The performance of music without electrical amplification requires precise acoustic engineering of its venues. A customer advises concerning large-scale renovations to existing cathedrals, concert halls and opera houses. Renovations may include decoupling the structure from its surroundings to minimize outside noise and redesigning the interior in order to optimize echo patterns.
Buildings are acoustically characterized with precision sound equipment. A sound Stimulus, S(t), is generated by a loudspeaker and the Response, R(t) is received by a microphone, where t is time. Usually, the loudspeaker and microphone are positioned at the musician and audience position, respectively. The Impulse function, I(t), is defined from the Stimulus and Response functions as:
Where F() is the Fourier Transform and F-1() is the Inverse Fourier Transform.
The Impulse Response Function is used by engineers to diagnose the venue. It is equivalent to the echo pattern that would be received by the microphone if an infinitely sharp spike of sound were generated by the loudspeaker. In practice, a toneburst of rising pitch, called a chirp, is used as a stimulus. A well-engineered venue has an I(t) that decays over about two seconds and has no sharp features that correspond to dominant echoes. For instance, the focusing effect of a domed roof leads to undesirable spikes in I(t) which correspond to undesirable loud reverberating echoes.
Typically, the effect of venue renovation is simulated in an exact model of about one-twentieth scale. Simulations must also reduce acoustic wavelengths by this same scale factor. Even audience members are simulated within the model by bits of carpeting, since they absorb a significant fraction of the sound. The model can be verified for exactness by comparing the shape of its I(t) with the I(t) measured in the actual venue. Renovations are then optimized in the model to provide the best possible I(t) before renovations of the actual venue begin.
The customer requires an integrated analog output/analog input system to generate acoustic stimuli and capture received response signals. The system must have one D/A channel and four A/D channels that will accommodate multiple microphones. Each A/D channel must simultaneously sample at up to 1 MS/s with at least 400 KS of memory per channel.
The GaGe solution is a GaGe Instrument Mainframe 586, containing a CompuGen 1100 D/A card with 512K of on-board memory plus two CompuScope 512 A/D cards, each with 1M of on-board memory. The experimental setup is schematically illustrated below:
Through an audio amplifier, the CompuGen 1100 generates a signal to a loudspeaker placed at the musicians' position within the model venue. The CG1100 is a true arbitrary waveform generator so that the excitation signal can be precisely controlled. Since excitation signals must contain frequency components up to about 50 kHz, on-board samples are clocked out of the CG1100 at a rate of 500 kHz so that at least 10 points are generated per signal cycle.
The 512K of CG1100 on-board memory allows up to one-second-long signals to be generated, exceeding the requirement. A new pattern can be uploaded to the on-board memory within less than one second through the ISA bus. The customer explicitly triggers the CG1100 from software to generate the pattern. Alternatively, the CG1100 can be set to generate on an external trigger or to endlessly loop its pattern in a perfectly seamless fashion.
The sound generated by the loudspeaker is received by four microphones strategically placed at audience positions. The microphone signals are amplified and connected to the four input channels of two CompuScope 512 A/D cards. The CS512 cards are connected in Master/Slave mode, wherein a bridge board allows the sharing of timing signals for true simultaneous sampling on all four channels. The input sensitivity of each channel can be independently selected from software so that the 12-bit resolution of the CS512 cards can be fully exploited. The CS512 cards are externally triggered by the SYNC output of the CG1100, which can be arbitrarily controlled to provide a trigger edge at any point during generation of the pattern.
The CS512 cards sample at 512 kS/s so that, as with the CG1100, their on-board memory can capture a record up to one second long on each channel. All data can be downloaded from the CS512 cards within 2 seconds. Excitation chirp signals are repetitively generated and successive records are captured by the CS512 cards and averaged to provide further improvement in signal-to-noise ratios.
The identical clocking rate of 512 kHz used to generate the CG1100 stimulus signal and digitize the CS512 response signal is very convenient for the calculation of the Impulse Response function, I(t). Since both stimulus and response have the same timebase, their Fourier Transforms occupy an identical set of discrete frequencies. If the timebases were different, interpolation of either stimulus or response would be required for the calculation of I(t).
During preliminary development, the customer was able to perform the experiment with GaGe's stand-alone software - CompuGen for Windows and GaGeScope for Windows. In this way, the customer was able to manually perform the experiment without writing one line of computer code. In order to integrate the control of the CompuScope and CompuGen cards together with calculation of I(t), the customer easily developed an integrated MATLAB utility using GaGe's MATLAB Software Development Kits (SDKs) for the CG1100 and CS512. The SDKs' sample programs provided a convenient starting point for the customer's utility. The customer required no knowledge of low-level software calls to the hardware and simply combined GaGe's sample programs with calculation of I(t). On the GaGePC 586, calculation of I(t) takes only a couple of minutes. Previously, rented stand-alone instruments provided no on-line feedback on I(t).
While this application is acoustic, an identical GaGe stimulus-response system can be used in many other applications such as impedance bridge measurements, ultrasonics, Radar or Lidar. With GaGe's wide array of instrument cards and expertise in multi-card configuration, similar systems can be constructed to provide different numbers of channels and sampling speeds. For high-speed stimulus-response experiments, GaGe provides an elegant, integrated, single-box replacement for collections of bulky and uncooperative discrete instruments.
We encourage you to contact us and discuss your research & development application in more detail with our engineering team. GaGe can provide tailored custom data acquisition hardware and software solutions to meet specific application requirements.