In professional audio
Even though it was created a decade ago the digital microphone is still to make its mark on the professional audio industry. It appears that the market penetration is more a matter of time as price drops and more manufactures take advantage of the technology. It will be shown in this paper what the AES42 standard represents and what advantages digital microphones offer the professional audio community. This paper is designed to educate the audio professional about the design and operation of digital microphones.
The research method uses the literature review to acquaint the reader with the implementation of the AES42 standard for digital microphones and the implementation of the standard by specific manufactures. This paper will examine the operation of digital microphones in the workplace, highlight the developments of new technologies and compare the benefits of this technology against traditional analogue devices.
The Audio Engineering Society (AES) design specific industry standards and protocols useful to manufactures so they can produce products that are compatible and synchronous with other products. These same standards are a key to the end user in understanding the state of the industry and good professional practice. Two standards concern us directly with digital microphones (1) AES3; the protocol used for two track digital streaming that is commonly called AES/EBU, which was adopted for commercial digital microphone manufacture and (2) AES42; the AES published AES42-2001 after AES3 had evolved. AES42 is designed specifically for digital microphone application and it is this standard that will be explained in detail. Microphones have been modelled on these standards outputting the AES/EBU protocol and can be used directly with recorders and mixing consoles fitted with this digital input.
Beyerdynamic GmbH implemented AES3 in microphone design but has since ceased manufacture of any digital microphones. As a historic benchmark and as an example of the implementation of AES3, Beyerdynamic GmbH will be quite a useful reference with regard to the development and implementation of digital microphone design.
To bring life to the practical application, a couple of manufacturers have been selected to put the protocol in perspective. The manufacturer Nuemann had an influential input on the writing of AES42 and produced a microphone with the implementation of this specification. As such this microphone will be investigated in detail. Samson presently advertises their digital microphone as the first to utilise USB, and has an application for the small studio consumer.
While digital microphones are taking hold of the computer market before infiltration into the professional audio market, this paper will not include the application of this technology. It is through this selective approach of the material that the paper can focus and portray a good overview of the current technology and how it is used in the professional recording environment.
AES3-1985, AES Recommended Practice for Digital Audio Engineering - Serial transmission format for two-channel linearly represented digital audio data.
AES3 was first adopted in 1985 with revisions made in 1992 and again in 2003. The 2003 revision was written in close cooperation with the European Broadcasting Union (EBU). The AES/EBU digital interface is quite familiar to those working as audio professionals. The standard is intended for use with shielded twisted-pair cable and can be used for distances of up to 100 metres. Preferred sample frequencies are derived from AES5 and synchronisation is stated in AES11; both of these standards are incorporated in AES3. The standard connection for the digital interface is the XLR or IEC 60268-12. Pin one being used as earth and pin two and three for signal, the relative polarity of the signal pins is not relevant in digital audio. The AES3 standard gave a framework for serial transmission of digital audio data and this has been implemented in the manufacture of digital microphones.
The Digital Microphone
To best preserve the analogue signal one must digitise the sound at the sound source, because analogue components degrade the signal by combining noise and other elements that were not in the original waveform and the only way to avoid this is to convert the sound at source. This is confirmed by Georg Neumann GmbH (Peus, Kern, p 3):
The technical objective is defined so that high-quality digitization of the signal output by the capsule occurs right in the microphone (in the first step of processing, as it were), and level adjustment or other processing steps only occur once the signal is in the digital domain. This provides ideal conditions for preserving the quality of the signal coming from the microphone.
Some research has been done in converting sound into digits without the use of a conventional transducer. One such device made by Sennheiser is an optical microphone developed by Niehoff (1999). It uses intensity modulation and transfers vibrations on its diaphragm to a beam of light before it is digitised. At present this process is in the early stages of development, but efforts are being made to decode the acoustic sounds directly into a digital bit stream. Research is still to determine whether this line of investigation has any advantages over using an analogue transducer to track the analogue sound before the digitisation process.
Beyerdynamic MCD 100 and MCD 101.
The Beyerdynamic MCD 100 and MCD 101 microphone provides a standard AES/EBU output with a sample frequency of 48kHz using 24bit technology. It provides an equal mono signal on the left and right channels of the AES/EBU output with a frequency range of 20 - 20,000 Hz, and a dynamic range of 115 dBa.
The microphone uses digital phantom power (DPP) 6 - 10 volts, 150 mA and can be sourced from an external power supply if the console is not equipped with DPP. The microphone is protected against the accidental use of conventional phantom power and normal microphone cables can be used up to 25 metres before AES/EBU cable must be used.
Pre-attenuation can be achieved by remote control from the external power supply or at the console, and synchronisation is achieved by using the console as a master clock. If master-clock is not available, the external power supply can act as master. This unit can also sync a number of microphones.
Beyerdynamic has ceased production, though stock of the omni-directional MCD 101 microphone is still available at the time of publication of this paper. We will now discover how the development of the digital microphone has progressed with the inclusion of software to enhance the options available to the recording engineer.
AES standard for acoustics - Digital interface for microphones.
AES42 is described as an extension of AES3 specifically for microphones. One implementation in the standard was the ability of the microphone to be a self-clocking device. This is identified as the asynchronous mode or mode 1 with mode 2 as the true synchronisation mode as it uses a voltage-controlled crystal oscillator and a control voltage from the remote control. The control voltage is processed by a simple phase-locked loop system generated by phase-frequency comparison between the master clock and the microphone. If the receiver does not support mode 2, the microphone will work in mode 1 automatically. As we will see later, the manufacturers warn professionals against the use of mode 1 operation.
In the specification, remote control of attenuation, polar pattern, limiting, equalisation, low-cut filter, MS-XY operation, balance width, mute, reset, ADC calibrate, test signal, light control, multiple sample frequencies, dither-noise shaping and mode synchronisation are all included. The light control here refers to the ability to be able to switch a recording or on-air light outside the studio on and off. Wireless microphone application parameters such as low battery indication, battery charge and squelch, forward error correction capacity, error concealment have not been excluded. Manufacture identification and model identification are also available in the specification. This will give a further indication of the creative and practical application of the remote control parameters. We can see from the remote control possibilities that this is not just another microphone or just an opportunity to convert the analogue sound into the digital domain at the first possible opportunity. The ability to limit the signal at source, change polar patterns and introduce equalisation is a powerful creative tool.
Neumann has developed a patent pending technology of seamlessly combining two processors to convert the analogue source into digital. It has one processor working on low level sound, and when the sound level goes above - 30 dB the second processor starts, keeping the noise floor down to less than the capsule noise itself. This means that the switch from one capsule to the other is non-critical which has technological advantages compared with other crossover designs. Peus and Kern (2001) states:
Traditional gain-staging methods of dual processors are far less affective. These devices use the first processor to encode the programme until their full limit is reached with the second processor recording the high level information. This creates a situation where the crossover point is in a critical range and it is extremely difficult to match without introducing distortion. Intricate software algorithms have been developed to delay the signal until the two processors can be time-aligned but with limited success (p.5).
Solution-D consists of a studio condenser microphone, a digital microphone interface and remote control software. The analogue-to-digital converter converts the capsule output to a 28bit signal. This is followed by DSP to implement some of the software features. The microphone supports sample rates of 48/96/44.1and 88.2 kHz synchronised externally by the master-clock in the remote control. This microphone model D-01 has 15 available polar patterns using a 30mm capsule diameter. Neumann (2004) warns the user about operating the microphone asynchronously or in mode-1:
The asynchronous mode of operation, which is also provided generally, requires the use of a sample rate converter (SRC). However this mode of operation should be used only if synchronization is not possible, since available sample rate converters markedly impair the signal quality, for instance in terms of dynamic range and latency time (p. 9).
A few words need to be said about the software in practice. The digital gain can be set from 0 to 63 dB in 1 dB steps without incremental noise. This is in addition to the analogue pre-attenuation of 0, -6, -12 and -18dB steps. The microphone does not yet implement the equalisation capabilities of the AES42 standard but has the limiter function. The limiter is frequency conscious, switchable between 2 kHz and 4 kHz, and has attack, release and threshold parameters. It has a graphic display of gain reduction and operating level. The phase is reversible from the remote device.
The software is updateable and can hold five versions of the software to be accessed by the user. Cable lengths of 100 m using high quality XLR cable can be used; for longer runs of up to 300 m, AES/EBU cable is used.
Digital microphones can utilise their own digital connector identified as a XLD described in AES14 with additional grooves and user-insertable coding keys. To be fully coded a connector must have both a groove and a coding key. A half-coded connector has a groove only. Figure 1 shows the mating possibilities allowed by fully coded, half-coded and non-coded XLD connectors. Remember that both these coding standards are optional for situations were interconnection with analogue equipment is to be avoided and the system makes it an impossibility. The digital connector can also have a zebra ring with a colour coding of black-white-black-white with bumps to facilitate tactile feedback for low lighting situations (Figure 2). The zebra ring indicates a digital signal (p. 32).
The above implementation is a serious innovation in microphone design and although it has many benefits, the industry has not yet been able to take advantages of the technology except in the high-end market. The reason for this is the microphone has a very large price tag and one would expect that it would take time for the technology to catch on enough to make it more affordable for general use.
Figure 2: The zebra ring for XLD connectors
USB Studio Condenser Microphone
This product is a very good example of the infiltration of AES42 into the project market. This microphone has a 19 mm diaphragm in a robust case and connects into your DAW via USB. Where you did have to buy a studio condenser microphone and preamplifier or mixer before having a sound source available at the computer, now the front-end is combined in the microphone. This is a dramatic cost saving and if we look at the functionality to the general public, it has become possible to plug a good quality microphone into a computer and record it without complication. The microphone has software that controls gain, low cut filter, input monitoring and phase. The cost of this product is very affordable and is die cast, 16 bit/48kHz, with a single hyper-cardioid pattern. For someone making demonstration recordings of compositions little else would be needed.
For professional users some questions needs to be raised beyond the tone of the microphone. The device is self-clocking or asynchronous and as such will not perform to the standard of their synchronous counterparts. Wells (2006) quotes Rupert Neve:
Significant trends in the search for quality seem to be still wrapped in 'mystery'. Various designs have provided switched input impedance, which, of course, does produce an effect-maybe beneficial in masking other deeper problems. A bit like adding the extra load of an RV to your sports car to slow it down! USB can only seriously limit available voltage, headroom etc. There seems to be a trend to sit back and let the wonderful computer do it all--so convenient! (p. 2).
In the right environment the cheaper digital microphones are going to do an excellent job, economically. The Samson product is one of these and the main purpose for its inclusion is that it provides a definite indication of where the professional audio industry is headed.
Looking at the advantages of the digital microphone I can only see it becoming more useful and less expensive. The expense of the high quality devices excludes its use for the average engineer but I can see this situation changing as the technology develops. In a digital world technology changes pretty fast, and evaluating what we have already gives a clear indication where technology could develop. For further research I could explore the possibility of designing a microphone application that uses the existing DSP (digital signal processing) technology to set the microphone level automatically. As the microphone level is automated, level sensing devices could be used to automatically reduce the microphone input by a user determinable amount with reference to full scale. For example the microphone could be set 6dB below the highest peak it read over a relevant test period and also inform the operator if the peaks exceed -6 dB. Other artistic applications of the technology will obviously outdo ideas derived at the present time. It will also be interesting to see how manufactures implement the equalisation capabilities of AES42, as this would add some quite creative aspects to the soundscape.
I am quite excited about the implementation of the limiter into digital microphone design as this stabilises the input which and is a necessary function for close microphone application. The SSL and other analogue consoles come with a compressor and equalisation on every channel and soon this could accompany most new microphones. The microphone stage does seem like the place to convert the sound into digital if we are working in a digital medium, but it's an artistic choice. We have quite a number of years integrating our analogue microphones with our new digital counterpart. I am supported by Wells in his article, The Low Down On The Front End (2006) that the digital microphone will have a marked presence in the life of the audio engineer in the near future:
Marc Benard, product manager for PC audio interfaces at Digigram, anticipates that there will be growth in the professional digital microphone market. "We believe that digital microphones that are compliant with the AES42 standard will become more and more important. Production studios, and as well broadcast studios, boast more and more digital equipment, and it makes sense to start the digital signal chain immediately at the microphone," he explains, offering a new outlet for the type of products Winkler noted as generating little market enthusiasm. "For this reason," says Benard, "Digigram will be the first sound card manufacturer to implement AES42-compliant inputs" on I/O cards to be introduced early in 2006 (p. 6).
AES42-2001: AES standard for acoustics – Digital interface for microphones. Audio
Engineering Society, Inc. 2001
beyerdynamic GmbH & Co. KG., (2006). MCD100 Operation Manual. Heilbronn, Germany: beyerdynamic GmbH & Co. KG.
Georg Neumann GmbH., (2004). Solution - D Operation Manual. Berlin: Georg Neumann GmbH. Retrieved May 12, 2006 from Georg Neumann GmbH, http://www.neumann.com/?lang=en&id=current_microphones&cid=solutiond_publications
Niehoff, W., (1999). A Simple Optical Microphone. Wedemark, Germany: Sennheiser Electronic GmbH & Co. Retrieved May 3, 2006, from Acoustical Society of America, http://www.acoustics.org/press/137th/lay_lang.html
Peus, S., Kern, O., (2001). The Digitally Interfaced Microphone, The last step to a purely audio signal transmission and processing chain. Berlin: Georg Neumann GmbH.
Wells, F., (2006). The Low Down On The Front End. Retrieved May 3, 2006. from http://www.prosoundnews.com/articles/article_3075.shtml