Comments on: Carbon nanotubes rock—literally! http://2020science.org/2008/11/26/carbon-nanotubes-rock%e2%80%94literally/ Providing a clear perspective on developing science and technology responsibly Thu, 29 Jul 2010 08:52:58 +0000 hourly 1 http://wordpress.org/?v=3.0 By: Andrew Maynard http://2020science.org/2008/11/26/carbon-nanotubes-rock%e2%80%94literally/comment-page-1/#comment-63 Andrew Maynard Thu, 18 Dec 2008 08:06:11 +0000 http://2020science.wordpress.com/?p=496#comment-63 Martin, Thanks for the clarification - in my haste I over-simplified! Of course, a typical 8-bit digital signal (or a compressed/encoded signal from a DVD/MP3 player etc) would not work here without conversion. As you point out, you would either need to convert this to a Pulse Width Modulated signal - or possibly a high sampling rate 1-bit bitstream. But because you are still using a digital signal, it is in principle possible for lossless conversion in the digital domain from AES or SPDIF to something that can be amplified (very efficiently) and fed direct to the loudspeaker--with the D/A conversion occurring directly within the transducer. Martin,

Thanks for the clarification – in my haste I over-simplified! Of course, a typical 8-bit digital signal (or a compressed/encoded signal from a DVD/MP3 player etc) would not work here without conversion. As you point out, you would either need to convert this to a Pulse Width Modulated signal – or possibly a high sampling rate 1-bit bitstream.

But because you are still using a digital signal, it is in principle possible for lossless conversion in the digital domain from AES or SPDIF to something that can be amplified (very efficiently) and fed direct to the loudspeaker–with the D/A conversion occurring directly within the transducer.

]]> By: martin Winsemius http://2020science.org/2008/11/26/carbon-nanotubes-rock%e2%80%94literally/comment-page-1/#comment-64 martin Winsemius Wed, 17 Dec 2008 21:12:38 +0000 http://2020science.wordpress.com/?p=496#comment-64 Your suggestion that thin film speakers could play digital signals directly demonstrates a misunderstanding of how digital audio signals are encoded and decoded. What you suggest could only possibly work from Pulse Width Modulated signals as are used in motor controllers. Audio is not encoded in PWM because it can not extract bit rate clock and also creates unwanted DC shifts during transmission. Therefore audio encoding is done in a manner that allows clock extraction and compensates for DC shifts in the signal as transmitted. AES for broadcasters and SPDIF for consumers are the encoding schemes used. Both are a special form of Pulse Code Modulation that can not be directly filtered for play through any analog amplifier or speaker. PCM must be processed through a special Digital Signal Processor circuit for conversion to an analog form (or digital-pulses-filtered-to-analog signals such as Class D, I, T) that can drive a loudspeaker. Your suggestion that thin film speakers could play digital signals directly demonstrates a misunderstanding of how digital audio signals are encoded and decoded. What you suggest could only possibly work from Pulse Width Modulated signals as are used in motor controllers. Audio is not encoded in PWM because it can not extract bit rate clock and also creates unwanted DC shifts during transmission.

Therefore audio encoding is done in a manner that allows clock extraction and compensates for DC shifts in the signal as transmitted. AES for broadcasters and SPDIF for consumers are the encoding schemes used. Both are a special form of Pulse Code Modulation that can not be directly filtered for play through any analog amplifier or speaker. PCM must be processed through a special Digital Signal Processor circuit for conversion to an analog form (or digital-pulses-filtered-to-analog signals such as Class D, I, T) that can drive a loudspeaker.

]]>