EYE on NPI - Renesas MRAM episode artwork

EPISODE · Jun 25, 2020 · 15 MIN

EYE on NPI - Renesas MRAM

from Adafruit Industries · host Adafruit Industries

This week's EYE on NPI takes a look at a new kind of memory called MRAM - and some fresh new QSPI memory chips from Renesas that let you try out MRAM for your embedded projects! (https://www.digikey.com/en/product-highlight/r/renesas/m1000-m3000-series-mram) Memory and memory management is one of the trickiest challenges for an engineer, when it comes time to optimize power, speed and performance because you bump up against many of the restrictions different types of memory have - and the reason there's so many types of memory is cause there's no one perfect type of memory storage. Just about every microcontroller has two kinds of storage - flash storage and RAM storage. RAM storage is just like the kind you have in your computer - except a few KB usually, instead of GB. RAM storage is pretty expensive, which is why a microcontroller that is only a few bucks has only a few KB. BUT RAM is really fast - you can read and write to any section of RAM within one or two instructions. There's one down-side to this speed - when the microcontroller is reset, or power is lost, the RAM is also lost. For that reason, your microcontroller has the other kind of memory - flash - to store the program itself. Otherwise, your program would just disappear when you unplugged it (not good!). Flash memory is pretty inexpensive, so you usually get much more flash than RAM. Once in a while you also get microcontrollers that have a separate flash memory area, often referred to as EEPROM. Usually this is used for storing configurations, calibrations, identifiers, etc. So normally, you've got non-volatile flash storage for code, volatile RAM storage for running the code, and maybe a smidge of EEPROM. Say you want to build a sensor data logger - you can't store the data in RAM, because you'll lose it when power is lost. You can't safely stick it in flash because that's where your code lives, and EEPROM is way too small. No problem, you can buy external non-volatile flash memory for microcontrollers! Again, for small amounts of storage, say less than 64K you can get low cost I2C EEPROM (https://www.digikey.com/products/en/integrated-circuits-ics/memory/774?k=i2c%20eeprom). There's one big downside though - while these SPI flash chips have great non-volatility, they have an annoying problem that when you want to change data in a block you have to erase that whole block first, then write the new changed data - and a block is 4KB to 16KB! And erasing takes at least a few milliseconds, if not longer. Same with SD cards by the way. THUS this week's EYE on NPI is interesting to us because it introduces a new kind of memory we've never used before - MRAM for magnetic random access memory. Where as flash memory stores memory by saving a bit by storing charge on a MOSFET gate, MRAM stores it on a ferromagnetic particle that can changed from N to S polarity to store 1 or 0. The core of Avalanche’s STT-MRAM cell is a proprietary perpendicular magnetic tunnel junction (p-MTJ) element which includes including a magnetic fixed layer, a dielectric barrier layer and a changeable ferromagnetic storage layer. The magnetic orientation of the fixed layer is fixed during the manufacturing process. During a programming operation, the magnetic orientation of the storage layer is electrically switched from a parallel state (low resistance state “0”) to an antiparallel state (high resistance state “1”), or vice versa, depending on the current direction through the p-MTJ element. Two distinct resistance states can thus be realized for data storage and sensing. The structure of the p-MTJ element can be tailored for different performance specs and application needs. (http://www.avalanche-technology.com/technology/mram-technology/) What's nice is that every bit/byte can be written or read at any time, no block erases, no delays! You can drive these chips with SPI or QSPI at 100+MHz and down to 1.8V. These chips will be more expensive than NV SRAM or FLASH, but when you need fast writes and reads without waits, these are a great option. Another nice bonus with MRAM is that it is naturally less susceptible to radiation bit-flips - something that folks working on cubesats are super experienced with as a bane of their existence - you send a microcontroller into LEO and the flash memory gets corrupted by high energy particles. While this part is not specifically rad-hard or space-ready, for DIY experimentation, its way way less expensive than official aerospace devices - as seen in the PyCubed dev board (https://circuitpython.org/board/pycubed_mram/) These chips are pin-compatible with most other SPI flash you may have designed in, so it's easy to try out MRAM! Check out the chips on Digi-Key available in 8-SOIC package for easy prototyping! https://www.digikey.com/products/en?keywords=800-M30082040054X0IWAY-ND%20

This week's EYE on NPI takes a look at a new kind of memory called MRAM - and some fresh new QSPI memory chips from Renesas that let you try out MRAM for your embedded projects! (https://www.digikey.com/en/product-highlight/r/renesas/m1000-m3000-series-mram) Memory and memory management is one of the trickiest challenges for an engineer, when it comes time to optimize power, speed and performance because you bump up against many of the restrictions different types of memory have - and the reason there's so many types of memory is cause there's no one perfect type of memory storage. Just about every microcontroller has two kinds of storage - flash storage and RAM storage. RAM storage is just like the kind you have in your computer - except a few KB usually, instead of GB. RAM storage is pretty expensive, which is why a microcontroller that is only a few bucks has only a few KB. BUT RAM is really fast - you can read and write to any section of RAM within one or two instructions. There's one down-side to this speed - when the microcontroller is reset, or power is lost, the RAM is also lost. For that reason, your microcontroller has the other kind of memory - flash - to store the program itself. Otherwise, your program would just disappear when you unplugged it (not good!). Flash memory is pretty inexpensive, so you usually get much more flash than RAM. Once in a while you also get microcontrollers that have a separate flash memory area, often referred to as EEPROM. Usually this is used for storing configurations, calibrations, identifiers, etc. So normally, you've got non-volatile flash storage for code, volatile RAM storage for running the code, and maybe a smidge of EEPROM. Say you want to build a sensor data logger - you can't store the data in RAM, because you'll lose it when power is lost. You can't safely stick it in flash because that's where your code lives, and EEPROM is way too small. No problem, you can buy external non-volatile flash memory for microcontrollers! Again, for small amounts of storage, say less than 64K you can get low cost I2C EEPROM (https://www.digikey.com/products/en/integrated-circuits-ics/memory/774?k=i2c%20eeprom). There's one big downside though - while these SPI flash chips have great non-volatility, they have an annoying problem that when you want to change data in a block you have to erase that whole block first, then write the new changed data - and a block is 4KB to 16KB! And erasing takes at least a few milliseconds, if not longer. Same with SD cards by the way. THUS this week's EYE on NPI is interesting to us because it introduces a new kind of memory we've never used before - MRAM for magnetic random access memory. Where as flash memory stores memory by saving a bit by storing charge on a MOSFET gate, MRAM stores it on a ferromagnetic particle that can changed from N to S polarity to store 1 or 0. The core of Avalanche’s STT-MRAM cell is a proprietary perpendicular magnetic tunnel junction (p-MTJ) element which includes including a magnetic fixed layer, a dielectric barrier layer and a changeable ferromagnetic storage layer. The magnetic orientation of the fixed layer is fixed during the manufacturing process. During a programming operation, the magnetic orientation of the storage layer is electrically switched from a parallel state (low resistance state “0”) to an antiparallel state (high resistance state “1”), or vice versa, depending on the current direction through the p-MTJ element. Two distinct resistance states can thus be realized for data storage and sensing. The structure of the p-MTJ element can be tailored for different performance specs and application needs. (http://www.avalanche-technology.com/technology/mram-technology/) What's nice is that every bit/byte can be written or read at any time, no block erases, no delays! You can drive these chips with SPI or QSPI at 100+MHz and down to 1.8V. These chips will be more expensive than NV SRAM or FLASH, but when you need fast writes and reads without waits, these are a great option. Another nice bonus with MRAM is that it is naturally less susceptible to radiation bit-flips - something that folks working on cubesats are super experienced with as a bane of their existence - you send a microcontroller into LEO and the flash memory gets corrupted by high energy particles. While this part is not specifically rad-hard or space-ready, for DIY experimentation, its way way less expensive than official aerospace devices - as seen in the PyCubed dev board (https://circuitpython.org/board/pycubed_mram/) These chips are pin-compatible with most other SPI flash you may have designed in, so it's easy to try out MRAM! Check out the chips on Digi-Key available in 8-SOIC package for easy prototyping! https://www.digikey.com/products/en?keywords=800-M30082040054X0IWAY-ND%20

NOW PLAYING

EYE on NPI - Renesas MRAM

0:00 15:19

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

API Intersection Stoplight Building a successful API requires more than just coding. It starts with collaborative design, focuses on creating a great developer experience, and ends with getting your company on board, maintaining consistency, and maximizing your API’s profitability.In the API Intersection, you’ll learn from experienced API practitioners who transformed their organizations, and get tangible advice to build quality APIs with collaborative API-first design.Jason Harmon brings over a decade of industry-recognized REST API experience to discuss topics around API design, governance, identity/auth versioning, and more.They’ll answer listener questions, and discuss best practices on API design (definition, modeling, grammar), Governance (multi-team design, reviewing new API’s), Platform Transformation (culture, internal education, versioning) and more.They’ll also chat with experienced API practitioners from a wide array of industries to draw out practical takeaways and insights you can use.H TV Podcast Industries Chris Jones, Derek O'Neill and John Harrison. TV Podcast Industries TV Podcast Industries is a podcast that provides discussions and reviews of various TV shows, including recent popular series like Alien Earth, The Sandman, The Last of Us, The Boys, and Daredevil Born Again. They also cover shows such as Ironheart, Star Trek: Picard, The Rings of Power, and many more, spanning both Marvel and DC universes, as well as other genres. Heart to Heart Podcast One on One / Next Level Studios In the Heart to Heart Podcast, we talk to some of our favorite & most interesting people in the entertainment industry so you can feel empowered and learn that even in the most challenging of industries, where there’s a will there’s a way.Whether you’re an actor, a writer, a casting director, a talent rep, or just someone interested in the behind the scene happenings of the entertainment industry, this podcast will have something for you. NOW, this isn’t just another How To podcast for actors. Plenty of those shows already exist. In Heart to Heart, every guest will share stories inspired by their sometimes winding path to success on their own terms. Revisionist History Pushkin Industries Revisionist History is Malcolm Gladwell's journey through the overlooked and the misunderstood. Every episode re-examines something from the past—an event, a person, an idea, even a song—and asks whether we got it right the first time. From Pushkin Industries. Because sometimes the past deserves a second chance.To get early access to ad-free episodes and extra content, subscribe to Pushkin+ in Apple Podcasts or at pushkin.fm/plus.iHeartMedia is the exclusive podcast partner of Pushkin Industries.

Frequently Asked Questions

How long is this episode of Adafruit Industries?

This episode is 15 minutes long.

When was this Adafruit Industries episode published?

This episode was published on June 25, 2020.

What is this episode about?

This week's EYE on NPI takes a look at a new kind of memory called MRAM - and some fresh new QSPI memory chips from Renesas that let you try out MRAM for your embedded projects!...

Can I download this Adafruit Industries episode?

Yes, you can download this episode by clicking the download button on the episode player, or subscribe to the podcast in your preferred podcast app for automatic downloads.
URL copied to clipboard!