 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a non-volatile counter, and more
precisely, relates to a non-volatile counter in which data is stored in
erasable and writable semiconductor non-volatile memory devices.
2. Description of the Related Art
A non-volatile counter in which a counter circuit for counting pulses and a
non-volatile memory device such as EEPROM or the like are arranged in
parallel to each other, wherein when the value of the counter is changed
that value is written in the EEPROM, is well known.
The non-volatile counter can be used as an electronic trip meter of a car
or a data collecting device in a plant or the like as it can hold data in
the memory after the electric power supply thereto is shut off.
The non-volatile counter provided with the non-volatile memory device such
as EEPROM or the like, however, has a problem in that a maximum counting
value to be counted is necessarily restricted to a certain value, due to a
limitation of the number erasing and writing operations of the EEPROM.
Note, since the EEPROM has an endurance characteristic, the number of the
erasing and writing operations thereof is naturally limited; for example,
when a binary counter is used, the number of changes of data is largest in
the least significant bit (LSB), and thus the upper limit of the counting
value is inherently decided by the capacity to change the number of data
per unit time in the least significant bit (LSB).
Further, when a deficit arises in a memory cell and when the bit in which
the deficit arises is a higher digit, the data obtained therefrom is
completely different to the correct data.
To overcome these problems, a counter in which a plurality of memory cells
for memorizing the count value, especially in a lower digit, is provided
in the EEPROM, and the count values for the lower digit are sequentially
written in one of the memory cells in turn, to thereby improve the upper
limitation of the count value, has been proposed, or a counter in which
the same counted values are stored in a plurality of the memory cells and
one data thereof is selected therefrom by applying a majority logic
thereto, to thereby improve a reliability thereof, also has been proposed.
Even when these counters are used, however, the problems above are not
necessarily completely resolved.
Note, when these counters are used, other problems arise in that the number
of the cells is naturally increased, and moreover, a monitoring operation
for checking a number of writing operations of the memory cells or an
operation for processing a majority logic to check the reliability of the
cells, must be carried out separately.
Further, when performing these operations, a micro-computer for controlling
these operations is usually required, to thereby make the construction of
the cells complicated, and accordingly, the reliability thereof will be
lost.
Further, in this case, when the micro-computer malfunctions, it becomes
difficult to read the count value.
Therefore, to overcome these problems, the present invention provides a
non-volatile counter having a simple construction which can improve the
upper limitation for the counting value, and further, improve the
reliability of the memorized counting value.
To attain the above object of the present invention, there is provided a
non-volatile counter which updates data stored in erasable and writable
non-volatile memory devices based on count data and performs a counting
operation, said memory devices being arranged as an array of cells having
a predetermined number of bits, wherein said counter is provided with a
selecting means for selecting a cell corresponding to a value of a
predetermined digit of said count from said cell array and a writing means
for writing data corresponding to the value of a digit different from said
predetermined digit in said selected cell.
3. Mode of operation
The non-volatile counter used in the present invention carries out a
counting operation by updating the data stored in the cell array having a
predetermined bit number and forming an erasable and writable non-volatile
memory device, based on count data.
Note, in the present invention, a cell corresponding to a value of the
predetermined digit of said count from said cell array is selected by the
selecting means and data corresponding to the value of a digit different
from said predetermined digit is written in the selected cell by the
writing means.
Namely, in the present invention, a plurality of cells are not simply
provided in the non-volatile memory device for storing the count value but
are arranged in a cell array, and thus a specific cell in the cell array
can be designated by a value of a predetermined digit of the count value.
Therefore, the number of writing operations for writing data and the
erasing operation for erasing same from each of the cells, can be reduced
in proportion to the number of cells arranged in the cell array, and
further, each cell can carry not only a value of a certain digit which the
cell has memorized but also a value of a certain digit with respect to a
position thereof in the cell array.
4. Effect of the Invention
According to the non-volatile counter used in this invention, cells
comprising an array of a predetermined number of bits, constituting a
volatile memory, are selected according to the predetermined digits of the
count and data corresponding to the value of another digit is written in
the cell, so the numbers of the erasing and writing operations per cell
can be remarkably reduced at the lower digits of the conventional simple
counter, and thus a superior and improved reliability of the count to be
stored is obtained.
Further, each cell comprising the non-volatile memory carries not only
information on the value of a certain digit stored by the cell but also
information on the value of a predetermined digit obtained by the position
of the cell in the array, so the volume of the information stored can be
increased compared to the case wherein a plurality of cells is used.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic diagram of an odometer of an automobile to which the
present invention is applied;
FIG. 2 is a block diagram of an IC circuit used in a control means shown in
FIG. 1;
FIG. 3 is a table of the timing of adjustment of the pulse counting
operation;
FIG. 4 is a flow chart of a program for changing the display value used in
this invention;
FIGS. 5 and 6 are charts illustrating examples for switching the display
value and relationships between each switching operation and the actual
number of the output pulses;
FIG. 7 is a block diagram of a device for displaying a physical measure of
this invention;
FIG. 8 is a table illustrating another example of the timing of adjustment
of the pulse counting operation;
FIGS. 9A, 9B, 9C and 9D collectively show one embodiment of a counter used
in this invention showing a non-volatile memory counter;
FIGS. 10(A) and 10(B) show a code system used with the counter shown in
FIG. 9;
FIG. 11 is one embodiment of a circuit for a row-decoder used in the
non-volatile memory counter shown in FIG. 9; and
FIGS. 12A, 12B, 13, and 14 are charts illustrating the embodiments used
with the coding system in the counter shown in FIGS. 10(A) and 10(B).
DESCRIPTION OF THE PREFERRED EMBODIMENTS
The present invention will be explained with reference to a digital
odometer for an automobile.
FIG. 1 is a block diagram schematically showing a digital odometer for an
automobile. In FIG. 1, 1 is a reference pulse generator, mounted on, for
example, a gear of a wheel of an automobile, which generates a pulse
(reference pulse) at every rotation of the wheel. The generator 1 is set
to generate 637 pulses per kilometer running.
Reference 2 is a control device which calculates the running distance of
the automobile in either units of kilometers or units of miles by counting
the clock pulses from the reference pulse generator 1. Reference numeral 3
is a switching means provided on the control device 2 for switching the
processing operation in the control device 2 between the units of
kilometers for automobiles used in Japan and units of miles for
automobiles used in the U.S. Reference numeral 4 is a display device which
displays the running distance of the automobile on a panel provided inside
the automobile cabin in accordance with the state of the switching means
3.
FIG. 2 is a block diagram of an IC circuit used inside the control device 2
shown in FIG. 1.
In FIG. 2, reference numeral 5 is a .mu.ROM including a control program for
commanding the transfer of data. Reference numeral 6 is a kilometer
display EEPROM which stores the running distance of the automobile in
units of kilometers. Reference numeral 7 is a mile display EEPROM in which
is stored the running distance in units of miles. The value of the EEPROMs
6 and 7 can be selectively displayed on the display panel inside the
automobile cabin by the display device 4. In this invention, a RAM may be
used instead of using the EEPROMs.
Reference numeral 8 is, a RAM for storing a count of the clock pulses
generated from the generator 1, 9 is a data ROM for previously storing a
predetermined mile display and a predetermined count of the clock pulses,
10 is an incrementor, 11 is a comparator, and 12 is an internal bus.
The basic principle for adding the running distance used in this invention
will now be explained. As explained above, the generator 1, which
generates one reference pulse corresponding to a length of one rotation of
a wheel of an automobile, i.e., the first physical measure, generates
exactly 637 pulses per kilometer. The technology for displaying the
running distance in units of kilometers from the reference signal pulses
is well known, so the explanation thereof will be omitted.
When it is judged necessary from the state of the switching device 3 to
display the distance in miles, as mentioned earlier, it is sufficient to
convert it using about 1025.152128 pulses as 1 mile (corresponding to the
second physical measure defined in this invention and a physical measure
not a whole multiple of the reference signal pulses).
However, counting reference clock pulses in decimal fractions requires a
complexed processing operation and additional new memories, so is
disadvantageous.
On the other hand, if one adds one mile at every 1025 reference pulses, as
the running distance increases, the value displayed will gradually become
larger than the actual running distance, so the problem arises that the
accuracy will increasingly deteriorate. For example, in such a system,
when 10250 pulses are counted, 10 miles will be displayed, i.e., there
will be an error of one pulse with respect to respect to 10,251 pulses.
In the embodiment of this invention, to overcome this problem, the count of
the clock pulses is adjusted just once at some one point before 10 miles'
running, whereby the actual number of clock pulses is made 10,251, though
only 10,250 pulses were really counted. Therefore, the accuracy can be
improved. This operation is repeated with every 10 miles' running by a
suitable program.
Even performing such an operation every 10 miles, the system displays 100
miles when 102,510 pulses are counted, so even in this case, the 100 miles
is displayed 5 pulses earlier than when an actual 102,515 pulses are
counted.
Accordingly, it is advantageous to adjust the count by five pulses before
each 100 miles is reached. That is, assuming that 1025.152128 pulses
correspond to 1 mile and using as the approximate number of pulses the
uppermost four digits of the same, i.e., 1025, it is sufficient to make
adjustments for the value 0.152128, below the decimal point, the same
number of times as the figures of each digit at a suitable timing.
The basic idea behind this timing is that, if the digits following the
lowest digit of the approximate number of pulses 1025 is the l th digit (l
is a natural number) lower than the lowest digit, adjustment of the
counting of the clock pulses the same number of times as the figures of
each digit before 1025.times.10 l pulses are counted, would enable
suppression of the conversion error to mile units to an extremely small
level. (Note that, in this embodiment, "adjustment" means to count extra
clock pulses).
For example, even by making adjustment to only up to the third digit
displayed (up to 1000 mile) and without making any adjustment to the
fourth digit on, the display error over the true value after even one
million miles is less than 0.2 mile.
Accordingly, in the present invention, a quite practical value for
indicating the actual running distance can be obtained.
On the other hand, the information indicating when such adjustment should
be carried out is stored in a predetermined address of a data ROM 9. One
example of the timing is indicated in (a) of FIG. 3. Note that (a)
represents a timing table indicating the times at which an adjusting
operation takes place.
FIG. 3 is a two-dimensional map and shows the state of storage in the data
ROM 9 of display data to be adjusted.
The vertical column shows the digits corresponding to the current displayed
value (units of mile) (with 0.1 mile as the lowest digit).
The horizontal row shows the value of change of the current displayed
values corresponding to each digit.
In FIG. 3, circles C, D, E, F, and G are given to certain addresses of the
table. Adjustment is performed when the current displayed value moves to
those addresses. At portions with no circles, the usual operation goes on
and the display value is changed at every 1025 pulses counted.
Each of circles C through G represent a time at which the adjustment should
be carried out for the 10 mile digit. Thus, when the 10 mile digit changes
from 1 to 2 as shown by reference C in FIG. 3, the adjustment takes place,
and the 10 mile digit number is changed from 1 to 2 after 1026 pulses have
been counted. Other circles D-G in FIG. 3 represent similar timings.
On the other hand, regarding to the digits below the decimal point,
normally, the display value is changed at every 102 pulses counted, but in
circle mark portions, the display value is changed at every 103 pulses.
For example, supporting that the current displayed value is 244.6 miles.
When this value is about to change to 244.7, adjustment is made since a
circle is stored at the address A in the horizontal column of FIG. 3 for
the 0.1 mile digits.
As explained above, in this situation, the current displayed value is
changed to 244.7 after 103 pulses are counted.
Suppose that the current displayed value is 243.9. When this value is about
to change to 244.0, adjustment will be carried out because a circle is
stored in the address B in the horizontal column in FIG. 3 of the 1 mile
digit. In this situation, the current displayed value is changed to 244.0
after 1026 pulses are counted.
The portion (b) of FIG. 3 shows the stored values for the count of clock
pulses in the RAM 8 when the current displayed value is changed. The
parenthesized numerals shows the actual number of output pulses of when
the adjustment is carried out.
The accuracy can be improved by uniformly dispersing the adjustments at
equal intervals, but it is also possible not to specify any particular
locations and set them by random numbers.
It is preferable that the storage addresses of the digits of the display
data to be adjusted should be kept from overlapping as much as possible.
In this invention, when the current displayed value is changed, the timing
of this is based upon the time when the value of the lowest digit of the
current displayed value changes from the predetermined value to a
successive value. However, it is preferable that the change from the
predetermined value to the successive value be from 9 to 0.
The operation of this invention will be explained with reference to the
flow chart shown in FIG. 4.
Note that all the following processing steps are previously stored in the
.mu.ROM 5.
FIG. 4 shows the processing in the program for changing the current
displayed value.
At step 100, it is discriminated whether a pulse generated by a reference
pulse generator 1 is input in this system. In other words, if a clock
pulse is recognized, the answer to the question shown in the diamond
representing step 100 is "Yes," and the program proceeds to step 150. If a
clock signal exists, the answer to step 100 is "No," and step 100 is
repeated. Each time a pulse is input, steps 150 on are executed.
At step 150, it is determined whether a flag F indicating the timing for
carrying out the adjustment is ON or not. When F=0 at step 150 (flag is
OFF), the process goes to step 200.
As step 200, data of a count n of the clock pulses stored in the RAM 8 is
transferred to an incrementor 10 through an internal bus. The count is
incremented by 1, and the value n+1 is set as a new count N.
At step 300, this n is transferred to a comparator 11. A count m (in this
case, m is 1025), previously stored in the data ROM 9 as an approximate
pulse number, is also transferred to the comparator 11. The value n and m
are compared.
If the value n reaches the value m (n.gtoreq.m), the process goes to step
400. If the value n has not yet reached the value m, the process goes to
step 700 and the instant value n is returned to the RAM 8 and stored
therein.
At step 400, the current displayed value indicated in units of miles stored
in the EEPORM 7 is transferred to the comparator 11, while the table
indicating the timing at which the adjustment of the clock should be
carried out and stored in the data ROM 9, is transferred to the comparator
11.
At step 500, utilizing the table, it is determined whether the values of
each digit of a display data to be adjusted and the current mile displayed
value coincide. When they coincide (that is, adjustment should be carried
out), the process goes to step 600. If not, the process goes to step 800.
At step 600, a flag F indicating the adjustment is ON, i.e., F=1. If F=1,
when the next pulse is input, the process goes to step 160 through step
150, whereby the flag F is turned OFF (F=0) and the process goes directly
to step 800 without going through steps from 200 to 500.
In accordance with this process, the adjustment of this invention is
carried out and the current displayed value is changed after an extra 1
pulse is counted.
At step 800, the count n stored in the RAM 8 is reset to zero, i.e., the
number n is cleared. At the following step 900, the displayed value in the
EEPROM 7 is incremented by 1.
After steps 600, 700, and 900, the process goes to the next routine.
According to the process as mentioned above, referring to the 10 digit
place, for example, the display value is changed along with addresses C,
D, E, F, and G in the 10 mile digits of FIG. 3 as 19 miles.fwdarw.20
miles, 39 miles.fwdarw.40 miles, 59 miles.fwdarw.60 miles, 79
miles.fwdarw.80 miles, and 99 miles.fwdarw.100 miles. In actuality, the
change is made after the automobile runs 1026 pulses instead of 1025
pulses.
FIGS. 5 and 6 show the relationship between the change of the display by
the display device 4 in the passenger compartment and the actual number of
pulses output from the reference pulse generator 1.
FIG. 5 shows the case in which the lowest unit of the display is one-tenth
of a mile, while FIG. 6 shows the case in which the lowest unit is one
mile.
According to this invention, the display of miles can be realized with a
high accuracy by just switching a switching device. In this embodiment,
further, the adjustment can be realized by amending the control program of
the .mu.ROM 5 and slightly increasing the capacity of the memory devices,
so production costs are almost the same as in the prior art where no
adjustment is performed.
Another embodiment of this invention will be explained below. This
embodiment is characterized by the method for storing the display data to
be adjusted in the data ROM 9. The constituent elements, circuits, etc.
are the same as those in the previous embodiment, so an explanation
thereof will be omitted.
In this embodiment, a reference pulse generator 1 which generates exactly
2548 pulses per kilometer and 4100.608512 pulses per mile is used.
Further, in the approximate number of pulses is set at 4100, and
adjustment is carried for three digits following the lowest digit of the
approximate number of pulses (i.e., 0.608).
Note, for example, the display data to be adjusted stored at the address P
in FIG. 8. When the current displayed value on the display device 4
changes from 299.9 to 300.0, the comparator 11 determines at step 500 that
the two values (display value and display data) coincide. This timing
corresponds to when the lowest digit of the current displayed value
changes from 9 to 10 (0).
In this situation, in FIG. 8, suppose a display data to be adjusted, is
stored in one or more the addresses S.sub.1, S.sub.2, and S.sub.3 where
the values of the digits (0.1, 1, and 10) of the digits of a display data
to be adjusted other than the highest digit (100) change from 9 to 10 (0).
In this case, at least 2 pulses are simultaneously adjusted at the timing
when 299.9 changes to 300.0.
Therefore, in this embodiment, display data to be adjusted are not stored
in the addresses S.sub.1, S.sub.2, and S.sub.3 as shown in FIG. 8, so 2 or
more pulses are never simultaneously adjusted.
It is preferable that the error between the actual running distance of an
automobile and the distance obtained by counting pulses fall within one
pulse. The adjustment of two or more pulses simultaneously is a problem
because of causes more than one pulse worth of error. In this embodiment,
however, 2 or more pulses are never simultaneously adjusted, so the error
is always kept within one pulse and thus the accuracy can be improved.
This invention is effective when the display data to be adjusted, is not
stored in addresses where the value of the 0.1 digit of the lowest digit
of the display data to be adjusted, changes from 9 to 10 (0). The timing
at which the comparator 11 determines coincidence is the same for
predetermined values even other than the timing where the value when the
lowest digit of the current displayed value changes from 9 to 10 (0).
In these embodiments, the adjustment is made by adding a predetermined
number of pulses, for example, one pulse, to the approximate number of
pulses at a suitable timing and at a suitable frequency, but when the
approximate number of pulses is set in a different way, the adjustment may
be made by subtracting predetermined number of pulses.
In this invention, a second physical measure and a third physical measure
may be displayed simultaneously or one of them may be selectively
displayed by a suitable switching means. In that case, the second physical
measure need not be a whole multiple of the pulses and the third physical
measure may be whole multiple.
For example, the second physical measure and the third physical measure may
be displayed in units of miles and kilometers respectively.
This invention is not restricted to only digital electronic odometers for
automobile and can also be applied to any usage in which the physical
measure is obtained by counting pulses of a predetermined interval and the
physical measure to be counted is not whole multiple of counted pulses.
Accordingly, in this invention, time, weight, or the like are included as
the first physical measure in addition to distance.
Another application of this invention is for a stage moved by a pulse
motor. If the moving distance of the stage is not a whole multiple of the
pulse or pulse interval, the distance can easily be obtained utilising the
present invention.
The invention is also useful when changing the tire diameter of an
automobile.
That is, when an automobile tire is changed from a standard size to a
different size, the number of pulses corresponding to 1 kilometer will no
longer be integral. This can be dealt with by the measuring system of the
present invention.
Further, if the display value to be adjusted is stored in an externally
rewritable semiconductor memory device, for example, an EEPROM, the
contents of the memory can be rewritten so as to deal with such a case
effectively when it occurs.
Another application of this invention is to quarantee the accuracy of the
oscillator such as a crystal used for a watch or the like or to change the
oscillator to one having a different oscillating frequency.
Below, an explanation will be made of a counter in accordance with this
invention and useful as a counter for a physical measure display device,
particularly a counter having non-volatile characteristics, and a coding
system used for the same.
In a display device such as one of the present invention, a tremendous
amount of calculations are always performed and therefore a large number
of memories must be provided. The size of the device thus becomes large
and the production costs thereof increase. Further, there is an adverse
effect on accuracy. In the prior art, when the same data processing as in
this invention is carried out, the accuracy of the measuring system has
been kept at a certain level by restricting the number of memories even at
the expense of some accuracy.
The embodiment of the present invention explained below improves on the
drawbacks of the prior art and, while restricting the number of memories,
uses a non-volatile counter which can impro | | |
