Electronic Test Instrument for Measuring the Speed and Timing Angle of an Internal Combustion Engine

The invention relates to the field of electronic internal combustion engine test equipment and more particularly to instruments for measuring timing advance and engine speed.

In timing internal combustion engines the typical practice has been to use timing lights such as stobe lamps to illuminate a marker on the engine flywheel or damper and to compare its position with a reference mark located on the engine body. Since the prior art techniques are of limited accuracy and dependent on the abilities of the observer and since greater requirements for fuel economy and pollution control call for engine timing standards of far higher accuracy, the need for an instrument to provide high accuracy reading of engine timing angle and speed has become apparent.

It is therefore an object of the invention to provide an instrument for directly displaying the timing angle of an internal combustion engine.

It is another object of the invention to provide a measuring instrument for internal combustion engines wherein a chosen parameter of an electrical signal representing the time duration of the timing angle of advance is divided by a chosen parameter of an electrical signal representing the time duration of one revolution.

It is a further object of the invention to provide a timing instrument for a diesel engine wherein an electrical pulse signal whose width is representing the time period from fuel injection to when the piston reaches a top dead center position is converted into either a digital signal or a voltage representing this time period and an electrical pulse signal, wherein the pulse width which is proportional to the time of one engine revolution is converted into either a digital or a voltage representing one engine revolution are used to generate signals representing timing angles and engine speed.

Referring to diesel engines in particular, fuel is injected into the cylinder slightly before its piston reaches its top dead center (TDC) position, i.e. in an instant which corresponds to a so-called "advance angle of injection." The measuring device should read this angle in angular degrees, i.e. to determine the time interval from the instant when the fuel pressure exceeds a predetermined level thus causing the injector to inject fuel in the cylinder to the instant when the piston arrives at the top dead center position in the cylinder. In most engines, a marker is located on the flywheel to indicate the position of the piston on the first cylinder. But to measure this interval alone would not be sufficient to determine the advance angle since the same time interval can correspond to different angles according to the engine speed. It is, therefore, necessary to also measure the time duration of one full revolution and to compare both these intervals. If we designate the first interval, i.e. that from the fuel injection to the top dead center position, as t and the duration of one revolution as T, then the advance angle in angular degrees is given by the formula:

fig. 1 is a block diagram of a timing advance angle measuring system:

FIG. 2 is a block diagram of a portion of the system of FIG. 1 containing digital circuits for computing advance angle;

FIG. 3 is a schematic diagram of a portion of the system of FIG. 1 containing analog circuits for computing advance angle;

FIG. 4 is a schematic diagram of a first advance angle calibration circuit;

FIG. 5 is a schematic diagram of a second advance angle calibration circuit;

FIG. 6 is a block diagram of a circuit for measuring injection duration;

FIG. 7 is a block diagram of a circuit for measuring engine speed; and

FIG. 8 is a block diagram of a circuit using a strobe light.

The timing system disclosed herein will be discussed in terms of a diesel engine although it should be understood that the general concepts would equally apply to the measurement of advance angles and speeds of an ignition type engine. As shown in FIG. 1 a pressure transducer 10 is connected to a fuel line 12 and generates a voltage V.sub.1 on line 14 proportional to the instantaneous fuel pressure. Thus the voltage V.sub.1 on line 14 will increase substantially in response to the increase in fuel pressure necessary to inject fuel into the cylinder. A second sensor is a magnetic pickup 16 which is placed close to the circumference of the engines flywheel 18 and generates a one cycle sine-wave shaped signal V.sub.2 on line 20, which crosses the zero axis in the instant when the center of a ferrous tooth 22 or groove on the engine flywheel 18 passes the center line of the pickup 16. Each sensor 10 and 16 is connected to a separate pulse shaping circuit, indicated within the dashed lines 24 and 26, which act as trigger pulse generators that generate square-wave pulses in response to the sensor signals. The first pulse shaping circuit 24 includes a clamping circuit consisting of capacitor 28, diode 30 and resistor 32 which clamps the lower part of the transducer waveform V.sub.1 approximately to a normally zero potential so that only a voltage V.sub.1 which exceeds the residual level is applied to the first input of a comparator 34. A second input 36 of this comparator is connected to a center arm 38 of a potentiometer 40 which provides an adjustable reference voltage. When the voltage on the first input exceeds the reference voltage, there is an abrupt change in the comparator 34 output which triggers a one-shot multivibrator 42. The square wave pulse triggered on line 44 by multivibrator 4 has a leading edge which coincides with the instant when the fuel pressure exceeds the predetermined level. The second pulse shaping circuit 26 consists of an amplifier 46 which preferably is equipped with automatic gain control and is followed by a differentiating network consisting of capacitor 48 and resistor 50 and by a second one-shot multivibrator 52. The amplifier 46 amplifies the pickup signal V.sub.2 on line 20 so that its zero-crossing portion becomes very steep. When this voltage is differentiated the zero crossing portion corresponds to a very narrow and high peaked signal which triggers the second one-shot multivibrator 52, which in turn generates a pulse on line 54 with a leading edge that coincides with the zero-crossing point of the pick-up signal. Since pulses in both circuits 24 and 26 are generated repeatedly as long as the engine is running, they actually produce periodic pulse trains.

Both these pulse trains are fed to a set and a reset input of a flip-flop 56 so that the flip-flop 56 is repeatedly set by leading edges of the first pulse train on line 44 and reset by leading edges of the second pulse train on line 54. Thus the flip-flop 56 output delivers, on complementary output lines 58 and 60, pulses of complementary polarities having a duration equal to the time interval t. At the same time one of the pulse trains, preferably the second one on line 54, is applied to the input of a binary counter 62 which changes its output state upon receipt of each pulse thus delivering complementary pulse outputs on lines 64 and 66 having a duration proportional to the duration of one engine revolution. In the event that the counter 62 is driven by pulses on line 54 which are derived from the pickup signal 20, the duration of output pulses on line 64 and 66 are simply equal to the duration of one engine revolution which we designated at T. In the alternate case where the engine is of the four-cycle type and counter 62 is driven by pulses derived from the pressure signal 14 which are generated every other engine revolution, the duration of output pulses of the counter 62 will be equal to the duration of two engine revolutions.

It is of interest to note that the desired angle information is already included in the first pulse train on line 58 in the form of its average value. It would be, therefore, possible to measure the timing angle by using a low-pass filter (not shown) which would suppress individual pulses and pass only their average value. However, this filter would result in a delay and still leave some ripple on the measured value which would affect the accuracy if a digital display is used. Therefore, the preferred method is described in the following paragraphs which gives more accurate and instantaneous readings.

As explained above, two pulse trains from flip-flop 56 and counter 62 are obtained: the first one 56 having a pulse duration equal to the time interval t and the second one 62 having a pulse duration equal, or at least proportional, to the duration T of one engine revolution. In order to obtain the desired angle .alpha. information it is necessary, according to equation (1), to divide the first pulse width on line 58 by the second on line 64 and to use required scaling factor 360. Since both intervals t and T essentially represent time and an electronic circuit can process only electrical quantities, it is necessary to convert these intervals in some proportional electrical quantities. The time intervals t and T are converted into this quantity in converters generally represented by boxes 68 and 70 having outputs proportional to the time intervals. In agreement with equation (1) output of the first converter 68 is connected to a dividend input 72 of a divider circuit 74 while output of the second converter 70 is connected to the divisor input 76. The divider output on line 78 generates a value which, according to equation (1), is equal to the desired timing angle .alpha. except for the scaling factor 360. This factor can be accounted for in different ways, according to the kind of applied electrical quantity, as discussed in the following paragraph.

There are various suitable quantities which generally can be divided into digital or analog values. FIG. 2 represents an embodiment of the digital type wherein pulses from a clock 80 having a frequency considerably higher than the frequency of the pulses on lines 58 and 64, are fed to the input of a first counter 82, via an AND gate 84, having a second input connected to the output 58 of flip-flop 56, so that AND gate 84 is enabled during the interval t and consequently the counter 82 reaches a certain final reading which, at a given clock frequency, is proportional to the time interval t. The same clock 80 is connected to the input of a second counter 85 via another AND gate 86 having a second input connected to the output 64 of counter 62 and is enabled during the interval T so that the final reading of counter 84 is proportional to the time interval T. These readings are fed as a dividend and divisor over lines 88 and 90 respectively to inputs of an arithmetic unit 92 which is under control of a processor 94 and acts as a digital divider. In order to obtain the correct value of .alpha. in degrees it is necessary to use the scaling factor 360 which can be done either by a subsequent digital multiplication, or preferably by a one-to-360 counter indicated by the dashed line 96 on the input line 98 of the counter 84. The processor 94 also controls all other related functions such as repeated resetting of both counters e.t.c. The output of the arithmetic unit 92 representing .alpha. can then be applied to either a digital readout display device or an analog meter represented by box 100.

FIG. 3 illustrates a representative embodiment of the converters 68 and 70 of FIG. 1 of the analog type. A first constant-current source 102 including FET 104 is controlled by a switching transistor 106 having its base connected to the complementary output of 60 of flip-flop 56. As a result, the constant current source 102 will charge a storage capacitor 108 through diode 110 during the interval t. In this way capacitor 108 is charged to a voltage having a final value proportional to the time t; after which its voltage is sampled by a first sample and hold circuit 112 including amplifier 115 and capacitor 117, and then discharged by a FET switch 150. Both the sampling function and the discharging are controlled by a sequencing circuit 114 which will be described later.

In a similar manner a second constant-current source 116 including a FET 118 is controlled by transistor 120 having its base connected to the output 66 of the counter 62, serves to charge a second storage capacitor 122 during the time interval T. The final voltage of the capacitor 122 is then sampled by a second sample and hold circuit 124 including amplifier 125 and capacitor 127 and then discharged by a second FET switch 129. These functions are controlled by a second sequencing circuit 128. Outputs of both sample and hold circuits 112 and 124 are connected to the dividend and divisor inputs of an analog divider 130 by lines 132 and 134 wherein the output of circuit 130 represents the angle .alpha.. The 360 scaling factor can be accounted for either by an attenuating or amplifying device on the divider output represented by the amplifier in dashed lines 136 or preferably by a suitable choice of values of both charging currents and/or storage capacitors 108 and 122.

In some cases it may not be possible to locate the magnetic pickup 16 of FIG. 1 exactly in the top dead center position so that it is offset a predetermined angle after TDC. The previously described circuitry would then display a timing angle which is, by this basic angular difference, larger than the correct timing angle and consequently requires a subtraction of the offset angle. This is easily compensated for both in the digital or analog embodiments of FIG. 2 and 3. In the digital version of FIG. 2 it is only necessary to program the arithmetic unit 92 to subtract the offset angle and display the remainder. In the analog version of FIG. 3 a summing network consisting of resistors 138 and 140, passive or preferably active, can be used on the divider output. One input of resistor 138 is connected to the divider 130 output while the other resistor 140 is connected to a correcting DC voltage -V of an opposite polarity. In this way the offset angle is subtracted and only the correct timing angle is displayed in the display device 100.

In actual operation the pressure waveform on line 14 of FIG. 1 usually varies slightly from stroke to stroke so that the position of the starting point varies and the resulting angle is consequently slightly different for each cycle. It is possible to obtain either a separate different for each cycle or an average reading over a certain number of cycles. Both these cases will be explained below in connection with the analog embodiment of FIG. 3.

The sequencing circuit 114 in FIG. 3 consists of several cascaded one-shot multivibrators, preferably of three one-shots 142, 144 and 146. The first one-shot multivibrator 142 is triggered by the trailing edge of the charging pulse on line 60 and generates a sampling pulse for the first sample and hold circuit 112. The trailing edge of this pulse triggers the second one-shot multivibrator 144, which is used merely for generating a space between the end of the sampling period and the beginning of the discharging process of the storage capacitor 108. At the end of this space, the third one-shot multivibrator 146 is triggered and generates a pulse on line 148 which turns on a discharging FET 150. This pulse starts, therefore, when the sample and hold circuit 112 already has the correct signal and will end before the arrival of the next charging pulse; therefore, it has a suitable time position for providing separate readings of individual cycles. The details of sequencing circuit 128 are not shown, however, it can operate in the same manner as seqencing circuit 114 in controlling FETs 126 and 129 of the sample and hold circuit 124.

In the event that a separate reading in analog format is desired, this pulse can be used as strobing pulse e.g. in an output buffer stage 136 connected to line 148. In the event that a reading in digital format is desired, an analog-to-digital converter 152 is used in the output portion of the measuring system of FIG. 3; in this case the pulse on line 148 is used as conversion command which starts the conversion process in synchronism with individual fuel injections so that separate readings for each cycle are obtained.

In case an average reading over a certain number of cycles is required, the sequencing circuit 114 can be adapted in the following manner: a digital counter shown by dashed lines 154 divides the number of incoming pulses in the required ratio. If, for example, an average reading over ten cycles is to be obtained, counter 154 will be a decade counter driven by trailing edges of the incoming charging pulses on line 66 and will trigger after ten pulses one-shot multivibrator 142. As a result the storage capacitor 108 is successively charged during ten subsequent cycles and its final voltage is then sampled by the sample and hold circuit 112 and discharged by the FET switch 150 after ten charging cycles. In order to obtain a reading which is equal to an average individual reading, it is necessary either to decrease, by a factor ten, the value of the charging current from 102, or to increase by the same factor the value of the storage capacitor 108, or to change both these values, e.g. to increase capacitance five times and decrease current to one half. In this way the capacitor is charged in a stepwise manner during ten cycles to a final value which represents the average value of all ten cycles. This value is then sampled and processed in the divider 130, instead of the original individual values.

Since the charging current and the capacitor value can vary slightly as a result of aging and temperature variations, it is often required to have an internal calibrating source which can simulate a predetermined angle .alpha., preferably one which corresponds to a full-scale output, which usually is in the vicinity of 40 to 50 degrees. According to the present invention this is achieved as shown in FIG. 4. An auxiliary pulse generator 160 generates a train of narrow, equally spaced pulses on line 162 connected to the input of a counter 164 which in turn generates output pulses having a width equal to a multiple of the repetition period of incoming pulses. If the counter 164 consists of four cascaded binary counters, its first output delivers on line 166 pulses whose width is equal to one repetition period while its fourth output on line 168 delivers pulses whose width equals eight periods. Upon depression of a calibration switch 170 the base of the transistor 106 which controls the charging process of capacitor 108 in FIG. 3 is disconnected from flip-flop 56 and connected to the first output 166 of the counter 164. At the same time the base of the second control transistor 120 is disconnected from the binary counter 62 and connected to the other output 168 of the calibrating counter 164. In this way pulses which control the charging process of the first storage capacitor 108 have a duration which is exactly one eighth of the duration of pulses which control charging of the second capacitor 122 of FIG. 3. This, according to equation (1), corresponds to an angle 45 degrees.

The function with an octal counter 164 was explained only as an example; a more convenient application of the same basic idea as illustrated in FIG. 5 contains a decade counter 172 that delivers pulses on line 174 with a width equal to one repetition period of incoming pulses while spacing between pulses equals nine of these periods. These pulses are then connected to the base of the first control transistor 106 while the second control transistor 120 receives the same pulses only with polarity inverted in an inverter 176. In this way a calibrating angle of 40 degrees is simulated.

The second parameter, measurement of which is sometimes required, is the injection duration angle, i.e. the angle during which the fuel pressure is higher than a predetermined level. Since the pressure waveform in transducer may have several peaks due to reflections, it can intersect the reference level several times, so the injection angle may consist of several separate sections. In order to measure this angle only, some minor changes in the previously described circuit are necessary which will be explained in conjunction with FIG. 6.

The output of the comparator 34 in which the pressure waveform is changed into a corresponding square-wave with its width equaling the time interval, or several intervals, during which the pressure is higher than a predetermined level is connected, by means of contacts of a switch 180 to the base of the first control transistor 106 of FIG. 3 and disconnected from the corresponding flip-flop 56. The input on line 60 of the sequencing circuit 114 of FIG. 3 which controls the sampling and discharging process remains connected to the output of the flip-flop 56. It would be possible to leave the reset input on line 54 of flip-flop 56 connected with the output of the second pulse shaper 26 but, in this case, only an injection duration shorter or equal to the timing angle could be measured which, in case of a small timing angle, would be insufficient. Therefore, an additional oneshot multivibrator 182 is preferably included in the reset path by means of another contact set 184 of the switch 180. This results in a certain delay in resetting of flip-flop 56 and in triggering of the sequencing circuit 114. This delay, of course, can have only a limited duration since the discharging process of capacitor 108 must be finished before the arrival of the next injection pulse. The same arrangement for measurement of individual angles or average angles, which was described in previous paragraphs, can also be used for the injection duration.

The portion of the disclosed system which is used for measuring the engine speed in revolutions per minute (RPM) will be discussed next. In the digital embodiment of FIG. 2 the number of clock pulses counted during one engine revolution by counter 85 can be defined as n.sub.2 ; so it is only necessary to compute the value

where f.sub.cl is frequency of the clock 80. This division and multiplication by a scaling factor can be accomplished in the same way as was explained in connection with the timing angle measurement, preferably using the same arithmetic 92 and processing 94 units.

In the analog embodiment FIG. 7, a voltage V.sub.3 is equal to kT (where k is a proportionality factor) of the second sample and hold circuit which in turn is proportional to the duration T of one revolution. In order to obtain a measure of the engine speed the value must be generated:

This requires the division of a reference voltage, V.sub.4, which represents the scaling factor 60k, by the output voltage V.sub.3 of the second sample and hold circuit 124. This is accomplished in an analog divider 186 which has as a dividend input on line 188 the reference DC voltage V.sub.4 from the resistive divider 189 and a divisor input from the second sample and hold circuit 124. The divider 186 then provides the display device 100 with a signal representing engine speed. The advantage of this measurement, as compared with the usual counting method, is the fact that the speed is shown immediately, without waiting for the counting interval to elapse. It is apparent that the divider 130 of FIG. 3, which provides the angle .alpha. measurement, can be used for the speed measurement since only one of its inputs needs to be switched to another point. For checking the accuracy of speed measurement a calibrating circuit based on a comparison with 60 Hz power line frequency can be used. Since such calibration is well-known and used e.g. for calibrating LF oscillators, this operation will not be described.

It is sometimes required to measure the timing angle of diesel engines which have no provision for accommodating magnetic pickup nor a tooth or groove on the flywheel 18 of FIG. 1, but only a timing pointer which coincides with a marker on the engine body when the piston of the first cylinder is in its top dead center position. According to the present invention the arrangement which has been described in the preceding paragraphs can be adapted to work with a strobe light that replaces the magnetic pickup. To use a strobe light for timing a diesel engine is, of course, a very well known procedure but the present invention can be adapted to use a strobe light in a manner that permits cooperation with the rest of the described circuitry. An illustration of the modifications required is provided in FIG. 8. The adjustment of the time delay for a strobe light 190 is done in the conventional way, i.e., by setting adjusting knob the strobe flash so that the pointer lines up with the TDC marker. When the time delay has been adjusted to a particular engine speed, it is no longer necessary to readjust it when the speed is changed since the delay will be changed automatically in response to changes in the engine speed.

Since no magnetic pickup is used, no pulses are coming from the output of the pulse shaper 26, so the input of the binary counter 62 must be disconnected by means of contacts 192 of a switch 194 from the pulse shaper 26 of FIG. 1 and connected to the output 44 of pulse shaper 24 which is driven by pulses of the pressure transducer 10 of FIG. 1. This results in the circuit, including elements 116, 120, 122, 124 and 128 operating in a similar manner as described in connection with FIG. 3.

Another contact set 198 of the switch 194 serves to disconnect input of the flip-flop 56 from the pulse shaper 24 and to connect it to the output 66 of the binary counter 62 so that flip-flop 56 is now driven by every other pressure pulse. When the incoming pulse on line 44 changes the output 66 of counter 62 so that it drops to zero, capacitor 122 starts charging until the arrival of the next pulse. In a similar way capacitor 108 starts charging but, since there is now no pulse from the magnetic pickup 16 which would come to the reset input 54 and stop the charging process, some other means must be used for this function. This means consist of a comparator 200 having as a first input 202 connected to the storage capacitor 122 and a second input 204 connected to the center arm of a potentiometer 206 which in turn is connected, preferably via an additional resistor 208, across the output 134 of the corresponding sample and hold circuit 128.

Since, at a given engine speed, the output of the sample and hold circuit 128 has a certain steady DC voltage, the second comparator input 204 has a certain DC voltage while the first input 202 follows the voltage ramp of the storage capacitor 122. At a certain instant the ramp voltage exceeds the steady voltage on the second input 204 and results in an abrupt change of the comparator 200 output voltage. This change is fed to the following pulse shaper 210, comprising resistor 212, inverter 214, and capacitor 216, which derives from it a pulse for resetting, via a contact 220 of switch 194, the flip-flop 56 and a triggering pulse for the strobe light 190. This pulse shaper can be made in various ways: it can consist e.g. of a differentiator 212 and 216 which supplies directly the reset pulse via contacts 220 and is connected also to the input of the inverting gate 214, which generates trigger pulse for the strobe light 190. Another possibility is to use an additional one-shot multivibrator (not shown) having complementary outputs that would deliver both pulses.

Since the sample and hold circuit 124 has an output voltage which corresponds to two revolutions, i.e. to 720 degrees, it is necessary, in the event that the timing angle is e.g. 45 degrees, in order to trigger the light in a correct instant, to the second comparator input with a part of the sample and hold output which corresponds to 720/45 or 1/16 of its output voltage taken from potentiometer 206. If the engine speed is changed, e.g. increased, then both the capacitor voltage and the sample and hold output voltage will decrease in the same ratio, so that the potentiometer 45 should remain adjusted to 1/16 of the full output. This demonstrates that once the potentiometer is correctly adjusted for a certain angle, it need not be readjusted when the engine speed is changed because the circuit automatically maintains a constant angular delay. Since the flip-flop 56 is receiving a reset pulses from the output of the pulse shaper 210, the remainder of the system works as previously described and delivers an output voltage on line 134 which is then processed by the divider 130 and defines the angle .alpha..

It is to be understood that the preceding description concerns only a typical embodiment of the invention and that various alterations are possible without departing from the scope of the invention. Particularly it should be understood that instead of the tooth or groove on the engine flywheel any intentionally arranged magnetic irregularity on the flywheel can be used for generating a signal in the magnetic pickup, that instead of the flywheel any other rotating part which is driven by the engine, can be used and that individual electronic units can be replaced by units of another type as long as their function remains the the same. In addition, even though the invention was discussed in terms of a diesel engine where the timing angle was defined as the angle between a fuel injection pulse and when the piston reached top dead center, the disclosed instrument would be equally useful in timing internal combustion engines using electrical ignition. For example the point where a spark is applied to the cylinder can be considered functionally equivalent to the fuel injection point for timing purposes wherein they both can be considered a firing event. Also it should be understood that the magnetic pickup unit 16 of FIG. 1 could be replaced by another type of pickup, e.g. by an electro-optical device for indicating the relative position of the flywheel 18.

The basic idea of this invention includes also cases in which the pressure pulse is started at some other point than in the instant when the pressure exceeds a predetermined level, e.g. in the instant when the pressure reached its peak value or when it dropped below a predetermined level.