JPH0456504A - Digital compensation method for oscillated frequency and digital compensation oscillator - Google Patents

Abstract

PURPOSE:To reduce a storage capacity of a compensation memory by dividing a frequency compensation characteristic into plural small regions and storing in advance a frequency increment/decrement between the small intervals as a relative compensation value being the result of comparison between preceding and succeeding small regions sequentially. CONSTITUTION:A compensation memory 9 stores data to compensate the characteristic in response to the temperature compensation characteristic shown in a cubic curve A, for example. In the temperature compensation characteristic, let a frequency compensation quantity in each of small regions T1-Tn be a prescribed value DELTAV, a compensation quantity of a small region Tn is compared with that of one preceding small region Tn-1 and an increment/ decrement in the compensation quantity is set as a relative compensation quantity of +1 or -1. An address section 10 and a compensation quantity decision section 11 decide sequentially the relative compensation quantity based on a reference compensation quantity as a start point and integrate sequentially the decided relative values to decide a compensation quantity Vs. In this case, the compensation data consists of the relative compensation quantity in which the increment/decrement of the compensation quantity between the small intervals is represented respectively as +1 or -1 and is made up of a minimum bit number in the unit of one bit. Thus, the storage capacity of the compensation memory is reduced.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention digitally analyzes frequency characteristics caused by physical quantities.
The field of application is a digitally compensated oscillator, and particularly relates to a digital temperature compensated oscillator in which the physical quantity is temperature and the storage capacity of the compensation memory is significantly reduced.

(Background of the Invention) A crystal oscillator has frequency-temperature characteristics (hereinafter referred to as temperature characteristics) due to the crystal resonator. For this reason, crystal oscillators whose load capacitance is changed according to temperature to compensate for temperature characteristics are becoming popular.For example, one such device is a so-called voltage sub* Type (VC
There is a temperature compensated oscillator that applies voltage to the voltage variable capacitance element (XO).

In recent years, digital devices have been attracting attention, in which temperature compensation characteristics corresponding to temperature characteristics are stored as digital information in a compensation memory, and a compensation voltage is thereby supplied.

(Prior Art) Reference numeral 145i1 is a diagram illustrating such a conventional example, where (a) is a schematic block diagram of a temperature compensation oscillator and (b) is a schematic block diagram of a temperature compensation circuit.

The temperature compensated oscillator has an oscillation circuit 1 of, for example, a Colpitts type, and a voltage variable capacitance element 3 connected to a crystal oscillator 2.
o. A temperature compensation circuit 4 for applying a compensation voltage to the voltage variable capacitance element 3 is connected thereto. The temperature compensation circuit 4 includes a compensation memory 5, a temperature detection section 6, and an A/D conversion section 7.
and a D/A converter 8. The compensation memory 5 stores compensation data of temperature compensation characteristics corresponding to the temperature characteristics of the crystal oscillator. As shown in FIG. 5, the temperature compensation characteristic is represented by the amount of compensation voltage corresponding to the amount of frequency compensation at each temperature. The compensation data includes temperature compensation characteristics in multiple microregions T++~. ). Then, the compensation value corresponding to each compensation voltage amount is made into a digital value and stored in each address of the compensation memory 5. The temperature detection section 6 is composed of, for example, a thermistor circuit network, and generates a detection voltage in response to the ambient temperature. The A/D converter 7 converts the detected voltage into a digital value and selects an address in the compensation memory 5 corresponding to the digital value. D/A conversion #8 converts the compensation value from the compensation memory 5 into a compensation voltage, and applies it to the voltage variable capacitance element 3 to compensate for the temperature characteristics (maintains the reference frequency).
.

(Origin between prior art) However, in such a thing, the minute area T (+ ~
. ) for each frequency compensation J1 (compensation voltage amount)
Since the data is basically stored as an absolute amount corresponding to , there is a problem that the number of bits is increased and a large-capacity compensation memory 5 is required. Note that when the temperature changes rapidly, the selected addresses of the compensation memory 5 become discontinuous, so it is necessary to use a random access type storage element with a high bit unit price as the compensation memory 5.

(Objective of the Invention) An object of the present invention is to provide a digital compensation method for an oscillation frequency and a digital compensation oscillator that can significantly reduce the storage capacity of a compensation memory.

(Solution Means) The present invention divides the frequency compensation characteristic into a plurality of microphysical regions, uses the frequency compensation amount in the reference physical quantity as a reference compensation value, and calculates the frequency increase/decrease between each microphysical region from the reference compensation value. , the compensation value for each physical quantity during the operation of the oscillator is stored in advance as a relative compensation value compared to the previous microphysical region sequentially, and the compensation value for each physical quantity during the operation of the oscillator is divided into a relative compensation value and a reference compensation value according to the increase/decrease in the physical quantity. The basic solution is obtained from the integrated value of . Hereinafter, a case in which the present invention is applied to a temperature compensated oscillator will be described in detail as an example.

(Embodiment) FIG. 1 is a block diagram of a temperature compensation circuit in a temperature compensation oscillator explaining an embodiment of the present invention.

As mentioned above, the temperature compensated oscillator uses a crystal oscillator as VCXO.
The temperature compensation circuit consists of a compensation memory 9, an address selection section 10, and a compensation value determination section 11. Each part operates in synchronization with clock pulses.

The compensation memory 9 stores in advance data for compensating for the temperature compensation characteristic, which is, for example, a cubic curve. The temperature compensation characteristic is expressed by a cubic curve (I [as shown in the schematic diagram in Figure 2).
Let the amount of frequency compensation (hereinafter referred to as compensation amount) within each minute region T (+-nl) be a constant value. Then, the increase or decrease between each minute region T + + ~.) is set to a constant compensation amount ( ΔV) is replaced by a stepwise curve (b). However, the compensation amount (V) is a voltage value corresponding to the frequency amount to be corrected. Compensation data Dz-n+ is based on the 5 Wi step upper curve (b).

Each minute area T. The compensation amount is compared with the previous minute region T0-4, and the increase or decrease in the compensation amount is formed as a relative compensation value of +1 or -1. In this embodiment, the reference temperature tiI (reference region Tl1) is set as the low end of the temperature compensation region where the amount of compensation is the smallest. Then, the compensation amount for the reference temperature and date and time is set as the reference compensation amount Vll, and this is set as the reference compensation value (in this case, 0). Further, relative compensation values are sequentially determined using this reference compensation value as a starting point.

In this example, the compensation data Dt+~. ), each relative compensation value is
At each temperature j+s, +~n), address A-5 of compensation memory 9
. In addition, compensation memory 9
It is assumed that the address A@ is included therein for convenience, but in reality it is not included.

Table 1 Temperature t@t+ t2°・ts tv t8°°t+@t
z t+2°' n-I Ln address A@AI A2...
A6 A7 A3-Al＠＾++ A-Electric 2...＾. −+
An value 0 1 2-- 6 7 8-10 11
12=n-1n theta Ds D+ D2・-D
s 07 Ds-・D+@ D++ D+2・-
Do-+ D-value 0 +1 +1...+1
-1 +l...-1-1-1...-+1+1 The address selection section 10 includes the comparison section 12. Temperature detection section 13-1! LD/A
It consists of a converting section 14 and a first counting section 15. The comparison unit 12 compares the ambient temperature voltage Vt corresponding to the ambient temperature from the temperature detection unit 13 and the operating temperature voltage v0 from the first A/D conversion unit 14. If the ambient temperature (voltage Vt) is higher than the operating temperature (voltage V.), set +1, and if lower, -
The judgment value for either one of 1 is output. First counting section 15
functions as a so-called integration counter, and its initial value is set to 1, for example, an 8-bit O value (binary number) in response to the reference address code. Then, the judgment values from the comparator 12 are integrated to determine the operating temperature value n together with the 1st address code n. However, the operating temperature t0 is the temperature t corresponding to the selected address A7 of the compensation memory 9 during operation of the oscillator. The first D/A converter 14 converts the operating temperature value n into an analog operating temperature voltage V e + t n +.

The compensation value determination section 11 includes a selection section 1, a code setting section 17, a second counting section 18, and a second A/D conversion section 19. The selection unit 16 includes, for example, a switch and a delay element, and includes a comparator 1
It operates based on the judgment value from 2. That is, the operating temperature t
If n is increasing and the judgment value is +1, the relative compensation value output from the compensation memory 9 is selected as is, and the opposite -
In the case of 1, the relative compensation value held in the delay element one clock ago is selected. This is the operating temperature t. This is to prevent the address An from being shifted by one when the determination value is +1 while falling. The sign setting section 17 operates based on the determination value in the same manner as described above, and determines the sign of the relative compensation value from the selection section 16. That is, when the determination value is +1, the sign of the relative compensation value is left unchanged. Conversely, in the case of -1, the sign of the relative compensation value is inverted and used as the determined relative value. The second counting section 18 functions as an integration counter similarly to the first counting section 15, and is set to, for example, an 8-bit 0 value (binary number) with an initial value as a reference compensation value. Then, the determined relative values from the sign setting section 17 are sequentially integrated to determine the compensation value. 2nd D
/A converter 19 converts the integrated compensation value into an analog compensation voltage V
, and is applied to the voltage variable capacitance element 3 of the vcxo.

Below, the operation of the temperature compensated oscillator with such a configuration will be explained as follows.
This will be explained in more detail with reference to the timing charts of FIGS. 2 and 3. 3(a) shows the first counting section 15 for the clock pulse P, FIG. 3(b) shows the compensation memory 9, FIG. 3(c) shows the sign setting section 17, and FIG. 3(d) shows the same. j[2 is the output waveform of the counting section 18.

In such a device, first, when the power is turned on, Ii! 1
and I! The second counting section 15.18 receives the 0 value (0
0000000). That is, it is an operating state where the operating temperature value is 0, and the subsequent clock pulses p++~. ). and.

Ambient temperature t+x+ is operating temperature tl! For example, if the room temperature is 25°C, the ambient temperature t8 at power-on is higher than the operating temperature voltage tI11. Therefore, the first clock pulse p1 causes the comparison unit 12 to set the judgment value to +1, and the first counting unit 15 integrates the Values (address code, operating temperature value) (
1). Then, compensation data D for address code (1)
I(+1) is sent to select 516. Since the selection unit 16 receives the determination value of +1 from the comparison unit 12, it outputs the compensation data DI (+1) as it is to the sign setting unit 17 and holds and stores it. Similarly, the sign setting unit F sends the compensation data D+(+1) to the tJ2 counting unit 18 as a determined relative value based on the judgment value +1 without changing the sign. Second
The counting unit 18, like the Irl counting unit 15, sets the integrated value of tl to (1) and uses this as a compensation value. On the other hand, 1st D/A
Transducer 14 receives the operating temperature value (1) and increases the operating temperature voltage. From then on, the operating temperature t is similarly synchronized with the clock pulse Pf2-1. is t2
As the ambient temperature increases from tll to just before t0,
The integrated compensation value is obtained by integrating each determined relative value (3).

Next, at the rising edge of clock pulse P+2, operating temperature 1++ is still lower than ambient temperature t8, so comparator 1
2 is in a state where the judgment value is maintained at +1.

Then, due to the rising edge of clock pulse p+2.

Since the operating temperature rises to tl2 and the 1 judgment value is +1, the selection section 16 sends the compensation data DI2 (-1) as is to the sign setting section 17. And sign setting #17
is directly determined as a determined relative value (-1) based on the judgment value +1, and the integrated value is made (2). Immediately after that, the operating temperature t+2 becomes i
Since the IR ambient temperature tx is exceeded, the judgment value is set to -1.

Then, the operating temperature is set to 1++ by clock pulse PI3.
descend to When the clock pulse PI3 rises, the operating temperature 1++ is still higher than the ambient temperature, so the comparator 12 maintains the judgment value at -1. Then, with the rise of the clock 6rus p+3,
The operating temperature drops to 1++. Also, since the determination value is -1, the selection unit 16 selects the compensation data D12 from one clock ago.
(-1) is sent to the code setting section 17. Then, the sign setting unit 17 inverts the sign based on the determination value -1, sets the determined relative value to +1, and sets the integrated value to (3). In this way 1
The ambient temperature is between the operating temperature 1++ and tl2 (temperature range T
12), the compensation value calculated by 11t is (2)
Or (3).

Then, a compensation voltage corresponding to this is applied to vcxo, and the oscillation frequency at the operating temperature 1++ and tl2, which are alternately selected, becomes a steady state.

As described above, in this type of device, the determined relative value in response to the ambient temperature and its increase/decrease direction is sequentially integrated to the reference compensation value at the reference temperature tl1, and the amount of compensation voltage is determined by the integrated compensation value. , allowing digital temperature compensation. Since the compensation data D consists of relative compensation values expressing the increase/decrease in the amount of compensation between each minute temperature region as +1 or -1, each is formed with the minimum number of bits of 1 bit. Therefore, The storage capacity of the compensation memory 9 is basically only required for the number of minute areas, so it is significantly reduced compared to the conventional example.

In addition, in this embodiment, since the reference compensation value is set to the low end of the temperature characteristic with the smallest amount of frequency compensation, the initial values of the first and second counting sections 15.18 can be set to O, and the integration counter It is advantageous in terms of hardware when configuring.

(Other matters) Note that in the above embodiment, the address selection section 10 is replaced by the comparison section 12,
It is composed of a temperature detection section 13, a first D/A conversion section 14, and a counting section 15, but the point is that it has a function of specifying the address of the compensation memory according to the ambient temperature and recognizing the direction of increase/decrease in the ambient temperature. The configuration is not limited to the above configuration. In addition, although the 1 selection section 16 and the sign setting section 17 are provided between the compensation memory 9 and the Wi2 counting section 18 to determine each compensation data Dl-0), they are substantially the same regardless of the location and method. Those with functions correspond to the selection section and code setting section here.

Moreover, each minute region Tz~ with step-like characteristics. ) is +1 or -1 compared to the previous minute area.
For example, if the temperature characteristic has a large slope and the increase/decrease cannot be covered by +1 or -1, that part can be expressed as a relative compensation value of +2 or 12 or more. . In this case, the compensation memory requires two or more bits for the small area, but even such a memory can produce the effects of the present invention. Further, if the relative compensation value is set to three values including O, for example +1.0 and -1, and the flat portion is made to correspond to O, it is possible to keep the compensation amount constant and reduce frequency fluctuation.

Further, although the reference compensation value is set to the low end where the amount of frequency compensation is the smallest, the same effect can be obtained even if the reference compensation value is set at, for example, room temperature. Further, the reference compensation value does not necessarily have to be 0, but may be a value other than O, for example, and the initial values of the first and second counting sections 15.18 may be made to correspond to this value.

In addition, the judgment value from the comparator 12 is a binary value of +1 or 1, but if the temperature is a constant value within the allowable range, for example, if the temperature is a constant value within the allowable range, the judgment value is set to 0 and the value between The compensation value may be configured not to change. For example, the determination value from the comparator 12 and the relative compensation value from the compensation memory 9 may be multiplied to produce the output of the sign determination I$17.

Note that according to the present invention, an inexpensive sequential type memory element can be used instead of an expensive random access type memory element. In this case, the first counting section 15 of the embodiment
The circuit can be configured by integrating the compensating memory 9 and the compensating memory 9, which greatly simplifies the circuit and has extremely high industrial value.

Furthermore, the present invention is not only applicable to a temperature-compensated oscillator using temperature as a physical quantity, but also can be used, for example, to compensate for oscillation frequency due to fluctuations in power supply voltage, and can be modified as appropriate without departing from the spirit thereof. It is.

(Effects of the Invention) The present invention divides the frequency compensation characteristic into a plurality of microphysical regions, uses the frequency compensation amount in the reference physical quantity as the reference compensation value, and calculates the frequency increase/decrease between each microphysical region using the reference compensation value. As a starting point, it is stored in advance as a relative compensation value sequentially compared to the previous microphysical region, and during the operation of the oscillator, the compensation value at each physical quantity is divided into a relative compensation value and a reference compensation value according to the increase/decrease in the physical quantity. Since the calculation is made from the integrated value of the oscillation frequency, it is possible to provide a digital compensation method for the oscillation frequency and a digital compensation oscillator that greatly reduce the storage capacity of the compensation memory.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a temperature compensation circuit in which an embodiment of the digital compensation oscillator of the present invention is applied to a temperature compensation oscillator. FIG. 2 is a schematic diagram illustrating digital temperature compensation of the temperature compensated oscillator of the above embodiment. FIG. 3 is a timing chart, in which FIG. 3(a) shows the first counting section for the clock pulse in the above embodiment, FIG. 3(b) shows the compensation memory, and FIG. 3(c) shows the sign setting section. Same figure (d)
is an output waveform diagram of the second counting section. FIG. 4(a) is a block diagram of a digital temperature compensation oscillator illustrating a conventional example, FIG. 4(b) is a block diagram of a temperature compensation circuit, and FIG. 5 is a schematic diagram illustrating temperature compensation of the same temperature compensation oscillator. Figure 1] 3 Figure 3 (α)%+<+, aif' (4r koJ' 叡r
(￥ FJL cliff phantom 1) Ku・2 Pai Shiiri P Enemy voice plastic Figure 4 Figure 5

Claims (7)

[Claims] (1) The frequency compensation characteristic corresponding to the frequency characteristic of the oscillator caused by the physical quantity is divided from the reference physical quantity into multiple microregions, and the frequency compensation amount in each microregion is expressed as a constant value, and the frequency compensation characteristic between each microregion is The increase/decrease in the frequency compensation amount is stored in advance as a relative compensation value having a sign compared with that in the immediately previous minute region, using the frequency compensation amount at the time of the reference physical quantity as a reference compensation value, and during the operation of the oscillator, Selecting the relative compensation value according to the increase or decrease in the physical quantity, and re-determining the sign of the relative compensation value according to the direction of increase or decrease in the physical quantity to make it a determined relative value,
A digital compensation method for an oscillation frequency, wherein the determined relative value is integrated with the reference compensation value to obtain an integrated compensation value for each physical quantity, and the frequency characteristic is compensated based on the integrated compensation value. (2) The digital compensation oscillator according to claim 1, wherein the physical quantity is temperature. (3) The frequency compensation characteristic for the frequency characteristic of the oscillator caused by the physical quantity is divided from the reference physical quantity into multiple minute regions, and the frequency compensation amount at the time of the reference physical quantity is used as the reference compensation value, and the frequency compensation amount between each minute region is a compensation memory that stores the relative compensation value in advance at each address; and the oscillator; an address selection unit that selects an address of the compensation memory according to an increase or decrease in the physical quantity during operation, and re-determines the sign of the relative compensation value stored at the address according to the direction in which the physical quantity increases or decreases; a compensation value determining unit that sequentially integrates the relative compensation value whose sign has been redetermined as a determined relative value on the reference compensation value and determines a frequency compensation amount corresponding to the accumulated compensation value. . (4) The digital compensation oscillator according to claim 1 or 3, wherein the relative value is +1 or -1. (5) The address selection section takes the address code for selecting the address of the compensation memory as an operating temperature value, compares the operating physical quantity based on the operating temperature value with the surrounding physical quantity, and outputs a judgment value;
4. The digital compensation oscillator according to claim 3, wherein the address code is determined by sequentially integrating the determination value according to an increase or decrease in the surrounding physical quantity. (6) The digital compensation oscillator according to claim 5, wherein the judgment value is +1 or -1. (7) The digital compensation oscillator according to claim 2, 3, 4, 5, or 6, wherein the physical quantity is temperature.