Some Principles of Violin Setup
by Joseph Curtin
The following article was featured in the November 1995 issue of the Journal of the Violin Society of America.
Saturday, November 11, 1995, 9:00 am
Albert Mell: We start the lectures today with a talk by Joseph Curtin, whose name is familiar if you have been reading the Strad lately. I first met Joseph - and he may not remember this - in Toronto, I think it was in 1981 on the occasion of the Viola Congress. At that time he was just at the beginning of his career as a maker. He'd been a student at the University of Toronto, had begun working with Otto Erdesz, and had made, I think, his first viola - with the indentation on the A-string side to facilitate playing in the upper positions for those of us violists who suffer from arthritis and other such conditions.
Then I lost track of Joseph. Later I found he'd gone to Cremona to continue his studies. The next thing I knew he had allied himself with Gregg Alf, and the firm of Curtin & Alf became known to all of us. In the years since, both have earned distinguished reputations as makers, for their individual instruments as well as for their collaborations on copies of old ones.
When first scheduled to speak here, Joseph was going to talk about something in relation to computers because he is one of those rare people who not only make violins but have an interest in science. He is, therefore, bridging the gap between scientists and violin makers. But then his lecture was changed to the one you see on our programs here, involving setup. But I found out secretly he's going to apply principles of acoustics and physics to violin setup, so he hasn't really changed his subject.
I would now like to introduce Joseph Curtin.
Joseph Curtin: The title of this lecture suggests a more rounded discussion of setup than I will actually present. My initial intention was to look at various elements of setup and how they work together. But I found myself sinking deeper and deeper into ruminations about the violin bridge. Late one night last week, I decided to concentrate on the bridge, more or less to the exclusion of everything else. I didn't realize Vahakn Nigogosian would be speaking on the same subject. I hope my thoughts will complement his own very useful and interesting observations.
A great deal of discussion about setup consists of conversations about measurements and workshop standards. The width of a violin bridge is 41 mm. The bass bar is placed one millimeter inside the bridge foot. The string spacing at the nut is 16.5 mm, and so on. This is all very important, of course. A thorough understanding of accepted standards, along with a sense of appropriate deviation, forms the basis for practical violin making. However, I would like today not to be preoccupied with measurements. Rather, I would like to attempt answers to questions that began to preoccupy me five or six years ago.
About once a year, Gregg Alf and I collaborate on an exact copy of an old violin, and five or six years ago we copied a Stradivari. When the copy was strung up we were not especially satisfied with the sound, which seemed rather dull.
Now it happened the client had wanted a new bridge on the original - I don't remember why - so we fitted one. And so it was that the original bridge, a rather beautiful one by a well-known British firm, was lying on my workbench one day as I adjusted the copy. On impulse, I tried this bridge on our instrument. It fit rather well, and what is more, the violin sounded quite brilliant, with a clarity of response I loved.
Now the bridge we made seemed well cut. The wood looked very good, a nicely aged Aubert deluxe blank. Why the difference in sound?
I dismounted the violin and started tapping the two bridges in various ways to see if there were any interesting differences. A physicist friend had recently asked whether we tuned our bridges. I said no, we simply used the best blanks we could find and tried to make them look right. I thought that if a bridge looked beautiful, it would work well. Actually, I thought this true for the whole instrument. But no matter how good one's craftsmanship or sense of style, I have become convinced that lack of understanding, in physical terms, of how violins work is what holds us back most as makers.
At any rate, I held each bridge near its feet, then tapped its edge with the handle of my bridge knife. The original bridge seemed to respond with a lower “click” than ours did. This surprised me: I'd heard stories of makers selecting bridges on the merit of the higher clicks they made when dropped on a hard surface.
I trimmed the waist of our bridge, calculating this might lower the tap tone to match the other. As it turned out, this happened rather quickly - hardly two-tenths of a millimeter of wood needed removing. But the effect on the instrument was startling. It seemed to bring out the brilliance and responsiveness which pleased me so much with the bridge from the original. I began tuning the bridges of other instruments around the shop, matching the tap tones to the “miracle” bridge. Sometimes it made as big a difference, sometimes almost none. It was time to try and figure out what was going on.
I soon discovered that a number of physicists, engineers, and musical acousticians had studied the violin bridge. I have relied on their thinking and research as well as my own speculations in developing the following ideas. A great deal of research still needs to be done. In the meanwhile, here are some working hypotheses. I hope they will not all be discredited by future research!
Static and Dynamic Functions
The bridge on a stringed instrument can be looked at as having two kinds of functions, which I will call static and dynamic. By static I mean the things the bridge is expected to do even when the instrument is not being played: holding the strings up, for example, and refraining from warping or falling over.
By dynamic, I refer to what the bridge is expected to do while the instrument is being played: transmitting vibration from the strings to the body of the instrument, at the same time acting as a stable support for the vibrating strings. These two functions, I will later try to show, create a rather paradoxical set of constraints for the violin maker.
The static functions of the bridge are easily described and understood. The strings must be held a certain distance apart along a prescribed arc, and they must be maintained at a comfortable height above the fingerboard.
The curve of the top of the bridge has been widely standardized. So has the bridge's slight backward tilt, which helps to avoid warping. To this end, it is often noted that the bridge should bisect the angle of the strings, in order that the downward force of the strings goes straight through the center of the bridge. In practice, this gives a larger-than-appropriate slant; the force of the strings, though indeed through the center of the bridge, would have an unfortunate tendency to make the feet of the bridge slip forward.
One way in which bridges vary significantly is in their height. In practice, the height of a bridge tends to be adjusted to wherever the neck happens to be. But in setting up an instrument from scratch, one has the chance to establish an ideal bridge height, then set the neck accordingly. But on what principles is the ideal bridge height established? What, exactly, is at stake? If we limit ourselves to the bridge's static functions, what is at stake is the amount of room the bow has to clear the treble C bout, and perhaps more important, the amount of string tension that will be transferred downward onto the belly of the instrument.
It is an interesting and sometimes surprising exercise to draw, full-scale or larger, the geometry of a violin you are working on, trying to get exactly right the angles created by the neck, the strings, the arching, and the bridge. Figure 1 shows a drawing prepared in our shop some years ago. I have drawn in a triangle formed by the playing edge of the upper saddle, the top of the bridge, and leading edge of the lower saddle. The question of downward force on the belly can be understood by studying this geometry (Figure 2).
If the strings traveled in a straight line across the bridge, there would certainly be no downward force exerted (except for that created by bow pressure or string vibration). If, on the other hand, one exaggerated the string angle with a very high bridge, most of the tension of the strings would be transferred downward on to the top (requiring it to be very substantial indeed).
What, then, is a normal range of heights for this triangle, and how much difference does it make in terms of the downward force? If bridge height ranges from about 31 mm to 35 mm, and archings from 12 mm to 20 mm, we have a spread of about 12 mm between a very low bridge on a very low belly and a high bridge on a high belly. In practice, this range is not found; lower archings are usually paired with higher bridges.
To determine the actual forces involved, we need to solve some calculations relating to both the string tension and the geometry of the setup (Figure 3). If we assume that the bridge is at a right angle to the base of the triangle (a close enough approximation for our purposes), then the force downward on the belly consists of T times H over L, plus T times H over L1, where T is the total string tension, H is the height of the triangle, L is the string length, and L1 is the distance from the bridge to the leading edge of the lower saddle. Expressed mathematically:
The combined tension of a typical set of strings on a full-sized violin is, according to Norman Pickering (whose book The Bowed String I highly recommend), about 51 pounds.
For the fairly typical violin illustrated in Figure 3, the string length L is 327 mm. A 15.5 mm arch and a 33.5 mm bridge give a value for H of 42.6 mm (which is not simply the combined heights of the bridge and the arching, as the base of the triangle rides somewhere above the plane of the top of the ribs). The distance behind the bridge, L1, is 160 mm.
Solving this equation, we find a downward force on the belly of 19.22 pounds, which is almost 40% of the total string tension. Recalculating for different values of H. each millimeter change in bridge height changes the downward force by about .47 pounds.
As a practical way of measuring the height of the triangle on finished instruments, Gregg Alf designed the jig pictured in Figure 4, based on some ideas given to us by Carl Becker.
How much does a change in the height of the triangle affect the sound of the instrument? By way of an anticlimax, I propose that it does not make much difference. A simple experiment, suggested by University of Paris physicist Xavier Boutillon, illustrates this. Dampen the upper three strings of a violin by wedging a soft cloth between strings and fingerboard. Play the G string and note its power and quality. Then, one by one, loosen each of the two middle strings, playing the G string between each change in string tension. (Leaving the E string tuned up helps keep contact between top and sound post, which tends to be compromised when the top three strings are released.) I think you will find very little change in the sound, considering the radical change in the downward pressure.
Bridge Height
At any rate, from the point of view of static functions, bridge height seems relatively straightforward. We can assume a normal bridge height produces a downward pressure that has proven sustainable over long periods of time by normally arched and graduated violin tops. If the arching is higher, the height of the bridge can be reduced. It can be raised in the case of an instrument with low arching. All this is common practice.
And yet, we are often requested to put a higher bridge on an instrument to “increase the tension” in the hope of greater power, projection, brightness - whatever is felt to be missing. How much of this is realistic? The question can be addressed at a number of levels.
The first concerns the oft-encountered misconception that raising the height of the bridge increases the tension of the strings. Using heavier strings will accomplish this, as greater tension is required to bring them up to pitch. Tuning the violin to a higher A will also increase the overall tension. I have heard it said that increasing the string length behind the bridge will increase the overall tension. Actually, the strings could extend for several miles behind the bridge and it would still take the same tension to bring the playing portion up to pitch. (In such a case, however, the pegs might have to take up several hundred yards of slack while tuning up!)
Raising the height of the bridge won't increase overall tension, but more of the existing tension will be transferred down on to the belly. If our experiment suggests that this alone doesn't much affect the sound, are there any valid tonal adjustments to be made by changing the bridge height?
I believe there are. But I think we have to look to what I have called the dynamic functions of the bridge to understand the basis of such adjustments.
Dynamic Functions
The bridge's dynamic functions can be conveniently divided into two categories, the first based on cases in which the bridge acts as an essentially rigid support for the strings, the second in which the bridge is seen as a flexible coupling.
It is interesting to note that if the bridge were truly rigid. it would not matter what it were made of. As long as the weight and dimensions remained the same, the choice of bridge blank and the shape of the cutouts would make no difference to the sound.
This, by the way, I believe is very largely true for the violin sound post. Research has shown that the sound post had no resonances falling within the important areas of the violin's range. This means that the post is acting as a rigid column and neither flexing nor compressing/expanding longitudinally. This implies it does not much matter whether the post is made from old wood or new wood, “good” wood or “bad” wood, providing,of course, that length, fit, and mass are optimal. Rene Morel, speaking from long experience, stated as much in a lecture he gave at the 1994 Guarneri exhibit at the Metropolitan Museum in New York. I myself have found it to be true, though on the basis of far less experience.
But back to the bridge, which at low frequencies does seem to act as a rigid unit. At higher frequencies its resonant properties become important.
When the strings are set into motion by the bow, the force they exert on the bridge tends, mainly, to rock it from side to side. If the bridge is acting as a rigid unit, then its ability to move in response to the urgings of the strings depends on the mobility of the top directly under the bridge feet.
This mobility, in turn, depends on such factors as the bass bar, sound post, the mass of the top, and ultimately of the resonant properties of the violin body. In fact, the mobility of the top varies greatly with frequency. If there is a resonance at a particular frequency, and the bridge is favorably placed to excite that resonance, it will move easily when excited at that frequency by the strings.
It might seem that what we want is a very mobile bridge, allowing a great deal of energy to be transferred from the strings, through the bridge, to the body of the instrument, from there to be radiated as sound.
Unfortunately, there is a complication. Imagine a rope tied to a wall. You hold the free end of the rope and swing it back and forth. It is relatively easy to produce a standing wave. Now imagine a rope tied to a rather flexible pole: a fishing rod, for example, planted in the ground. The rope will no longer have a stable support and it will be more difficult - and in certain cases impossible to produce a standing wave.
The same is true for a violin string, which depends on the upper saddle and the bridge to remain relatively immobile. As the bridge begins to move, string vibration becomes harder to control. An extreme case of this is the wolf note.
A wolf note is caused by a strong resonance in the body of the instrument. This allows the bridge to move freely when excited at the frequency of that resonance. As the bridge motion increases, it ceases to be a stable support for the string. At this point the regular musical vibrations of the string break down. The bridge then stabilizes itself. The cycle begins again. Thus we get that uncontrollable pulsing so characteristic of a wolf note.
So we seem to have a paradoxical set of constraints. We want the bridge to move as much as possible in order to set the body in motion. At the same time, we want the bridge to be as stable a support for the strings as possible, ensuring rapid and predictable response. I believe that much of the art of setup - and indeed the design and construction of the instrument itself - is a kind of balancing act played out between these two limits.
What has this got to do with bridge height? I hope Figures 5 and 6 will clarify this. Let's again, for the moment, consider the bridge a rigid body. It can then be treated as a lever. The longer the arm of the lever, the more mobile the end of this arm becomes. The complication is that, in the case of the bridge, the fulcrum shifts with frequency. For example, at low frequencies, experiments have shown that the sound post more or less immobilizes the treble bridge foot and becomes a kind of fulcrum. The bridge rocks around this fulcrum, moving the bass bar side up and down. At higher frequencies, the situation becomes considerably more complicated. Depending on which combination of resonances is being activated, the bass bar side may remain immobile while the treble foot drives the sound post (and thus the back) up and down. In this case, the bass foot of the bridge is acting as a fulcrum.
But as long as the fulcrum is somewhere near the bottom of the bridge, the higher the bridge compared to its width, the greater its immobility - that is, the greater the deflection of the top of the bridge will be for a given string force. A cello bridge is much higher in relation to its width than is a violin bridge. This helps explain the cello's greater tendency to have wolf notes.
Ideal Bridge Heights
How do we determine the ideal bridge height for a given instrument? This is rather like having to choose a single gear in which to drive a car for an entire journey. If the car is heavy or acceleration is especially important, one might opt for first gear. Top speed would be limited by this choice. Given the same motor in a lighter car, one might choose a higher gear and get the same amount of acceleration and a higher top end.
If you have a violin that is relatively heavily built - dense wood or thick graduations - a high bridge may work well. It is for this reason, I believe, that many new instruments are best set up this way.
A more lightly built instrument may respond best to a lower bridge. This provides increased resistance to the strings, and so good response, at the same time allowing greater dynamic range than a higher bridge might allow.
If the above suggests lighter instruments have an advantage over heavier ones, I believe this to be true. But there are so many factors at work it is difficult to draw simple conclusions. For example, anyone who heard the extraordinary sound of the heavily built Cannone (ex-Paganini) Guarneri del Gesu played in recital at the 1994 Guarneri exhibition might well be tempted to explore thicker graduations.
Longitudinal Vibrations
If wolf notes are the result of overly mobile bridges, they can be relieved by some of these strategies: lowering the bridge; making it wider (this of course is not always practical, given the internal placement of bass bar and sound post); making it heavier; using lighter strings. Each of these strategies has differing effects on the sound. Wolf note eliminators can be employed, but a discussion of these is outside the scope of this talk.
I would now like to look at bridge height from a different point of view. When the string vibrates from side to side, it creates a force tending to pull and release its end supports. This is easy to visualize if we imagine holding a cord between our outstretched arms. If someone pulls the rope at the middle of its length, our arms will be pulled together. As the rope is released, our arms can move apart again and take up the slack.
During a single vibratory cycle, the bowed string will be pulled off center twice - once to each side - and this produces a shortening and lengthening of the string an octave above the fundamental frequency. I will refer to this as the longitudinal vibration of the strings, in contrast to their more familiar side-to-side, or lateral, vibrations.
How important are these longitudinal vibrations? This depends on at least two factors. The first concerns the string angle, which we looked at earlier. If the strings passed in a straight line across the bridge, the longitudinal vibration would tend to pull the top of the bridge back and forth. But this motion would have almost no affect on the belly of the instrument, the leverage being so poor.
However, as the string angle increases with either a higher bridge or higher arching, more and more of the longitudinal vibration is transferred downward to the belly directly through the bridge, just as the static force of string tension was.
The second factor is how loudly the instrument is being played. According to Arthur Benade, the force of these longitudinal vibrations increases by the square of the lateral ones.* Therefore, twice the bow speed will double the amplitude of lateral vibrations, but quadruple that of the longitudinal vibrations. Benade concludes that the effect of the longitudinal vibration is negligible at pianissimo, but becomes more important at higher dynamics. This may be part of the reason for the change in tone color as the instrument is played more loudly. (A recording of a violin played very softly will not, when the volume is turned up, sound like a violin played loudly.)
If this is so, then a higher arching/bridge height would give a greater shift in tone colors across the instrument's dynamic range. The increased brightness claimed for sharper string angles may be due to the increased transmission of longitudinal vibrations. This needs exploration at an experimental level.
*Arthur H. Benade, Fundamentals of Musical Acoustics (New York: Dover Publications), 530.
Flexible Coupling
I would now like to look at the bridge in its role as a flexible coupling between strings and violin. Allow me to approach this through a series of digressions.
Psychoacousticians have mapped out the sensitivity of the human ear and found it to be far from even. Although we can, especially when young, hear frequencies from below 50 hz up to and beyond 15k hz, the area of maximum sensitivity tends to be around 3000 hz. This is a higher pitch than one might expect (almost three octaves above A 440) if one considers that the long-term average spectrum for both full orchestra and the speaking voice shows peak amplitudes at about 450 hz (just above concert A). Presumably this is satisfying to our ears, otherwise instrumental music would have evolved differently.
But let's for a moment put ourselves in the position of a soloist performing with an orchestra - an operatic tenor, for example - who must often struggle simply to be heard above the orchestra. Because the spectrum of the normal singing voice, like the speaking voice, falls within that of the orchestra, the singer is in danger of being masked by the orchestra.
Opera singers, in contrast to choral singers (whose role is to blend in), have developed an interesting and effective strategy for holding their own above the orchestra. They learn to use their vocal apparatus to produce what has become known as the “singer's formant.” This formant is effectively a means of producing overtones in the 3000 hz area. The formant has the double advantage of being well above the orchestral spectrum's center, and of lying in the ear's most sensitive range.
How does the violin manage to be effective as a solo instrument? Its body is not especially effective at amplifying sounds in the 3000 hz range. This is where the bending of the bridge becomes important. In a typical bridge the relationship between the mass of the top portion of the bridge and the stiffness of the waist and legs creates a resonance at about 3000 hz. When string partials in this range excite the resonance, the upper part of the bridge begins to rock back and forth, helping to transfer string vibrations in this range to the body of the instrument.
This resonance is the lowest, and I believe the most important, of many resonances found in the bridge. What factors affect it? Its frequency is determined by the mass of the top portion of the bridge in relation to the stiffness of the waist area. Clearly, the waist must flex if the top portion is to move independently of the bottom. The stiffer the waist (including any surrounding areas that flex when this resonance is excited), the higher the frequency of the resonance. And the lighter the mass of the top portion, the higher the resonance. So to raise the pitch of the resonance. we could either remove wood from somewhere in the top portion, or else shave wood away from the waist area.
Now, the overall mass of the top portion will affect the resonance's ability to move the top. (Imagine the effect of a heavy man jumping from foot to foot on a wooden floor, compared to that of a small boy.) It will also affect the strings' ability to move the bridge. Establishing the optimal bridge mass for a given instrument is obviously a matter of trial and error and experience. I know of no experimental work yet done in this area.
In Figure 7 I have tried to show a relationship between the height of the upper portion of the bridge and the effective leverage of each string in exciting the resonance. It can be seen that the higher the top section of the bridge, the more equal the leverage between inner and outer strings. If the top section of the bridge is low, there is a significant difference in leverage.
This difference may be part of why the outer strings tend to be more brilliant and soloistic than the inner. (Another reason is that the outer strings can be bowed at a steeper angle, thus more directly driving the top in an up-and-down motion.) I suspect the practice of keeping the feet of the bridge as low as possible, thus maximizing the height of the upper portion of the bridge, helps maintain a good balance between inner and outer strings.
It must be remembered that the effectiveness of the bridge resonance in amplifying sound in the 3000 hz range also depends on the violin body's willingness to cooperate. If the peak of a bridge resonance overlaps with a peak of a body resonance, good coupling, and thus good amplification, will occur. But resonances in this area of a violin's spectrum vary greatly from instrument to instrument. They cannot be controlled by the maker the way, let's say, the first few free-plate resonances can. And so different violins may require differently tuned bridges for optimal performance.
Damping
The last point I will make concerns the damping of the bridge. A well-damped resonance will tend to have low amplitude and a broad peak. A relatively undamped resonance will tend toward higher amplitude and a narrower peak. This means a well-damped resonance will be more effective at amplifying a broader range of frequencies than an undamped one, but the amount of amplification will be less. Which is better in the case of a violin bridge? I don't know yet, and it is quite possible different amounts of damping are appropriate for different violins. As the damping depends mainly on the wood used and how it is treated, one begins to understand why the sorts of things Mr. Nigogosian discussed - the quality and preparation of the bridge blank, etc. - are so important.
I hope that further experimentation with bridges will lead to practical guidelines for best matching bridge to instrument. In the meantime, tapping bridges, comparing them with each other, keeping track of weight, height, and proportion, finding what works best for any given instrument have, for me at least, been an excellent way of developing a sense for this crucial element of setup.
That about concludes the formal portion of this talk. I'll now open the floor to questions.
Bill Dolittle: Do you thin the outer edges of the bridge in the area of the kidneys to tune the bridge?
Mr. Curtin: We start with a standard cut which has something of a lens-shaped curve, and so it does get thinner toward the edge. If the pitch of the bridge seems higher than you want it. thinning the area you suggest would certainly work.
Ralph Rabin: Could you talk a little about the decoupling effect of the two cuts above the feet of the bridge?
Mr. Curtin: The bridge has a number of resonances and those cuts may affect one which involves the bridge as a whole bouncing up and down, I believe at around 6000 hz. But I haven't studied this yet, and so I can't answer the question.
Tom Croen: In regard to Ralph's question, if you thin down the width of the ankles, you lower the pitch of the bridge radically.
Mr. Curtin: If it lowers the stiffness of the bending portion of a resonance, it would certainly lower the pitch of that resonance. But I haven't yet studied the resonances affected by the ankles, or how they affect the sound of the instrument.
Question: Doesn't the bridge vibrate back and forth, in the direction of the fingerboard and tailpiece, and doesn't the bridge itself radiate sound?
Mr. Curtin: An interesting question. The bridge would have to move in the way you suggest in order to radiate sound, and it would have to be at fairly high frequencies. Does anyone have an answer to this?
Norman Pickering: It does move and it does radiate at very high frequencies.
Mr. Curtin: In what range?
Mr. Pickering: In the useful range of 3000 to 5000 cycles
Mr. Curtin: OK. Something I didn't know. This vibration of the bridge, if it radiates directly, may be important. Otherwise, I don't see it as being effectively coupled to the instrument, because the instrument is, longitudinally, very stiff, and because the effective leverage is poor.
Mr. Pickering: The bridge motion is far more complicated than you have described, and I'm sure you know you have simplifled. For example, the bridge waves in peculiar ways - like putty sometimes - and it does actually make sounds independently of the belly of the instrument.
Joseph Regh: Somehow the energy from the strings has to be transmitted through the bridge to the body of the instrument. You seem to ignore the lower half of the bridge and the springiness of the legs.
Mr. Curtin: As Norman said, I've very much simplified things. Oliver Rodgers has done some wonderful work in computer modeling the bridge, and has published the results in the CAS Journal. He shows graphic representations of the different ways the bridge flexes and bends.
Mr. Regh: Do you in fact tune the sides?
Mr. Curtin: No. We adopted a more or less standard cut which comes out about right. We then do the adjusting around the waist. But in principle, the horizontal sections are important. In my own work, I try to do one little thing at a time, many times over, and so get a sense of which changes make a difference. Casual inspection suggests the waist area is the most sensitive, so I have concentrated there.
Mr. Nigogosian: The two kidneys affect the mobility very strongly.
Question: What is the significance of the string biting the tip of the bridge? Some people say the string should bite the bridge strongly, some that it should be lubricated.
Mr. Curtin: I would tend to lubricate it with a little graphite, so it's easier for the player to straighten the bridge. I can't imagine that affecting the sound.
Question: What about the fit of bridge foot to belly?
Mr. Curtin: It doesn't look good with gaps! In principle, the fit might affect the coupling, though given the softness of varnish and top wood, I'm not sure a very slight misfit affects the sound much. Some people like to leave the feet slightly hollow. This creates, to my mind, an increased tendency of the feet to dig into the varnish and belly.
Mr. Nigogosian: If the fit is a little hollow, and you put a little moisture on it, it will fit better. The outer edges will become compressed by the pressure of the strings, and so become stronger.
Gregg Alf: An interesting part of your talk concerned the effect of bridge height on the effective leverage of the lower and higher strings. As we all know, the bridge is lower on the treble side because of the differing string heights, and also because of the rotation of the neck. Can you comment on how this affects things?
Mr. Curtin: If increasing the bridge height increases brightness, lowering the treble side should decrease it, and this may create a better balance between outer strings.
Question: I don't think you can really make a standard pattern for a bridge. Different wood requires a different cut.
Mr. Curtin: We use Aubert blanks and more or less cut them to identical dimensions. I must say, they all come remarkably close in the kind of tuning I've been discussing, so very little tuning is necessary. This suggests that, by and large, workshop specs over the years have developed designs that work, and you don't have to start with a tabla rasa with each bridge. The arena for this kind of adjustment can be very small and still be effective. Of course, style varies from maker to maker. I've seen bridges by Morel that are relatively high and thinly cut. This may have tonal advantages in some cases. It also has structural implications.
Mr. Nigogosian: He also spreads his strings a little more, which affects the sound.
Dennis Topper: In reference to the string biting into the bridge: with basses, they often use a little metal ferrule over the top of the bridge, and this makes a big difference in sound.
Mr. Curtin: I assume this is due to a change in damping more than anything else, just as, at the other end, the open strings sound brighter because the ebony of the upper saddle dampens the strings less than the fingers do.
Mr. Pickering: You are very much exaggerating the longitudinal motion of the bridge. It is microscopic. Furthermore, the strings will not slide on a wooden bridge; they will dig in. If you look under a microscope, you can see the ridges from the windings. The top nut is a different story; the strings have to slide very freely. But at the bridge they are locked infirmly.
Mr. Curtin: That's what I would assume.
Mr. Pickering: When you tune the strings, the whole bridge moves over.
Mr. Curtin: Are we correct in assuming if we put a piece of soft material under the string, it will change the damping?
Mr. Pickering: Sure. It won't transmit higher frequencies to the bridge. It will act as a little filter. If you put a piece of rubber there, you'll hear quite a difference.
John Wodowski: This is a neck-set question. On the East Coast a lot of people rotate the neck toward the treble side half a millimeter or so, and on the West Coast, they rotate it the other way. I was told this makes the E side much lower than the G, the first way, and more equal, the second.
Mr. Curtin: I think this has to do with the relative differences in the rotation of the earth on the East and West Coasts. Seriously, at our shop we tilt it to make the treble side lower on the violin. This makes the playing angle lower. If you start hitting the treble C bout on a wide viola, for example, you might want to even things out. Gregg, you do that on violas, don't you?
Mr. Alf: On cellos as well.
Mr. Nigogosian: I am against it. On viola and cello, the C tends to be soft and the A metallic.
Joe Martin: Isn't it generally accepted that the top strings have a little more tension, and so are lowered to equalize the tension?
Mr. Curtin: I think that's probably right. The rotation helps the playing angle and the sound a little. I think this is what Nigo is saying. Thank you for your questions and attention.