Internet Explorer is not a supported browser for TI.com. For the best experience, please use a different browser.
Video Player is loading.
Current Time 0:00
Duration 9:38
Loaded: 0%
Stream Type LIVE
Remaining Time 9:38
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected

[MUSIC PLAYING]

Hello, and welcome to the TI Precision Lab video discussing the temperature accuracy and power consumption specifications commonly found in temperature sensor data sheets. In this chapter, we will learn about how to interpret these different accuracy and power consumption specs to ensure system requirements are being met, and show how to find these values in different data sheets.

First, let's begin by discussing temperature accuracy. For this example, we'll look at TI's TMP117 sensor data sheet. Here is the first page of the data sheet. In most cases, the temperature accuracy spec is one of the first values listed. Zooming in, we can see that the temperature accuracies listed are designated as maximum accuracies. This means for a given temperature range, the corresponding temperature accuracy is the absolute worst case error that a sensor should have.

This list is plotted in two graphs, also shown on the first page of the data sheet for two different package types. The maximum and minimum temperature accuracies of the TMP117 are represented by the stepped red line. Let's look at the first accuracy specification listed of plus or minus 0.1 degrees Celsius, from -20 to 50 degrees Celsius.

This spec means that between -20 to 50 degrees Celsius, no TMP117 device should report a temperature value that deviates from the true temperature value by more than plus or minus 0.1 degrees Celsius. These maximum temperature accuracies are warranted across the sensor's operating temperature range, here shown to be -55 to 150 degrees Celsius. This range defines the temperatures at which the sensor is rated to operate safely and provide accurate measurements.

Now that we have learned about maximum temperature accuracy, we will discuss the typical temperature accuracy, which is also a common value reported in temperature sensor data sheets. On page 6 of the TMP117 sensor data sheet, we can again see the max and min temperature ranges of the three TMP117 devices that were listed on the first page of the data sheet, as well as the test conditions used to achieve these accuracies.

For example, this first line shows the plus or minus 0.1 degrees maximum accuracy from -20 to 50 degrees Celsius. But what about the values listed in the middle as the typical temperature accuracy? The typical temperature accuracy is the accuracy you would most likely encounter if you selected a random unit from a batch of temperature sensors and used it to start taking measurements.

It is not the worst case temperature error, like the max accuracy, but rather, the most commonly seen error in a normal distribution curve from a population of temperature sensors. The typical temperature accuracy will always be a smaller absolute value than the max temperature accuracy, since the max temperature accuracy will be a certain number of standard deviations away from the typical accuracy, depending on what worst case temperature error was determined to still be allowable for operation.

The graph on the right shows that at an ambient temperature of 25 degrees Celsius, the temperature accuracy of the majority of the units tested fell between plus or minus 50 milliC, or 0.05 degrees Celsius, which is thus the typical temperature accuracy reported for many of the tested temperature ranges. As the temperature range at which the sensor is being tested widens, the typical temperature distribution error curve will also widen, which is why we see typical accuracies of 0.1 degrees Celsius at wider temperature ranges.

Now that we have discussed temperature accuracy, let's look at another important parameter found in temperature sensor data sheets-- power consumption. This value will also commonly be found on the first page of a data sheet, and it is important to consider when designing an efficient system.

When we look at power consumption for the TMP117 temperature sensor, we see that 3.5 microamps are consumed for a 1-Hertz conversion cycle, and 150 nanoamps are consumed in shutdown mode. A 1-Hertz conversion cycle means that one temperature measurement is made every second. However, let's look further into different power consumption values.

On page 6 of the TMP117 sensor data sheet, many types of current consumption for this device are listed. Just like temperature accuracy, a max or worst case and typical or most commonly seen value are provided. However, four different types of current are listed on the left-hand side of the table. In the next slide we will discuss the differences in these current types.

Let's first start with the quiescent current during an active conversion. Quiescent current during active conversion can sometimes be called the active current in other data sheets. This active current is consumed while a digital temperature sensors on and making a temperature measurement, and therefore, draws the highest amount of current. The time spent at this active current level is called the conversion time, which is how long it takes for the digital temperature sensor to take a measurement.

The graphs below show the two possible modes to measure temperature in-- one-shot mode and continuous conversion mode. Although these graphs show current values for the TMP102 device, the TMP117 sensor records temperature in either one of these two modes as well. The quiescent current during active conversion is shown here by the blue lines.

The next type of current we will discuss is shutdown current. Shutdown current is consumed while a digital temperature sensor is completely idle. The device does not make measurements in this mode, and the alert and trip outputs will not activate. Different shutdown currents are shown in this table for different test conditions and can represent the leakage through the IC when in a disabled state. When in shutdown mode, the power consumed is almost zero microamps and represented by the blue lines between conversion cycles on the left graph.

Standby current is the quiescent current consumed in between active temperature conversions. If the device is not configured for shutdown mode, then it must count passing time to make regular measurements as defined by conversion rate settings if there are any. This idle current represents current consumed by an onboard oscillator and digital counter.

Many temperature sensors power on in a continuous conversion, because providing regular updates makes for ease of use. However, the standby current will always be higher than shutdown current, and it is represented by the blue line shown on the continuous conversion graph on the right.

The final current listed in the TMP117 sensor data sheet is for quiescent current. Although not listed in every data sheet, quiescent current is the amount of current consumed by the sensor when it is powered on and ready to perform a conversion, but it is in a non-operating state. This quiescent current can be affected by factors such as what frequency is chosen for how many temperature measurements are taken per second, and how long the conversion time is.

Finally, we will discuss how to compare temperature sensor accuracy and power consumption specs across data sheets. Let's compare the accuracy and power consumption specs found on the first page of the TMP117 sensor data sheet to values found in another digital temperature sensor data sheet.

In the data sheet on the right, we see that it lists 0.1 degrees Celsius accuracy, but not whether the accuracy is a max or typical value. Additionally, we see a 600 micro amp typical operating supply current. Is this current the active current during a conversion cycle, or an average current?

Oftentimes, to be able to compare temperature accuracy and power consumption across data sheets, more digging is needed beyond the first page. In the electrical characteristics section, we can better compare temperature accuracy. Here, we see that the 0.1 degrees Celsius specified on the first page of the data sheet is a maximum temperature accuracy. But it is only warranted in the very narrow temperature range of 37 degrees Celsius to 39 degrees Celsius. Therefore, it is always important to double-check if the temperature accuracy provided is maximum or typical, and if your desired accuracy is within all possible conditions your system could be subjected to.

Now let's compare current. The active quiescent current is the value given first in both data sheets. This value can usually be identified as the highest current value on the data sheet. These devices both utilize a one-shot mode in order to reduce overall current usage. So the shutdown current is also listed.

Finally, conversion time, or the time that the sensor is actively taking a temperature measurement, is also provided in the electrical characteristics. With these three values, an average current used over the conversion cycle time can be calculated. These currents can only be comparable if both devices are in the same mode, in this case, one-shot mode, and if the conversion cycle time is the same. In this case, a 1-Hertz cycle time is used, or one temperature measurement is made per second.

Using the values in the data sheets under the given conditions, this gives an average power consumption of about 2.24 microamps for the TMP117 device, and about 28 microamps for the device in the data sheet on the right. Therefore, for a system using these conditions, the TMP117 sensor would consume a smaller amount of power than the digital temperature sensor on the right used for comparison.

In summary, this training video showed how to interpret maximum and typical temperature accuracy, different types of power consumption, and how to compare these values across different data sheets for temperature sensors. Thank you for your time.

This video is part of a series