-
www.cypress.com Document No. 001-70698 Rev. *H 1
AN70698
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
Author: Praveen Sekar and Todd Dust
Associated Project: Yes Associated Part Family: All PSoC 3 and
PSoC 5LP Parts
Software Version: PSoC Creator 2.2 SP1 and higher
Related Application Notes: AN75511, AN66477, AN60590
AN70698 explains the theory of temperature measurement using an
RTD, and then shows how to do so with a single PSoC 3 or PSoC 5LP
without the need for external ADCs or amplifiers. To make it easy
to calculate temperature from ADC readings, PSoC Creator provides
an RTD Component. Four example projects are included to demonstrate
operation with both low and high levels of accuracy and
resolution.
Contents
Introduction
.......................................................................
2 RTD Theory of Operation
............................................... 2 RTD Resistance
Measurement Methods ........................... 3
Two-Wire Measurement
............................................... 3 Three-Wire
Measurement ............................................. 3
Four-Wire Measurement
............................................... 4
Reference Resistor Method
............................................... 4 Reference
Resistor Selection ....................................... 5 Offset
Error Cancellation .............................................. 5
Gain Error Cancellation
................................................ 5
RTD Resistance-to-Temperature Conversion ................ 5
Positive
Temperatures.................................................. 5
Negative Temperatures
................................................ 6 Choosing the
Right Polynomial Order .......................... 6 RTD Component
.......................................................... 6
RTD Temperature Measurement with PSoC ..................... 7
Hardware Used CY8CKIT-025 EBK .......................... 7
Project Description
............................................................ 8
Firmware Flow
.............................................................. 9
Testing the Project
..................................................... 10
Interfacing Multiple RTDs
................................................ 10 Broken RTD
Reconfiguration .......................................... 10
Performance Measures
................................................... 10
Temperature Resolution
............................................. 10 Temperature
Accuracy ............................................... 11 List of
all Errors
.......................................................... 14 Test
Results
...............................................................
14
Summary
.........................................................................
15 About the Author
.............................................................
15
Appendix A
......................................................................
16 Appendix B
......................................................................
18
Broken RTD reconfiguration
....................................... 18 Worldwide Sales and
Design Support ............................. 26
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 2
Introduction
Temperature is one of the most frequently measured environmental
variables. Temperature measurement is typically done using one of
four sensors: resistance
temperature detector (RTD), thermocouple, thermistor, and diode.
Table 1 compares these four sensor types.
Table 1. Comparison of Temperature Sensors
Parameter RTD Thermocouple Thermistor Diode
Temperature range 200 to +850 250 to +2350 100 to +300 50 to
+150
Sensitivity at 25 C 0.387 /C 40 V/C (K-type) 416 /C 250 V /C
Accuracy High Medium to High Medium Low
Linearity Good Fair Poor Good
Typical cost (US $) $3 $80 $3 $15 $0.2 $10
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 3
Equation 1 and Equation 2 are also referred to as CallendarVan
Dusen equations, and A, B, and C coefficients are also known as
CallendarVan Dusen coefficients; the term is named after the
scientists who discovered and refined the equations.
The values of A, B, and C for PT100 RTD are specified in IEC
60751 for standard industry-grade platinum and are:
412
27
13
10*183.4
10*775.5
10*9083.3
CC
CB
CA
and the resistance at 0 C,
1000R (PT100 RTD)
Plotting the resistance versus the temperature yields a nearly
linear curve, which is shown in Figure 1. However, it is not a
perfectly straight line. Figure 1 also shows the straight line
approximation imposed on the RTD curve.
Figure 1. RTD Resistance versus Temperature
RTD temperature measurement involves two steps:
1. Measure the RTD resistance accurately
2. Convert the measured resistance accurately to
temperature.
The following sections describe the two tasks.
RTD Resistance Measurement Methods
The techniques commonly used to measure resistance, and their
effectiveness for RTDs, are described in the following section.
Two-Wire Measurement
In a two-wire measurement, a current is passed through the RTD
and the voltage across the RTD is measured, as shown in Figure
2.
Figure 2. Two-Wire RTD Temperature Measurement
Rw1
IsRRTD
VRTD
Rw2
In Figure 1 the Linear Approximation is made by assuming
the RTD changes by 0.385/C. A 1-ohm error in the resistance
measurement leads to a temperature error of about 2.6 C(1/.385).
Therefore, while measuring RRTD, the RTD resistance has to be
determined with an accuracy of
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 4
Figure 3. Three-Wire RTD Temperature Measurement
IsRRTD
VRTD
Vwire
Rw1
Rw2
Rw3
1
2
Caveat
Although this method is better than the previous method, the
three-wire method gives accurate results only if Rw1 = Rw2.
To avoid all error associated with the wire resistances, use the
four-wire measurement described next.
Four-Wire Measurement
The four-wire measurement, as shown in Figure 4, greatly reduces
any error caused by wire resistances.
Figure 4. Four-Wire RTD Temperature Measurement
Rw1
ISRRTD
VRTD
Rw2
Rw3
Rw4
In this method, a known constant current is passed through the
RTD. The voltage across it is measured using a separate sensing
path. The separate sensing path ensures that the voltage drop
across the wire resistances, Rw1 and Rw4, does not affect the RTD
voltage measured.
There will be little voltage drop across resistances Rw2 and
Rw3, which are in the ADC measurement path, because there is
negligible current flow into the high-input impedance terminals of
the ADC.
The RTD resistance in this method is given by Equation 7.
RTD
RTDRTD
I
VR Equation 7
To find the RTD resistance, the current source and the ADC
measuring the voltage must be accurate. Specifically, the current
source and the ADC should be free from offset, gain and
non-linearity errors. Even a small error in voltage measurement can
result in a large temperature error at higher temperatures.
To overcome the gain/offset error caused by the ADC and the
current DAC (IDAC), add a reference resistor to your design.
Reference Resistor Method
The reference resistor makes the measurement error independent
of both the current source accuracy (gain error,), and the ADC
accuracy (gain error). The measurement error depends primarily on a
reference resistance.
The schematic of this method is shown in Figure 5. A constant
current is passed through a known accurate reference resistance in
series with the RTD. The voltage, Vref, across the reference
resistor and the voltage, VRTD, across the RTD are measured. The
RTD resistance is given by Equation 8:
ref
ref
RTDRTD R
V
VR * Equation 8
Figure 5. RTD Temperature Measurement
IDAC
ADC
RRTD
Rref, 100 ohm0.1%
0
1
Rw1
Rw2
Rw3
Rw4
PSoC
The current path is shown by the bold blue line, and voltage
measurement paths are shown in red. The current does not flow
through wire resistances Rw3 and Rw4, because of the high-input
impedance of the ADC. The rest of the application note discusses
this method and the associated project based on it. The following
sections discuss how to choose a reference resistor, and how the
offset error and the gain error of the ADC and IDAC become
null.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 5
Reference Resistor Selection
Choose a small reference resistor that doesnt load the IDAC.
This project uses a reference resistance of 100 ohms. The IDAC has
a compliance voltage of VDDA-1V. This means that the voltage at the
IDAC cannot exceed VDDA-1V. If the maximum RTD resistance is 400
ohms, the reference resistor is 100 ohms, the internal PSoC routing
is ~400 ohms and a 1-mA current is passed through the three in
series, then a voltage of 900 mV will be produced (1 mA * 400 ohms
+ 100 ohms + 400 ohms). This is well below the compliance
voltage.
In addition, choose a reference resistor that uses the same ADC
range as the RTD. This will reduce the time it takes to measure the
RTD, because the ADC will not need to be reconfigured. Again, if
the maximum RTD resistance is 400 ohms, and 1 mA is passed through
it, the RTD will produce a voltage of 400 mV. The DelSig ADC has an
input range of 512 mV. Keep the voltage across the reference
resistor in this 512-mV range. If it is outside this range, you
will have to reconfigure the ADC every sample; reconfiguration
takes time.
Offset Error Cancellation
In this application note, offset cancellation is done by
correlated double sampling (CDS). In CDS, the offset is measured
and then, in firmware, it is subtracted from the other voltage
measurements. Using this method reduces the ADC sample rate by 50%,
because offset is measured every other sample.
You can use CDS to measure offset in various ways. See AN66444
PSoC
3 and PSoC 5LP Correlated Double
Sampling for details.
For this application note, offset is measured by passing zero
current through the IDAC and measuring the voltage directly across
channel 0 or 1.
With CDS, the equation for resistance measurement becomes:
RTDref
ref
RTD RVV
VV*
0
0
Equation 9
Where, VRTD is the voltage measured across the RTD
Vref is the voltage measured across the 100- resistor
V0 is the offset voltage measured when the IDAC current is set
to zero.
The offset current of the IDAC does not cause any error because
it would be nulled by the difference in Equation 9.
VRTD = (I+I0) * RRTD + ADC offset
V0 = I0 * RRTD + ADC offset
Subtracting V0 From VRTD removes ADC offset. The difference
nulls the I0 term, which is caused by the IDAC offset. If CDS is
done regularly, then offset drift is also canceled out.
Gain Error Cancellation
Assume that the ADC has a gain error of k and the DAC has a gain
error of k. These errors reflect as multiplicative factors in the
voltage measurements, VRTD and Vref. Because Equation 9 includes a
ratio, the multiplicative errors k and k cancel out.
RTD = ref
ref
RTD RVVkk
VVkk*
)(**
)(**
0
'
0
'
Now the error depends primarily on the accuracy of the reference
resistor, Rref.
This method also removes any errors associated with gain drift,
because the ratio metric measurement is being taken every time.
Using the reference resistor method, you can determine the RTD
resistance accurately. The next step is to convert the RTD
resistance to temperature. This conversion is discussed in the
following section.
RTD Resistance-to-Temperature Conversion
The straightforward method to obtain temperature from resistance
is to use the CallendarVan Dusen equations. But Equations 1 and 2
show resistance in terms of temperature; you need to know
temperature in terms of resistance. The solution is given
below.
Positive Temperatures
Solving Equation 1 for T,
B
R
RBAA
T
T
2
140
2
Equation 10
The other solution of the quadratic equation is eliminated using
the known point (T = 0; R =100).
Using Equation 10 involves a square root, which requires a math
library and about 5800 8051 CPU cycles for computation. This is a
lot of time to spend doing a single calculation.
Instead of Equation 10, use a polynomial to calculate
temperature. Compute the polynomial by first constructing a
resistance versus temperature table and then using curve-fitting
techniques on the table. Excel is a good tool to use for curve
fitting.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 6
Polynomial computation, which does not require a math library,
executes faster. The temperature accuracy increases as the order of
the polynomial increases. Using a fifth-order polynomial
[1] can reduce the conversion error
to
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 7
For example, if you are using PT100 RTD, your range is -100 C to
400 C, and you require 0.05 C temperature calculation accuracy,
select PT100 RTD, enter the maximum and minimum temperatures in
their respective text boxes, and select 0.05C temperature accuracy.
The
component automatically chooses the appropriate polynomial for
you.
Figure 8 shows that a third-order polynomial meets that
requirement.
Figure 8. Customizer for 100 C to 400 C Range
In your code, use the RTD_GetTemperature(int32 res) API to
calculate temperature. The API takes resistance (in milliohms) and
computes the temperature using the polynomial coefficients and
returns the temperature (in 1/100
th of C).
The temperature error depends on many factors in addition to
resistance-to-temperature conversion error. The RTD customizer
shows only the error due to resistance-to-temperature conversion.
To accommodate other errors, ensure that the
resistance-to-temperature conversion error is less than one-tenth
of the total error budget. The other errors are discussed in
Accuracy of the Measurement section.
RTD Temperature Measurement with PSoC
This application note has four example projects (RTD_HighEnd,
RTD_MidEnd, RTD_LowEnd, and Broken RTD reconfiguration) to showcase
RTD temperature measurement using PSoC. The first three projects
display the RTD temperature on an LCD. These projects have the same
signal chain. This section describes the RTD_MidEnd project in
detail. Differences between the projects are explained in detail in
the Performance Measures section. The fourth project, Broken RTD
Reconfiguration, is explained in Appendix B.
Hardware Used CY8CKIT-025 EBK
The PSoC Precision Analog Temperature Sensor EBK (CY8CKIT-025)
is used in the example project. The kit provides four
sensorsthermocouple, thermistor, RTD, and diodefor measuring
temperature. In addition, interface slots let you plug in your own
thermocouple, thermistor, RTD, and diode. You can connect the EBK
to the CY8CKIT-030 PSoC 3 Development Kit (DVK), or to the
CY8CKIT-050 PSoC 5LP DVK. Figure 9 shows the kit and Figure 10
shows the RTD portion. For more details on the kit, go to
www.cypress.com/go/Cy8CKIT-025.
Figure 9. PSoC Precision Analog Temperature Sensor EBK
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 8
Figure 10. RTD Section of the EBK
Project Description
The RTD_MidEnd PSoC Creator project associated with this
application note displays the RTD temperature directly on an LCD.
The schematic of the project, which uses the CY8CKIT-025 PSoC
Precision Analog Temperature Sensor EBK, is shown in Figure 11.
This figure is similar to Figure 5 except that the reference
resistance is in a separate path, and is not in series with the
RTD.
The CY8CKIT-025 does it this way to share the reference
resistance for diode temperature measurement. Because the reference
resistance is not in series with the RTD, the current through the
reference resistance can differ from the current through the RTD.
If the currents through the reference resistance and RTD are not
equal, you can get a temperature error. Manual analog routing is
used to avoid a significant error. See Appendix A for a detailed
explanation.
If you are designing your own board for RTD measurement, use the
reference resistor in series with the RTD, as shown in Figure 5.
You can avoid the 0.7 C error at 850 C (see Appendix A). You
require only one pin from IDAC, and the current mux shown is not
required (see Figure 11). Use an IDAC direct connection (P0_6,
P0_7, P3_0, P3_1) to connect the series combination of RTD and
reference resistor to the IDAC. The PSoC Creator schematic for that
project is shown in Figure 12.
Figure 11. PSoC Creator Top Design Schematic
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 9
Figure 12. PSoC Creator Top Design Schematic for Reference
Resistance in Series with RTD
Firmware Flow
The following chart shows the firmware flow for the low- and
mid-end RTD temperature measurement projects associated with this
AN. The projects differ only in the ADC resolution and the filter
attenuation factor used.
Read voltage across reference
resistor at zero current for
Correlated Double Sampling (CDS)
Connect IDAC current mux to
Reference resistor and measure
voltage V1.
Connect IDAC current mux channel
to RTD and measure voltage V2.
Calculate temperature using the
RTD_GetTemperature(uint32 res)
API of the RTD component
Start IDAC and set to source 1-mA
current
END
CDS V1 and filter (IIR Filter)
CDS V2 and filter
Measure RTD resistance
from V2 and V1RRTD = (V2/V1)*Rref
Read voltage across RTD sensor
at zero current for CDSStart
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 10
If the reference resistance is in series with the RTD, the
firmware flow remains as shown, except that the IDAC output is not
multiplexed.
Testing the Project
1. Plug CY8CKIT-025 into PORT E of the CY8CKIT-030 or
CY8CKIT-050.
2. For the CY8CKIT-030, choose the CY8C3866AXI-040. For the
CY8CKIT-050, choose the CY8C5868AXI-LP035
3. Build the attached project and program the PSoC device.
4. The LCD displays the RTD temperature.
Interfacing Multiple RTDs
The easiest way to interface to multiple RTDs is to wire them in
series with one another. However, make sure you dont violate the
compliance voltage of the IDAC, as discussed earlier.
Alternatively, you can easily interface as many as four RTDs
with PSoC 3 / PSoC 5LP using the four IDACs in the device.
Broken RTD Reconfiguration
If one of the RTD wires is broken, you can reconfigure the
four-wire RTD to a three-wire RTD and continue to show the
temperature without a considerable loss of accuracy. Appendix B
describes the broken RTD reconfiguration in detail. A project
(Broken RTD reconfiguration) is also associated with this
application note.
Performance Measures
The RTD-based temperature-sensing market can be categorized into
three performance segments: high end, mid end, and low end (see
Table 3).
Table 3. RTD Performance Ranges
Market segment Resolution* (C) Accuracy* (C)
High 0.01 0.1%
Mid 0.1 0.2 0.5%
Low >0.1C >0.5C
* Resolution is generally specified only for temperatures >
0C. Accuracy doesnt include the sensor accuracy and is usually
specified with a fixed offset, such as 0.1% or 1C, whichever is
greater.
In this section, we will see how to use PSoC 3 and PSoC 5LP to
address all three segments. Unit test results appear at the
end.
Temperature Resolution
The temperature resolution depends on three factors:
The temperature range to be measured The IDAC current The ADC
resolution High End
Lets calculate the resolution required for measuring -200 C to
850 C with a 0.01 C resolution. First, we need to translate the
temperature resolution to voltage resolution. From Equations 1 and
2, you can calculate the resistance change per unit change in
temperature .
A change from 850C to 851C causes a 0.292- change in
resistance.
A change from -200C to -201C causes a 0.432- change in
resistance.
Take the smaller of the two: 0.292 for the calculations. Use an
IDAC current of 1 mA; the current used in the associated project,
for the calculations.
The voltage resolution required for 1 C = 0.292 * 1 mA = 292
V.
If you can resolve 2.92 V, you will achieve a 0.01 C
resolution.
This means the IDAC (operating at 1-mA current) and the ADC
together should have a noise of less than 2.92 V.
Next, calculate the full-scale voltage range of the RTD:
Resistance at -200 C = 18.52 (using Equation 2)
Resistance at 850 C = 390.481 (using Equation 1)
Voltage output at -200 C = 18.52 * .001 = 18.52 mV
Voltage output at 850 C = 390.481 * .001 = 390.481 mV
The ADC has an input range of 512 mV; this range fits well with
the output voltages of the RTD.
Using a 20-bit ADC with an input range of 512 mV, each bit
represents 1.024 V/2^20 = 0.9765 V.
According to the DelSig ADC datasheet, the ADC has an RMS noise
of ~1 count in the 512-mV range. This indicates the noise is below
the required 2.92 V. However, when the voltage is close to a
temperature transition, the noise causes flicker in the output.
To eliminate flickering bits, the attached project adds a small
software-based IIR filter to the code. See AN2099, Single-Pole IIR
filter for a detailed description of a software-based IIR
filter.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 11
Mid End
For mid-end applications, the resolution requirement is 0.1 C.
Using the same logic we used before, we need to resolve only 29.2 V
to achieve a resolution of 0.1 C. For the mid-end project, the same
ADC configuration is used as the one in the high-end project. The
only difference is that the filter attenuation is lower in the
mid-end project. Low End
For low-end applications, a resolution of 1 C is often
sufficient. Based on our earlier calculations, we need to resolve
only 292 V. Using a 12-bit ADC in the 512-mV range gives a voltage
resolution of 250 V. The 250 V is below the requirement of 292 V.
Using a 12-bit ADC allows us to move into lower-cost PSoC devices.
The project RTD_LowEnd associated with this application note
demonstrates how to use a 12-bit ADC to measure an RTD.
Temperature Accuracy
You can calculate the temperature accuracy by summing all
possible individual errors, which fit into one of two
categories:
1. Error due to the measurement system
2. Error due to RTD
Error Due to the Measurement System
The error due to the measurement system is due to the circuit
shown in Figure 5. Consider Equation 8, which is used to obtain RTD
resistance.
ref
ref
RTDRTD R
V
VR *
VRTD is the voltage measured across the RTD
Vref is the voltage across the 100- resistance
Rref is the reference resistance
As discussed, the only major source of error using this method
is the accuracy of the reference resistance. The offset error is
nulled by the difference, and the gain error is nulled by the
ratio. The other source of measurement error is non-linearity in
the ADC.
Error Due to ADC Integral Non-Lineari ty
The integral non-linearity (INL) of an ADC at any point is the
difference between the ideal ADC count and the actual ADC count at
that point after gain and offset corrections have been completed.
The datasheet specifies the maximum INL of all points across
process, voltage, and
temperature (PVT). PSoC 3 ADC has an INL of 32 LSb; 32 LSb
corresponds to 64 V for 20-bit resolution and 1.024-V range. The
ADC INL has the same analog value, 64uV, in 0.512 V range. Lets
calculate the error due to INL at 850 C for IDAC current of 1
mA.
mVRIV RTDRTD 481.390481.390*1* (The RTD resistance at 850 C =
390.481, using Equation 1)
mVRIV refref 100100*1*
Using equation 8
100*100
064.0481.390100*
100
481.390errorResistance
064.0
Substituting the worst-case INL at the numerator, we get 0.064-
resistance error corresponds to a temperature error of 0.16
C(.064/.385). The INL is at its worst when it applies only to the
numerator.
Note that we have taken the worst-case INL across PVT and
substituted worst-case positive INL at the numerator. In practical
application, the error due to INL is much lower. For example, for
an INL of +8 LSb at the numerator and 0 LSb at the denominator, the
temperature error at 850 C is 0.054 C.
Error Due to Reference Resistance Tolerance
In Equation 8, we substituted 100 for the value of the reference
resistor, Rref. But the actual value of Rref will change because of
its tolerance and temperature coefficient. Therefore, the value
Vref, which is measured across the reference resistance, will be
erroneous. Assume that the tolerance of Rref is 0.1 percent and the
temperature coefficient is 10 ppm/C.
RTDRTD RIV *
refref RIV *
Substituting the value of Rref with the tolerance and
temperature coefficient yields Equation 11.
)25(*00001.0001.01(100* sysTempIVref Equation 11
The measured value of the RTD resistance, Rmeas, is:
))25(*00001.0001.01(
sysTemp
RR RTDmeas
Equation 12
where the 0.1 percent resistance tolerance contributes to the
additive factor 0.001, and the temperature coefficient
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 12
(10 ppm/C) contributes to the additive factor
0.00001*(sysTemp-25). Note that the temperature coefficient is
usually specified with reference to 25 C. Therefore, the effect of
the temperature coefficient is zero at 25 C, but it increases as
the temperature deviates from 25 C.
The measured temperature can be calculated from the measured
resistance using Equations 1 and 2.
Figure 13 shows the temperature error due to the reference
resistor error at different RTD temperatures and ambient
temperatures. Ambient temperature (see legend) is the temperature
of the reference resistance on the printed circuit board (PCB)
while RTD temperature (x axis) is the actual temperature to which
the RTD is exposed.
Figure 13. Temperature Error Due to Reference Resistor Accuracy
and Temperature Tolerance
The orange line shows that for a 0.1 percent reference resistor,
the error is
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 13
Figure 15. Temperature Error due to RTD Interchangeability
Error
At 25 C, the worst-case temperature error = 0.3 + 25 * 0.005 =
0.43 C
At 800 C, the worst-case temperature error = 0.3 + 800 * 0.005 =
4.3 C.
If this error is not acceptable, the RTD has to be calibrated.
For most users seeking a high-performance RTD (0.1% or 1C),
calibration will be required.
Ideally, the resistance at 0 C, R0, is expected to be 100 for a
PT100 RTD. But, practically, it will have a tolerance that causes
the interchangeability error. Going back to Equations 1 and 2, note
that a deviation in the value of R0 causes a scale error in the
measured temperature. Therefore, performing a scale correction
corrects the RTD interchangeability error.
For RTD calibration, perform the following steps.
1. Make sure that all other sources of error are nulled (Offset
should be nulled by correlated double sampling, and the gain error
is automatically nulled by the ratio).
2. Adjust the RTD to a known temperature, T1, for this method T1
should be close to 0 C. Validate and measure that temperature, T1,
with an accurate thermometer.
3. Measure the RTD resistance, Rmeas, at T1.
4. Calculate the actual RTD resistance at T1, Ractual, using
Equations 1 and 2.
5. The scale error is given by
meas
actual
R
Rscale Equation 13
6. Multiply this scale error by the measured RTD resistance.
This process is done in the project RTD_HighEnd. When the
project starts, it asks the user if he wants to calibrate. If he
does, he must press SW2.
Next, the project asks for the calibration temperature. The
CapSense
slider on the CY8CKIT-030 or CY8CKIT-050
can be used to adjust the temperature by 1 C. The CapSense
buttons can be used to adjust the temperature by 0.01 C. After the
temperature has been entered, press SW2.
The firmware will calculate the resistance for the entered
temperature. Next, it will measure the resistance of the RTD.
Finally it will compute the ratio and store it in EEPROM.
If an accurate and stable temperature cannot be achieved,
replace the RTD with a known resistor. Using a known resistor will
calibrate out only the error due to the reference resistor. It will
not calibrate out the RTD interchangeability error.
Note that the reference resistance tolerance also causes a fixed
scale error in the measured resistance. Hence, when the above six
steps are completed, the reference resistance also is calibrated.
The scale error value can be stored in PSoC EEPROM and retrieved
each time the device goes through a power cycle.
The 100 comes because you measured at 0C and 100C,
for a difference of 100. Now multiply all temperatures by this
scale factor.
This process is not implemented in the project.
Figure 15 and Table 4 show that the temperature error has an
offset and a gain. Thus, it may become necessary to perform a
two-point temperatue calibration. For example, place the RTD at 0 C
and measure the temperature (Toffset). Subtract this measured
temperature from subsequent temperature readings.
Second, place the RTD at 100 C and measure the temperature
(TGain). Next, compute a scale factor using the equation.
Scale = TGain Toffset / 100.
The 100 comes because you measured at 0 C and 100 C, for a
difference of 100. Now multiply all temperatures by this scale
factor.
This process is not implemented in the project.
RTD Self -Heat ing Error
An RTD self-heating error is the increase in temperature of the
RTD due to the current flowing through the RTD.
As a result of self-heating, the RTD can show a temperature
slightly higher than ambient. This error can be found in the RTD
datasheet. For PTS080501B100RP100, the value is specified at
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 14
List of all Errors
Table 5 shows the temperature error due to various components at
150 C. As seen from the table, RTD interchangeability and reference
resistance tolerance are the biggest sources of error. By
comparison, the other errors are negligible.
Table 5. Possible Errors in RTD Temperature Measurement at 150
C
Error Source Error Value at 150 C (0.1% Reference Resistor,
class B
RTD)
Error Value at 150C (Both Reference Resistor and RTD
Calibrated)
Signal Chain Error
Offset Error/drift 0 C 0 C
Gain Error/drift 0 C 0 C
ADC INL* 0.2 C 0.2 C
RTD self-heating error (PTS080501B100RP100 SMD RTD)
< 0.13 C
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 15
Summary
When high accuracy is critical in measuring temperatures, an RTD
is the sensor of choice. PSoC 3 / PSoC 5LP have the necessary
hardware integrated in the device to achieve high accuracy. The
PSoC Creator RTD Component makes designing with RTD easier.
About the Author
Name: Praveen Sekar
Title: Applications Engineer
Background: Praveen holds a Bachelors degree in Electronics and
Communication from the College of Engineering, Guindy, Chennai. He
focuses on analog modules in PSoC
Contact: [email protected]
Name: Todd Dust
Title: Applications Engineer Staff
Background: Todd holds a Bachelors degree in Electrical
Engineering from Seattle Pacific University.
Contact: [email protected]
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 16
Appendix A
This section describes the effect of having the RTD reference
resistance in a separate path (not in series with the RTD).
The IDAC current droops by about 1 percent as the IDAC load
voltage increases from 0 V to Vdda-1 V. Each of the four IDACs in
PSoC 3 / PSoC 5LP can connect to one pin directly and to other pins
through the analog globals. The direct connection is through a
single switch, which has a resistance of about 200 . When the IDAC
connects to a pin using an analog global, it connects through two
switches, each with a resistance of about 200 . For a detailed
description of the PSoC 3 or PSoC 5LP analog routing, see AN58827
Internal Routing Considerations for PSoC
3 and PSoC 5LP Analog Designs.
For IDAC3, the direct connection is to pin 3_1. The RTD on the
EBK is connected to pin 3_1 of PSoC, and the calibration channel is
connected to pin 3_4 of PSoC (see Figure 16).
The Top Design schematic, shown in Figure 11 requires separate
IDAC multiplexer channels: the RTD channel and the calibration
channel. In building the project, PSoC Creator can route the RTD
channel through the direct switch and the calibration channel
through the analog global. In this case,
IDAC load resistance in RTD channel = 200 + RTD resistance
IDAC load resistance in calibration channel = 500 (400 +
calibration resistance 100 ).
The different load resistances result in different IDAC load
voltages, which produce different IDAC currents in both channels.
The error in RTD resistance caused by the different currents is
calculated by Equation 16.
f
RTDRTD
I
IRR
Re
1* Equation 9
The resistance error increases as the RTD resistance
(temperature) rises. If the ratio, IRTD/IRef, is equal to 1, the
resistance error falls to zero, as expected.
Lets calculate the error in temperature at 25 C due to the
different IDAC currents.
The IDAC current used = 1 mA
The IDAC load voltage across the RTD channel = Load resistance *
IDAC current = (200 + 109.732) * 1 mA = 309.732 mV
The IDAC load voltage across the calibration channel = 500 * 1
mA = 500 mV
As the load voltage rises from 309.732 mV to 500 mV, the IDAC
current drops by about 0.5 A (See IDAC user module datasheet
graph)
From Equation 16,
055.0
5.999
10001*734.109R
This corresponds to a temperature error of 0.14 C at 25 C.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 17
Figure 16. IDAC to Pin Routing Resistances
IDAC3
Pin 3_4
Pin 3_1
AGR 4 AGR 5
200
200
RTD
Calibration
Resistance
Temp Sensor EBK
PSoC
200
200
200
To avoid that error, use manual analog routing to force the
router to use the analog global path for both channels. Now the
load resistance in both channels remains about the same. The only
factor that would affect the load resistance is the change in RTD
resistance due to temperature. At 850C, the RTD resistance would be
400 , while the calibration resistance would be 100 . That 300-
difference causes a 300-mV difference in the load voltages. For
that voltage difference of 300 mV, the current change would be
about 0.5 A. (The current droop with load voltage is available in
the IDAC user module datasheet) As a result, the worst-case error
would be 0.7 C at 850 C.
Also, you should force the IDAC to IDAC3 by using a force
placement directive in the directives tab of the design-wide
resources (.cydwr) file, as shown in Figure 17.
Figure 17. Force Placement Directive on IDAC
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 18
Appendix B
Broken RTD reconfiguration
If one of the four wires of the RTD breaks, PSoC can
automatically detect a broken wire and reconfigure four-wire RTD
connection to three-wire connection and display temperature with
minimal degradation to accuracy (< +/- 0.5C). Broken RTD
reconfiguration involves three steps.
1. Detect broken wire connection
2. Reconfigure the analog routing to change four-wire connection
to three-wire connection
3. Compensate for the additional wire resistance due to the
three wire mode and display temperature.
The PSoC Creator project (Broken RTD Reconfiguration) associated
with this application note demonstrates this feature.
Detect ing Broken RTD wire
A four-wire RTD connection to PSoC is shown in Figure 18. The
current is passed through pin 3_1 into RTD wire 1 and it is
grounded through RTD wire 4 (RTD wire 4 is not connected to a PSoC
pin). The ADC differential inputs are connected to pins 4_0 and
4_1, which are connected to wires 2 and 3 of the RTD. The pin
choices are made according to the connections in CY8CKIT-025 PSoC
precision analog temperature sensor EBK
Figure 18. Four-Wire RTD Connected to PSoC
IDAC
ADC +-
PSoC Kit025External 4-wire
RTD
R1
R2
R3
RTD
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4
A broken wire can be detected by using PSoCs GPIO structure.
PSoC GPIO can be configured to source vdd through a pull up
resistor while simultaneously sensing the pin state through its
digital input buffer. For example, pin 3_1 can be configured as
shown in Figure 19.
Figure 19. Pin Configured in Resistive Pull-Up Mode and Digital
Input Mode
VDDIO
Pin
State
> 3.5k
PSoC
To analog
global
Digital Input
Buffer
To analog
Mux Bus
To detect the broken RTD wire connected to pin 3_1, pin 3_1 is
configured in the resistive pull-up mode and the pin state is
sensed back.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 19
When wires 1 and 4 are not broken, the RTD resistance forms a
resistor divider with the internal pull-up resistor and the voltage
across the RTD is sensed back as the pin state.
The pull-up resistance has a minimum value of 3.5 k and the RTD
can have a maximum resistance of 390 (at 850 C). Assuming wire
resistances (Rw1 and Rw4) = 5 each, we get a maximum value of 400
.
Figure 20. Detecting Wires 1 and 4 for Breakage
VDDIO
Pin
State
> 3.5k
External 4-wire
RTD
RTD
Rw1
Rw2
Rw3
Rw4
Pin 3_1
PSoC
To analog
global
Digital Input
Buffer
To analog
Mux Bus
Voltage sensed by pin = Voltage across RTD+Rw1+Rw4
=
pwwRTD
wwRTD
RRRR
RRR*VDDIO
21
21
= 3900
400*VDDIO
< 0.1 * VDDIO
VIL of the pin < 0.3 * VDDIO
Therefore, when pin 3_1 is configured as resistive pull up (with
a high voltage forced through the pin) and when no RTD wire is
broken, the pin state will be low. When either RTD wire 1 or 4 is
broken, the pin state will be high.
Similarly, we can detect if RTD wires 2 and 3 are broken by
configuring the respective pins to resistive pull-up modes and
reading the pin state back.
To find which RTD wire is broken, follow these steps:
1. Disconnect pins 3_1, 4_0 and 4_1 from ADC and DAC
2. Configure pin 3_1 in resistive pull up mode
3. Drive high through pin 3_1
4. Read the pin state of pin 3_1
5. Repeat steps 2, 3, and 4 for pin 4_0 and pin 4_1
Let the pin states of pin 3_1, pin 4_0 and pin 4_1 be stored in
variables A, B, and C respectively. Based on different values of A,
B, and C, we can have eight states, as shown in the following
table.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 20
A B C Result
0 0 0 No Wire Broken
0 0 1 Wire 3 broken
0 1 0 Wire 2 broken
0 1 1 Wires 2 and 3 broken
1 0 0 Wire 1 broken
1 0 1 Wires 1 and 3 broken
1 1 0 Wires 1 and 2 broken
1 1 1 Wire 4 broken
The previous table also shows the result of each combination of
A, B, and C.
If any of A, B and C is equal to 0, then wire 4 is not
broken.
If A, B and C are all equal to 1, wire 4 is definitely broken.
Apart from wire 4 any other wire can also be broken. But in such a
case reconfiguration is not possible. Reconfiguration is possible
only if one of the wires is broken.
Reconf iguring four-wire RTD to three-wire
After detecting the broken RTD wire, we have to reconfigure the
four-wire RTD to three-wire RTD eliminating the broken wire. The
flexible analog routing structure of PSoC makes it very easy to
reconfigure four-wire RTD to three-wire RTD eliminating the broken
wire. The reconfiguration routes are shown below.
RTD Wire 1 Broken
If RTD wire 1 is broken, the current path is opened. To close
the current path, the routing is reconfigured such that the current
is forced through the ADC pin 4_0 as shown in Figure 21.
Figure 21. RTD Wire 1 Broken
IDAC
ADC +-
PSoC Kit025
R1
R2
R3
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4External 4-wire
RTD
RTD
R1 = Routing resistance from IDAC to pin
R2 = Routing resistance from ADC (positive) to pin
R3 = Routing resistance from ADC (negative) to pin
In this case, the IDAC current flows through R1, Rw2, through
RTD to ground. Since Rw2 is in the measurement path of the ADC, the
RTD resistance and wire 2 resistance is also measured. This wire
resistance can be eliminated through calibration explained in one
time wire resistance computation section.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 21
RTD Wire 2 Broken
If RTD wire 2 is broken, the path from RTD to ADC positive
terminal is opened. To close the path, we connect ADC positive
terminal to pin 3_1 as shown in Figure 22 .
Figure 22. RTD Wire 2 Broken
IDAC
ADC +-
PSoC Kit025External 4-wire
RTD
R1
R2
R3
RTD
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4
In this case, the ADC measures RTD wire resistance 1 in addition
to the RTD resistance.
RTD Wire 3 Broken
If RTD wire 3 is broken, the path from RTD to ADC negative
terminal is opened. To close the path, we connect ADC negative
terminal to ground as shown in Figure 23.
Figure 23. RTD Wire 3 Broken
IDAC
ADC +-
PSoC Kit025External 4-wire
RTD
R1
R2
R3
RTD
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4
In this case, the ADC measures RTD wire resistance 4 in addition
to the RTD resistance. Also, any difference in potential between
the two grounds (kit-025 ground and the chip internal ground) adds
to measurement error. One time offset correction eliminates both
the wire resistance error and the ground difference error.
RTD Wire 4 broken
If RTD wire 4 is broken, the current path from RTD to ground is
opened. The ADC input terminal is Hi-Z and no current flows through
the ADC input. To close the path, we provide the ground path by
configuring the pin in open drain low mode as shown in Figure
24.
Figure 24. RTD Wire 4 Broken
IDAC
ADC +-
PSoC Kit025External 4-wire
RTD
R1
R2
R3
RTD
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4R4
In this case, the ADC measures RTD wire resistance 3 in addition
to the RTD resistance.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 22
One-t ime Wire Resistance Computat ion
When reconfiguring a four-wire RTD into a three-wire RTD, wire
resistances affect the RTD temperature measurement accuracy. 1-ohm
wire resistance can cause 3C error in measured temperature. To
eliminate the error due to the wire resistances, we perform a
one-time wire resistance computation before making the RTD
measurements.
The steps followed to compute the wire resistances are given
below
1. Configure the RTD in four-wire mode as shown in Figure 18 and
find the RTD resistance (R0)
2. Configure the RTD in three-wire mode (wire 1 broken) as shown
in Figure 21 and calculate the resistance (R1)
3. Compute additional wire resistance, CompRes1 = (R1 R0). When
wire 1 breaks and the RTD is reconfigured as shown in Figure 21,
CompRes1 should be subtracted from the measured resistance.
4. Configure the RTD in three-wire mode (wire 2 broken) as shown
in Figure 22 and calculate the resistance (R2)
5. Compute additional wire resistance, CompRes2 = (R2 R0). When
wire 2 breaks and the RTD is reconfigured as shown in Figure 22,
CompRes2 should be subtracted from the measured resistance.
6. Configure the RTD in three-wire mode (wire 3 broken) as shown
in Figure 23 and calculate the resistance (R3)
7. Compute additional wire resistance, CompRes3 = (R3 R0). When
wire 3 breaks and the RTD is reconfigured shown in Figure 23,
CompRes3 should be subtracted from the measured resistance.
8. Configure the RTD in three-wire mode (wire 4 broken) as shown
in Figure 24 and calculate the resistance (R4)
9. Compute additional wire resistance, CompRes4 = (R4 R0). When
wire 4 breaks and the RTD is reconfigured as shown in Figure 24,
CompRes4 should be subtracted from the measured resistance.
Project Descript ion
The Project (Broken RTD reconfiguration) has been built to work
with CY8CKIT-025. The project detects a broken RTD wire and
automatically reconfigures a four-wire RTD connection to a
three-wire RTD connection and continues to display temperature with
a broken alert. The project also performs one-time wire resistance
compensation so that the temperature displayed is accurate.
The PSoC Creator schematic of the project is shown in Figure
25.
Figure 25. PSoC Creator Schematic of Broken RTD Reconfiguration
Project
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 23
This project is similar to the RTD temperature measurement
project except that there are more IDAC and ADC channels to support
four-wire to three-wire reconfiguration. Manual analog routing has
been used to ensure that the ADC positive connection always happens
through analog mux bus. This way we ensure that the ADC and IDAC
connections are through two different paths until the pin. Each pin
connects to the analog network through two paths: Analog mux bus
and Analog globals (see Figure 19 and Figure 20). See PSoC3
Datasheet I/O System and Routing section for a detailed description
of pin structure. By making sure that the ADC connects to the pin
through the analog mux bus and that the DAC connects to the pin
through one of the analog globals, we eliminate any additional
resistance that can add to RTD resistance.
Figure 21 shows separate IDAC and ADC paths until pin. Figure 26
shows IDAC and ADC paths merged before pin. In this we have
resistance R4 in addition to Rw2. This additional resistance will
also get subtracted out when we perform resistance compensation.
But having separate current and voltage paths until the pin is a
better way of doing a three wire RTD. If the compensation fails due
to some reason, the additional resistance error will be in the
order of ohms when you have separate IDAC and ADC paths whereas it
will be in the order of hundreds of ohms when you do not have
separate IDAC and ADC paths until the pin.
Figure 26. Wire 1 Broken IDAC and ADC Paths Merged Before
Pin
IDAC
ADC +-
PSoC Kit025
R1
R2
R3
Pin 3_1
Pin 4_0
Pin 4_1
Rw1
Rw2
Rw3
Rw4External 4-wire
RTD
RTD
R4
Firmware Flowchart
The firmware flowchart is shown in Figure 27. There is a step in
which the firmware reconfigures the routing to four-wire mode on a
switch press. Whenever a wire is broken, the RTD reconfigures the
routing and remains in that routing even if the broken wire is
fixed. The user can press the Switch, SW2, in the DVK and the
firmware reconfigures the routing to four wire mode. When one of
the RTD wire breaks, the RTD resistance will drop down to less than
10 . This condition is used in the project to trigger the checking
of wires.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 24
Figure 27. Firmware Flowchart
Start
Find additional wire resistances in
each of the four cases, wires 1, 2, 3,
and 4 broken
Calculate RTD resistance, R, and
subtract additional wire resistance
Is R
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 25
Document History
Document Title: AN70698 PSoC 3 and PSoC 5LP Temperature
Measurement with an RTD
Document Number: 001-70698
Revision ECN Orig. of Change
Submission Date
Description of Change
** 3458038 PFZ 12/12/2011 New Application note
*A 3490797 PFZ 1/12/2012 MEH Review Feedback in CDT#116240
*B 3520653 PFZ 2/8/2012 Updated project. No change to
document
*C 3689958 PFZ 08/09/2012 RTD component includes support for
PT100, PT500 and PT1000 RTDs
A new section on broken RTD reconfiguration has been added
Other minor changes
*D 3740378 PFZ 09/11/2012 Updated associated project files.
*E 3818484 PFZ 11/21/2012 Updated title to read as PSoC 3 and
PSoC 5LP Temperature Measurement
with an RTD.
Updated Associated Part Family as All PSoC 3 and PSoC 5LP
Parts.
Updated Related Application Notes as AN75511, AN66477,
AN60590.
Updated Introduction.
Updated RTD Resistance-to-Temperature Conversion (Updated
Choosing the Right Polynomial Order (Updated description), updated
RTD Component (Updated Figure 6, Figure 7, Figure 8)
Updated Project Description (Updated Figure 11 and Figure
12)
Updated Appendix B (Updated Broken RTD reconfiguration (Updated
Project Description (Updated Figure 25))).
Replaced PSoC 5 with PSoC 5LP in all instances across the
document.
*F 4057734 TDU 07/11/2013 Added two projects and discussed
different performance ranges.
*G 4152296 TDU 10/09/2013 Updated attached Associated
Project.
Completing Sunset Review.
*H 4202789 TDU 11/26/2013 Fixed formatting errors.
-
PSoC 3 and PSoC 5LP Temperature Measurement with an RTD
www.cypress.com Document No. 001-70698 Rev. *H 26
Worldwide Sales and Design Support
Cypress maintains a worldwide network of offices, solution
centers, manufacturers representatives, and distributors. To find
the office closest to you, visit us at Cypress Locations.
Products
Automotive cypress.com/go/automotive
Clocks & Buffers cypress.com/go/clocks
Interface cypress.com/go/interface
Lighting & Power Control cypress.com/go/powerpsoc
cypress.com/go/plc
Memory cypress.com/go/memory
Optical Navigation Sensors cypress.com/go/ons
PSoC cypress.com/go/psoc
Touch Sensing cypress.com/go/touch
USB Controllers cypress.com/go/usb
Wireless/RF cypress.com/go/wireless
PSoC Solutions
psoc.cypress.com/solutions
PSoC 1 | PSoC 3 | PSoC 5LP
Cypress Developer Community
Community | Forums | Blogs | Video | Training
Technical Support
cypress.com/go/support
PSoC and CapSense are registered trademarks of Cypress
Semiconductor Corp. All other trademarks or registered trademarks
referenced herein are the property of their respective owners.
Cypress Semiconductor 198 Champion Court San Jose, CA
95134-1709
Phone : 408-943-2600 Fax : 408-943-4730 Website :
www.cypress.com
Cypress Semiconductor Corporation, 2011-2013. The information
contained herein is subject to change without notice. Cypress
Semiconductor Corporation assumes no responsibility for the use of
any circuitry other than circuitry embodied in a Cypress product.
Nor does it convey or imply any license under patent or other
rights. Cypress products are not warranted nor intended to be used
for medical, life support, life saving, critical control or safety
applications, unless pursuant to an express written agreement with
Cypress. Furthermore, Cypress does not authorize its products for
use as critical components in life-support systems where a
malfunction or failure may reasonably be expected to result in
significant injury to the user. The inclusion of Cypress products
in life-support systems application implies that the manufacturer
assumes all risk of such use and in doing so indemnifies Cypress
against all charges. This Source Code (software and/or firmware) is
owned by Cypress Semiconductor Corporation (Cypress) and is
protected by and subject to worldwide patent protection (United
States and foreign), United States copyright laws and international
treaty provisions. Cypress hereby grants to licensee a personal,
non-exclusive, non-transferable license to copy, use, modify,
create derivative works of, and compile the Cypress Source Code and
derivative works for the sole purpose of creating custom software
and or firmware in support of licensee product to be used only in
conjunction with a Cypress integrated circuit as specified in the
applicable agreement. Any reproduction, modification, translation,
compilation, or representation of this Source Code except as
specified above is prohibited without the express written
permission of Cypress. Disclaimer: CYPRESS MAKES NO WARRANTY OF ANY
KIND, EXPRESS OR IMPLIED, WITH REGARD TO THIS MATERIAL, INCLUDING,
BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
FITNESS FOR A PARTICULAR PURPOSE. Cypress reserves the right to
make changes without further notice to the materials described
herein. Cypress does not assume any liability arising out of the
application or use of any product or circuit described herein.
Cypress does not authorize its products for use as critical
components in life-support systems where a malfunction or failure
may reasonably be expected to result in significant injury to the
user. The inclusion of Cypress product in a life-support systems
application implies that the manufacturer assumes all risk of such
use and in doing so indemnifies Cypress against all charges. Use
may be limited by and subject to the applicable Cypress software
license agreement.