JPH0261715B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Technical Field of the Invention The present invention relates to an improvement in a method for measuring time intervals, particularly delay times, in signal waveforms.

(b) Background of the technology Traditionally, oscilloscopes or counters have been widely used to observe the waveforms of analog or digital signals, or to measure time intervals between multiple signals or signal change points, that is, delay times at subsequent signals or change points. It is used.

(c) Conventional technology and problems Oscilloscopes usually use a cathode ray tube (CRT) to apply an input signal voltage to one deflection means to make the bright spot swing in the Y direction, and at the same time deflect it in the X direction in synchronization with the input signal. The waveform of the input signal voltage is observed by sweeping with the deflection means.

It has a built-in input signal amplifier, time axis generator, various adjustment circuits, etc., and is widely used as a means of observing signals ranging from DC to several GHz. In addition to multi-phenomenon observation means that simultaneously measure multiple signals using a CRT having multiple sets of input signal deflection means or means for extracting instantaneous values using narrow sampling pulses synchronized with input signals, Since the device is equipped with a means for starting the sweep when triggered, the sweep is started by the change point of the input signal, and the time interval, for example, the delay of the circuit under test, is determined by observing other subsequent input signals or change points. Since the preceding signal that serves as a trigger for time measurement is not necessarily a sharp falling edge or falling pulse, which is the ideal condition for triggering, the reference point may float due to fluctuations in the trigger level, or the value of the delay time may had the disadvantage that measurement accuracy could not be obtained especially when the delay time was large. Another counter method uses a preceding input signal or change point to open the counter gate and start counting the reference clock, and a subsequent input signal or change point closes the gate to measure time. In addition to easy reading through display, it has excellent accuracy and resolution of, for example, 10 ppm. However, since the gate trigger of the counter is usually based on the signal level, it is difficult to detect the desired change for signals with gradual or complex changes. It is inferior to visual confirmation using an oscilloscope in that there is no guarantee that the points will be accurately captured.

(d) Purpose of the Invention The purpose of the present invention is to provide a highly accurate and easy-to-handle delay time measuring means that eliminates the disadvantages of conventional time measurement using an oscilloscope and takes advantage of the visual confirmation provided by the oscilloscope. It is something.

(e) Structure of the invention Means for generating a pulse waveform with a constant reference repetition period; counting means for applying the reference pulse and outputting a long-period pulse having a repetition period n times;
A means for delaying the long-period pulse and a timing observation means using an oscilloscope are provided, and the rise of the secondary pulse obtained by the delay means is made to coincide with an arbitrary rise position of the reference pulse, and is used as a zero reference for time measurement. The pulse under test obtained by applying the pulse to the circuit under test is displayed along with the reference pulse on the same time axis on the screen of the timing observation means, and the delay time of the pulse under test is set to 1 of the reference pulse.
This can be achieved by providing a delay time measuring method which counts cycles as a unit and obtains fractions less than one cycle by enlarging the screen or observing means.

(f) Embodiment of the invention An embodiment of the invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a delay time measuring method according to an embodiment of the present invention, and FIG. 2 is a time chart thereof. In the figure, 1 is the reference oscillator, 2
is a counting circuit, 3 is a delay circuit, 4 is an oscilloscope, 5 is a circuit under test (DUT), and SW is a changeover switch.

The reference oscillator 1 is basically a crystal oscillator with an accuracy of, for example, 100 MHz error ±3 PPM, which is used as a reference for time measurement, and the reference pulse Pa obtained by waveform shaping is applied to the output of the crystal oscillator. Therefore, the required number of pulses shall be sent. Therefore, the repetition period τ of the reference pulse is 10 ns. The counting circuit 2 inputs the reference pulse Pa, performs n-ary counting, and sends out a long-period pulse Pb with a repetition period τ×n. In the example of Figure 2b, n=
10, but any counting circuit with n=2 ⁿ obtained by cascading a plurality of flip-flops may be used, for example, any counting circuit that does not interfere with the measurement of delay time as will be described later. Next, the long-period pulse Pb from the counting circuit 2 is input to the delay circuit 3 and its delay time is adjusted to obtain a secondary pulse Pc in which the rising edge of the output pulse coincides with the rising edge of the reference pulse Pa, for example, delayed by one cycle. Depending on the delay time t ₀ of the long-period pulse Pb, it is sufficient to match the rising edge of the primary pulse Pa which is further delayed. Although not shown, the secondary pulse Pb at any rising position of the primary pulse Pa is generated by the counting circuit 2.
This can be similarly achieved by using the long-period pulse as it is and obtaining it by separately delaying the reference pulse Pa. The above is a reference pulse Pa and a changeover switch for the oscilloscope 4, which has a multi-phenomenon display function.
It is assumed that the SW is set to the calibration side, the secondary pulse Pc is input, and the pulse waveform is observed by sweeping along the common time axis. Next, the changeover switch SW is set to measurement, the secondary pulse Pc is input to the DUT 5, and the obtained pulse to be measured Pd is compared and observed with the reference pulse Pa.
Of course, if the display capacity of the oscilloscope 4 has sufficient capacity, the reference pulse can be displayed without going through the changeover switch SW.
The secondary pulse Pc may be observed simultaneously with Pa and the pulse to be measured Pd. The pulse to be measured Pd is obtained on the screen of the oscilloscope 4 as shown in Figure 2 (d), so the delay time of the DUT 5 is displayed as _t3 , but the upper _t1 of the delay time _t3 is the repetition of the reference pulse ta. The period is 30 ns from 3 cycles, which can be easily determined by counting. Fractional time t ₂ less than 1 cycle, e.g. 5
~6ns can be read directly on the screen of the oscilloscope 4 or, if necessary, in combination with the magnification and delay functions of the oscilloscope 4, allowing the reading on the screen of the oscilloscope 4 to cover a narrower range in a shorter time axis sweep than before. By setting and reading in the time range, a highly accurate delay time t ₃ =t ₁ +t ₂ can be obtained as t ₃ =35.5 ns, for example, from t ₁ =30 ns and t ₂ =5.5 ns.

(g) Effects of the Invention As explained above, according to the present invention, a long-period pulse with a delay time longer than that of the conventional method is used as a measurement pulse, and the rising edge of the reference pulse and the rising edge of the long-period pulse can be By matching, applying the accuracy of the reference pulse to most of the delay time, which is easy to operate, and setting the range of the oscilloscope to read fractions more accurately than before, an excellent method of measuring delay time can be obtained. Therefore, it is useful.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a delay time measuring method according to an embodiment of the present invention, and FIG. 2 is a time chart thereof. In the figure, 1 is a reference oscillator, 2 is a counting circuit, and 3 is a reference oscillator.
4 is a delay means, 4 is an oscilloscope, and 5 is a circuit to be measured.

Claims (1)

[Claims] 1. Means for generating a pulse waveform with a constant reference repetition period, counting means for applying the reference pulse and outputting a long-period pulse having a repetition period n times, means for delaying the long-period pulse, and timing observation using an oscilloscope. The pulse to be measured is obtained by applying the secondary pulse to the circuit under test, using means as a zero reference for time measurement while making the rise of the secondary pulse obtained by the delay means coincide with any rising position of the reference pulse. is displayed along with the reference pulse on the same time axis on the screen of the timing observation means, and the delay time of the pulse to be measured is counted in units of one cycle of the reference pulse.
A method for measuring delay time, characterized in that a fraction less than a cycle is obtained by an enlarged display operation of the screen or observation means.