U.S. patent application number 12/642426 was filed with the patent office on 2010-05-06 for battery monitor.
This patent application is currently assigned to Stragent, LLC. Invention is credited to Roger Green Stewart.
Application Number | 20100114511 12/642426 |
Document ID | / |
Family ID | 37397023 |
Filed Date | 2010-05-06 |
United States Patent
Application |
20100114511 |
Kind Code |
A1 |
Stewart; Roger Green |
May 6, 2010 |
BATTERY MONITOR
Abstract
A circuit tracks the total amount of time that a host device has
spent in its high power "activated" state (an optionally idle and
hibernate states) and thereby can estimate the total power consumed
by the tag. A remote device can query the state of a counter
storing a value representing this time to accurately determine how
much of the energy has been consumed by the host device and how
much time is left and/or how many more operations can be performed
before the host device's battery is exhausted.
Inventors: |
Stewart; Roger Green;
(Morgan Hill, CA) |
Correspondence
Address: |
The Caldwell Firm, LLC
PO Box 59655, Dept. SVIPGP
Dallas
TX
75229
US
|
Assignee: |
Stragent, LLC
Longview
TX
|
Family ID: |
37397023 |
Appl. No.: |
12/642426 |
Filed: |
December 18, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11124487 |
May 6, 2005 |
7643951 |
|
|
12642426 |
|
|
|
|
Current U.S.
Class: |
702/62 ; 377/16;
702/63 |
Current CPC
Class: |
G06K 7/0008 20130101;
G01R 31/36 20130101; G06K 19/0707 20130101; G06K 19/0723 20130101;
G06K 19/0712 20130101 |
Class at
Publication: |
702/62 ; 702/63;
377/16 |
International
Class: |
G01R 21/00 20060101
G01R021/00; G06F 19/00 20060101 G06F019/00 |
Claims
1. A host Radio Frequency Identification (RFID) device, comprising:
a subsystem for estimating power consumption thereof, the subsystem
comprising a mechanism for tracking at least an amount of time the
RFID device is in an active state; and a mechanism for generating a
value representing at least the amount of time the RFID device is
in the active state based on the tracking; and a mechanism for
sending the value or derivative thereof to a remote device.
2. The host RFID device of claim 1, wherein the mechanism for
tracking an amount of time the host RFID device is in the active
state is a first oscillator
3. The host RFID device of claim 2, wherein the first oscillator
operates at a fraction of a rate of a primary oscillator of the
host RFID device.
4. The host RFID device of claim 3, wherein the first oscillator
operates at no greater than one sixteenth the rate of the primary
oscillator of the host RFID device.
5. The host RFID device of claim 2, wherein the first oscillator is
idle when the host RFID device is in an inactive state.
6. The host RFID device of claim 2, wherein the first oscillator
operates at a first frequency when the host RFID device is in the
active state, wherein the first oscillator operates at a second
frequency when the host RFID device is in a state other than the
active state.
7. The host RFID device of claim 2, wherein the first oscillator is
a voltage controlled oscillator, a frequency of the first
oscillator varying depending on a power dissipation of the host
RFID device.
8. The host RFID device of claim 2, wherein a frequency of the
first oscillator is calibrated.
9. The host RFID device of claim 1, wherein a reference current
source is calibrated.
10. The host RFID device of claim 1, wherein a rate that the value
increases varies independently of a data transmission rate of the
host RFID device, wherein the mechanism for tracking the amount of
time is dependent upon a rate of a primary oscillator of the host
RFID device.
11. The host RFID device of claim 1, wherein a rate that the value
increases varies depending on a power dissipation rate of the host
RFID device.
12. The host RFID device of claim 1, wherein the mechanism for
generating the value representing the amount of time the host RFID
device is in the active state is a counter, the value being a
cumulative count stored in the counter.
13. The host RFID device of claim 12, wherein the speed of the
counter is digitally controlled.
14. The host RFID device of claim 1, wherein the mechanism for
tracking an amount of time the host RFID device is in the active
state is a first oscillator, the first oscillator operating at
least two frequencies, wherein the first oscillator operates at a
first frequency for tracking the amount of time the host RFID
device is in the active state, wherein the first oscillator
operates at a second frequency for tracking an amount of time the
host RFID device is in a state other than the active state.
15. The host RFID device of claim 14, wherein the frequency of the
first oscillator is controlled by an analog control signal.
16. The host RFID device of claim 14, wherein the oscillator
operates continuously.
17. The host RFID device of claim 1, wherein the mechanism for
tracking an amount of time the host RFID device is in the active
state is an oscillator, the oscillator operating at about a
constant frequency, wherein at least one frequency divider operates
between the oscillator and the mechanism for generating the
value.
18. The host RFID device of claim 17, further comprising a
calibration matrix of digitally controlled diodes for calibrating a
bias current of the oscillator.
19. The host RFID device of claim 18, wherein the calibration
matrix is automatically calibrated upon detection of an event.
20. The host RFID device of claim 18, wherein the subsystem
self-calibrates from a single reference frequency.
21. The host RFID device of claim 20, wherein the reference
frequency is derived from a signal from the remote device.
22. The host RFID device of claim 17, wherein the oscillator
operates continuously.
23. The host RFID device of claim 17, wherein the oscillator
operates on a current of less than about one microampere
(.mu.A).
24. The host RFID device of claim 17, wherein the oscillator
operates on a current of less than about 5 nanoamperes (nA).
25. The host RFID device of claim 1, further comprising a switched
capacitor precision resistor for controlling a bias current to the
mechanism for tracking the amount of time the host RFID device is
in the active state.
26. The host RFID device of claim 1, wherein the mechanism for
tracking an amount of time the host RFID device is in the active
state is a first oscillator, and further comprising a second
oscillator, the second oscillator tracking an amount of time the
host RFID device is in a state other than the active state.
27. The host RFID device of claim 1, wherein the remote device uses
the value to estimate usage of power in the host RFID device.
28. A host Radio Frequency Identification (RFID) device having a
subsystem for estimating power consumption thereof, the subsystem
comprising: a mechanism for tracking at least an amount of time the
RFID device is in an active state; and a mechanism for generating a
value representing at least the amount of time the RFID device is
in the active state based on the tracking.
29. A method for estimating power consumption, comprising: tracking
an amount of time a host Radio Frequency Identification (RFID)
device is in an active state; generating a value representing the
amount of time the host RFID device is in the active state; and
sending the value to a remote device.
30. The method of claim 29, wherein the remote device uses the
value to estimate power consumption in the host RFID device.
31. A system for estimating power consumption of a host device,
comprising: an oscillator for tracking at least an amount of time a
host device is in a first state; a counter receiving output from
the oscillator for generating a value representing a power
consumption of the host device based on output from the oscillator,
wherein the counter operates at different speeds depending on the
state of the host device, wherein the host device has a battery,
wherein the host device calculates an estimation of a remaining
battery life of the host device based on the power consumption
value.
32. A system for estimating power consumption of a host device,
comprising: a mechanism for tracking at least an amount of time a
host device is in an active state; a mechanism for generating a
value representing at least the amount of time the host device is
in the active state based on the tracking; and a mechanism for
sending the value or derivative thereof to a remote device.
33. A method for estimating power consumption, comprising: tracking
an amount of time a host device is in an active state; generating a
value representing the amount of time the host device is in the
active state; and sending the value to a remote device.
Description
RELATED APPLICATIONS
[0001] The present invention is related to U.S. patent application
Ser. No. ______ entitled "ACCURATE PERSISTENT NODES" and filed
concurrently herewith, and which is incorporated by reference.
FIELD OF THE INVENTION
[0002] The present invention relates to power consumption
monitoring circuitry, and more particularly, this invention relates
to circuits that monitor power consumption in a host device.
BACKGROUND OF THE INVENTION
[0003] Radio Frequency Identification (RFID) technology employs a
radio frequency ("RF") wireless link and ultra-small embedded
computer circuitry. RFID technology allows physical objects to be
identified and tracked via these wireless "tags". It functions like
a bar code that communicates to the reader automatically without
requiring manual line-of-sight scanning or singulation of the
objects. RFID promises to radically transform the retail,
pharmaceutical, military, and transportation industries.
[0004] Several advantages of RFID technology are summarized in
Table 1:
TABLE-US-00001 TABLE 1 Identification without visual contact Able
to read/write Able to store information in tag Information can be
renewed anytime Unique item identification Can withstand harsh
environment Reusable High Flexibility/Value
[0005] As shown in FIG. 1, a basic RFID system 100 includes a tag
102, a reader 104, and an optional server 106. The tag 102 includes
an integrated circuit (IC) chip and an antenna. The IC chip
includes a digital decoder needed to execute the computer commands
the tag 102 receives from the tag reader 104. The IC chip also
includes a power supply circuit to extract and regulate power from
the RF reader; a detector to decode signals from the reader; a
back-scattering modulator to send data back to the reader;
anti-collision protocol circuits; and at least enough EEPROM memory
to store its EPC code.
[0006] Communication begins with a reader 104 sending out signals
to find the tag 102. When the radio wave hits the tag 102 and the
tag 102 recognizes the reader's signal, the reader 104 decodes the
data programmed into the tag 102. The information can then be
passed to a server 106 for processing, storage, and/or propagation
to another computing device. By tagging a variety of items,
information about the nature and location of goods can be known
instantly and automatically.
[0007] The system uses reflected or "backscattered" radio frequency
(RF) waves to transmit information from the tag 102 to the reader
104. Since passive (Class-1 and Class-2) tags get all of their
power from the reader signal, the tags are only powered when in the
beam of the reader 104.
[0008] The Auto ID Center EPC-Compliant tag classes are set forth
below:
Class-1
[0009] Identity tags (RF user programmable, maximum range .about.3
m)
Class-2
[0009] [0010] Memory tags (8 bits to 128 Mbits programmable at
maximum .about.3 m range) [0011] Security & privacy
protection
Class-3
[0011] [0012] Battery tags (256 bits to 64 Kb) [0013] Self-Powered
Backscatter (internal clock, sensor interface support) [0014]
.about.100 meter range
Class-4
[0014] [0015] Active tags [0016] Active transmission (permits
tag-speaks-first operating modes) [0017] Up to 30,000 meter
range
[0018] In RFID systems where passive receivers (i.e., Class-1 tags)
are able to capture enough energy from the transmitted RF to power
the device, no batteries are necessary.
[0019] In systems where distance prevents powering a device in this
manner, an alternative power source must be used. For these
"alternate" systems (also known as active or semi-passive),
batteries are the most common form of power. This greatly increases
read range, and the reliability of tag reads, because the tag
doesn't need power from the reader. Class-3 tags only need a 10 mV
signal from the reader in comparison to the 500 mV that a Class-1
tag needs to operate. This 2,500:1 reduction in power requirement
permits Class-3 tags to operate out to a distance of 100 meters or
more compared with a Class-1 range of only about 3 meters.
[0020] One concern with powered tags is the life of the battery.
Battery-powered RFID tags draw very little power when silent, but
draw orders of magnitude more power when active. If the tag has
been forced into activated states many times, the battery will be
used up more quickly than a tag activated less. Because certain
tags are active more often than others, it is hard to estimate the
battery life of a tag. The current method is to replace all tags
when the battery on one of the tags dies, as it is likely others
will die soon as well. However, many other tags may still have a
long life remaining. Thus it would be desirable to estimate the
remaining life of a battery in an RFD tag.
SUMMARY OF THE INVENTION
[0021] A circuit tracks the total amount of time that a host device
such as an RFID tag has spent in its high power "activated" state
(an optionally idle and hibernate states) and thereby can estimate
the total power consumed by the tag. A remote device such as an
RFID reader can query the state of a counter storing a value
representing this time to accurately determine how much of the
energy has been consumed by the tag and how much time is left
and/or how many more operations can be performed before the tag's
battery is exhausted. The circuit includes oscillators wherein the
speed (e.g., frequency) of the oscillators is controlled by
reference currents and current mirrors. The circuits may also
include fixed-frequency ultra-low power oscillators with multiple
or variable frequency dividers that can divide the fixed-frequency
output by varying amounts depending on whether or not the tag is in
a higher-power activate state, a lower-power inactive state, or
other power state. The battery monitor circuit is designed to
consume only a small fraction of the power of the chip itself so as
not to significantly shorten the battery life of the tag.
[0022] In one embodiment, a counter circuit includes an ultra low
power oscillator that initiates when the tag wakes up, runs when
the tag is active, and stops when the tag turns off. The oscillator
count is then added to a value in a register. The reader can
interrogate the register storing the count and estimate how much
life is left in the battery, such as by comparing the count to a
benchmark count.
[0023] Optionally, the oscillator can run during the tag's inactive
states, thereby keeping track of total time.
[0024] In another embodiment, a second oscillator is run during tag
inactive state to estimate off-time usage. Note that this may
require a second register to store the counts from the second
oscillator. Then the reader can query the registers and estimate
the remaining battery life using both active and inactive
times.
[0025] In a further embodiment, the system has one clock that is
run at two different speeds, or a varying speed based on power
consumption. In the former case, the clock runs at a fast speed
when the tag is active and at a slow speed when the tag is
inactive. In the latter case, the clock runs proportional to power
dissipation. The more power being used, the faster the clock cycle
(higher count). When the tag is idling or hibernating, the clock
runs more slowly. In any case, the reader can compare the total
count to a benchmark value representing tag life. For instance, the
reader can use an experimental average life count of, say 7
million, and compare it to the total count in the tag to estimate
remaining battery life.
[0026] Other aspects and advantages of the present invention will
become apparent from the following detailed description, which,
when taken in conjunction with the drawings, illustrate by way of
example the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0027] For a fuller understanding of the nature and advantages of
the present invention, as well as the preferred mode of use,
reference should be made to the following detailed description read
in conjunction with the accompanying drawings.
[0028] FIG. 1 is a system diagram of an RFID system.
[0029] FIG. 2 is a system diagram for an integrated circuit (IC)
chip for implementation in an RFID tag.
[0030] FIG. 3 is a system diagram of a system that provides
intermittent battery monitoring.
[0031] FIG. 4 is a circuit diagram of a circuit for continuous
battery monitoring.
[0032] FIG. 5 is a circuit diagram of a circuit for continuous
battery monitoring with automatic calibration.
BEST MODE FOR CARRYING OUT THE INVENTION
[0033] The following description is the best embodiment presently
contemplated for carrying out the present invention. This
description is made for the purpose of illustrating the general
principles of the present invention and is not meant to limit the
inventive concepts claimed herein.
[0034] The following specification describes systems and methods
for tracking the total amount of time that a host device spends in
a high-power "Active" state and optionally, low-power state(s), to
thereby allow estimation of the total power consumed by the host
device. Active states can include those states where the device is
actively using power, or where power consumption is above what
would be considered a low-power state. Note that the host device
can have multiple active states, and operation of the system can
vary for each active state. Low power states typically include
"hibernate", "idle", or "wait/listen for an activation command"
states. A remote device can query the device to accurately
determine how much of the battery energy has been consumed by the
host device, how much life is left in the battery, and/or how many
more operations can be performed before the host device's battery
is exhausted.
[0035] Many types of devices can take advantage of the embodiments
disclosed herein, including but not limited to Radio Frequency
Identification (RFD) systems and other wireless devices/systems;
pacemakers; portable electronic devices; remote controllers for
televisions, audio devices, and other electronic devices; smoke
detectors; etc. To provide a context, and to aid in understanding
the embodiments of the invention, much of the present description
shall be presented in terms of an RFID system such as that shown in
FIG. 1. It should be kept in mind that this is done by way of
example only, and the invention is not to be limited to RFID
systems, as one skilled in the art will appreciate how to implement
the teachings herein into electronics devices in hardware and/or
software. Examples of hardware include Application Specific
Integrated Circuits (ASICs), printed circuits, monolithic circuits,
reconfigurable hardware such as Field Programmable Gate Arrays
(FPGAs), etc. Further, the methodology disclosed herein can also be
incorporated into a computer program product, such as a computer
disc containing software. Further, such software can be
downloadable or otherwise transferable from one computing device to
another via network, nonvolatile memory device, etc.
[0036] A preferred embodiment is preferably implemented in a
Class-3 or higher Class RFID tag, but will function with any type
of module or class of RFID tag. FIG. 2 depicts a circuit layout of
a Class-3 module 200 according to a preferred embodiment for
implementation in an RFID tag, and is presented by way of example
only. This Class-3 module can form the core of RFID modules
appropriate for many applications such as identification of
pallets, cartons, containers, vehicles, or anything where a range
of more than 3 meters is desired. As shown, the module 200 includes
several industry-standard circuits including a power generation and
regulation circuit 202, a digital command decoder and control
circuit 204, a sensor interface module 206, a C1V2 interface
protocol circuit 208, and a power source (battery) 210. A display
driver module 212 can be added to drive a display.
[0037] A battery activation circuit 214 is also present to act as a
wake-up trigger. The battery activation circuit 214 includes with
an ultra-low-power, narrow-bandwidth preamplifier. The battery
activation circuit 214 also includes a self-clocking interrupt
circuit and may use an innovative 32-bit user-programmable digital
wake-up code as described in US Patent Appl. entitled "BATTERY
ACTIVATION CIRCUIT" and having Ser. No. 11/007,973, filed on Dec.
8, 2004, and which is herein incorporated by reference. The battery
activation circuit 214 draws less power during its sleeping state
and is much better protected against both accidental and malicious
false wake-up trigger events that otherwise would lead to
pre-mature exhaustion of the Class-3 tag battery 210. A battery
monitor 215 monitors power usage in the device. The information
collected can then be used to estimate a useful remaining life of
the battery.
[0038] A forward link AM decoder 216 uses a simplified
phase-lock-loop oscillator that requires an absolute minimum amount
of chip area. Preferably, the circuit 216 requires only a minimum
string of reference pulses.
[0039] A backscatter modulator block 218 preferably increases the
backscatter modulation depth to more than 50%.
[0040] A pure, Fowler-Nordheim direct-tunneling-through-oxide
mechanism 220 is present to reduce both the WRITE and ERASE
currents to less than 0.1 .mu.A/cell in the EEPROM memory array.
This will permit designing of tags to operate at maximum range even
when WRITE and ERASE operations are being performed.
[0041] The module 200 also incorporates a highly-simplified, yet
very effective, security encryption circuit 222 as described in US
Patent Appl. entitled "SECURITY SYSTEM AND METHOD" and having Ser.
No. 10/902,683, filed on Jul. 28, 2004 and which is herein
incorporated by reference.
[0042] Sensors to monitor temperature, shock, tampering, etc. can
be added by appending an industry-standard I2C interface to the
core chip.
[0043] Extremely low-cost Class-2 security devices can be built by
simply disabling or removing the wake-up module, pre-amplifiers,
and IF modules from the Class-3 module core.
[0044] Several embodiments are described below with reference to
specific speeds (frequencies), voltages, amperages, etc. to
exemplify the preferred low power aspects of the embodiments. It
should be kept in mind that these are provided by way of example
only, and any suitable speed (frequency), voltage, amperage, etc.
can be used, as will be apparent to one skilled in the art.
Intermittent Battery Monitoring:
[0045] FIG. 3 illustrates one embodiment of a system 300 having an
RFID tag battery monitor 302 in which a mechanism, e.g., dedicated
oscillator 304, tracks the total amount of time that the tag has
spent in its high-power "Activated" state. Another mechanism, e.g.,
counter 306, generates a value based on the tracking, where the
value can thereby be used to estimate the total power consumed by
the tag. A remote device such as a reader (see FIG. 1) can query
the state of this counter 306 to accurately determine how much of
the battery energy has been consumed by the tag, how much life is
left in the battery and/or how many more operations can be
performed before the tag's battery is exhausted.
[0046] In a simple embodiment, the battery monitor oscillator 304
operates only when the tag is active. In this circuit 300, the
battery monitor oscillator 304 is, using a frequency divider 312,
divided down by the ratio of 16:1 from the tag's internal
oscillator 308 (typically a 40 KHz oscillator that might be used in
the C1G2 specification) that the tag uses to parse and decode data
it receives from the reader. In this way, the battery monitor
oscillator 304 operates accurately at 2.5 KHz since this frequency
is ultimately derived from the reader's oscillator 310 or
accurately generated on the RFID chip to an accuracy of better than
.+-.15%. If the tag is commanded by the reader to download data at
a faster rate than 40 Kbps, then the tag may increase the divider
ratio for the battery monitor 302 above 16:1 (as shown in FIG. 3)
to make sure the battery monitor oscillator 304 continues to run at
2.5 KHz. In this way, the battery monitor oscillator 304 can
provide a consistent tracking of the battery usage, regardless of
data speed.
[0047] The battery monitor oscillator 304 drives the "Battery
Monitor Counter" 306 that increments when the oscillator 304 is
running and continues to store the current cumulative count during
periods when the tag is inactive or unpowered. The contents of the
battery monitor counter 306 can be read in a pre-assigned location
of the optional user memory bank 314 of the tag. In this example,
the oscillator 304 drives the counter 306 at 2.5 KHz. The counter
306 is preferably at least 32-bits long to make sure it never
overruns its storage limit, but only the state of the 16 MSBs have
to be addressable by the reader.
[0048] The divider 312 and the resulting slow 2.5 KHz oscillation
frequency of the battery monitor oscillator 304 ensure that the
power dissipation of the battery monitor circuit 302 is negligible
compared to the rest of the active power dissipation of the tag.
Typical power dissipation will be only a few nanoamperes.
[0049] Those skilled in the art may replace the simple digital
frequency divider 312 with current mirrors and reference current
sources. In a similar manner, the battery monitor 302 need not
operate at a fixed frequency but may be made to vary in proportion
with the variations in the power dissipation of the RFID chip
assuming that the power dissipation may vary as a function of the
forward data rate, whether the tag is writing data to memory or
not, whether or not the tag is operating a sensor, etc.
[0050] In a variation on the above, a second oscillator can be run
during other tag states to estimate off-time usage. Note that this
may require a second register to store the counts from the second
oscillator. Then the reader can query both registers and estimate
the remaining battery life using both active and inactive
times.
Continuous Battery Monitoring with Analog Control:
[0051] In a more complex illustrative embodiment shown in FIG. 4, a
tag battery monitor 400 may also include a fixed-frequency
ultra-low-power oscillator 402 with multiple or variable frequency
dividers that can divide the fixed-frequency output by varying
amounts depending on whether or not the tag is in a higher-power
activate state, a low-power inactive state, or other power state.
As in the circuit 302 of FIG. 3, the battery monitor circuit 400 of
FIG. 4 is designed to consume only a small fraction of the power of
the chip itself so as not to significantly shorten the battery life
of the tag.
[0052] In this illustrative embodiment, assume the RFID tag
contains a temperature sensor and supports 3 different operating
modes: [0053] a hibernate or "inactive" mode in which it consumes
only 0.1 .mu.A in power; [0054] an "activated" mode in which it
exchanges data with the reader and consumes 10 .mu.A [0055] a
"sense" mode in which it takes and records a temperature reading
and consumes 30 .mu.A
[0056] While the amount of time the tag will spend in each of these
modes varies widely, the circuit 400 monitors the time spent in
each mode and accumulates the effect this will have on battery
life.
[0057] The circuit shown in FIG. 4 contains a precision calibrated
Ultra-Low-Power ("ULP") oscillator 402 that runs at a frequency
rate of only about 500 Hz and consumes only about 3 nA of power.
This oscillator 402 can run continuously which permits more
accurate measurement of power consumption since it now monitors and
measures inactive "hibernate" power consumption in addition to the
active power consumption measured in FIG. 3. Addition of the ULP
oscillator 402 also facilitates other important tag functions like
enabling a real time clock, enabling logging of temperature and
other sensor data, etc.
[0058] The ULP oscillator 402 in this embodiment consists of a
Voltage Controlled Oscillator ("VCO") formed with three inverters
(I1, I2, I3) connected with feedback to form a ring oscillator. The
frequency of this oscillator is controlled by matched pairs of
current mirrored transistors P5/N5, P6/N6, P7/N7, and a capacitor
connected to the output of I1. The current flowing in these
transistors is in turn controlled both by the "2 nA reference
current" flowing in P3 and the analog current multiplier circuit
formed by N1, N2, N3, and N4.
[0059] In the lowest-power "Inactive" mode, the tag is neither in
the "Sense" nor "Active" state, and the negative-sense inputs
"NSense" and "NActive" are both high. This effectively connects
transistors N1, N2, and N3 in parallel at the drain of N3 for a
total conductivity of 1x+2x+297x=300x. The "1x" or "297x" here
refers to the relative "size", "conductivity", or more accurately
"g.sub.m," of the transistors. Following standard current-mirror
design practice, the effective mirror ratio is now 30x/300x which
sets the currents flowing through P4 and N4 (plus each the other
current mirrored transistor pairs P5/N5, P6/N6, and P7/N7) at 0.2
nA. An additional 2 pf of low-leakage non-junction capacitance is
also added to the output of inverter I1 to reduce the oscillator
frequency to about 500 Hz. Note that in this "inactive" mode the
power dissipation of this entire battery monitoring circuit totals
only about 5 nA, which is much less than the power consumed by the
tag itself.
[0060] However, when the tag enters the "active" state, transistor
N2 is disconnected from the mirror circuit. This changes the
current multiplier ratios as follows: the total conductivity is now
3x(1x+2x), the ratio is now 30x13x, the P4 current has increased to
20 nA, and the VCO oscillator frequency has increased from 500 Hz
to 50 KHz. Note that when the tag power dissipation increased 100x
from 0.1 AA to 10 .mu.A, the battery monitor circuit responded by
increasing the counter frequency by 100x also. Also, note that
although the monitor power dissipation increased to about 0.1
.mu.A, it remains less than 1% of the total power dissipation of
the chip itself.
[0061] Finally, when the tag enters the highest-power "sensor"
state, transistors N2 and N3 are both disconnected from the mirror
circuit. This changes the current multiplier ratios as follows: the
total conductivity is now only 1x, the ratio is now 30x/1x, the P4
current has increased to 60 nA, and the VCO oscillator frequency
has increased to 150 KHz. Note that when the tag power dissipation
increased 300x from 0.1 .mu.A to 30 .mu.A, the battery monitor
circuit responded by increasing the counter frequency by 300x also.
Also, note that although the monitor power dissipation increased to
about 0.3 .mu.A, it remains less than 1% of the total power
dissipation of the chip itself. It is well known to those skilled
in the state of the art that well-designed VCO oscillators like the
one shown in FIG. 4 or described in U.S. Pat. No. 4,236,199, can be
accurately controlled over frequency ranges of 10,000:1 or
more.
A Calibrated Ultra-Low-Power Current Source:
[0062] While the preceding discussion shows how the battery monitor
circuit will work with a 2 nA reference current source, no such
current sources have ever existed in the IC chip world. For
example, just trying to scale a conventional PMOS transistor to
source only this much current (with it's gate grounded and it's
source at 1.2V) would require the channel length to be scaled to
over 100,000 microns--hardly a practical design. And in any case,
the accuracy and stability of any 2 nA current source would be
extremely poor without a method for accurately calibrating this
current. FIG. 4 therefore includes a practical circuit for
generating and calibrating the 2 nA current-source. Again, it
should be stressed that the 2 nA current source specification is by
way of example only, and higher and lower currents can be used with
the circuit.
[0063] In FIG. 4, the calibrated current source is controlled by a
replica of the first VCO that is used to drive the accumulating
counter 404, except that this reference VCO 406 runs at a constant
frequency independently of the Activate or Sense modes of
operation. In this example the VCO 406 operates at 5000 Hz, but
again, it could operate at either a higher or lower frequency. The
output of the reference VCO 406 is buffered and clipped to form a
square wave and used to drive P1 and P2. P1, P2 and their
associated capacitors form a "switched capacitor" precision
resistor 408. The bias current flowing through this network is
nominally: I=(C1).times.(f).times.(.DELTA.V)=20 ff.times.500
Hz.times.0.5V=50 pA. Nominally, this 50 pA bias current also flows
through the calibration matrix and induces an offset voltage across
the calibration matrix of 0.75V--assuming that 3 of the bypass
calibration transistors are turned off. With a nominal P threshold
voltage of 0.44V, then both P3 and P13 will be biased at exactly
0.01 V above their thresholds and will in theory each inject
exactly 2 nA into the circuit.
[0064] In practice P3 and P13 (and the other mirrored transistors
in the circuit) can operate either just above or below their
respective threshold voltages, i.e. both can be operated in their
sub-threshold region if necessary to keep the reference current
low. Also in practice none of these nominal variables are well
controlled, so without calibration, the resulting reference
currents and oscillation frequencies might vary greatly from their
nominal values due to variations in threshold voltage, temperature,
and sub-threshold characteristics of both the diodes and
transistors.
[0065] Accurate calibration of the current source and the
oscillator frequency is therefore achieved as follows. The reader
issues a "Calibration" command and sends a 5000 Hz reference tone
to the tag. The tag uses a simple PLL circuit 412 to compare the
reader reference frequency with it's own reference oscillator
frequency and adjusts the 5 digital inputs to the calibration
matrix as necessary to force it's own reference oscillator
frequency to match that of the reader. Once set, the digital
calibration settings are permanently stored in memory, e.g., either
EEPROM or static RAM until a reader tells the tag to re-calibrate
itself.
[0066] The calibration matrix shown in FIG. 4 is digitally
adjustable. The nominal maximum voltage across the calibration
matrices is 1.25 V (250 mV forward bias at 50 pA for each
"1x-sized" diode). Coarse adjustments of about 250 mV are made by
shorting out completely one or more of the "Cal" diodes. As shown
in FIG. 5, the circuit of FIG. 4 can be extended as necessary to
provide digital adjustments as fine as 1 mV.
Continuous Battery Monitoring with Digital Control:
[0067] FIG. 5 shows another embodiment 500 of the invention
including a single calibrated ULP oscillator 502 running at 1 KHz
and two digital dividers 504, 506 to monitor time in each of the
three illustrative tag power modes (Inactive, Active, Sensor) and
to total the cumulative impact of operating in each of these modes
on the remaining battery capacity. When operating in the "Inactive"
mode the 1 KHz oscillator output is divided 300:1 before the output
is fed into the 40-bit accumulating counter 508. In the "Active"
mode the reference oscillator is divided 3:1, and in the "Sensor"
mode it passes directly to the accumulating counter 508. In this
circuit 500, changes to the reference oscillator frequency are
minimized, and a constant 10 pA bias current passes through P1 and
P2 to the calibration matrix.
Auto-Calibration:
[0068] The calibration matrix shown in FIG. 5 is digitally
adjustable with a combination of both fine and coarse adjustment
bits. In this example, use of an ultra-low-bias current of only
about 10 pA reduces the offset voltage across the parallel
combinations of the four diodes (a,b,c,d which total 22.1x in size)
to only about 200 mV. In one preferred embodiment, the tag will
calibrate itself with the following simplified algorithm.
[0069] Initially all 24 of the calibration transistors are turned
on and the calibration matrix is shorted out completely. The tag
reference frequency will then initially exceed that of the reader
reference frequency and this fact is detected by the PLL oscillator
510 shown in FIG. 5. In response, the calibration logic starts
turning off each of the N1a, N2a, N3a . . . transistors in sequence
until the PLL detects that the tag frequency has dropped below that
of the reader reference oscillator (or until N1a through N6a are
all off). Each disconnected "a-series" transistor increases the
voltage to the gate of P3 by 200 mV. If and when the PLL detects
that the tag frequency has dropped below the reader reference
frequency of 1000 Hz, the calibration circuit turns the last two
"a-series" transistors that it had switched off, back on. The
"a-series" coarse calibration sequence is now complete.
[0070] Next, the tag begins the "b-series" calibration sequence by
turning off each of the N1b, N2b . . . transistors in sequence
until the PLL detects that the tag frequency has dropped below that
of the reader reference oscillator (or until N1b through N6b are
all off). Each disconnected "b-series" transistor decreases the
size of the diode from 22.18x to 2.18x, and this increases the
voltage to the gate of P3 by 60 mV. This is because the
forward-current/junction-area of the diode is an exponential
function of the forward voltage with a slope of about 60 mV/decade
at room temperature. If and when the PLL detects that the tag
frequency has dropped below the reader reference frequency of 1000
Hz, the calibration circuit turns the last two "b-series"
transistors that it had switched off, back on. The "b-series"
calibration sequence is now complete.
[0071] Next, the tag begins the "c-series" calibration sequence by
turning off each of the N1c, N2c . . . transistors in sequence
until the PLL detects that the tag frequency has dropped below that
of the reader reference oscillator (or until N1c through N6c are
all off). Each disconnected "c-series" transistor decreases the
size of the diode from 2.18x to 1.18x, and this increases the
voltage to the gate of P3 by 20 mV, based on the equation
.DELTA.V=(log.sub.102.1/1.18)(60 mv/decade)=20 mV. As before, if
and when the PLL detects that the tag frequency has dropped below
the reader reference frequency of 1000 Hz, the calibration circuit
turns the last "c-series" transistor that it had switched off, back
on. The "c-series" calibration sequence is now complete.
[0072] Finally, the tag begins the "d-series" calibration sequence
by turning off each of the N1d, N2d . . . transistors in sequence
until the PLL detects that the tag frequency has dropped below that
of the reader reference oscillator. Each disconnected "d-series"
transistor decreases the size of the diode from 1.18x to 1.0x, and
this increases the voltage to the gate of P3 by 5 mV, based on the
following equation:
.DELTA.V=(log.sub.101.1/1.0)(60 mv/decade)=5 mV
[0073] The calibration circuit then stops and locks the digital
inputs to each of the calibration transistors in either EEPROM or
static memory until the tag receives another "Calibration" command
from the reader. The full auto-calibration sequence is now
complete.
[0074] If necessary, even finer adjustments could be made by
connecting even more diodes of different sizes in parallel thereby
controlling the forward drop by increments as small as 1 mV: The
net effect is to adjust the voltage across the calibration matrix
such that at the nominal 10 pA bias current, there is just the
right combination of diodes so that the input voltage to P3 is
exactly what is necessary to produce the 2 nA reference current.
The negative feedback employed during the calibration sequence
ensures the tag will calibrate itself accurately despite the
variability in threshold voltages, leakage currents, etc.
[0075] Once calibrated, the bias voltage on the gate of P3 is
maintained by negative feedback through P1 and P2. If, for example,
the P3 gate voltage were to decrease, then the current through P3
would increase and the reference frequency would also increase.
This would increase the current flowing through P1 and P2 which
would fully restore the P3 gate voltage to the original value set
by the calibration sequence.
[0076] While circuits like those shown in FIGS. 4 and 5 can achieve
initial frequency and current calibration accuracies of better than
.+-.10%, this accuracy may be degraded by changes in temperature or
by subsequent variations in the power supply voltage. The best
results are achieved by minimizing the variability of the power
supply voltage using either a band-gap regulator or a battery power
supply. In addition, the regulated power supply could compensate
for both the temperature effect on the threshold voltage of P3/P13
and for the 2 mV/degree variation of the calibration diodes. If
necessary, the accuracy can also be further improved by periodic
re-calibration of the tag.
Additional Considerations:
[0077] To estimate the remaining life of the tag, the reader can
query the tag for the value stored in the counter (or derivative of
the value) and compare that value (or derivative thereof) to a
benchmark value representing tag life or battery life. For
instance, the reader can use an experimental average life count of,
say 7 million, and compare it to the value in the counter to
estimate power usage and remaining battery life.
[0078] If the user changes the tag battery, the count can be reset
from the reader. Or the count can automatically reset if a battery
is removed and replaced. The reader can send an alert by email,
integrated display screen, computer link, etc. if a tag life is
nearing its end. Similarly, the tag can include an integrated
display screen, activated at user request, to show an estimated
power consumption/life remaining.
[0079] Note that the method and circuit herein could also apply to
measuring the service life of the tag, measuring tag activity,
estimating a remaining useful life of the tag (e.g., if the
circuitry is prone to a limited lifetime), etc.
[0080] There have thus been described circuits and systems that
permit accurate monitoring of the total charge consumed by an IC
chip and thereby permit one to accurately predict the end of life
of the battery. In addition, there has been described an accurate
calibrated ULP oscillator that can operate continuously at power
levels as low as 3 nA, and frequencies in the range of 10 Hz to 100
KHz without the use of external capacitors or other external
components.
[0081] While various embodiments have been described above, it
should be understood that they have been presented by way of
example only, and not limitation. Thus, the breadth and scope of a
preferred embodiment should not be limited by any of the
above-described exemplary embodiments, but should be defined only
in accordance with the following claims and their equivalents.
* * * * *