British Aircrew Headset Impedance Testing
Anyone who's ever looked into this will know it's a bit of a headache to understand what's meant by terms like impedance matching, ohms (Ω), DC resistance, and the difference made by wiring speakers in series or parallel.
What is impedance?
To quote Wikipedia:
Electrical impedance is the measure of the opposition that a circuit presents to a current when a voltage is applied.
While this statement is not a million miles away from direct current (DC) resistance, is also takes into account the effects of devices in a circuit on alternating currents (AC), which will vary at different frequencies.
In the context of aviation headsets, and other speakers alike, the electrical signals sent to the headset speakers are AC, and the coil of the speaker presents an opposition to the flow of current which can vary with AC frequency due to speaker design. So the term AC impedance (or just impedance) is relevant to speakers, not DC resistance.
How much does it matter?
Aircraft radio sets are designed to operate with headsets within a certain range of impedances. There can be a fairly wide range of allowable impedances as radio / intercom systems will be designed to support either just one headset, or several headsets. By connecting two headsets in parallel, you halve the overall impedance, connecting four headsets results in a quarter of the impedance of a single headset being presented to the radio / intercom.
For example, the amplifiers types A1961 and A1961M are designed to be able to operate between 1 and 10 headsets before running into difficulties. Where 1 headset = 150Ω, and 10 headsets in parallel = 1.5Ω so we can see the amplifier has an output impedance range of 1.5Ω to 150Ω.
If a headset impedance presented to the radio or intercom amplifier is higher than the designed range, this could lead to distorted sound quality as the amplifier circuit runs out of voltage to drive adequate current into the speaker, and "clipping" can occur to the top and bottom of the waveform. If the headset impedance is significantly lower than the radio or intercom amplifier's design, it could end up damaging the amplifier circuit by drawing too much current from it.
The other issue is that if in a multi-seat aircraft a mix of headset impedances is used, a radio volume level that may sound loud to one person, may be too quiet for another person (but other factors will also have involvement here such as speaker sensitivity and efficiency.
In a nutshell, you should always try to use headset speakers of an impedance that the radio or intercom amplifier has been designed for, and try not to mix and match headset types.
Motivation
As mentioned earlier, impedance isn't a fixed figure, but varies with the frequency of the AC signal applied. Household speaker manufacturers usually quote speaker impedance as measured at 1kHz, but the nominal impedances I've found for aviation headset impedances don't always state the frequency measured at, so I set about measuring the headset speaker impedance of real headsets and flying helmets, and how they behaved at different frequencies, to see if they are similar or show different patterns.
Impedance testing
As we're dealing with varying performance with AC frequencies, checking with a multimeter set to Ω isn't going to help us greatly, instead we have to devise a way of taking measurements conducive to figuring out the impedance at various frequencies.
The human ear can generally hear sounds between 20Hz and 20kHz (decreasing with age), so this defined the range I was interested in testing over.
Test equipment
The test apparatus is fairly straightforward: an AC signal generator (variable frequency sine wave); a shunt resistor to help measure the current drawn by the headset; and an AC voltmeter, capable of very accurate small AC voltage measurements.
Methodology
The end result of impedance for a given frequency is to be calcualted by using ohms law, taking the voltage across the speakers and the current through them. The following guide shows how to do it.
Add the shunt resistor in series with the AC output connection of the signal generator, and connect the headset speakers under test between the free resistor leg and the ground / return to the signal generator. As the headset draws current from the signal generator via the shunt resistor, a voltge develops across the shunt resistor directly proportional to the current. We can use the AC voltmeter to measure the voltage developed across the shunt resistor, and knowing the resistance of this resistor (measure with multimeter for better accuracy than the marings), we can calculate the current flow using Ohms law I=V/R, and record.
Without touching the signal generator frequency or voltage, we then measure the voltage across the speaker coil(s), and record. Re-arranging Ohms law to R=V/I, we can then calculate the speaker impedance at the frequency applied.
Repeat the above steps at various frequencies, and plot the results on a graph of Frequency Hz (x axis) against Impedance Ω (y axis).
Test results
Testing of a standard headset used in a WW2 RAF C-Type Wired / D-Type Wired / E-Type Wired helmets, which have two 10A/13466 speakers connected in series, compared to two speakers from a Mk4 helmet connected (unconventioally to attain a higher impedance) in series, and for comparison, a David Clarke General Aviation H10-30 headset.
The results were actually quite surprising... The military headset speakers behaved exactly as anticipated, however the David Clarke headset have a much shallower increase in impedance against frequency, but also surprised me as they are 150Ω, the same as the standard military headset at 1kHz. Numerous websites and GA radio and intercom datasheets state 600Ω as the standard GA headset impedance, so why is David Clarke selling headsets with a lower impedance?
I believe the answer lies in the fact that the David Clarke headset comes with a volume control which simply adds a variable resistance in circuit with the headset. When the volume is turned up (as in this experiment) the resistor is effectively out of the circuit, and we see the true headset impedance. With the volume turned down, resistance is introduced, which offsets the impedance upward to roughly 1000Ω. So with the volume control set roughly in the middle range, the David Clarke headset will present the radio / intercom with roughly a 600Ω impedance (comprised of both inductive and resistive elements).