss

My wife and I don't think alike. She donates money to the homeless and I donate money to the topless! ... R. Dangerfield

[This message was edited by Rick Aiello on 20 February 2005 at 08:39 AM.]

If you do not already know, I believe that McGill University has done some work of a similar nature..GOOGLE it if interested.

The pickups will be profiled re Electrical Impedance Vs. Hz..

The excited strings (strummed and individual)
will be profiled as pickup Hz & magnitude output vs. time.

The whole structure will be profiled as Pickup output Hz and magnitude as a function of external structure excitation (poor man's anechoic chamber) via swept frequency, and other.

I do not have calibrated hammers etc.

Anything beyond the pickup electrical output (essentially unloaded) is ignored for this series.

I now understand that your focus will be the electrical output of the pickup. I know that some people use the acoustic "ring" test to determine the sustain of a steel guitar. This is done by raking the strings and listening to the overtones and the duration of each overtone. This could certainly be looked at in a spectrograph. But it got me thinking about the difference between the acoustic "ring" test and the pickup's output. That's where the transfer function comes in.

I use the SIM3 measurement system for transfer functions as well as SMAART LIVE for spectrograph.
http://www.meyersound.com/products/integration_tools/sim3/02.htm
http://www.siasoft.com/images/smaart/RTA%2BSpectro_lg.jpg

Also, I do play my '69 Sho Bud Professional everyday and it does sound good. Hmmmm, I wonder why it sounds so good...Doh, there I go again, better cut that out.

quote:

---

mental masturbation

---

Is there another kind ?

------------------

Aiello's House of Gauss

My wife and I don't think alike. She donates money to the homeless and I donate money to the topless! ... R. Dangerfield

DD: Already gone for, ..got a lot of data...just trying to figure out the better way to format and send it. After Dallas.

[This message was edited by ed packard on 22 February 2005 at 12:04 PM.]

Ed Packard, who happens to be an alright guy and my friend, has continued that trend with the development of his new guitar. I know how much study and research Mr. Packard performed in the development of that guitar and I highly encourage his further efforts. Ed put me on the list, I'll listen, even if I don't understand each and every item. BTW I've got a couple of guitars that sound super, but I don't believe I want to live in a box.

fred

------------------
"From Truth, Justice is Born"--Quanah Parker-1904

Had a few moments, so I just emailed the first FSA blurb to those that asked, and a few who did not. If this is not your "cup of tea", just ask and I will remove you from the list.

If you want on the list, email me and I will put you on.

Ed, if you could add me to your mail list, I would really appreciate it. I spent 13 years making quartz vibrate in a well behaved manner, and to this day, I have trouble not picking up things to see how they ring.

------------------
Bill

------------------

[This message was edited by Lyle Clary on 25 February 2005 at 10:52 AM.]

The first FEA send shows the modal vibration frequencies in a simple top plate for a PSG. The top plate is solved for several widths and thicknesses. This will give the base against which the results of adding skirts, end plates, ribs for holding the cross shafts, the neck block, the changer, the tuner block, the tensioned strings, etc..

That will be followed by the static loading deformation of the body and the attendant frequency change in the string frequencies.

[This message was edited by ed packard on 25 February 2005 at 05:03 PM.]

FEA 2 shows the effect upon the resonant frequencies of adding the skirts and endplates to a top plate solved for in FEA 1. This is moving toward the effect of body/mechanism components on the Tone and Sustain of the finished instrument.

The first part (FSA-1-1) covers the strings strummed at fret 36. The results are the resulting frequency spectrum at 0, 4, and 8 seconds (Sustain) for openers. These, when double smoothed at a cutoff of .333 (1/6th octave) provide some insight to the underlying resonances from the body and mechanism.

Individual string responses at 0, 4, 8 seconds follow, showing why we like the 7 tone diatonic and, 11 tone chromatic scale, as well as our preferred chord structures.

We started the FEA with just top plate for a PSG, and looked at it's first 10 modes of vibration for frequency (Hz).

Then we added skirts, end plates, and string terminator blocks, and looked at several nodes of vibration to determine Hz and magnitude.

A more detailed approach would be to get the Hz and magnitude of each of the individual pieces, then show how the same pieces would react as an assembled unit ...we did not go that far yet; we jumped from the top plate to the assembled unit.

That data was supplied to those on the email list. It shows the resulting damping and Hz shift from combining top plate, skirts,end plates et al. in graphic form.

How the top plate, skirts, etc. are combined (screws, screw tightness, screw location, other than screwed, etc.)will determine the resonances and magnitudes of the combined parts. Each attached part will dampen the vibrations in the other parts.

Then we look for the resonance's to show up in the output from the pickup for exciting the strings. These "underlying" resonsnces are then compared to the resonance's of the FEA model for coincidence.

So far the correlation of the FEA results for the simple PSG model is good enough to encourage putting more detail into the model to see if we can find the source of the more obscure resonances.

Static body assembly deflection in length, width, and thickness are also part of the modeling data; these can relate to "cabinet" distortions from activated changes. Body bending as a function of strings and their attendent tensions is considered.

So far we have not added the neck block to the assembly as I don't use one on the BEAST, but it is one of the things to be done.

Keep in mind that the Damping coefficient of Aluminum is less than that of wood(s).

The FSA-2 series looked at the effect of string tension on the harmonic content and harmonic content vs. time of the string(s)/pickup output = tone & sustain.

The next series is FSA-3. FSA-3 will look at two different pickups, at two different neck locations on the same instrument. These pickups are both two coil pickups, and are by two different manufacturers. Their fequency response will be profiled for each of the tap settings into both the >5 Megohm, and 500k (standard pot volume pedal)conditions.

The maximum frequency response settings for both pickups will then be compared in both neck positions ...at least that is the plan as seen from here.

It would be nice if each of these experiments were accompanied by an audio file so that what is seen could be heard ...maybe after this "fire for effect" run that will happen.