I know you can pay a lot of money for a great mic preamp, but when buying audio interfaces at the more 'budget' end of the market, how do I tell which preamps are good and which aren't? Are there any particular specs I should I look out for?
SOS Technical Editor Hugh Robjohns replies: Few modern mic preamps are 'bad' — even the simplest implementation in low-cost equipment can deliver a level of clean performance that easily matches the state-of-the-art consoles of the '60s and '70s, as shown by our mic preamp comparison — www.soundonsound.com/reviews/pick-preamp — in SOS October 2012. Still, preamps can also be judged by how they sound when 'pushed' towards overload and more elaborate and sophisticated designs often do better in this respect, as well as boasting more facilities and higher-quality controls and construction.
But what about the technical specifications? Let's start with gain. Most manufacturers provide the gain ranges for the mic, line and (if applicable) DI inputs, the vast majority settling on 60dB as a practical maximum gain, but some providing only 40dB and some over 80dB. For insensitive vintage ribbon mics or quiet, distant wildlife sounds, 80dB of gain can be useful, but when recording loud sources with high-output capacitor mics 40dB may be more than enough. In practice, 60dB covers most typical studio-recording scenarios, even if more may occasionally be helpful. The minimum gain is also important. When placed close to a loud source (for example, a powerful amp, or a drum), many capacitor mics can generate signals at line levels. If the preamp's minimum gain is 20dB, it could be overloaded. To address this, some preamps include a switchable 20dB pad, though you can also buy separate inline pads.
A preamp's operating bandwidth is linked to its gain. Good designs maintain a consistent frequency response of, say, 10Hz to 30kHz (+0/-3 dB) at all gain settings. But in others the HF response is curtailed at the highest gain settings, and this is rarely revealed in the specs! With lower-cost preamps, the gain also tends to build very slowly at low settings of the control, before delivering large amounts in a great rush towards the maximum. This 'gain bunching' can make fine setting and matching gains between channels frustrating and, again, the specs won't mention this. More expensive designs often use mechanical or electronic switching, which overcomes this.
High-end preamps typically have more headroom than budget designs, which often translates into less distortion and a more 'open', 'effortless' sound character. A high-end preamp might have a maximum output capability of +32 or even +36 dBu, whereas a budget one might manage only +16dBu. Many stand-alone A‑D converters are calibrated to need +24dBu to achieve 0dBFS, and it's not helpful if the preamp runs out of steam well before that level is reached.
At the opposite end of the scale is the noise floor. Most manufacturers specify this in terms of the Equivalent Input Noise (EIN). Every signal source has a defined output impedance, and the resistive element of that generates noise called Johnson noise. For a 150Ω source (typical of most mics), the self-noise figure is -130.9dBu (at 20 degrees C, with a flat measuring bandwidth of 20Hz-20kHz). Given a theoretically perfect, noise-free amplifier, the lowest the EIN can be (for a 150Ω source) is -131dBu, but real amps aren't noise-free, and their noise varies with gain.
The EIN figure can be manipulated to make a preamp look better... So do watch out for unusual or 'massaged' test conditions!
To record the preamp's noise contribution, its input is terminated with 150Ω and the output noise measured, usually at a gain of 60dB. The gain is subtracted from the noise level to derive an EIN figure. If the noise floor measured, say, -65dBu then the 60dB of preamp gain is subtracted to produce an EIN of -125dBu. In practice, preamp EIN values range between about -120 and -129 dBu. The larger this negative figure, the quieter the preamp — I regard -125dBu as the minimum acceptable for professional results. However, the EIN figure can be manipulated to make a preamp look better. Some manufacturers choose a much lower source resistance (or even shorted inputs), or use A‑weighting to discount the noise contribution from the high and low ends of the spectrum, especially if much of the 'noise' comes from the mains-related artifacts of the power supply. So do watch out for unusual or 'massaged' test conditions!
The difference between the noise floor and the headroom limit defines the preamp's dynamic-range capability, and figures typically range from 90 to 130 dB. Other things being equal, bigger numbers are better — while few mics have a range larger than 130dB, it's nice if the preamp can cope with all the mic can deliver.
Mic preamp input impedances are well standardised these days, with most in the 1.5 to 5 kΩ range. The lower figure is around 10 times higher than a nominal mic impedance and ensures adequate 'voltage matching' conditions. Some preamps offer alternative settings, as low as 50 or 30 Ω (to 'impedance-match' vintage designs) or as high as 30kΩ (to minimise the loading on ribbon mics). This setting can affect the tonality of passive dynamic mics, but it rarely affects that of capacitor mics or dynamics with active electronics.
Often, high-end mic preamp specifications include myriad other figures: slew rates, total harmonic and intermodulation distortions, phase responses, common mode rejection ratios (CMRRs), and more. But as they'll rarely print these unless they're proud of the figures, you can probably ignore them. Of these, the CMRR is probably the most important, since it denotes how effectively the preamp can reject external interference — something budget preamps can be poor at doing.