Optimise system energy consumption with MRAM

A 1?F capacitor allows the writing of 50B (46 data bytes) on the SPI bus at 40MHz with the MRAM consuming 27 mA. This calculation is the source of using 46B for comparisons.

Energy used by MCU and LDO: For the MCU, I assumed that it takes 100?s to wake up, make a measurement, and communicate the result to non-volatile memory and any housekeeping required. During this time, I assumed an active current consumption of 500?A (typical of small microcontrollers running at approximately 5MHz). This gives us an energy consumption of 3.3 V x 500?A x 100?s = 0.165?J per data acquisition. In addition to the energy to make an acquisition, I added the energy required to keep the microcontroller active during the non-volatile memory write.

When not acquiring or storing data the MCU was in a sleep state, consuming 5?A.

The power supply was assumed to be an LDO that consumes 1?A during all phases of operation, active and sleep.

Results summary
Working through the calculations, I get the figures shown in table 4. The calculations are shown for systems that make 10 and 100 acquisitions per second (4B per acquisition). For 100 acquisitions per second, EEPROM required 0.5 seconds just in the write time of the EEPROM. A larger number of acquisitions would have to look at caching the acquisition data and performing block writes.

Table 4: Summary of energy consumed per acquisition.

The results are graphically summarised in figure 3.

Figure 3: Total energy by type and number of acquisitions.

Conclusions
The write time of a non-volatile memory significantly affects the total energy consumption of a system. For systems with a low duty cycle the effects are less pronounced and become more pronounced as the rate of acquisitions increase.

The write times of EEPROM and Flash significantly increase the energy consumption of the MCU because they cause the MCU to be active for longer. Energy consumption could be reduced if the MCU was in a sleep mode while the writes to EEPROM and Flash complete. However, the energy consumed by the EEPROM or Flash is the majority of the energy consumed by the system, so having the MCU in sleep mode will not affect the overall consumption significantly.

It is clear that the lowest energy consumption can be achieved with a fast-write, non-volatile memory that is power gated.

Why power gating has more of an effect on energy consumption than sleep mode: EEPROM has low standby power consumption, so operating it with V_DD always present could be considered. However, the write energy of just one write operation to EEPROM is equivalent to it being in standby mode for about 15 seconds. Again, the write energy is dominant.

Power gating really applies to MRAM where the fast write time significantly reduces the amount of energy used. The energy consumption of an acquisition using MRAM is 1.54?J. This is the same as a 3.3 V EEPROM with a standby current consumption of 1?A being in standby for 0.46 s.

The write energy is dominant over standby energy.

About the authorDuncan Bennett is a Product Marketing Manager at Everspin Technologies Inc., the only company currently delivering production MRAM into the merchant market. Prior to Everspin he was the Product Marketing Manager for Ramtron International Corporation. He has held similar positions with Cyan Technology and Cygnal Integrated Products, and served as a field applications engineer with Dallas Semiconductor. Duncan holds aBachelor of Computing and Communications Engineering with honours from Bradford University. He is stationed in London.

?? 1???2???3