You are at: Elements
- Sensors - Introductin
Return
to Table of Contents
Sensors are devices that are used to measure physical variables like temperature, pH, velocity, rotational rate, flow rate, pressure and many others. Today, most sensors do not indicate a reading on an analog scale (like a thermometer), but, rather, they produce a voltage or a digital signal that is indicative of the physical variable they measure. Those signals are often imported into computer programs, stored in files, plotted on computers and analyzed to death.
Sensors come in many kinds and shapes to measure all kinds of physical variables. However, many sensors have some sort of voltage output. There are a number of implications to that. Here are some.
The LM35 spec sheet lists an internal resistance of 0.1W. Since the LM35 produces .01v/oC, you can compute how much current you can draw for a specified change in reading.
Say you are measuring 50oC. Then the output voltage
should be 0.5v. To change that output voltage to 0.49v - and error
of one degree - you would have to have a voltage drop of .01v across the
internal resistance of 0.1W.
would take .1A or 100mA. That would be a lot of current to draw from
the LM35 and is beyond the limit of what it can supply.
A
More Advanced Model - Using System Dynamics
Often sensors have time constants. (You can refer, for example, to the charts in the LM35 spec sheet.) If you take the sensor time constant into account, you would not expect the output voltage to change immediately when the temperature changed suddenly. A circuit/system model that takes that into account is shown below.

If you are knowledgeable
about Laplace transforms, the denominator quantity indicates that the input
(Measured Variable) is processed by a system with a single time constant
(one pole). If you use this sensor in a block diagram, you should
take that transfer function into account.