About Us About Us Web Zine Back Issues What they say about us






NXT Loudspeakers

peteraczel | 09 August, 2006 19:27

A Note on NXT Distributed Mode Loudspeakers 

This was an attempt to investigate the high-fidelity possibilities of a radically new and different transducer technology.

NXT is a fairly young but rather large company based in Huntingdon, England. They are responsible for the development of the Distributed Mode Loudspeaker (DML), which is a flat-panel transducer that can assume innumerable different shapes and forms, both opaque and transparent, and is not based on the piston concept but on bending-wave physics. A DML panel vibrates in a large number of modes instead of moving back and forth as a rigid piston-like unit. I have never seen a formal mathematical analysis of the basic design, but there are literally hundreds of products out there, large and small, using the NXT patents, from “talking” TV screens to cheap little surround-sound satellites. Everything except high-end audio—and I wondered why. (Actually, I witnessed a CES demo of a high-endish prototype a few years ago, but it never flew.)

I managed to get my hands on a two-foot by one-half-foot transparent NXT panel, which had been part of a discontinued mass-brand compact music system. I know that this panel does not represent the state of the art in DML technology, but it was intended to reproduce music and therefore promised to give me a minimal indication of the hi-fi possibilities of the DML concept. I connected it to my bench amplifier (no EQ, as there may have been in the commercial system) and set it up to run a frequency-response curve with the MLS (quasi-anechoic) method. The result is shown in Fig. 1.

Fig. 1: Frequency response at 1 meter on axis (amplitude blue, phase green).

As can be painfully seen, the response is horrible—so horrible, in fact, that I do not believe it. It is possible that the bending-wave/multimodal sound propagation cannot be accurately measured with the MLS method. I am just speculating because, as I said, I have not seen a formal analysis of the system. Maybe some kind of averaging measurement at a large number of points would be more valid. I just don’t know. My reluctant conclusion, tentative as it is at this juncture, has to be that the hi-fi potential of the DML concept is extremely limited and that the absence of high-end applications is inevitable. I hope to change my mind if and when contrary evidence becomes available to me.

General | Next | Previous | Trackbacks (0)