Specification sheets are important when we need to measure, analyze and describe the sound system and its components. They help in planning, and selecting components.
There are two types of variables found on a specification sheet - dependent and independent. Dependent variables change, and vary over time. Independent variables have fixed values. Most graphs and charts show the relationship between two variables, one dependent and one independent.
In general, the horizontal axis or x-axis represents the independent variable, while the vertical axis or y-axis represents the dependent variable.
Even though we see a joint line across most 2-dimensional plots and graphs, the values on each axis are actually sampled and measured at points called "data points". The points are joined by a line just for ease of reading.
Common graphs an audio engineer will come across include:
Y-Axis---------------------------------X-AxisEach graph shows a Y-value as a function of an X-value.
Amplitude--------------------------Frequency
Impedance--------------------------Frequency
Directivity--------------------------Frequency
Phase--------------------------------Frequency
Amplitude----------------------------Time
Level-----------------------------------Time
When the independent variable on the X-axis is frequency, the Y-value is said to be "frequency dependent". When the independent variable is time, the Y-value will be "time dependent". This means that all the variables on the Y-axis do not have a fixed value, varying and depending on the X-value at which it is measured.
An example of a frequency-dependent variable is relative level. On a graphic equalizer, we can see a graph or plot on the face of the device. the horizontal axis represents frequency while each adjustment made to the settings is changing relative level at whichever frequency band.
Other variables that are time-dependent include loudness, temperature and background noise.
Graphs show trends in data, giving us a visual aid in understanding the characteristics of the equipment. Graphs give us measured values so that we have a reference. It still is up to the engineer or technician to analyze that data, understand it, and apply it.
Another important thing to think about is how precise the data is, and at what resolution the data points were measured. The resolution needs to be appropriate for the data, such that all significant data points are represented, and yet we do not need to go too far with the resolution as there will be a point of "diminishing return", where the points are insignificant and do not serve our purposes.
Graphs and plots provided by manufacturers do not fully describe the product, they describe certain aspects of the equipment such that we can understand its behavior and compare it with similar products.
"One-number" ratings often over-simplify data, for example, instead of providing the user with the full graph, only the value measured at a frequency of 1kHz is given. If the product is purchased based solely on this information, the user will later realize that at all other frequencies outside of 1kHz, the performance of the equipment is unpredictable and perhaps not up to expectations.
Spectrum plots alone are still not enough to fully describe the behavior of equipment. For example, the loudspeaker response depends on frequency, but the amplitude x frequency graph only shows us the response measured on-axis. Depending on the Q value, with Q = 1 meaning omni-directional, the intensity on-axis will be the Q-value times the average radiated intensity. The higher the Q-value, the higher the axial intensity in relation to the average radiated intensity. Q is interchangeable with DI or directivity index, derived by DI (dB) = 10 . log Q.
Specification sheets help us determine if a certain product suits our needs for a certain application. We still need to measure and listen in order to evaluate the performance of the product, and purchases should only be made after doing both.
No comments:
Post a Comment