Beyer MCD100

Beyerdynamic have become the first mainstream manufacturer to offer the audio industry a practical digital microphone. Hugh Robjohns investigates the MCD100 to see if it represents the future of recording technology.

Ever since the digital audio revolution started, almost 20 years ago, there has been steady and consistent progress towards the all‑digital studio. Already a number of digital consoles are available, spanning a very wide price range and designed to suit all applications, from the small home studio through to the largest of professional establishments. Likewise, digital recording is de rigeur for many these days, whether based around tape formats like the Tascam DTRS and Alesis ADAT machines, or disc systems such as MiniDisc, CD‑R/CD‑RW, the various Magneto‑Optical (MO) formats, and the enormous range of hard‑disk workstations like Pro Tools and SADiE.

Digital signal processing is commonplace in any studio setup these days too, with digital I/O capability being high on the list of desirable features. It is even possible to buy so‑called 'digital loudspeakers' in which D‑A converters are integrated with line‑level crossovers and power amplification to make digital‑input active loudspeaker systems. Although only a few such designs exist, and those are mainly to be found in the domestic hi‑fi industry, the concept will inevitably move into the semi‑pro and professional audio environments sooner or later.

That only leaves the opposite end of the chain as far as digitisation is concerned — the microphone — and Beyerdynamic have recently released their first digital microphone, the MCD100.

Digital Or Analogue?

The MCD100 without its cover, so you can see the bits inside...

The first point I must address is whether the MCD100 is really a digital microphone or an analogue one. That probably sounds rather odd, given that Beyer themselves describe this device as a "digital condenser microphone", but I think there is an important distinction to be made here.

Consider for a moment a hypothetical specialist compressor/limiter. Something a bit 'tweaky', perhaps, with hand‑made valves using filament wires derived from the original teeth braces of the Osmond family (younger readers should ask their mums) and encapsulated in twelfth‑century stained glass. If such a device used traditional analogue 'tube' circuitry but was only fitted with digital inputs and outputs would it be called a digital signal processor? Well, not by me, but it probably all depends on the marketing approach...

Perhaps you can now see the dilemma I have over the Beyer Digital Condenser Microphone. Sure, it generates a digital output conforming to the standard AES‑EBU format with a sample rate of 48kHz and 22‑bit resolution, so in that way it would be reasonable to describe it as a digital microphone. However, the actual capsule is a traditional analogue device, a standard pressure‑gradient condenser element providing a cardioid polar response. The advance Beyer have made over other microphone manufacturers with the MCD100 is that the output from the usual internal head amplifier is connected directly to an A‑D converter which, with the benefits of surface‑mount technology, can now be incorporated into the microphone body instead of into an external unit. Apart from the miniaturisation and integration, there is little difference from a conventional microphone, so I'd prefer to call this a hybrid digital microphone — but maybe that's not so easy to squeeze into a snappy marketing campaign, and the title probably wouldn't be so appealing to potential purchasers!

To me, a true digital microphone is one where diaphragm movement in response to sound waves is captured by some means which generates a digital signal directly. For example, a potential technique which many R&D labs have already spent some time investigating involves projecting a set of optical interference rings onto the diaphragm surface and using some kind of light sensor to detect the presence or absence of rings as the diaphragm moves, thus quantifying the diaphragm excursion directly and numerically. Needless to say, no‑one has yet managed to make a commercially viable microphone in this way, although I'm sure this, or something like it, will appear in the not‑too‑distant future.

In the meantime, the closest we have is this new 'hybrid' digital microphone. Let's take a closer look at it.

Singing Digits

The microphone itself looks much like a standard Beyer 740 with a grey satin metallic finish. It weighs just over 300g, and is 190mm long and 40mm in cross‑section. The mesh grille is slightly trapezoid in shape, and a useful degree of protection from popping and blasting of the capsule is afforded by a fairly dense foam liner. Although it's difficult to see clearly, the capsule element appears to be roughly 20mm square (yes, I do mean square), and the front of the mic is indicated by the Beyerdynamic name and logo on the body.

Beyer offer to tailor the frequency response of the microphone (for an extra charge) to anything the purchaser requires.

Although Beyer do not supply print‑outs of the polar and frequency responses with the microphone, the user manual states it has a cardioid response, offering a rear‑rejection ratio of 22dB at 1kHz, with a nominally flat frequency response between 20Hz and 20kHz. However, Beyer offer to tailor the frequency response of the microphone (for an extra charge) to anything the purchaser requires, by providing an exchange EEPROM which re‑programs the DSP with the desired characteristics. Enter the world of personalised microphones!

Looking at the microphone, there is absolutely no clue to its hybrid nature. The audio output is provided on a standard male 3‑pin XLR which carries a standard AES‑EBU signal, with the microphone output encoded on both channels. As the familiar XLR connector is used, there would be nothing to stop anyone inadvertently connecting the microphone to a normal 48V phantom power supply. If it's possible, sooner or later someone is going to do it, so in the interests of investigative journalism I tested the microphone before and after connecting it to a standard analogue mic channel with phantom power enabled. I'm pleased to report that it survived without problems!

Power Points

The MCD100 can't be used alone — it requires a separate power supply, and the user manual refers to a variety of suitable units that Beyer can provide. The MPD50 provides facilities for two digital microphones and has two separate AES/EBU outputs; the MPD100 and the MPD200 (the latter was supplied with the review microphone) also cater for two mics but issue a stereo output. The third model is the MPD800, which accepts eight microphones and provides four stereo AES/EBU feeds. All these units supply the necessary microphone powering arrangements, called 'digital phantom power' or DPP. This is a new mic powering and signalling system which operates at between 6 and 10V and supplies a generous 150mA of current. The manual suggests that some digital consoles and recorders provide this facility directly, although I am unaware of any as yet. However, such a system would be necessary for future digital microphones and Beyer's scheme seems as valid as any, so perhaps it will eventually be incorporated into the AES specifications documents for the AES‑3 digital interface.

The MPD200 is a small, lightweight, rectangular black plastic box carrying an assortment of connectors on three sides, and a pair of buttons with an associated collection of LEDs on the fourth. For those who like to know the statistics, it measures 145 x 85 x 42mm and weighs around 350g. Unfortunately, a large part of the reason why the control unit is so small and light is that there's no internal power supply — an external 12V DC unit is needed and the Uniross model supplied for the review was almost the size of a small house brick! To be fair, the MPD200 requires around 350mA of supply current at between 12 and 15V, so the external power supply does need to be reasonably meaty. Although I'm not a fan of external power units, I have to accept their necessity. But when the co‑axial power connector is so poor that the supply becomes intermittent and the unit is forever rebooting itself, I have to question why a decent power supply could not have been incorporated properly into the control unit , making the whole thing substantially more professional and reliable.

In terms of sound quality, it's hard to find fault with this microphone.

As already mentioned, the control unit provides connections for two MCD100 microphones via a pair of female AES/EBU 3‑pin XLR connectors on one end of the box. The (two) digitised microphone signals are then synchronised and combined onto a single AES/EBU output connector at the other end of the box. There is provision for word clock signals in and out of the control unit (via a pair of BNCs on one side panel), as well as the co‑axial DC power input (alongside the AES/EBU output socket).

On the other side of the box from the clock connectors are two large round push‑buttons, a row of nine yellow LEDs, and a pair of red LEDs. Starting with the last items first, the red LEDs illuminate if one of the inputs is not receiving a valid digital signal (either because a microphone has not been connected or the cable is too long or unsuitable). The row of yellow LEDs performs two functions. The three lights on the left indicate the status of an analogue pre‑attenuator (pad) in the microphone, which can be set to 0, ‑10 or ‑20dB. (This pad in the microphone head is controlled via pulse signalling sent back up the microphone's cable).

The remaining lights show the level of digital gain applied to the microphone signal by the control box (from +6 to +36dB in 6dB steps). The two push‑buttons increase or decrease the amount of digital gain and analogue pre‑attenuation as required.

Technology

Taking a quick peek inside the microphone reveals a pair of heavily populated double‑sided circuit boards connected back‑to‑back and running the full length of the microphone's shaft. Surface‑mount components have been used throughout, of course; the most easily recognisable elements include an Analog Devices 2115 digital signal processor with associated EEPROM, and a 'True Match' A‑D converter from the German company Stage Tec.

Stage Tec have yet to become a well‑known name in the UK audio market, but in Germany and around Europe they are already very highly regarded. The company manufacture a range of very sophisticated digital routers and consoles, and their products have established a very good reputation — most notably their microphone preamplifier/A‑D converters.

The MPD200 control unit contains even more surface‑mount circuitry. Again, the easily recognisable components include a pair of standard Crystal AES/EBU receiver chips and a single transmitter. There's also a pair of Analog Devices 1890s which perform sample rate conversion and digital synchronisation functions. A lone jumper inside the control unit determines whether the system runs as a stand‑alone device with a fixed 48kHz sample rate, or accepts and slaves to an external word clock signal (of between 32 and 48kHz).

Setting Up

Rigging the microphone is very straightforward. A standard mic cable connects the microphone to the control unit, and another takes its output onto a digital desk or recorder. About five seconds after the external power supply is hooked up, the gain/attenuator LEDs flash sequentially (while the system runs through its boot‑up and self‑diagnostic procedure) and the system is fully functional. The pre‑attenuators and digital gain default to 0dB after power‑up or when the microphone connection is interrupted (which would appear to be a software revision, as the manual suggests it will default to ‑10dB of attenuation). As I had only one microphone to connect to the control box, the second channel reported the absence of a signal via the appropriate red 'Error' LED.

The manual suggests that any ordinary microphone cable up to 25 metres can be used with the MCD100, but that "special AES/EBU cables" should be used for longer runs. I made use of ordinary Musiflex cables and had no problems whatever running the mic through an entire cable drum of the stuff.

With the control system up and running, the effective sensitivity of the microphone can be set to suit the conditions. If the mic is placed close to a particularly loud source, the head amplifier can be protected by remotely inserting the 10dB or 20dB pads. In its natural state, the microphone's internal converter is set to clip at 130dB SPL, (although the pre‑attenuators can be used to raise that to 140 or 150dB SPL). The claimed 22 bits of coding resolution should theoretically provide the microphone with a dynamic range of 132dB, although the user manual quotes an A‑weighted range of 'only' 115 dB. That being the case, the microphone's self‑noise effectively equates to around 15dB SPL — not equal to the very best‑performing analogue microphones, perhaps, but still extremely good, particularly when you bear in mind that this includes the mic preamp and converter noise figures.

Whether or not the MCD100 really qualifies as a true digital microphone is a rather philosophical argument.

When the MCD100 is used in a distant placement, or with quiet sources, the control box allows the digital signal level to be stepped up so that the signal peaks can approach 0dBFS (ie. maximum level on the digital mixer or recorder). This facility is very important, because if the mic was used, for example, to record the spoken voice, its output could easily be 60dB or more below the digital clip level. Most digital mixers are not equippedto introduce anything like that amount of gain into a signal, so the facility to boost the AES‑EBU output by up to 36dB directly from the MPD200 control box is extremely welcome.

When the control unit is set up as a 'Master' unit (via the internal jumper), the output is always at 48kHz, and if more than one microphone is connected to the control unit they are synchronised to each other before their signals are combined into a common AES‑EBU data stream. However, the unit can be set up as a 'Slave', and if the control box is provided with a word clock signal the digital output can be synchronised to an external reference clock source at any rate between 32 and 48kHz (see the 'Clocks & Microphones' box for more details on digital synchronisation).

As the MCD100 always runs from its internal 48kHz clock, the MPD200 control box is continuously re‑synchronising the signal according to its own internal or external clock signal, possibly with some sample rate conversion thrown in. However, in professional circles sample rate conversion is often frowned upon and generally avoided, because 'golden ears' claim to hear detrimental effects from the extensive signal processing involved. With the MCD100, the DSPs in the control box are always active and there appears to be no way to disable this facility. The only solution would be to provide some means of sending a clocking signal to the microphone itself so that its A‑D converter could lock to an external reference, which would require a bi‑directional interface. Quite why Beyer have not followed that route is a mystery, as I would have thought it preferable to having a sample rate converter permanently active in the signal path.

Practical Exercises

The MCD100 is very easy to set up and use. I had no problems whatsoever with synchronising the unit to the house word clock syncs, and it seemed very tolerant of longish cable runs. The control box allows the gain and/or attenuators to be set quickly and easily, and provides a very good visual indication of the status of the microphone.

In terms of sound quality, it's hard to find fault with this microphone. I was impressed by how open and extended it sounds — the frequency response seems to be wider and smoother than many similar microphones. It really doesn't seem to add much in the way of its own personality to the sound unless the source is sufficiently close to give rise to a proximity effect. In general, it yielded an accurate and neutral quality with the various sound sources I used. The polar pattern is a reasonably tight cardioid at high frequencies but seems to open out considerably at lower frequencies. Indeed, a quick voice test suggested it has strong tendencies towards hypercardioid or even figure‑of‑eight characteristics at very low frequencies. However, this is not uncommon with large‑diaphragm mics.

Conclusion

Whether or not the MCD100 really qualifies as a true digital microphone is a rather philosophical argument. The fact is that it's the first commercially available microphone to generate an AES‑EBU digital output, and Beyerdynamic should be applauded for that. Personally, I'm not sure whether the MCD100 offers any real practical advantages over the conventional arrangement of microphone, mic preamplifier and separate A‑D converter. Certainly the MCD100 is a neater, smaller and better integrated package than the collection of rackmount boxes you'd normally have to use to get the same result, and some people will no doubt appreciate that, but I fear that others will feel constrained by the fact that they can no longer use their favoured mic preamplifiers and the most esoteric of A‑D converters and dithering algorithms. Although there must be specific applications where a compact, integrated solution to sound acquisition is beneficial, I can't think of many, and for the majority of real‑world microphone users, I don't believe the MCD100 has usurped the conventional approach of separate microphone preamplifiers and A‑D converters — indeed, the ability to mix and match mics, preamps and converters is often considered an essential ingredient in the art of sound recording. However, the MCD100 does sound very nice, the A‑D converter is obviously extremely good, and I can appreciate the accomplishment of integrating the technology into such a compact package.