The ribbon microphone relies on the phenomenon that an electrical voltage is generated when a wire is moved in a magnetic field. The most efficient way to make use of this effect is to design a system of magnets, electrical conductors, and a mechanical force to move the conductor, so everything is at right angles. (If you are having recollections of the physics "right hand rule" at this point, then you are on the right track). Such an arrangement for use in a ribbon microphone is shown in Figure 1. Everything is at right angles: the magnetic field is from left to right, the electrical conductor (in the form of a freely suspended ribbon) is from top to bottom, and the mechanical force into the page.
An alternating mechanical force, in the form of sound waves impinges on the ribbon and makes it move in and out of the page. And the voltage developed at either end of the ribbon is directly proportional to the velocity of the ribbon in the gap. Double the speed of the ribbon, and you double it's output. If you double the strength of the magnetic field, or the length of the ribbon in the field, then the output is also doubled.
At low frequencies the ribbon moves very easily. In fact the ribbon, which is usually an aluminium alloy so light that a slight blast from a vocalist's breath (a signal close to DC) can "pop" it right out of the gap, destroying the microphone. As the frequency increases, the ribbon needs more and more force to move it back an forth at the same velocity. So, as you can imagine, it is very important for the ribbon to be made as light as possible to get good treble response. Wrong - the previous sentence is wrong, wrong, wrong.
The mass and thickness of the ribbon has no effect on the high frequency response. Normally the ribbon is made from an aluminium alloy (eg: duralumin), about 2 to 4 microns thick (0.002mm to 0.004mm). But you could changed this to 10 microns of duralumin, or even 20 microns of brass and get the same frequency response (excluding minor effects from internal damping properties of the material used).
I found this to be an astounding feature of ribbon microphone design, as it went against my intuition. And it is the root cause of the technical beauty I referred to before.
It is true however, that as the frequency of the impinging sound waves is increased, the force required to act on the ribbon (for a constant output voltage) increases. Every time the frequency is doubled (ie: every octave), twice the force is required, regardless of the mass of the ribbon. If the applied force was not doubled every octave, then the microphone would have great bass, with a frequency response falling at 6dB per octave as the frequency increaes (ie: the output would halve for each octave increase in sound frequency).
Where does this increase in force come from? The short answer is that it comes about quite naturally, with no effort at all on our behalf, up to an upper frequency limit.
The force that moves the ribbon comes about from differences in sound pressure between the front and the back of the ribbon. The sound waves arriving at the front of the ribbon will push the ribbon back, by the inherent pressure in the sound wave. Likewise, the same sound wave, moments later, arrives at the back of the microphone, this time pushing the back of the ribbon forward.
At a glance it might be thought that the two forces would exactly cancel, and the ribbon would go no-where, stopped in its tracks by two equal and opposing forces. But with the sound-source located on-axis to the ribbon, the two forces are slightly unequal. For example, with a sine-wave sound source, as the pressure reaches its maximum on the front of the microphone, the pressure waveform acting on the back of the ribbon is still rising, as it is lagging in time by some micro-seconds (due to the longer path around to the back). Thus, in the example above, the total pressure would result in the ribbon being pushed back at this instant in time.
The above illustrates where the force actuating the ribbon comes from - it comes from the difference in pressure between back and front. (This is why ribbon microphones are sometimes referred to as pressure-gradient microphones. I am told that if one cares to study this further - or even take a short-cut and ask Jim Meandue - the velocity of the air in a sound wave is directly proportional to the pressure-gradient or pressure-difference, and this is where the more popular term "velocity microphone" comes from, as the output voltage is determined by the air velocity in the sound wave.)
Now we are close to finding the source of the doubling-in-force required for a flat frequency response.
The path length from front to back is a fixed dimension, in inches if you like, determined by the size of various objects in the microphone. But in an acoustic sense it halves every octave. For example, consider a design with a physical path length of 1- inch from the front to back of the ribbon. At 1kHz this is about one-tenth of a wavelength. At 2kHz this becomes two-tenth's of a wavelength, even though it is still 1- inch. This means that at 1kHz, the pressure wave at the back is one-tenth of a wavelength behind that at the front. But at 2kHz, the pressure wave at the back is two- tenth's of a wavelength behind. Shown in Figure 2, it can be seen that the doubling of frequency, results in a doubling of the acoustic path, which in turn doubles the net pressure on the ribbon. So the force on the ribbon doubles each octave, giving the microphone a ruler-flat response from about 20Hz to around 10kHz.
What determines the top frequency limit of the microphone then? As the frequency in the example above increases, a frequency is reached at which the acoustic path length from front to back is one-whole wavelength. With the 1-inch physical dimension quoted, this would occur at 10kHz. Then the pressure wave arriving at the back of the microphone is 360 degrees out of phase, and the microphone output drops to zero. Before this frequency is reached, the response starts to fall. In this case, the 1-inch path difference is rather too big to give high fidelity results, and the output would be starting to fall at 5kHz, with a null at 10kHz.
More typical designs would have a -3dB point between 10kHz and 20kHz. This is achieved by making the path-difference from the front of the ribbon to the back of the ribbon as small as practical. This is where physics no longer has beauty, and the designer is boxed in by conflicting requirements.
To make a sensitive microphone (one that gives adequate output voltage) the designer requires a large magnetic field, and a light ribbon. Halving the ribbon mass potentially doubles the output voltage, but as we have seen, it does not affect frequency response. The large magnetic field can be satisfied by a big magnet. But to get the magnetic flux to flow across the face of the ribbon, large, thick pole pieces of soft iron are needed. Because the magnetic poles pieces are adjacent to the ribbon, they represnet an obstacle for the sound waves to travel around on the journey from the front of the ribbon to the back. This means they contribute to additional path-length, and so directly contradict the requirements for a a wide bandwidth mircophone.
Some practical magnet configurations for classic RCA ribbon microphones like the RCA-44 and RCA-77 are shown in Figure 3.
In one case, flux enters slender pole pieces from both top and bottom so that sensitivity and frequency response targets are met. In another design, the magnet at the bottom has its circuit completed by pole pieces that resemble "flying buttresses". The apertures in the buttress or pole pieces allow sound waves a short path from front to back, while still retaining magnetic efficiency. A more modern design using rare-earth magnets to eliminate pole pieces altogether and obtain wide bandwidth is shown in Figure 4.
One other attractive feature of ribbon velocity microphones is the figure-of-eight sensitivity pattern. If the sound source is located at 90 degrees (in either the vertical or horizontal plane) to the main lobe, then the sound arrives simultaneously at the back and the front of the ribbon – so there is no pressure difference, and the microphone output drops to zero. As mentioned, this was put to use in the movie industry, and A.D. Blumlein recognised is significance for co-incident microphone stereo recording techniques. Other microphone types can also achieve the figure of eight pattern, although not with the same frequency-independent accuracy as a ribbon microphone. In the critical horizontal plane, a ribbon type can exhibit the same polar pattern at 10kHz as that at 100Hz (mainly because the ribbon can be easily made so that it's width is less than one wavelenth at 20kHz) where as capacitor types often show a sharpening in the sensitivity lobe at 10kHz (making the stereo imaging frequency dependant).
Harry Olson of RCA pioneered the development of high quality ribbon microphones from the late 1920's onwards. The work was driven by requirements for movies with sound. The attractive figure-of eight directional properties of velocity ribbon design allowed mechanically noisy movie cameras to be positioned in the null of the microphone pattern, and actors placed in the sensitive forward lobe.
From the early 1930's flat response to 10kHz was achieved by Olson's RCA designs. Unfortunately, it would be the late 1940's before recording systems using tape could record such bandwidths at low distortion and high signal to noise ratio.
By 1950, the Neumann U47 capacitor microphone was becoming recognised in the industry as the microphone of choice. Despite it's non-flat rising response above 5kHz, the U47 swept aside the ruler-flat RCA designs rapidly during the 1950's.
Other famous ribbon microphones were made by Shure, Western Electric, Beyer, STC, Tannoy, Marconi and locally Zephr Products (the latter two appearing similar to the classic 44-BX design from RCA).
Ribbon designs are out of favour these days, except amongst a few cult-enthusiasts. There are some modern recordings that have used classic ribbon designs, one example being "In the Digital Mood" on GRP. Short of this, try some of the classic Capitol Collectors Series discs to get your fill of ribbon-microphone sound.
© Steve Spicer, July 1999 Steve Spicer
|This is a framed page so if you can't see a Contents page to the left then you need to Click Here in order to navigate this site.|