• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 31 subscribers
  • Views 454 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Stellaris LaunchPad (EK-LM4F120XL): suitable for µA application experimentation?

Joe Desbonnet
Intellectual 355 points
Other Parts Discussed in Thread: TLV803

I've got a very low power application (hibernates most of the time) currently served by a 8 bit MCU and I would like to migrate to a more flexible 32 bit ARM based MCU. The Stellaris Launchpad ((EK-LM4F120XL) seems to be a great way to try out the Cortex M MCUs from TI.

I realize I can prototype the application in the lab and measure application current by removing the V3,3/MCU_PWR (H24 - H25) jumper and bridging with an ammeter, but it's not clear if I can configure the LaunchPad board to run under low power. I really would like to use the Launchpad to demonstrate the Stellaris MCU working in the real application and not just in the lab.

I though it would be sufficient to disconnect the H24-H25 jumper and supply 3.3V via the boosterpack pin but that doesn't work for some reason. If I supply 5V via the boosterpack pin and keep H24-H25 connected it does work. *BUT* the green power LED (D4) is on... obviously it is not going to be very low power with that on.

So can the Launchpad board be configured to consume just a few µA while the MCU is hibernating?

Thanks,

Joe.

  • 0
    Genius 5695 points

    Joe:

    I ran through the course in the training wiki... and they do walk you through some tests...

    http://processors.wiki.ti.com/index.php/Getting_Started_with_the_Stellaris_EK-LM4F120XL_LaunchPad_Workshop?DCMP=stellaris-launchpad&HQS=stellaris-launchpadtraining

    Both on Stellaris and the MSP430 I got the results claimed..

    Maybe just give it a try and see if you can get the same numbers roughly. My meter will measure microAmps -- so it made it easy.

    In case you haven't noticed the link yet which leads you to the courses...

    http://www.ti.com/ww/en/launchpad/stellaris_head.html

    I think the claims are credible -- but then again your app may bring out the worst -- I guess you have to try.

  • 0 in reply to Dave Robinson
    Intellectual 355 points

    Dave,

    Thanks for the reply. Yes, I can see just a few µA across the H24 - H25 jumper when the device is hibernating: I have no doubt about the MCU's low power capability.  

    But my problem is this: how can I deploy a prototype in the field using the actual Launchpad board using nothing but a small solar panel power harvesting circuit to power it. With the power conditioning in the debug (top) part of the Launchpad board (eg the green power LED (D4) on)... there is no way the board is going to remain inside the power budget.

    Joe.

  • 0 in reply to Joe Desbonnet
    TI__Intellectual 1185 points

    Hey Joe,

    You should be able to supply 3.3V at any point past the regulator that steps the 5V signal down to 3.3V for the lm4f120 and not have any of the debug circuitry on the top of the board powered.  The easiest way to do that would be by supply a regulated 3.3V to the right leg of the jumper labeled VDD.  The downside to this is there's a reset regulator on the debug portion that will hold U1 in reset if the 5V line is not present.  You can get around this by depopulating D2, which should prevent the assertion of the reset line, thus allowing you to run from an external 3.3V source.

    Would that work?

  • 0 in reply to Stellaris Jordan
    Intellectual 355 points

    I think that sounds like a promising solution. I found a similar hack here:

    http://blog.psten.de/post/fuel-for-the-stellaris-rocket

    His solution is to cut the trace between the cathode of D2 and the RESET line of the TLV803 voltage supervisor IC. It's a pity I need to do something destructive to get there, but at $13 each I can't complain. (Perhaps this mode of operation is something worth supporting in a future hardware revision?)

    Thanks,

    Joe.