US2261346A - Electric carillon - Google Patents

Description

Nov. 4, 1941. G, w DEMUTH 2,261,346

ELECTRIC cARILLoN Filed May 31, 1940' Stwentor Patented Nov. 4, e;

met

ELECTRIC CARILLON Galan W. Demuth, Moorestown, N. J., assigner to Radio Corporation of America, a corporation of Delaware Application May 31, 1940, serial No. 338,088

(ci. :i4- 405) 3 Claims.

This invention relates to an electric carillon and particularly to the construction of the vibrating bars thereof. Electric carillons have been constructed in which a vibrating member or bar is struck by an appropriate mechanism and the vibrations thereof are picked up by an appropriate electromagnetic or electrostatic means, amplified and reproduced through loudspeakers. The greatest difficulty in the construction of such devices lies in the pleasing production of bell tones.

It is well known that bells are made of many different types and have correspondingly dinerent tone qualities. One of the most important qualities in bells used in a carillon is that they shall have true pitch, or at least an apparently true pitch. In the construction of the conventional type of carillon, this is extremely dimcult due to the fact that the size of the bell determines its fundamental, which can be tuned with fair accuracy by turning oil appropriate amounts at the proper places on the bell. One overtone may also be controlled with reasonable accuracy, but any attempt to control more than one overtone is so likely to throw the fundamental oil of pitch that it ls rarely attempted.

In the case of these large bells, the investment in the cast bell is so great as to necessarily limit the amount of tuning which may be accomplished without risking the possibility of recasting the bell. 'I'he usual form of bell, however, although containing some overtones which are slightly dissonant, does not have a. large number of loud dissonant overtones and the problem is correspondingly simplified.

In the case of electric carillons, a relatively small vibrating member is used and it is accordingly practical to tune theovertones more closely as the loss in material in case the vibrating member is tuned too sharp and has to be scrapped is considerably less than the labor involved in the tuning. 'I'he disadvantage is partly offset by the fact that the vibrating bars ordinarily used in electric carillons have a large number of dissonant and fairly prominent overtones.

In the method and apparatus of the present invention, I have provided a novel way of getting two overtones of a vibrating bar accurately in tune with each other, suppressing the dissonant an improved type of vibrating bar for electric musical instruments.

Another object of the invention is to provide a vibrating bar in which the fundamental or apparent frequency is produced as a resultant of two accurately tuned harmonics thereof without the fundamental being present as a mode of vibration of the reed.

Another object of the invention is to provide a vibrating bar in which the fundamental or apparent frequency is produced as a resultant of two accurately tuned harmonics thereof.

Other and incidental objects of the present invention will be apparent to those skilled in the art from consideration of the following specification in connection with the accompanying drawing, in which Figure 1 is an end view of a carillon mechanism showing the striking mechanism and mounting of one bar with its associated pickup and indicating the presence of additional bars and Figure 2 is a front view of one of the carillon bars showing the construction thereof.

Referring first to Fig. l, which shows a striking mechanism somewhat analogous to that described and claimed in my application Serial No. 248,542, filed December 30, i938 (RCV docket 6515), appropriate keys such as indicated at I0 are provided. 'I'hese keys, when struck, strike an appropriate stop such as a felt strip I I. When the key strikes the strip I I, the member I 2, which is rigidly connected to the key I0, is so located that the hammer I4 is about an eighth of an inch from the vibratable bar 2li. The hammer H is carried by the spring member I3 and swings forward by its momentum, striking the bar 20 and being immediately withdrawn therefrom both by the reaction from .the blow and by the action of the spring I3, thereby preventing any damping of the vibration of the bar 2U by continued contact of the hammer. The hammer Il is preferably made of wood, but may have a small metal insert for striking the bar if a sharp impact is desired. A massive support member I5 is provided which should be of sufficient weight so that it will not vibrate appreciably from thereaction of the bars thereagainst. This prevents damping of the vibration of the bars by transmission of the vibration to other members of the apparatus.

overtones and producing the fundamental as a Eachbali Securely fastened, Preferably by hard resultant or subjective tone from the accurately tuned overtones.

One object of the invention is to provide an improved form of musical instrument.

soldering, into a metallic block I6. These blocks may be made of steel, for example. and are about one-quarter of an inch wide, that is, in the direction of the length of the bar, and are of sub- Another object of the invention is to provide stantially the shape shown, being of sumeient thickness to permit the soldering of the bar into a slot in the block. The blocks i are secured to the massive member Il by appropriate screws l1. In proper cooperative relation to each of the bars 2l, there is mounted a magnetic pickup including a winding 22 and a magnetized core 13. 'I'he windings of a whole series of these pickups may be connected together and connected to the input of an appropriate amplifier adapted to operate an appropriate number of loudspeakers. Other appropriate types of pickups, such for example, as an electrostatic pickup may be used.

The bars 2l are preferably made of piano wire having a diameter of the order of .045" and are tuned as hereinafter described.

It is an inherent characteristic of a vibrating bar of uniform cross-section, clamped at one end and free over the remainder of its length, that its vibrating frequencies are as follows from the fundamental to the 11th partial .Y

It willbe noted from the foregoing table that the fifth and sixth partials indicated as fs and la above are almost related by the ratio of 2:3. Notes having frequencies in the proportion of 2 to 3 are quite usual in musical tones and. if sounded together, give the simulation of also producing a fundamental having the proportional frequency of unity. These two overtones of the bar come closer to having such a ratio of their frequencies than any of the others and I have found it possible to so modify the bar as to bring these two partials accurately in tune in the ratio of 2:3. Further, these two partials have an intensity near the maximum of that of all the partials and, therefore, have a correspondingly great part in producing the apparent tone of the bar.

I find that by grinding a notch such as indicated at 2| in Figs. 1 and 2 in the bar, I am able to tune the said fifth and sixth partials accurately to the ratio of 2:3. It seems that the grinding of the notch has the effect of spreading the series of partials given above so as to accom- `plish the accurate relative tuning of these two partials. After the notch has been ground, further tum'ng is accomplished by grinding off the lower end of each bar until the proper pitch is secured.

In the following table, the dimensions of a series of reeds over a range of two octaves is given for a diameter of .045" and with the notches ground to a radius of three-eighths of an inch and to such a depth that the remaining thickness of the bar at the bottom of the notch is .0250" with a tolerance of .0002" either way. The length L given in the table is the length from the bottom of the supporting block to the end of the reed to which the wires are cut. These lengths will leave, each note very slightly flat sd that they may be brought accurately into tune by grinding, as described above. The frequency indicated as 2f may be tuned with the assistance of an oscillograph, and when this frequency is the notch Aa little deeper and retuning the bar, but ordinarily it is less expensive to discard the bar and substitute a new one of proper dimenq sions.

Tuned frequencies Non 2/ a/ p L 196.00 mo ssao ses 13M 901.65 41s.: m9 51s 13M moo 440.0 oeao so: 121,4

mos 4cm 099.2 m 1214 m94 49:4.9 14o.s u1 129s.

%1.63 523.3 784.0 M7 111Mo 399.63 ma 988.9 soo 101%:

.1.49.21 was 1041.1 4s4 10% 3&.90 740.0 1110.0 408 91%0 moo 184.0 1116.0 .441s 9114.

411:40 me mso .45a 9% 44o seo 1390 4:11 916 406.16 932.3 1a9as 4:1 81A 587.3 1174.6 1762.0 4M 71ia was 1318.5 1911.3 :191 114 MAG 1396.0 2195.4 .391 7%:

730.90 14&).0 2211.0 .N1 71de It should be noticed that the resultant pitch is not tuned directly and does not appear as a partial of the bar, but it is produced as a difference frequency between the tuned frequencies 2f and Il. If an oscillograph is connected to the pickup magnet, it will show'the frequencies tabulated as 2f and as 3f, and it will also show that the frequency f is not physically present in the vibration of the bar, although the frequency f is the one which is apparent to the ear.

It will be apparent to those skilled in the art that the invention is not limited to the use of the fth or sixth normal partials of a vibrating bar tuned to two and three times the frequency of the resultant note which is to be produced, but that I may use other partials of the bar tuned to produce other true harmonics of the apparent fundamental frequencies desired. For

example, the second and third partials might be tuned to the ratio of one to three; or the third and fourth partials might be tuned in the ratio of one to two; or the seventh and eighth partials might be tuned in the ratio of three to four and so on. The selection of these other partials would, of course, require a corresponding change in the position of electrical pickup so as to bring out the particular frequencies which are desired and suppress others and in some instances would also require the design of special amplifiers to suppress the undesired frequencies. In the preferred form of the invention described above, the fundamental of the bar, which appears as a relatively strong vibration, lies below the range eectively amplified by commercial amplifiers and is not readily picked up at the position of the pickup described. In general the selection of low partials of the bar renders the elimination of the fundamental of the bar difficult and also renders the elimination of the dissonant higher harmonics diLcult due to their high amplitude and ready amplification, Likewise, the choice of partials which are too high in the series produced by the bar renders dimcult the elimination of dissonant lower partials which fall well within the range of audio amplifiers.

The pickup should be located at a position on the bar which is at a distance from the block II of twice the dimension D, as this gives the best pickup of thev particular frequenciesdesired. The position of the hammer Il is notsoiimportant, .although it. operates. the best when'flocated at approximately the relative position indicated in the drawing. `If -thehammer is moved much',

closer to the point ofs'upport of the bar,.itrmakes the sound produced more ,tinkly, while if it is moved much farther dowr'ionthe bar, it produces point of support the notch mustbe'made morev shallow, Conversely, if the `notch is m'oved far'- ther from the point of support it must be made deeper. Either of these' changes adds .to the difficulty of production. The bar has a tendency to be slightly. annealed where it is hard soldered into the supporting block and if the notch falls within the portion `of the bar` where the vhardness is affected, the dimensionsof the notch will not be uniform but must be adjusted according to the hardness of the part thereby requiring a separate tuning operation for the notch. Conversely if the notch is moved farther down the bar and is made deeper, the quantity of material remaining in the bar at the notch becomes less and the tolerances in the dimensions of the notch become correspondingly more critical.

Byv a comparison of the table with the length of natural frequencies of a vibrating bar given above, it will also be noted that the fundamental frequency of each unnotched bar is 1%66 of the fifth partial, which is given as 2f in the table, and the fundamental of the bar is, therefore, not more than 1%83 of the frequency ffwhich is the apparent or resultant fundamental. For example, for the note g having a resultant fundamental of 196jcycles per second, the fundamental of the bar itself is 6.95'cycles per second, which is well below the audiblerange, Even for the corresponding note tw'o octaves higher, or 783.99 cycles per second, the fundamental of the unnotched bar is only 27.7 cycles per second, which is substantially inaudible, particularly i when the electromagnetic pickup is located as' above described at a distance from the point of J diameterin order` to facilitate tuning. vIf the y support of only of the lengthof theI bar, and v.this frequency is lowered by the notch. It will e also be noted thatthe length of-'thevibratable bar' is` in all instances considerably more than 1GO times the diameter' thereof,y VVAlthough this is not -essential vto .the production of the desired tones, the vbar should'- be long inv relation to its respondingly great dificulty in tuning vand greater increase in wastage.

jI claim as my invention: Y 1. Incombination, a vibratable bar of subystantially uniform circular cross-section, supstantially uniform circular cross-section having a length more than one hundred times its diameter, supporting means lfor said bar, said bar having an arcuate notch therein at a distance of approximately one-twentieth of its length from the supported end thereof, said notch being substantially half Way through said bar, means for striking said bar at a predetermined point, and electrical pickup means located at a distance of approximately one-tenth of the length of the bar from the supported end thereof.

3. In combination, a vibratable bar of substantially uniform circular cross-section, supporting means for said bar at one end thereof, said bar having an arcuate notch therein at a distance of the order of one-twentieth of its length from the supported end thereof, means for striking said bar at a predetermined point, and electrical pickup means located at a distance of the order of one-tenth of the length of the bar from the supported end thereof whereby said pickup means will pick up two partials of the said bar having a frequency ratio which can be expressed in small whole numbers.

GALAN W. DEMUTH.

Images