RFEE 1 Posted November 6, 2014 Share Posted November 6, 2014 Hi All, My company is interested in developing passive sensors that are capable of harnessing energy from the environment, so basically the sensor are completely dead and without power until there's either light, temperature difference, vibration, or RF illumination. So I do not care about high current draw in sleep mode. In this type of environmentwhat would work best? MSP430FR? or M0+? Current draw during start-up on the MSP430 is also a problem, which is roughly 1 ms. Looking at the EMF32 Zero Gecko, Its list as 400us with a 24MHz clock. It seems to me that Zero Gecko would be a better choice, but from uA/MHz FRAM based MSP430 is a better choice. I don't need to do any crazy computation here, all I need is for the ADC to measure a time decay, and send it back out through a wireless tx module. Quote Link to post Share on other sites
username 198 Posted November 6, 2014 Share Posted November 6, 2014 Hi All, My company is interested in developing passive sensors that are capable of harnessing energy from the environment, so basically the sensor are completely dead and without power until there's either light, temperature difference, vibration, or RF illumination. So I do not care about high current draw in sleep mode. In this type of environmentwhat would work best? MSP430FR? or M0+? Current draw during start-up on the MSP430 is also a problem, which is roughly 1 ms. Looking at the EMF32 Zero Gecko, Its list as 400us with a 24MHz clock. It seems to me that Zero Gecko would be a better choice, but from uA/MHz FRAM based MSP430 is a better choice. I don't need to do any crazy computation here, all I need is for the ADC to measure a time decay, and send it back out through a wireless tx module. Low power is a fun nut to crack. The current usage figures in datasheets are generally marketing figures. I'd be best to grab a dev board for both processors your interested in and play around with a uA meter and the software features you know you need. Keep in mind that RF usually requires a decent amount of power for any type of range. Powering the processor may be peanuts compared to completing a transmission with any decent range. Quote Link to post Share on other sites
zeke 693 Posted November 6, 2014 Share Posted November 6, 2014 If solar is an option then you could harvest enough energy to charge a LiFePO4 battery and then run anything you wanted after that. I've done this myself already. If you must select an MSP430 then I would choose a new FRAM based part. TI's FRAM FR5969 LaunchPad is using a supercap to store charge and to run from. They do have demos of it operating as an RFID powered unit. It's worth considering anyways. Quote Link to post Share on other sites
igor 163 Posted November 6, 2014 Share Posted November 6, 2014 What sort of wireless TX module - can you combine that with the processor? Quote Link to post Share on other sites
RFEE 1 Posted November 6, 2014 Author Share Posted November 6, 2014 What sort of wireless TX module - can you combine that with the processor? I thought about that too, but right now I'm trying to determine whether M0+ or MSP430FR consumes less while it's on (Not Sleep). Then from there I can compare whether a MCU+wireless will consume less or more. Quote Link to post Share on other sites
rockets4kids 204 Posted November 6, 2014 Share Posted November 6, 2014 Based upon what you are asking, it seems as if you don't understand sleep modes or energy harvesting or both. Unless you have a self-contained energy harvesting system that will only power up an external device once it has harvested a sufficient amount of energy you are going to need to manage this. But in any case, the low power numbers provided by manufacturers are largely useless and I have found very little in the way of publications comparing performance across vendors. Most development shops do this internally -- and or course their results become part of their intellectual property. You are really only going to be able to find the best chip for your application by building and benchmarking with all potential parts. Alternatively, you can define your energy requirements and then see if your particular favorite chip can meet them. If it does, you are good to go. If not, then you start trying others. Quote Link to post Share on other sites
RFEE 1 Posted November 6, 2014 Author Share Posted November 6, 2014 As I mentioned before, I don't care about sleep mode current draw, and I already built a energy harvester with 40%+ efficiency. Please do not assume what I understand, it has nothing to do with the question. All I'm asking is which architecture is more efficient with sleep modes aside. Quote Link to post Share on other sites
rockets4kids 204 Posted November 6, 2014 Share Posted November 6, 2014 If you are not aware that going into sleep modes *while* you are communicating with your RF device is the key to low power operation then you know nothing about how to optimize microcontrollers for low power. Furthermore, I suspect the differences in power consumption between an MSP430 and M0+ to be trivial compared to the power consumption of your RF device. In fact, I suspect entire power consumption of the MCU may be trivial compared to RF power requirements. (unless you are doing something stupid like not going into sleep modes while letting the MCU's UART talk with your RF device.) Quote Link to post Share on other sites
spirilis 1,265 Posted November 6, 2014 Share Posted November 6, 2014 TBH this question may be better suited for the MSP430 section of TI's E2E forum - http://e2e.ti.com - due to the fact that you're (1.) Trying to make a determination around a potential commercial product and (2.) Are looking at competitive numbers that are probably best served by TI employees (and employees of the competing M0+ vendors you have in mind). Sent from my Galaxy Note II with Tapatalk 4 Quote Link to post Share on other sites
greeeg 460 Posted November 7, 2014 Share Posted November 7, 2014 @@rockets4kids While I tend to agree with alot of your points, we should assume that @@RFEE has done enough background work on his design to know which parameters he is using as a metric. This statement makes sense with where he is coming from with thew startup times. the sensor are completely dead and without power until there's either light, temperature difference, vibration, or RF illumination. Take the vibration for example. the system might be completely dead, i.e. not enough enough power to sleep. then when a large vibration occurs the system needs to wake up, and transmit some signals using only the energy it gathered from that shock. Ignoring a few extra things, like the RF system requiring energy to wake up, and any sensors etc. TI give a great table in their datasheets, Table 5-10 "Typical wake-up charge" Taken from the FR5969 datasheet Pg 31. Q_wake-up-reset: 75 nAs (2) 2.Charge required until start of user code. This does not include the energy required to reconfigure the device. A*s or Ampere Seconds is just Coulombs, the standard unit for charge. This is literally the energy required for the IC to power-up before it will execute any code. Looking at EnergyMicro's M0+ they don't have any comparable values. We can compute a value to try and compare. however during a MCU start-up not everything is running as it is in active mode, so these figures will be larger than if tested. Gecko zero M0+ typical startup time 163us (Device datasheet, table 3-5 power management) Startup speed 14Mhz (reference manual pg 106) Active mode current: 117uA/Mhz * 14Mhz = 1638uA (device datasheet, operating current) Startup charge = 266nAs Personally, I would build your device using both MCUs and personally test the difference. Asking about product comparisons on a forum about one of the products is inevitably going to return biased results. Full disclosure I have never used any energyMicro devices. I just spent a few minutes on google and reading datasheets. ike, tripwire and RFEE 3 Quote Link to post Share on other sites
rockets4kids 204 Posted November 7, 2014 Share Posted November 7, 2014 @greeeg: If there is one thing I know from 20+ years in the consulting business, it is that what clients ask for is only rarely what they actually need. In this case, I have a strong suspicion that because we're dealing with radio transmission of data, the power consumption of the microcontroller is going to be a non-issue by comparison. But even assuming this is not the case, you are *still* going to want to build a test jig to measure your *actual* power consumption. This is absolutely an area where you want to verify your datasheets before beginning any actual development. greeeg 1 Quote Link to post Share on other sites
zeke 693 Posted November 7, 2014 Share Posted November 7, 2014 I firmly agree with @@rockets4kids. You have to test the unit to be absolutely certain what the current consumption will be. I also firmly agree that current consumption will dwarf any consumption that a processor will have. I've worked with the Atmel ATMEA128RFA1 RF processor and the current consumption was surprising. During idle, it consumed between 1 and 12 mA. During RF reception, it consumed between 60 and 80 mA. During RF transmission, it consumed between 180 and 220 mA. When I was working with analog RF modulated radios operating in the 915MHz band, we had a saying "1 Watt of RF power will take 1 Amp of current". I believe this power ratio to be true no matter what modulation scheme is utilized. As a result, discussion of processor current consumption seems to be superfluous. We honestly do not know yet what @@RFEE wants to do with the processors. He's got a plan but hasn't shared it with us. I think we should invite @@jpnorair to get involved in this discussion. He regularly works with M0 processors and has past experience with CC430 processors. In the meantime, I imagine RFEE has many questions to form before he gets these questions answered. Quote Link to post Share on other sites
zeke 693 Posted November 7, 2014 Share Posted November 7, 2014 Hi All, My company is interested in developing passive sensors that are capable of harnessing energy from the environment, so basically the sensor are completely dead and without power until there's either light, temperature difference, vibration, or RF illumination. So I do not care about high current draw in sleep mode. In this type of environmentwhat would work best? MSP430FR? or M0+? Current draw during start-up on the MSP430 is also a problem, which is roughly 1 ms. Looking at the EMF32 Zero Gecko, Its list as 400us with a 24MHz clock. It seems to me that Zero Gecko would be a better choice, but from uA/MHz FRAM based MSP430 is a better choice. I don't need to do any crazy computation here, all I need is for the ADC to measure a time decay, and send it back out through a wireless tx module. @@RFEE, to directly answer your question, pick the MSP430FR part. It's a superior part. greeeg 1 Quote Link to post Share on other sites
Rickta59 589 Posted November 7, 2014 Share Posted November 7, 2014 It is pretty obvious that if you ask which is the better part in an msp430 forum you are going to get the answer that of course the msp430fr59xx chip is the better choice. greeeg and tripwire 2 Quote Link to post Share on other sites
rockets4kids 204 Posted November 7, 2014 Share Posted November 7, 2014 Most (if not all) of the FRAM parts are extreme overkill for this application. As an aside, I have yet to see a single demonstration of the FRAM parts showing appreciably less power consumption than non-FRAM parts for a simple application like this. Quote Link to post Share on other sites
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.