Working on the venting of a drum resulting in a split construction automatically draws questions like : now that the integrity of the shell is compromised, how will this affect the sound of such an instrument ? Let’s try and get a feeling for the contribution of the shell to the global sound, from a physical standpoint.

Shell construction & design

In the world of drum building, shells are usually made out of wood with various construction techniques, or out of metal:


  • Ply : the most common construction technique : thin layers of wood are glued together then undergo a specific curing process to ensure geometry and consistency of the shell. This method is the one using the largest quantity of glue for a given shell dimension.
  • Stave : Pieces of wood are prepared with beveled faces to assemble into a polygon, then lathed into round shape.
  • Single Ply : One thick plank of wood is steam bended into round shape, presenting one single glued joint.
  • Solid : a fairly rare method : a tree trunck is hollowed on a lathe or by other technique, absolutely no glue is used.


  • Steel : is the most common metallic material used for snare drum, it is usually rolled then welded.
  • Aluminum : rolled or (rarely) cast then machined, described as having a darker sound than steel
  • Copper, Brass, Bronze : can be either rolled or die cast
  • Titanium


  • Acrylic
  • Concrete
  • Glass

Tested snares :

  • 13×7 Stave Oak : is a 13” diameter, 7” deep drum, and has a shell built by stave technique : alike barrels, pieces of wood are beveled and assembled together then lathed to perfect roundness. This is said to minimize the amount of glue vs ply shells and therefore, the damping of the shell and the sustain of the sound.
  • 14×6,5” “Free Floating” : is a 14” Diameter,6,5” deep drum with a rolled sheet metal steel shell (thickness = 1mm). The particularity of this drum is to feature no tensioning devices (lugs) or strainer on the shell itself, hence letting the shell free to resonate as claimed by the manufacturer.
  • 14×6,5” Mahogany : is a 14” diameter, 6,5” deep drum, with a shell made out of mahogany plies. 4 layers of mahogany are glued together, and reinforcement rings at the top and bottom of the shell guarantee roundness and strength.
  • Repercussion Orchestral snare : is a 14” diameter, 6,5” deep prototype drum, with a 15 ply maple shell (11mm thickness). Its main feature is a ra
    dial ported vent in its center resulting in a “split shell” architecture (2 half shells).IMG_1382
  • Repercussion 1O slots vent : is a 14 “ diameter,  6,5” deep prototype drum with a 8 ply (6mm) maple shell, featuring ported vent close to the top head, between each lug.IMG_1408
  • Repercussion Standard Snare radial vented : is a 14” diameter, 6,5” deep drum with a radial ported vent, made from a 8 ply (6mm) maple shell.IMG_1384

Method :

In order to get a feel for the effect of the various shell construction, an “inverse technique” has been used : a volume velocity controlled source has been placed in the anaechoic room, and vibration measurements have been performed on all the snare drums  described above (with snares ON).

The principle being : when the instruments vibrates under the force of a hit, it  radiates sound – reciprocally, if the instrument is loaded with an acoustic field it will vibrates. If the acoustic field “force” (called volume velocity, and using specially designed sources) is controlled and the vibration measured, then the ratio between the two (sound emitted and vibration measured) can help understand how the instrument radiates sound.

Limits :

The method below is using a constant sound excitation (broadband noise) allowing for time averaged measurements, therefore the “contributions” evaluated below are more to be read as “maximum potential of radiation” for each element considered. In reality the signal IS a transient, and the relative contributions of each component might vary over time. In particular, if the shell does not show overall contributions of more than 20%, the accumulation of vibratory energy in the heavier shell structure than in the light heads might leave a larger relative contribution over time.

The other limitation being the very principle of reciprocity based on the linearity of the system –  which is questionable on a drum, with mechanisms such as the head tension being affected by the hit itself and therefore varying over time, and the snare action.

Nevertheless, it is interesting to see how different shells material and constructions affect those global readings.


The graph below apportions the “ability to radiate” from 3 elements on all 6 studied snares – globally shell readings oscillate between 20 to 25% – not that much but let’s keep in mind that those values can change during a real transient load case.


A crude technique for operational estimation :

In the same measurement campaign, vibration and sound pressure have been simultaneously recorded under a calibrated stick hit (constant energy). Using the results above as transfer functions a crude evaluation on contributions of both heads and shell under “normal loading” has been assessed: the results are presented below :


Up to 500-600 Hz, the energy is mainly distributed between the heads with a split point at 200 Hz. Then shell contribution starts significantly after 500Hz

Global summary of the contributions are given in the graph below.


Conclusions :

Vented or not, all drums used in this study tend to show that from a physical standpoint not much is actually coming from the shell, and top head is the driver for sound radiation hence the effect of any device or system (such as venting) changing his vibration will have an immediate effect on the sound.

Shell contribution for all drums is significant after about 500Hz, the energy below being mostly in head motions.

This being said, bear in mind all the limit of the method, in particular the fact that the transfer functions are obtained in a steady state loading instead of transient, and therefore the contributions over time might be slightly different.

From a perceptive standpoint we can imagine that some partials coming from the vibrating shell, have “stored” more energy than the vibrating heads (more motion, but less mass) and are less damped (metal shells in particular) resulting in lasting components in the sound spectrum long after heads have returned to resting position, leading to different tone perception.