GB2459530A

GB2459530A - Transducer with an over-voltage protection circuit and voltage regulator

Info

Publication number: GB2459530A
Application number: GB0807682A
Authority: GB
Inventors: Ashley Michael Rose; Stephen Hughes
Original assignee: Penny and Giles Controls Ltd
Current assignee: Penny and Giles Controls Ltd
Priority date: 2008-04-28
Filing date: 2008-04-28
Publication date: 2009-11-04
Also published as: GB0807682D0

Abstract

A transducer comprises a sensor with a sensor input voltage and a sensor output signal, an input protection circuit receiving a supply voltage 3 and operable to prevent the application of excessive voltages and possibly negative supply voltages to the transducer, a voltage regulator 34 which regulates the supply to the sensor input 20 at a predetermined voltage during excessive or negative voltage supply; and an output for providing the transducer output signal. Preferably the input protection circuit comprises a pair of transistors 14,16 (e.g. MOSFETS) and a voltage sensing arrangement 26 to deactivate the transistors in the event of excessive voltage. Additionally the circuit may feature a linearization circuit configured to linearise the output signal from the transducer. The sensor may also be programmable, in this case there may be an over-ride input configured upon application of an over-ride voltage to disconnect the sensor from the input voltage and but allow a microcontroller still connected to the input voltage to program the sensor.

Description

Transducer

Field of the Invention

The present invention relates to a transducer. More particularly the present invention relates to a transducer having circuitry for conditioning input and output signals of a sensor in the transducer. The invention is particularly, but not exclusively, concerned with a transducer for position sensing.

Bacicg round Many transducers, including many position-sensing transducers such as HaM-effect sensors and potentiometers, are designed to produce an output voltage that is ratiometric to a supply voltage. Although many such devices are electrically robust and can withstand a wide range of voltages, including reverse voltage supplies, ratiometric transducers containing integrated circuits can be damaged by over-voltages or reversed input connections. Such situations can arise accidentally during the installation of the transducers, and can lead to their destruction.

Hall-effect sensors are used in positions sensing transducers, particularly where it is not possible, or not practical, to use a contacting sensor. The Hall-effect sensor detects position by detecting a change in a magnetic filed, usually provided by a permanent magnet.

GB2,418,083 describes a protection circuit suitable for use with such ratiometric transducers, which protects the sensor from over-voltages or a reversed input voltage at the supply. The protection circuit described includes a first and a second metal oxide semiconductor field effect transistor (MOSFET). The source connections of the MOSFETs are electrically connected to each other to provide a common source and the drain of the first MOSFET receives the supply voltage. The circuit includes a voltage detector, which turns off the second MOSFET upon detecting that the input voltage exceeds a predetermined threshold, so that the supply voltage is no longer * supplied to the sensor. The MOSFETs each include an intrinsic body diode, ****** * p which allows conduction through the MOSFET in one direction even if the MOSFET is turned off. However, the circuit is configured so that when the voltage supply to the drain of the first MOSFET is negative, the body diode preients the negative voltage from being transmitted through the first MOSFET to the sensor. The load is therefore protected against negative voltages, which may occur, for example, when the supply is inadvertently connected the wrong way round. When the input voltage is positive and higher than the predetermined voltage threshold, the voltage sensor turns off the second MOSFET so that this higher voltage is not transmitted through the second MOSFET to the load, the body diode of the second MOSFET is reverse biased and so does not conduct in this situation.

A problem with this input protection circuit, is that every time the supply voltage exceeds the threshold, even if only by a small amount, the supply to the sensor is cut off. This means that where the device is being used for monitoring, such as in a position sensor, as described above, there will be no output from the sensor whenever there is an excursion in the supply voltage above the threshold. If the sensor is being used for control purposes, this could have serious consequences, but even if only being used for monitoring there will be a gap in the monitored data.

It is often desirable for the output signal of the sensor to have a linear relationship to the parameter being measured, for example displacement or any other physical or chemical variable. A linear relationship makes it very easy to interpret the sensor output or connect it to some sort of read-out display. Unfortunately many sensor types do not provide an output that has a linear response to the measured parameter. Consider, for example, ::::. temperature sensors where Platinum Resistance Thermometers (PRTs) have a slightly non-linear response; thermocouples are even less linear, and negative temperature coefficient (NTC) thermistors are very non-linear, exhibiting an exponential characteristic. Another example of a sensor that provides a non-linear response is a Hall-effect sensor of the type described above.

Where the output from a sensor is non-linear, then various electronic methods may be employed to correct for this. Where the non-linearity is slight, or the shape of the non-linearity is repeatable, then analogue electronic circuits can often be used to re-shape the signal, presenting a linear characteristic for display purposes.

Another method is to digitise the signal using an Analogue-to-Digital converter (ADC) and perform a linearisation process on the digital signal. The linearisation process will typically use a mathematical calculation or a table of correction values. If an analogue output signal is still required then a Digital-to-Analogue Converter (DAC) will also be required. This method has great flexibility because it can be applied to almost any shape and magnitude of the non-linearity to be corrected. Also, it is possible to linearise sensors individually. However, although the original analogue signal may be non-linear, it is continuous, meaning that even the tiniest change in the measured parameter can be detected. The resolution of the measurement is, to all intents and purposes, infinite. However, digitising the signal reduces its resolution so that the minimum resolvable change in the signal is limited to 1-bit. For example, if an 8-bit ADC is able to convert to digital form the signal from a temperature sensor with a range of 0 to 256°C, then the smallest step will be 1°C. To improve this to better than 0.1°C, an ADC having at least a 12-bit capability would be required. The same resolution arguments apply to the DAC used for converting the digitised signal back to analogue.

Clearly, digitising the signal always results in a loss of resolution. If only moderate performance is required then 8 or 10-bit conversion may be acceptable and low-cost ADCs and DACs may be used. However, if the required resolution is 12-bit or higher, then it is likely that a separate ADC and DAC will be required, and this becomes more expensive. As the required resolution rises further to 14-bits, 16-bits and beyond, the components can become very expensive indeed. S.

The present invention has been conceived with the foregoing in mind.

*..I..

S S

Summary of the Invention

According to a first aspect of the present invention there is provided a transducer comprising a sensor having an input and an output. The input receives a sensor input voltage and the output provides a sensor output signal responsive to a physical stimulus. An input protection circuit receives a supply voltage and prevents application to the sensor input of a voltage in excess of a threshold. A voltage regulator regulates the voltage applied to the sensor input at a predetermined level when the input protection circuit determines that the supply voltage has exceeded the threshold. An output provides a transducer output signal dependent on the sensor output signal.

It is an advantage that a voltage supply to the sensor is maintained, but at a safe, regulated voltage even when the input protection circuit determines that the supply voltage is too high. Most voltage regulators require a supply voltage greater than the regulated output voltage to operate. For example a 5 volt regulator would require a supply voltage of greater than 5 volts. Thus, it would not be possible to use the voltage regulator on its own to provide the full functionality of the protection circuit. For example, there may be situations where only a 5Volt supply line is available, in which case the 5 volt regulator would not function. Also there is frequently a need for the transducer to operate ratiometrically to the supply, from a low voltage up to, say, 5 volts, for example for error checking.

In embodiments of the invention, the input protection circuit comprises a pair of transistors and a voltage sensing arrangement configured to turn off one of the pair of transistors on sensing that the supply voltage exceeds the threshold. One, or both, of the transistors may be a metal oxide semiconductor field effect transistor, MOSFET. Alternatively, any P-type transistor may be used, or alternatively a N-type transistor could be used with a negative input voltage. *... S...

The input protection circuit may also be configured to prevent application of a negative supply voltage to the sensor. 5I*S

In embodiments of the invention, the transducer further comprises a linearization circuit providing the output signal of the transducer as a linearized function of the sensor output signal. The linearization circuit may comprise: an analogue to digital converter for providing a digital representation of the sensor output signal; a correction signal generator providing a correction signal from the digital representation; and a summing circuit configured to receive, as inputs, the sensor output signal and the correction signal and to add these to produce the linearized transducer output signal.

It is an advantage that, by combining the correction signal and the sensor output signal in a summing circuit, a much higher resolution of corrected output signal can be obtained. For example, 10-bit or higher resolution may be obtained when using an 8-bit analogue to digital converter.

The summing circuit may comprise an operational amplifier, biased around a mid-supply point of the input voltage. The summing circuit may further comprise a second operational amplifier so as to re-invert an inverted signal from the first operational amplifier.

The correction signal generator may comprise a processor for applying a mathematical formula to the digital representation of the sensor output signal. Alternatively, the correction signal generator may comprise a look-up table for providing a digital correction signal from the digital representation of the sensor output signal. The correction signal generator may further comprisea digital to analogue converter. The digital to analogue converter may comprise a Pulse Width Modulation (PWM) circuit and a filter. 0S*

The analogue to digital converter and correction signal generator may be comprised in a microcontroller. * S...

According to a second aspect of the invention there is provided a transducer comprising a programmable sensor for providing a sensor output signal responsive to a physical stimulus. An output conditioning circuit comprises a microcontroller, and a voltage input providing a supply voltage to both the sensor and the microcontroller. The transducer further comprises an * over-ride input whereby application of an over-ride voltage to the over-ride input disconnects the sensor from the input voltage to allow programming of the sensor with the microcontroller still connected to the input voltage. The over-ride input may be connected to a transistor such that application of the over-ride voltage switches the transistor to interrupt the voltage applied to the sensor input. The transistor may be a MOSFET.

Features of the first or second aspect may be used in combination with features of the other aspect.

In embodiments of the first or second aspects of the invention the sensor may be a position sensor. The position sensor may be a Hall-effect sensor. Alternatively, the sensor may be a temperature sensor, for example one of, a Platinum Resistance Thermometer (PRT), a thermocouple, and a negative temperature coefficient (NTC) thermistor. Alternatively, the sensor may be any other type of sensor.

Description of Drawings

Figure 1 is a schematic diagram illustrating the principal electronic components of a transducer according to an embodiment of the invention.

Figure 2 is a detailed circuit diagram of a supply voltage system of the transducer of figure 1.

Figure 3 is a detailed diagram of circuit incorporating a programmable sensor of the transducer of figure 1.

Figure 4 is a schematic representation of a linearization circuit.

: Figure 5 is a graph showing a non-linear raw input signal to the linearization circuit of Figure 4.

:... Figure 6 is a graph showing the raw input signal of Figure 5 inverted.

Figure 7 is a graph showing a correction signal for the raw input signal S..

of Figure 5, using the circuit of Figure 4. * * **. * *

Figure 8 is a graph showing an attenuated form of the correction signal of Figure 7.

Figure 9 is a graph showing an inverted linearised output signal from the circuit of Figure 1.

Figures 10 and 11 are circuit diagrams of a signal conditioning circuit including a linearization circuit, for the transducer of figure 1.

Figure 12 illustrates a Hall effect position sensing transducer constructed in accordance with the principles of the invention.

Figure 13 illustrates a magnet housing for use with the position sensing transducer of figure 12.

Figure 14 is an elevation in cross-section of the position sensing transducer of figure 12.

Detailed Description of Embodiments

Referring to Figure 1, a transducer 10 has three electrical contacts 1, 2, 3. Electrical contact I delivers an output voltage signal, while contact 2 is connected to a ground voltage and contact 3 is provided with an input supply voltage. The transducer includes circuitry that is shown schematically in three sections -an input voltage management circuit 4, details of which will be described in more detail hereafter; a sensor circuit 5 that provides an output signal in response to a physical stimulus, an example of which will be described in more detail hereafter; and an output signal conditioning circuit 6, details of which will also be described in more detail hereafter. Each of the three circuits is preferable mounted on a circuit board, which is enclosed in a sensor device, an example of which will also be described hereafter.

Figure 2 is a circuit diagram of the input voltage management circuit 4.

As described above for figure 1, an input voltage is supplied to electrical S...

: contact 3, with electrical contact 2 connected to ground. A standard noise-S...

S....' filtering arrangement 12 is provided before the supply voltage is applied at line :.". 13 to a voltage protection circuit. The details and operation of this part of the circuit are described in GB2,418,083. S.. * ..I

I

* **..* * I Briefly, the input voltage is supplied to the drain 14a of a first P-type metal-oxide semiconductor field-effect transistor (MOSFET) 14. The source 14b of the first MOSFET 14 is connected to a source 16b of a second P-type MOSFET 16 via a common source connection 18. The gates 14c, 16c of the first and second MOSFETS 14, 16 are each connected to the ground line 2 via resistors 22, 24. A voltage sensing arrangement 26 is connected to the sources 14b, 16b of the first and second MOSFETs 14, 16, to the gate 16c of the second MOSFET and to the ground line 2. The drain 16a of the second MOSFET 16 outputs a voltage to the output 20 of the voltage management circuit 4. The voltage sensing arrangement 26 includes a Zener diode 28, a transistor 30, and a resistor 32.

In use, where a positive voltage is supplied to the drain 14a of the first MOSFET 14, this voltage is transferred to the common source connection 18.

The first and second MOSFETS 14, 16 are turned on and the input supply voltage is directly connected to the output 20. In order to protect the sensor circuit 5 of the transducer 10 (as shown in figure 1) from receiving a voltage that is too high, the voltage at the output 20 is. limited to a maximum predetermined value Vmax, which is higher than the nominal operating voltage but lower than a voltage that might damage the sensor circuit 5 of figure 1.

For example, if the transducer is designed to operate at a nominal voltage of 5V, Vmax may typically chosen to be 6V. When the voltage sensing arrangement 26 senses a voltage at the common source connection 18 that exceeds Vmax, it short circuits the gate 16c of the second MOSFET 16 to its source 16b, and turns it off. Thus, the voltage at the drain 16a of the second MOSFET is OV. In the voltage sensing arrangement 26 shown in figure 2, when the input supply voltage exceeds a threshold level equal to the breakdown voltage of the Zener diode 28 plus the base-emitter voltage of the S...

: transistor 30, the transistor 30 turns on. This provides a short circuit between

SS

the source 16b and the gate 16c of the second MOSFET 16, which turns the second MOSFET 16 off. * S ***S S. S. * S..

S

S..... * S

In addition, the first and second MOSFETs 14, 16, each include an intrinsic body diode (not shown). If the input supply is connected in reverse, so that a negative voltage is supplied to the input 3, the drain 14a of the first MOSFET 14 will be negative with respect to the ground line 2. The negative input voltage will not be transmitted through first MOSFET 14 because the intrinsic body diode will only pass a positive voltage. The second MOSFET 16 will therefore also remain off, with the voltage at the common source 18 and at the drain 16a being OV.

However, although the circuit 4 described thus far, and in GB2,418,083, protects the transducer against incorrect supply voltages, which includes all negative excursions of input voltage, as well as excessive positive voltage excursions, this would mean that, in the presence of such excursions, there will be no output from the transducer 10 at the output electrical contact 3. Depending on the application of the transducer, there may be circumstances in which such interruptions in output could not be tolerated, or would cause problems or be inconvenient. To overcome this, the circuit 4 includes a voltage regulator 34. The voltage regulator 34 is of a type that will provide an output at a regulated voltage when it receives input voltages above certain minimum levels at two different inputs. An example of this type of voltage regulator is the LTRTM Low Dropout Micropower Linear Regulator supplied by Linear Technology Corporation. The voltage regulator 34 has a first input 36 connected to the common source 18, and a second input 38 connected to the emitter of the transistor 30. The voltage regulator 34 has an output 40 connected to the output 20 of the voltage management circuit 4.

Thus, when the voltage sensing arrangement 26 switches the transistor 30 because it has sensed an excursion above the threshold maximum supply voltage, both the inputs 36 and 38 of the voltage regulator 34 are connected : to the supply voltage. In such cases, the voltage regulator will provide its S....' stipulated output voltage at its output 40. *. .. * . .

It will therefore be seen that there are two voltage supply situations that arise, one, which is unregulated when the supply voltage is within the protection limits of the circuit 4, and a regulated voltage when the protection limits are exceeded. However, both these situations are provided in a circuit having a single output pin 20. This circuit avoids the disadvantages of cost and weight in comparable transducers, where four or more wire connections are required.

Figure 3 illustrates a circuit 50 for an exemplary transducer sensor, The circuit 50 includes a sensor circuit 52 (such as the circuit 5 of figure 1), but is shown connected to a microcontroller 54. The microcontroller 54 may be part of an output signal conditioning circuit, such as the circuit 6 of figure 1, an example of which is a linearization circuit described hereafter. The sensor circuit 52 includes a programmable sensor, such as a Hall-effect sensor, having a programmable sensor chip 56. The sensor Ohip 56 has an input terminal 58a for connection to an input supply voltage 60 and an input terminal 58b for connection to ground. An input supply voltage is provided across the circuit 52. The input supply voltage 60 may, for example, be provided from the output 20 of the voltage management circuit of figure 2.

The programmable sensor chip 56 has an output 62, which is provided as an input to the microcontroller 54 with a resistor 64, and capacitor 65 providing a filter to prevent output from the sensor chip 56 damaging the microcontroller 54. The microcontroller 54 receives an input supply voltage at an input terminal 66 from the supply voltage 60.

However, problems can arise, at least with certain types of programmable sensor chip 56, in that it is not possible to program the chip while its input terminal 58a is connected to the same supply voltage as the microcontroller 54. To program the sensor chip 56 it is necessary to disconnect the input 60 from the input terminal 58a. This does not present a S...

: major difficulty for prototyping purposes, but when manufacturing transducer S...

products it introduces a time-consuming manual intervention into the manufacturing process, for example by soldering in a link to the supply after the sensor chip has been programmed. Also, in a solid state transducer S...

product, after the terminal 58a has been reconnected to the voltage supply 60, ::.... the sensor cannot be reprogrammed without removing the link.

S.....

To overcome this problem, the circuit 52 includes a transistor, for example a MOSFET, 68 with its source and drain connected between the voltage supply 60 and the input terminal 58a. The gate of the MOSFET 68 is connected to an over-ride input 70. When an over-ride voltage (for example a significantly higher voltage than the supply voltage 60) is applied to the over-ride input 70, the MOSFET 68 is switched, breaking the line from the supply to the input terminal 58a. The sensor chip 56 can then be programmed while the over-ride voltage is applied to the over-ride terminal 70. After the sensor chip 56 has been programmed, the over-ride voltage is removed from the over-ride terminal 70, switching the MOSFET 68 to reconnect the supply to the input terminal 58a.

The output of most sensors is not perfectly linear. Thus, particularly where accuracy is required in control or monitoring applications, any non-linearity in the output from the sensor needs to be taken into account. Figures and 11 illustrate a circuit for use in linearizing the output signal from the output of a sensor such as the sensor chip 56 of figure 3. The principles of this circuit will first be described with reference to figures 4 to 9.

Referring to Figure 4, a linearization circuit 100 has an input 102 for receiving a non-linear input signal from, for example, a sensor (not shown).

The raw sensor signal is provided to the input of a microcontroller 104, in which an analogue to digital converter (ADC) converts the signal into a digital form. In this example, the microcontroller 104 is a low-cost microcontroller with an on-board an 8-bit ADC and includes a memory and a processor. The microcontroller 104 may comprise discrete components such as the ADC 106 and DAC 110. Alternatively, the microcontroller 104 may be embodied in a : single integrated circuit. The memory stores data in the form of a look-up table 108. The look-up table 108 includes a pre-determined set of data that : **. account for the non-linearity of the raw sensor signal, and provides a :...:. correction value as an output. The output correction value is in digital form, **..

and is supplied to a digital to analogue converter DAC 110, which provides an ** * S..

*..... * *

analogue correction signal as an output from the microcontroller 104. In this example the DAC 110 is an 8-bit DAC.

The raw input signal received at the input 102 is also fed directly to one input of a summing circuit 112 through a first resistor RI. The analogue correction signal output from the microcontroller 104 is supplied to a second input of the summing circuit 112 via a second attenuating resistor R2. The summing circuit 112 includes an operational amplifier 114, which is biased about the mid-supply voltage, and includes a feedback resistor RE. The summing circuit 112 has an output 116, which, as will become apparent from the discussion below, provides a linear, but inverted output signal.

Let us assume that the non-linear output from a sensor is as shown in Figure 5. For convenience, the maximum deviation from the best straight line is assumed in this example to be +1-12.5%. This means that if the best straight line ranges between from 0 to 5V, then the maximum deviation is 0.625V. In this example we assume that the supply voltage to the circuit is also 5V.

Now the raw sensor output is applied to one input of the summing circuit 112 based on an operational amplifier 114, biased around the mid-supply point. Thus, with no correction applied, the output of the summing circuit 112 will be simply an inverted version of the input as shown in Figure 6.

The raw sensor output is also measured and digitised by the 8-bit ADC 106 in the microcontroller 104. From the value of the sensor signal, the microcontroller 104 is able to determine the required correction value, using the look-up table. In an alternative embodiment, the microcontroller processor *.

.. : may apply a mathematical formula to the digitised input signal to determine the correction value. S. * * .

The DAC 110 converts the digital correction signal to an analogue *.SS signal and the microcontroller 104 then outputs the correction signal in analogue form. As an alternative to the DAC 110, a pulse-width modulated S..... * S

(PWM) output with a following filter may be used to convert the signal to analogue form. The magnitude of the correction signal is arranged to utilise the full 5Voutput dynamic range of the DAC 110, as shown in Figure 7.

Because the correction signal is derived from an 8-bit digital value it will have steps of 11256th of 5V, which is equal to 19.53mV.

The analogue correction signal is then attenuated by a factor of 4 around the mid-supply point, as shown in Figure 8. This means that the maximum deviation of the correction signal from the mid-supply voltage is now +1-0.625V and the digital step size is now 4.88mV.

This attenuated correction signal is finally added to the raw sensor signal to provide the inverted, but linear output shown in Figure 9. Since the step size of the correction signal is 4.88mV over 5V signal range, this is equivalent 1 part in 1024, or 10-bits.

In effect, we have linearised the sensor signal to a resolution of 10-bits using a microcontroller 104 employing an AUC 106 and DAC 110 with only 8-bit capability.

An obvious consequence of the summing circuit 112 is that the linearised output signal has been inverted with respect to the original sensor signal. In many cases this will be of no consequence but, where it is, a further operational amplifier inverter may be employed.

in general, the number of bits of resolution improvement is inversely proportional to the maximum deviation of the sensor signal from the best straight line. Thus, in the case above where the non-linearity is +1-12.5%, or 25% peak-to-peak, the magnitude of the correction needs only be a quarter of sensor's range. The inverse of % is 4, which can be represented by 2-bits, so we can get 10-bit resolution with an 8-bit digital system. * *

If the sensor were much more linear, with a maximum deviation of only +1-1.56%, or 3.12% peak to peak, then the correction would be only 1132nd of *S..*. * I r

range. A factor of 32 is represented by 5-bits, so an 8-bit digital correction would yield an overall resolution of 13-bits, or one part in 8192. On a 5V system, this would mean a resolution of less than lmV, which is typically below the output noise specifications of many sensor types.

The magnitude of the non-linearity that needs to be corrected dictates the amount of attenuation required between the analogue correction signal from the microcontroller 104 and the summing circuit 112.

Figures 10 and 11 show a transducer signal conditioning circuit, such as the circuit 6 of figure 1, for providing the transducer output signal.

Referring to Figure 10, the first part of the signal conditioning circuit, includes a microcontroller 120, connected to a supply voltage 122, which may be provided from the output of a voltage supply management circuit such as the circuit 4 of figures 1 and 2. The microcontroller 120 also receives a signal from the output of a sensor at a terminal 123. It will be seen that the microcontroller 120 is connected in a similar manner to the microcontroller 54 of figure 3, or the microcontroller 104 of figure 4. The microcontroller 120 performs an analogue to digital conversion to the sensor signal, for example converting this to an 8-bit binary number as described above with reference to figure 4. This is then referenced to a look-up table in the memory of the microcontroller 120 to determine a corrected value. The correction voltage signal is then output from the microcontroller through a pulse-width modulated (PWM) output 124 on board the microcontroller. The output from the microcontroller is then fed to the circuit 126, which is a Sallen and Key filter that converts the digital signal back to an analogue signal. The converted analogue correction signal is fed through resistor 129 to a connection 130.

The sensor output is also fed, bypassing the microcontroller 120, through I...

: resistors 127 and 128 to be combined with the converted analogue correction signal at the connection 130. *. S. * * S

Referring to figure 11, which shows the second part of the linearisation S...

circuit the combined output signal at 130 is fed to a summing circuit, which is similar to that described above with reference to figure 4. The combined *5*SSS S * signal is fed into an operational amplifier 132, which is biased around the mid-point by the use of the resistors 134 and 136 forming a potential divider between the supply voltage 122 and ground. The output V0. from the circuit is provided to the transducer output pin I (see figure 1). The other unreferenced components are for noise reduction and output protection to protect against reverse voltage conditions.

In practice, attenuation of the correction signal is achieved by making the resistance value of the input resistor R2 to the summing circuit 112 of figure 4 (equivalent to the resistor 129 in figure 10), an appropriate multiple of the resistance value of the resistor Ri connected to the raw sensor output (equivalent to resistors 127 and 128 in figure 10). The resistance value of the feedback resistor (RF in both figure 4 and figure 11) of the summing circuit is unimportant to the linearization scheme and may be selected to provide output scaling as required. Thus, although we have assumed in the example of figures 4 to 9 that the operational amplifier 114 of the summing circuit 112 has a gain of -1 with respect to the raw sensor signal, other gains may be employed. The output voltage is defined by: V0=-RF(V1-2.5 -V2-2.5) +2.5 ( Ri R2) where VI is the raw sensor output signal and V2 is the correction signal.

For the case where the linearization is applied to the output of a programmable sensor such as the programmable Hall effect sensor referred to above, the inversion of the voltage is not a problem because the sensor can be programmed to provide a high or low output depending on its proximity to either a South pole magnetic field, and vice-versa for a North pole magnetic field. For other sensor types a simple circuit may be employed to re-invert the output signal. S... * . S. * S..

S

Figure 12 illustrates a Hall effect position sensing transducer 150 constructed in accordance with the principles of the invention and incorporating features described above. The transducer 150 has a three-wired cable 152 for external connection. Each of the wires corresponds to one of the pin terminals 1, 2, 3 shown in figure 1. That is to say that one wire is for connection to pin 1 and carries the transducer output voltage signal, another wire is for connection to pin 2 and is a ground connection, and the third wire is for carrying the supply voltage. The transducer has a housing 154, a sealed cable connection 156, and a pair of mounting screws 158. The housing 154 has a sealed end face 160, behind which the Hall effect sensor is mounted.

Figure 13 illustrates a magnet housing 162 for use with the position sensing transducer 150. The magnet housing 162 has a permanent magnet (not shown) mounted inside it and includes a pair of mounting screws 164.

In use the magnet housing 162 is mounted to a first component and the Hall effect transducer is 150 mounted to second component that is moveable in close proximity relative to the first component. For example, the Hall effect transducer body 150 may be mounted to a stationary component, while the magnet hosing is mounted to a moving component, although they could be used the other way round if desired. The Hall effect sensor inside the transducer 150 produces an output signal that is dependent on the magnetic field in which it resides, and so any relative movement of the first and second components will result in a change in the output signal.

Figure 14 shows a cross-section of the Hall effect transducer 150 of figure 12. Inside the housing 154 is mounted a circuit board 166. Mounted at one end of the circuit board is a programmable Hall effect sensor chip 168, such as.a HAL815 chip, supplied by Micronas GmbH. In production, after the circuit board 166 has been inserted into the housing 154, the sensor chip 168 . is located close to one end of the housing, and this end is then potted over S...

with an epoxy or similar compound to provide a smooth and preferably blemish free domed shape for the sealed end 160.

S.....

Circuitry such as that described above with reference to figures 1 to 4, and 11 is built onto or into the circuit board 166 to provide a unitary sealed and programmed transducer having all the benefits of signal input voltage protection and output signal conditioning described above. S... * .. S. S **.. * S *55S *S ** * . * * S S... * . S... *5 * S..

S

5*SS*. * 5

Claims

Claims: 1. A transducer comprising: a sensor having an input and an output, the input receiving a sensor input voltage and the output providing a sensor output signal responsive to a physical stimulus; an input protection circuit receiving a supply voltage and operable to prevent appUcation to the sensor input of a voltage in excess of a threshold, and a voltage regulator operable to regulate the voltage applied to the sensor input at a predetermined level when the input protection circuit determines that the supply voltage has exceeded the threshold; and an output for providing a transducer output signal dependent on the sensor output signal.
2. The transducer of claim 1 wherein the input protection circuit comprises a pair of transistors and a voltage sensing arrangement configured to turn off one of the pair of transistors on sensing that the supply voltage exceeds the threshold.
3. The transducer of claim 2 wherein one or both of the transistors is a metal oxide semiconductor field effect transistor, MOSFET.
4. The transducer of any preceding claim wherein the input protection circuit is configured to prevent application of a negative supply voltage to the sensor.
5. The transducer of any preceding claim, wherein the sensor is a programmable sensor, the transducer further comprising: S... : a microcontroller having an input coupled to the sensor input; and **5* S.'. an over-ride input, :.". whereby application of an over-ride voltage to the over-ride input disconnects the sensor from the input voltage to allow programming of the *..S sensor with the microcontroller still connected to the input voltage.I.....S
6. The transducer of claim 4 wherein the over-ride input is connected to a transistor such that application of the over-ride voltage switches the transistor to interrupt the voltage applied to the sensor input.
7. The transducer of claim 6 wherein the transistor is a metal oxidesemiconductor field effect transistor, MOSFET.
8. The transducer of any preceding claim further comprising a linearization circuit providing the output signal of the transducer as a linearized function of the sensor output signal.
9. The transducer of claim 8 wherein the linearization circuit comprises: an analogue to digital converter for providing a digital representation of the sensor output signal; a correction signal generator providing a correction signal from the digital representation; and a summing circuit configured to receive, as inputs, the sensor output signal and the correction signal and to add these to produce the linearized transducer output signal.
10. The transducer of claim 5 wherein the summing circuit comprises an operational amplifier, biased around a mid-supply point of the input voltage.
11. The transducer of claims 9 or claim 10, wherein the summing circuit further comprises a second operational amplifier so as to re-invert an inverted signal from the first operational amplifier.
12. The transducer of any of claims 9 to 11, wherein the correction signal generator comprises a processor for applying a mathematical formula to the digital representation of the sensor output signal. * * *
13. The transducer of any of claims 9 to 12, wherein the correction signal **** generator comprises a look-up table for providing a digital correction signal from the digital representation of the sensor output signal.

*.*..I * *
14. The transducer of any of claims 9 to 12, wherein the correction signal generator further comprises a digital to analogue converter.
15. The transducer of claim 14 wherein the digital to analogue converter comprises a Pulse Width Modulation (PWM) circuit and a filter.
16. The transducer of any of claims 9 to 15, wherein the analogue to digital converter and correction signal generator are comprised in a microcontroller.
17. A transducer comprising a programmable sensor for providing a sensor output signal responsive to a physical stimulus, an output conditioning circuit comprising a microcontroller, and a voltage input providing a supply voltage to both said sensor and said microcontroller, wherein the transducer further comprises an over-ride input whereby application of an over-ride voltage to the over-ride input disconnects the sensor from the input voltage to allow programming of the sensor with the microcontroller still connected to the input voltage.
18. The transducer of claim 17 wherein the over-ride input is connected to a transistor such that application of the over-ride voltage switches the transistor to interrupt the voltage applied to the sensor input.
19. The transducer of claim 18 wherein the transistor is a metal oxidesemiconductor field effect transistor, MOSFET.
20. The transducer of any preceding claim wherein the sensor is a position sensor. S... * . .S
21. The transducer of claim 20 wherein the sensor is a Hall-effect sensor. *S ** * * .
22. The transducer of any of claims I to 14 wherein the sensor is a S...::... temperature sensor, for example one of, a Platinum Resistance Thermometer S.....(PRI), a thermocouple, and a negative temperature coefficient (NTC) thermistor. S... * *. * S * . S... S *S * . S * S * S *5SS * *..S *5**.S