Hope College Department of Physics and Engineering Research Experiences for Undergraduates Summer 2001 Project Summary

Project Title: Building a Better Speaker
Student Name: Jonathan Spaulding
Student’s home institution: Johns Hopkins University
Research Advisor(s): Dr. John Krupczak
Source of Support (NSF-REU, or other): NSF-REU
 
 

Several of the New Speakers

This research was designed to improve the laboratory for Dr. Krupczak’s course, titled "Science and Technology in the Everyday World." The lab deals with simple and complex machines that are in use in everyday life, reducing them to simple concepts that students can examine and understand by seeing them in action, and by building a working model. My job this summer was both to design a more efficient speaker and to find out as much as possible of the behavior of the speaker in various fashions, from resonance to magnetic field. This speaker must be as efficient as possible so as to play music when powered by a coil radio amplified with a transistor amplifier. The radio outputs a signal much too weak to use, and the amplifier still only outputs 20-50 mA. This signal is barely enough to run a speaker, but an efficiently designed speaker can use this signal to play audibly.

My assignment this summer was to research speaker efficiency, the geometry of the magnetic field of the magnet assembly used in the speaker, and to optimize the speaker. I accomplished this by changing the speaker type from a cone type speaker to a horn type speaker, which is far more efficient in the mid frequency range. I also discovered that the magnetic field was strongly peaked in an area of about 3 mm height, and so determined that the speaker coil should be entirely within that area. The pictures below illustrate several of the ways this data was determined.

Much work went into computer models of the magnet assembly. The optimum configuration of the washers and magnets was important, but so also was the degree of correlation between ANSYS and measurements. Only with a good correlation could ANSYS be relied upon to accurately model the assembly. Several discrepancies caused great problems with calculations. Without a correlation, no hard data was available from ANSYS, and only comparisons could be used.

The goal of my research was to create a new type of speaker that could play the signal from a coil radio. This meant a more efficient speaker that responded well to voice-level signals. The end result of my research is a new speaker that is far more efficient than the old type, even though it is more difficult to build. The new speaker I designed fulfilled all of these requirements. It is possible for the students to build in lab and will perform as needed.
 
 

Field lines shown dramatically by iron filings

Field lines plotted by ANSYS

A graph showing field strength inside the assembly.
 

Slide show of Jonathan Spaulding's work