Horn Loudspeaker

[Guest]

Hi there. We have constructed a matching pair of horn loudspeakers as published by T. Giffard in the Elektor Electronics magazine in May 1990 (pages 20-23). We have recently tested the speakers with a frequency analyser and found the low frequencies roll off at around 315 Hz, somewhat higher than expected and higher than the response indicated in Fig 2 of the article. Are you able to provide clarification of the following:

1. In Fig 2 of the article there are two response curves. Which one is the measured response of this unit? What is the other curve for? This figure isn’t referred to in the article.

2. Has there been any modifications or amendments to this design submitted or published that might answer our problem with lack of low frequency response?

Any suggestions or information would be very much appreciated.

[Elektor Editor]

We do not know how you did the measurement, but this design of a horn loudspeaker must give an low frequency roll off at about 50 Hz. The measurement in figure 2 shows (a) the frequency response of the complete unit at a distance of 1 m, with superimposed a second curve which shows the pressure right in front of the horn. You should add these two to get an impression of the total frequency response. And (b) shows the impedance curve of the complete loudspeaker system.
With a low frequency roll off at about 300 Hz you should hear no bass at all when playing music and we cannot imagine that's the case. Something else must be wrong, for instance a wrong value for a filter component (the coil in series with the woofer). Also the choice of the right bass speaker is important. We assume you have not used the original McFarlow speakers, since they are not available anymore. You have to use a speaker with the same Thiele-Small parameters, otherwise the coupling of the speaker to the horn will not be correct and you do not get enough bass output.

Kind regards,
Harry Baggen
editorial staff Elektor