There is an easy and cost-effective way to accurately measure a product’s temperature while also recording and tracking how your customers are using the product over its lifetime.
These days, increasing a product’s operational value to customers, while protecting your investment, is critical. Here I will explain the significant value and importance of EEPROM memory usage contained in digital output (I2C protocol) local temperature sensors as well as give examples regarding why and how to set up a simple write sequence to the EEPROM for possible design solutions – as well as a few other things along the way!
I2C digital temperature sensors are silicon-based integrated circuits (ICs) that have sold billions of devices since their market introduction over 20 years ago. Since then, these devices have been a popular thermal monitoring solution in seemingly countless electronic products and systems. They have helped system designers measure and control their product’s real-time temperatures and effectively react to any over or under temperature limit violations.
From a high-level view, digital temperature sensors are a complete temperature monitoring solution that measure their own internal temperature and then convert that temperature into a digital value that can be easily read via standard I2C protocol communication. These sensor outputs digitize temperature data, eliminating the need for any external components such as analog-to-digital converters or data post-processing components.
You may be interested to know that a digital temperature sensor is factory calibrated by the manufacturer to meet well-defined temperature accuracy requirements within a given temperature range. It is not uncommon today to see various suppliers offering +/- 1oC maximum accuracy (or better) over a wide temperature range like -40oC to 125oC. However, what is uncommon is to see integrated EEPROM memory designed into the temperature sensor so that users can store critical application specific data to improve the robustness of their end products. The EEPROM memory is a dedicated non-volatile memory array of various sizes dedicated for user storage. EEPROM stands for Electrical Erasable Programmable Read-Only Memory and one of its key features is it can be written to over one million times to every memory byte location, making it perfectly suited for high frequency data change transactions and for almost all products and applications.
Why is it important to use the integrated EEPROM memory?
There are three important reasons.
The first is that the integrated memory use provides the capability to record critical data locally from the temperature sensor. An example of critical data is logging temperature extremes of the end product that could force system shutdowns or failures, record temperatures, date/time, fault counts, power-on duration and many other conditions we’d like to avoid.
Consider the impact to future product improvements if you could review and study customer product failure returns by simply reading the integrated memory inside the temperature sensor to learn things like how many times the product reached a critical temperature and the date and time it occurred. This data alone could tell you if the customer operated the product in a consistently harsh temperature environment, perhaps voiding the warranty or any warranty claims due to customer misuse. The integrated memory could be viewed almost like the crash data recorders found in cars and planes today. The stored data could help improve future product development and/or product improvements by relating the stored temperature violations to, perhaps a wrong component selection where its performance degraded over time.
Was the product overclocked or abused? Consider the impact to the product analysis process when a customer returns a product for a failure analysis. The stored data could help in determining the root cause of the product’s failure mode and, on the opposite side, also could invalidate the return because the stored data reveals customer misuse and abuse of the product thereby reducing a product liability exposure. Having the integrated memory storage also minimizes data tampering and the stored data helps determine if warranty claims are justified.
The second reason is that the integrated memory enables the capability to track how end customers are using a product’s features or functions to determine if customers actually use the features or not. Perhaps a product has new features or specific new functions and you want to see if customers actually use them or not. How could this be done or implemented in a simple write sequence?
It could, for example, be implemented simply by dedicating two EEPROM byte locations to store two different associated values and knowing in advance that those two locations store that product’s specific feature or function. The first byte location could store the incremented value for each time the feature was used. The second byte location could be used as a date and time stamp to know the last time the feature or function was used. How valuable would it be to have real-time data from customers’ use to know if and when they actually used what you thought was a high-value feature only to find out that feature was rarely used if at all. These results would be exponentially more valuable than a customer survey or focus group!
The third reason to use integrated memory is for critical data storage for factory, system configuration settings and parametric data. The integrated memory could be used to store critical factory testing information about how the product was tested to certain test program revision. The test setup and equipment location, along with its parametric data, would help on customer product returns by providing unbiased information on how the product was factory tested. In addition, the integrated memory could be used to store critical system configuration data that helps the power up and initialization process to configure the system or product. Future product upgrades could be done by simply reprogramming the previous configuration data in a simple programming operation.
An I2C temperature sensor with integrated EEPROM, such as the AT30TSE758A from Microchip, enables manufacturers to take advantage of the three core benefits discussed above. Precise temperature monitoring devices, like the one shown below in Figure 1, combine a high-precision digital temperature sensor, programmable high and low temperature alarms, integrated nonvolatile registers and 8 Kbit EEPROM into a single compact package. These features make it applicable for use in a wide variety of applications that require measurement of local temperatures as an integral part of the system’s function and/or reliability. Figure 1 shows a basic architecture diagram for these products.
Basic block diagram for a temperature sensor with integrated EEPROM, such as Microchip’s AT30TSE758A
The temperature sensor shown above can measure temperatures over the full -55oC to +125oC temperature range and has a typical accuracy as precise as +/-0.5oC from 0oC to +85oC. The results of the digitized temperature measurements are stored in one of the internal registers, which is readable at any time through the device's I2C serial interface.
Another way to maximize thermal system management is to use a temperature sensor with nonvolatile registers. They will enable the devices to store and retain the configuration and temperature limit settings even after the device has been power cycled. This feature eliminates the need for the device to be reconfigured after each power-up operation, thereby allowing the device to run self-contained and not rely upon a host controller for device configuration streamlining the power-up initialization sequence.
Hopefully you now have a better understanding the importance of using integrated EEPROM memory that complements your thermal management endeavors and provides possible design considerations. Products like the AT30TS758A allow designers to bypass thermal design roadblocks and stay in the fast lane on the road to success!