TRANSMITTER INSET No. 21

BT Maintenance Information No. 030 (June 1984)

INTRODUCTION
This Maintenance Information paper gives general information on the various types of Microphone Inset No. 21 currently available, their applications and some of the problems that may occur under certain circumstances. There are Three manufacturers currently producing Microphone Insets No. 21, these may be identified by case design and manufacturers code as shown below.

          A.P. BESSON                                       GEC                            RATHDOWN INDUSTRIES

MANUFACTURERS
A.P. Besson Ltd. - (manufacturers code DAE).

G.E.C. Ltd. - (manufacturers code GEN).

Rathdown Industries Ltd  - (manufacturers code RIA).

SUPPLIER
BTE Consumer Products.

EQUIPMENT
Microphone Insets No. 21 are designed for use in telephone instruments that normally use Transmitter Insets 16, they incorporate Non-Carbon microphone units for improved transmission and reliability. There are three manufacturers versions currently available, all three have been coded `Microphone Insets No. 21A'.

The A.P. Besson and GEC models each contain an electret microphone capsule with line powered, Integrated Circuit amplifier. Rathdown Industries models contain a moving coil microphone insert with line powered, transistor amplifier, the only difference between the two types of Rathdown Industries Microphones is that the red bodied model contains an extra capacitor intended for Radio Frequency Interference suppression.

The Rathdown Industries model with the RFI capacitor will be separately available from April 1984 onwards as the `Microphone Inset No. 21B and will have a grey coloured body.

FIELD OF USE
Suitable for use in 700 type and other telephones that normally use Transmitter-Inset No. 16.

Transmission problems can arise however if telephones with Microphone Inset No. 21 are mixed with carbon microphone instruments on the arrangements shown below:-

• Simple plan arrangements where customers are in the habit of conducting three party conversations or answering a call at one extension and continuing it at. another.

• Extension to Extension calls or Operator to Extension calls on small PMBXs with non-divided transmission feed (e.g. 2/PMBXs)

• Intercom calls on Extension Plan 105/107 arrangements.

In these arrangements the telephones are connected in parallel and transmission current must divide equally to energise both instruments microphones/transmitters.  The line powered amplifier in the Microphone Inset No. 21 presents a higher impedance to line than a carbon microphone and requires approximately 4 Volts to 'switch on'.  When a carbon microphone telephone is used in parallel with a Microphone Inset No. 21 telephone, most of the current will pass through the carbon microphone causing distortion or transmission cut off from the Microphone Inset No. 21.  The problem may be overcome by connecting two 4.7 Volt Zener diodes (e.g. CV 7409) anode to anode in series with any carbon microphones on the system.  This will enable current to divide more equally, energising the transmission circuits of both instruments.

KNOWN PROBLEMS
There were problems with early production units made by A.P. Besson, these resulted in terminal posts coming loose from the circuit board and transmission cut-off on line reversal.  Quality control has been improved to overcome these problems.  GEC models were found to hum if used near mains powered equipment, these models (marked `S' on the front cover) have now been modified to overcome this.

All types of Microphone Inset No. 21 may be affected by radio frequency interference (RFI), this may be overcome by changing for a different type of Microphone Inset No. 21 or as a last resort, by replacing with a Transmitter-Inset No. 16.

Certain types of press button telephone rely on a D.C. path to energise the keypad and provide a transmission path, these are telephones marked:-

Telephone No's 3/756 GNA 79/1, or 756 GNA 79/4, or 756 GNA 80/4.

Whenever a Microphone Inset No. 21 is fitted in one of these instruments a 1,000 ohm resistance e.g. a Resistor No. 91 EF, 1K should be fitted to telephone terminals T3 and T10 (i.e. across the Mic.) to provide a D.C. path.

Bessons or GEC type Microphone Insets may cause `High Side Tone' when fitted in Herald or Ensign terminals, Microphone Insets No. 21B or Transmitter Insets 16 may be used in these instruments instead.

CONNECTIONS
Microphone Insets No. 21 have screw terminals suitable for terminating 0.25 in. spade terminals.

SCHEMATIC DIAGRAM

POWER REQUIREMENTS
Microphone Insets No. 21 incorporate line powered amplifier circuits that requires approximately 4 Volts to `switch on'.  They are designed to work satisfactorily from 100mA down to 20mA Line current and with possible degraded transmission between 20mA and 10mA.

SPARES
Microphone Inset No. 21A (Item Code 436189)
Microphone Inset No. 21B (Item Code 877034)

LINE TESTING
Telephones with Microphone Insets No. 21 fitted may result in high resistance telephone loop tests, this is due to the Line powered amplifier circuitry of the microphone not switching on at low voltages.

Transmitter Inset No. 21
Each variant opened up

A.P. BESSON                                      GEC                           RATHDOWN INDUSTRIES

An extract from
Post Office Electrical Engineers Journal
Volume 72, April 1979

The Electret: A Possible Replacement for the Carbon Microphone

The British Post Office (BPO), in common with other administrations, has for some time been seeking an alternative to the carbon-granule microphone for use in telephones. This article describes the design, construction and performance of an electret microphone developed in the BPO Research Department as a possible drop-in replacement for the Transmitter Inset No. 16.

INTRODUCTION
The Rev. Henry Hunnings of Yorkshire is generally credited with the first patent, taken out in 1878, for a carbon-granule microphone. The British Post Office (BPO) Transmitter Inset No. 1, shown in Fig. 1, was a development of the Hunnings transmitter1. The critical feature is its solid back; that is, the back electrode is flush with the back of the granule chamber. Thus, when the microphone is turned on its face there is a possibility that none of the granules will be in contact with the electrode and the microphone is open circuit. In the early part of this century there were designs of telephone in use that had handsets and used solid-back transmitters (the Edwardian instruments that are becoming fashionable again). The fact that the microphone may have been an open circuit while the handset was being picked up was of little consequence since the microphone did not form part of the signalling loop and obtained its feed current for transmission from a local battery. Signalling was, of course, provided by a magneto.

When systems that used a central battery for signalling and transmission came into use, it was necessary to use a telephone in which the microphone was likely to remain in a near-vertical plane throughout the duration of a call. The microphone obtained its feed current from the central battery and also provided the loop across the line. This led to the almost universal use of "candlestick" telephones in the 1920s.

FIG. 1 - Solid-back transmitter (BPO Transmitter Inset No. 1)

In 1929, a number of sources came up with the same solution to the problem of disconnexions: both electrodes of the microphone were immersed in the carbon granules. Fig. 2 shows a design, by Messrs. Siemens, from which the BPO Inset No. 10 was derived2. In this design, the granule chamber surrounds the electrodes. This simple change made it possible to revert to telephones with handsets. There were detailed changes over the years, but basically the design of carbon microphone used by the BPO did not change until the Inset No. 16 was introduced in 1964 (see Fig. 3) and, in slightly modified form, this is the item that is still in use today. The principle change from earlier designs that is immediately apparent is the use of hemispherical electrodes enclosing a granule chamber that is about 8 granules deep. In fact, a vast amount of theoretical study and practical experience went into the design of this microphone, which resulted in significantly improved performance.

FIG. 2 - Siemens transmitter - 1929

FIG. 3 - Transmitter Inset No. 16, Mark 1

Although the microphonic action of granular carbon has been the subject of study for as long as anyone can remember, the present position is that no one theory of operation has gained universal acceptance. Also, full explanations do not appear to be available for the various degradations in performance to which carbon microphones are prone: increase in noise, loss of sensitivity, and change of resistance. The authors can only offer their personal opinion that the full electrochemistry of carbon granules has yet to be discovered.

PERFORMANCE OF CARBON MICROPHONES
Over the years, many articles have been published, in this Journal and elsewhere, on the performance of carbon microphones. By the nature of its operation, a variable-resistance microphone causes non-linear distortion, and microphonic action of the carbon adds its own contribution to this. Non-linear distortion has become an accepted part of the performance of carbon microphones, but there are other characteristics that cause more trouble. The sensitivity increases with increasing feed current, therefore the microphone gives greater output on short lines. There is a wide spread to all the characteristics, which also change with time. Finally, the microphone generates noise and can become increasingly noisy with age and use, although specimens in carefully controlled experiments have also been known to become less noisy with use.

It is perhaps too well known that carbon microphones cause noise. This makes it difficult to be sure of microphone fault statistics. A customer may complain of noise on the telephone. When the complaint is investigated, the noise is not found; it has cured itself. The source of this intermittent noise could have been in the exchange, or in the line plant, or it could have been in the telephone. So the prudent maintenance engineer changes the microphone. Why not? The item is cheap, much cheaper than the cost of another maintenance visit.

But, with all these disadvantages, the carbon microphone is still in almost universal use because it is the only suitable microphone that does not require an amplifier in the telephone. It is cheap and it is robust. Yet the BPO, in common with most other administrations, has for some time been seeking a replacement for the carbon microphone.

The replacement must operate satisfactorily for very long periods, sometimes under arduous conditions of temperature and humidity, and it must include an amplifier. It is this last condition that has been the stumbling block because there does not appear to be any way of making a microphone plus amplifier for the same price as a carbon microphone.

The first step is to distinguish two different applications for linear microphones in telephones. Firstly, the drop-in replace-merit with a linear microphone and its amplifier in a similar case to that of the particular carbon microphone in use: this can be used to replace existing microphones in telephones, and to equip new production, without any other alteration to the telephone or to the rest of the system. Secondly, the use of a linear microphone in completely new designs of electronic telephone containing integrated circuits (ICs) that perform many other functions.

In the second case, the economics of a linear microphone may be fairly straightforward because the necessary amplifier can form part of one of the ICs at little extra cost. Moreover, a carbon microphone is likely to be out of place in an electronic environment. But the drop-in replacement can be difficult to justify until account is taken of the maintenance costs incurred by carbon microphones. There is, too, an unquantifiable benefit to be gained from increased customer satisfaction from improved transmission performance and fewer faults.
So, a replacement for the carbon microphone may be justifiable, provided that the replacement is cheap and highly reliable.

CHOICE OF A NEW MICROPHONE
There is no perfect microphone for telephony. The feasibility of designs employing electrodynamic (moving coil and moving iron), piezo-electric and electret elements has been demonstrated. A particular design using an electret is described in this article, both because the item has been developed by the BPO Research Department and because its principles may be less generally understood. Before giving that description, however, it should be mentioned that there is another new microphone just coming out of the research phase in many laboratories; this device uses a piezo-electric foil as a diaphragm. The foil is Polyvinylidene Difluoride (PVDF). It may prove more suitable than electrets for other forms of transducer (receivers, keypads) and, so far, no failure mechanism has been discovered for it.

But, to return to the electret; its merits as a microphone include:-

1. simplicity of construction,

2. low cost,

3. relatively high output,

4. low diaphragm mass, and therefore insensitivity to handling noises on telephone handsets,

5. easy control of parameters, and

6. mechanical robustness.

THE ELECTRET MICROPHONE
An electret is a material which, after receiving an electrical charge, retains that charge (or a proportion of it). The term was coined many years ago by Oliver Heaviside, who saw the properties of an electret as being analogous to those of a permanent magnet, despite the fact that no practical electrets could at that time be demonstrated.

Electrets can now be made using polymer films only a few micrometres thick, ideally suited for use as a microphone diaphragm. The simplified cross-sectional diagram, Fig. 4, shows that a microphone using the electret principle is essentially a very simple device comprising a metallised electret diaphragm located a defined distance (in this instance 70um) in front of a conducting backplate. in operation it closely resembles a capacitor microphone, but without the need for an external polarizing voltage.

Operating Principles
The electret charge, qe, has associated with it a capacitance to ground, Ce, and an equivalent polarizing voltage, Ve, related according to the equation

Ve = qe/Ce.

The proximity of the electret film to the backplate results in a modification of capacitance Ce. determined by the distance d between the backplate and the electret film. Flexure of the electret film, which is also the microphone diaphragm, results in the value of C. changing in sympathy with the movement of the diaphragm, which is of the order of nanometres. There is a corresponding change in voltage Ve, which appears as an EMF in series with a capacitance. It is necessary to pre-stress the diaphragm in order to oppose the electrostatic attraction which exists between it and the back plate. A back volume of air is also included to provide compliance, against which the diaphragm acts under the influence of a sound wave.

Since the charge and the capacitance are both proportional to the area of the diaphragm, Ve and changes in Ve are, to a first order, independent of the diaphragm area.

Microphone Construction
The success of an electret microphone hinges on its design and the assembly techniques employed, and of paramount importance is the electret itself. There are, at present, 4 main methods of charging an electret.

The Thermo-Charging Process
In the thermo-charging process, the film is placed between two electrodes and heated almost to softening point as the voltage across the electrodes is increased. This is followed by a controlled cooling cycle.

The Electron-Beam Method
In the electron-beam method, the film is charged by electron bombardment.

The Corona-Discharge Process
In the corona-discharge process, the film is subjected to corona from appropriately shaped electrodes.

The Knife-Edge Method
In the knife-edge method, the film passes in intimate contact with a knife edge. This injects charge directly by virtue of the high electric field existing across the film.

Evidence to date suggests that all of these methods give good results, but research is continuing at the BPO Research Centre and elsewhere into the physics of electrets in order to produce a better product. The BPO has chosen to use the knife-edge charging method because it can be carried out continuously at room temperature and requires no special conditions, such as high vacuum. Fig. 5 shows the uncharged film from the supply spool being passed slowly over the knife edge, which is maintained at a potential of about 700V and then taken up on the take-up spool drum. The film is 12.7um thick FEP Teflon, aluminized on one surface. The design of the charging machine is such that the aluminized surface is kept in contact with earthed rollers, and on take-up the charged surface comes into contact with the earthed aluminized surface. This allows any unwanted electrostatic charge to leak away, leaving only the electret charge; the potential is now about 250V.

FIG. 5 - Block diagram of knife-edge charging apparatus

The construction of the microphone achieves repeatable dimensional and electro-acoustic performance, long-term stability and ease of manufacture. Of particular interest is the method of locating and protecting the electret film. The metallised side of the film is cemented onto a brass ring using a Cyanoacrylate adhesive and is held against the backplate by the fitting of the conductive body moulding. The backplate is moulded from conductive polypropylene and has on its front face a number of small pimples and an annular knife edge, both of very closely controlled dimensions. The knife edge, in conjunction with the brass ring, ensures uniform tensioning of the film, and the pimples serve to maintain a constant distance between the film and the backplate.

Studies into the behaviour of electret materials indicate that, if protection from extremes of heat and humidity is provided, the lifetime can be very long indeed. In practical terms, only the humidity need be considered because the upper temperature ranges necessary to shorten the life would result in damage to other parts of the telephone. Effective protection has been incorporated in a simple and inexpensive manner using a Melinex moisture-barrier in front of the aluminized surface of the diaphragm.

Microphone Size
Electret microphones can be made quite small and most of the work by the BPO Research Department has been concentrated on designs of 15mm diameter. There is, however, some evidence to suggest that an increase in size to about 25mm diameter could be advantageous because this gives higher capacitance and, therefore, lower impedance; increasing the size of the electret does not necessarily increase its sensitivity.

FIG. 6 - Electret microphone (drop-in replacement for Transmitter No. 16)

Microphone Amplifier
The output of the electret microphone is some 20dB below that of a typical carbon-granule microphone and so compensating amplification has to be provided. The impedance of the microphone demands the use of an amplifier with an input impedance of the order of 10Mohms. This is most readily achieved using a field-effect transistor (FET) input stage, although bipolar designs can be used. The input stage is followed by further amplification, which can form part of a comprehensive electronic transmission circuit in a telephone. Alternatively, the complete unit of microphone plus amplifier can be made as a physical and electrical drop-in replacement for a carbon microphone for use in existing designs of telephone. Performance details presented here refer to a microphone of this type.

A complete microphone, made up as a drop-in replacement for the Transmitter Inset No. 16, is shown in Fig. 6. There are two points of special interest in the construction of the microphone: the microphone body moulding is made of conductive polypropylene, and forms one connexion to the electret; the amplifier uses discrete components rather than an I.C.

During the course of research and development on this microphone, the construction of the amplifier and its associated components (poling bridge, surge protection) has come full circle. It started in discrete components; then an IC was specially designed and constructed at the BPO Research Department for the application. But it was not possible to include some of the larger capacitors and higher-dissipation resistors on the chip, and it became apparent that the entire circuit was more expensive than the original breadboard. So thick-film circuits were investigated and models made. But, in the end, the design reverted to discrete components because this form of construction appeared to offer the least costly solution.

If this microphone is adopted for large-scale production it may well be that alternative forms of construction will be adopted for the amplifier.

Performance
One of the attractions of the electret microphone is that it is relatively easy to control the frequency response characteristic, by both electro-acoustic and electrical means. The frequency response curve shown in Fig. 7 is the mean, with standard deviations superimposed, of six drop-in replacement electret microphones measured in a standard telephone handset in accordance with internationally agreed standards. The limits to which these microphones were designed, which differ slightly from those currently employed, are also shown.

FIG. 7 - Sensitivity/frequency-characteristic of the Mark 3 electret microphone

FIG. 8 - Harmonic distortion produced by carbon and electret microphones

In addition to the good control of frequency response, distortion is greatly reduced by using an electret in place of a carbon microphone, as is shown in Fig. 8. Other characteristics of this microphone are given below:-

1. Noise level: -80dB (ref 1V psophometrically weighted).

2. Operating current range: 10-110mA.

3. Output impedance: 80ohms (nominal at 1kHz).

LONG-TERM STABILITY
As mentioned, it is of paramount importance that any replacement for the carbon microphone has extremely good reliability. The design described here has been subjected to extensive and severe environmental tests and has proved entirely satisfactory.

CONCLUSION
The move away from carbon microphones in telephones is gathering momentum throughout the world. Designs are now available that offer more stable characteristics, less distortion and improved reliability; the latter is the vital ingredient in a cost-effective solution. For the immediate future the electret is a strong contender. It remains to be seen whether, in the longer term, PVDF foil microphones offer significant advantages.