Back in the 1930s, the transition from purely acoustic sound recording to electrical recording and reproduction brought with it a dependency on a source of power.
The electricity grid back then was rather simple and small. Many parts of the world did not have access to electricity, and many commercial activities relied on other sources of power, such as steam. Technological progress resulted in the expansion of the electricity grid and the interconnection and synchronization of multiple power generation stations, and the evolution of grids into gigantic supergrids. The power requirements became much greater, and with great power came great responsibility. The various types of equipment connected to the grid became ever more diverse, as did the various types of generation stations.
Audio amplification (not just power amplifiers, but all amplification devices, including preamplifiers and the active stages and devices in all audio equipment) requires direct current to power components, which then modulate that DC with audio signals. In other words, audio in the electrical domain is simply a modulation of the power supply. More often than not, the quality of the power we feed our equipment with has a direct impact on its performance. This is not only audible, but measurable. And not only does the quality of the power affect our audio electronics, but also equipment such as turntables and tape machines, which use electric motors to transport our music at the correct speed and ideally with no unwanted artifacts.
Since the dawn of electricity generation, due to the nature of rotating machinery used in generating stations, the inevitable variations in load caused by varying consumption throughout the day would cause voltage and frequency variations of the generated alternating current. As grids grew larger, consumption patterns would become less predictable. Some types of loads would produce electromagnetic interference (EMI, not to be confused with the record label of the same name). The widespread application of wireless communications subsequently introduced radio frequency interference (RFI), which affected the power distribution grid.
Then came the switch-mode power supply units in audio (and electronic) components, replacing the older linear power supplies in most applications, and injecting high-frequency noise into the power lines.
More recently, distributed generation models of power production have added an entire set of new challenges.
Distributed generation refers to small-scale power generation, usually by means of renewable energy systems, in a large number of locations. The power is created using wildly different equipment, with some of this generation only producing power during sunny days and other types only kicking in when it is windy. In this modern Wild West of electricity generation, standards are often alarmingly low. Many of these devices produce highly distorted waveforms that are being fed into the grid, spreading distortion products over large distances. Tolerances for voltage and frequency stability have loosened up over the decades to the point where synchronous AC motors (often used to power turntables) will noticeably change the pitch of the music, while remaining entirely within the allowable range of operation of the grid. Voltage variations and distortion are likely to introduce audio distortion, while EMI and RFI are contributing to the noise floor.
At this point, I should perhaps take the time to point out that I am entirely in favor of renewable energy systems and believe them to be the direction we should be moving towards. My own home and my business are entirely powered by renewables. All of my listening to music and even my recording/mastering work is done with renewable energy systems providing the power. At the same time, I believe that the way renewables are currently implemented in the distributed grid generation model is neither the most efficient, nor the most environmentally friendly use of this technology. From an engineering standpoint, there are far better ways of transitioning towards renewable energy, but as with most things, best engineering practice is unfortunately not ranking very high in decision-making processes.
Raw power: the solar panels at Agnew Analog Reference Instruments.
In short, the current state of the electricity grids around the world renders them entirely unsuitable for powering any kind of audio equipment, if sound quality is of any concern. The same applies to video equipment, laboratory measurement instruments and any other kind of critical infrastructure that could be negatively affected by power of questionable specifications. In professional environments where equipment in the aforementioned categories is used, it has long been common practice to not power the sensitive equipment directly from the electricity grid.
Even in domestic environments, power regeneration or conditioning equipment is often found in audio and home cinema system applications. An example of this type of regeneration equipment is the PowerPlant range by Copper publisher PS Audio. The largest offering, the P20, is capable of delivering 3,600 watts of peak power, clean and free from the disturbances of the grid, by internally generating a 50 Hz or 60 Hz sine wave (depending on geographic location), unaffected by grid conditions. Such devices are highly effective against voltage and frequency fluctuations of the grid and shield against distortion, EMI and RFI. They are also encountered in professional environments, but there are still two types of disturbances they are unable to offer insurance against.
PS Audio PowerPlant 20 power regenerator.
One of these is power outages, an increasingly common occurrence nowadays in many parts of the world. These can occur due to poor maintenance of the grid, faults in the generating equipment, severe weather, disruptions in the fuel supply chain or even preventative outages (brownouts or blackouts) when the demand exceeds the capabilities of the grid or the available fuel. In one example I personally documented, a grid operator in Europe used a tactic of regular but brief power outages, lasting approximately 10 seconds. Such a power outage would trigger the automated protection systems of industrial machinery, heat pumps and air conditioning systems, which typically remain off for a period of 10 minutes following a power outage to protect the systems against damage from repetitive stop/start conditions. This gained the operator valuable time, which allowed the inadequate transformers used in this particular power grid to cool down with the low draw of the 10-minute off time following the power cut. This kept them going for a few years, preventing bankruptcy or widespread infrastructure damage, without having to upgrade the grid infrastructure.
One of the battery banks at Agnew Analog. The tubes visible are part of an automated electrolyte replenishing system.
While a brief power outage is not of major concern in a domestic environment, unless there are health issues and life support equipment involved, the situation can become much more complicated in a sound recording or disk mastering facility. Should the power fail in the middle of an outstanding performance that is being recorded, that performance would be lost forever. The next attempt at capturing that performance may not be as good, and it may be extremely difficult to get that group of musicians together again on a separate occasion. Even worse, if the power goes out halfway through cutting a record, there is a very high risk of major equipment damage, on top of the wasted lacquer disk that would have to be discarded.
In professional environments, in addition to power regeneration equipment we also use systems that can keep the power running even in the event of a grid outage. Such systems have traditionally used grid power, when available, to charge a massive high-voltage battery bank, usually the size of a big refrigerator or three. The batteries offer storage and supply direct current (DC) to an alternating current (AC) generator, which generates the clean power that will run the sensitive equipment. The capacity of the battery bank will define for how long the facility can remain operational in the absence of grid power.
These systems are highly effective in providing clean, stable power and can additionally protect against grid outages, making them a very popular feature in the more serious audio/visual facilities and laboratories. However, there are two risks that the aforementioned systems cannot offer any protection against: long-term grid outages, and extreme inflation in energy prices.
Over the past couple of years, I have designed and commissioned the first few professional audio facilities where the charging of the battery banks is accomplished without the grid, by using solar and wind power, along with a backup generator for emergencies if needed.
Severe weather phenomena and natural disasters in some parts of the world are known to cause such extensive damage to centralized grid infrastructure that it often takes months for the repairs to be completed and the grid to become operational again.
The economic consequences of a business being forced to remain closed for several months due to lack of power can be catastrophic. Similarly, instability in energy prices can throw businesses off their business plan calculations, often with severe repercussions.
In times of economic and political instability, the option of generating power on-site, with a known and predictable upfront investment, is often deemed preferable to unpredictable energy costs for the business. Especially considering the requirement for controlled temperature and humidity conditions in a large recording facility, the energy expenditure can be astronomical. It is common that by far the largest contribution to the running expenses of a professional recording facility are the energy costs for the HVAC system and lighting.
In all of these cases, the power reaching the audio equipment is generated on-site and is often referred to as technical power, remaining separate and isolated from non-audio systems, which may also be supported by battery banks, but the electrical supplies are kept separate to ensure that nothing unwanted reaches the audio power system. In a well-designed system, THD+N is typically under 2 percent (at the Agnew Analog Precision Engineering Laboratory, the technical power has a measured THD+N of 1.3 percent), whereas the amount of THD+N on the grid is typically in the double digits in some parts of the world. The voltage and frequency regulation offered by our custom-built gear are at least an order of magnitude better than typical grid conditions measured at the location of various recording and mastering facilities around the world. Immunity from power outages and inflation keeps these facilities operational against all odds.
The quality of the power reaching our equipment is of utmost importance if we expect it to perform at its fullest potential, both in the professional world and at home.. In one of the isolated rural locations where I had lived in the past, the grid voltage measured 203 VAC in a 230 VAC region. It would drop as low as 175 VAC and I recorded peaks as high as 275 VAC. Lightning strikes on power lines were known to take out appliances on a regular basis. I simply didn't dare to plug anything of significant value to that grid! The power reaching your audio system is one of the most important components of that system.
Header image: solar panels on one of the rooftops at Agnew Analog Headquarters. All images courtesy of Agnew Analog Reference Instruments (except the PowerPlant 20 image, courtesy of PS Audio).
