I hope this
paper does not upset too many people; because like Jack Nicholson
told Tom Cruise in a Few Good Men, some people "Can’t handle the
truth."
You would think that the way some of us approach Resistance Temperature
Detectors or RTDs, that the technology train left these very accurate
sensors at the station. Let's start with why it is necessary to
understand
the “Truth about RTDs”.
The first truth is that RTDs are growing at a much higher rate than
their thermocouple counterparts. It is expected that RTDs will grow at
a compound annual growth rate (CAGR) of 4.2% reaching a volume in the
Americas of 200 million US dollars in 2011. This will surpass
the thermocouple which is expected to have a negative CAGR of 2.0% and
dollar volume of 185 million during the same time frame.
What can affect this growth? Well, basically two fundamental issues;
price and technology. Let’s talk about price first. The truth is
that a ¼” RTD purchased for a replacement in a thermowell is less
than 5% more expensive than its thermocouple counterpart. Gone are
the days of a 25% price differential. Today’s instrumentation plays an
important role as well. Users can replace the existing sensor with
the sensor of their choice with little or no hassle.
Technology is a more complicated and detailed issue. It involves a
myriad of points and counter points which due to their complexity and
length prevent this paper from addressing all areas. It is accurate to
state that the RTD inherently can never measure up to the thermocouple
when it comes to response time and mechanical strength. It is safe to
say that today’s RTD has improved to a point that it can handle most
process applications. The real reason the RTD is the sensor leader
is its’ accuracy and stability characteristics and this paper will deal
with the truth in those areas.
Stability is defined by ASTM as the sensors ability to “remain within
the
tolerances” for a four week period. The RTD has a big advantage here.
Stability and accuracy are very dependent on chemical composition. The
tolerance for most RTD element manufacturers is for a purity of
platinum
that is 99 point 8 places pure. While it is relatively easy to control
this degree of purity for an RTD, it is not so for a thermocouple. For
instance a type K thermocouple consists of 90% nickel and the rest is
made up of chromium, manganese and silicon. It is difficult to repeat
the proper mix of these components and consequently the thermocouple is
less stable and less accurate than the RTD.
In the process industry it’s a given that the RTD is more linear and
more
accurate than its thermocouple counterpart. But hidden facts are the
RTDs’
accuracy options. Table 1 provides the accuracy options for 100 ohm
platinum RTD. Table 2 provides the formula for determining the
uncertainty
at a specific temperature point. You will note that the most accurate
point is at 0° C. There is a reason for this. Most RTDs today
are
of the thin film construction. They are made by depositing platinum on
a substrate and trimming to 100 ohms at 0° C.
Although Table 1 only goes up to 200°C, you can safely use a Class A or
B RTD as high as 600°C with an accuracy of ±1°C at 400°C for the
Class A.
Now for the truth that may shock you. These accuracies are element only
and do not include any errors associated with the manufacturing
process.
Virtually all RTD manufacturers quote the accuracies in the above
tables
with the assumption that they will remain within the uncertainty
tolerances
after the manufacturing process is completed. For typical process needs
this is a good assumption. But what if you wanted improved accuracy and
tolerance certainty post manufacturing?
The chart below provides accuracy options.
The truth is only Smart Sensors can give you guaranteed post
manufacturing
accuracy guarantees. We urge you to visit our website at www.smartsensors.com
and see how we provide Temperature
Measurement the right way.
