• State Verified Answer
  • Locked Locked
  • Replies 16 replies
  • Subscribers 64 subscribers
  • Views 5459 views
  • Users 0 members are here
Attachments (0)
Nordic Case Info
  • Case ID: 100991
Options
This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Extremely high current consumption during advertising - broadcast mode(ble_app_temp)

Rafal

I have written application similar to ble_app_temp (posted by Ole Morten) and during the test I noticed that current consumption is extremly high. I used brand new coin battery that was included in evaluation kit and after ~10 minutes battery status was only 40% ( I checked this also with ble_app_hrs app). At the beginning I though that it was caused by the data included in advertisment packet ( I added 10 bytes of data to manufacturer data - adv interval was 1 second), but then I tested my battery with original app (ble_app_temp) and the effect was the same. My question is: did anyone also encountered such a problem with high current consumpion in case of broadcaster mode? I will test different battery today but I am afraid that is not case of battery.

Thanks in advance for your help.

ble_app_temp.zip

  • 0 in reply to Rafal

    38 mA shouldn't be possible for the chip to draw no matter what you do, so I believe there is something very strange about your measuring setup if you see such currents. Always remember to follow my other comments above, and making sure that there are no LEDs or anything that draw excessive currents through the GPIOs.

    In general, ble_app_proximity for PCA10001 should be a useful application to measure current on.

  • 0

    I have followed most of your comments (except cutting Uart lines). What's more, 38mA that I mentioned is measured based on voltage drop in the moment of radio transmition (so it really doesn't matter if there is some other constant current consumption, I am just measuring the voltage difference (from peak to constant consumption). I have also checked current consumption during transmition in case of ble_app_proximity and it looked similar - the peak to peak voltage was almost the same. I will be back from delegation on Friday and then I will perform more accurate tests, and of course share informations with you:)

    Best regards, Rafał

  • 0 in reply to Rafal

    I'm sorry, but there is simply no way that the chip itself can consume 38 mA. The only way I can see how such currents could even flow into the chip would be if there is something drawing a lot of current from GPIO, which if so would be outside spec (max 15 mA total, 5 mA per GPIO).

    The peak current consumption of the chip itself should be ~17 mA (13 mA RX + 4.4 mA CPU + ~3 µA for the RTC). You can see the expected current curves in the S110 Softdevice specification, sections 8.1 and 8.2. Does your curve look similar? If so, where exactly do you see this 38 mA peak? https://www.nordicsemi.com/eng/content/download/13238/213065/file/S110_SoftDevice_Specification_v1.2.pdf

  • 0

    Thank you for your reply. I decided to compare my app with ble_app_proximity (one that you suggested). I removed SB9, 10, 11, 12 bridges, compiled both apps using same settings (O3 optimization), turned off led indicators in both apps (I am not using GPIO ports in my app) and then I measured voltage drop using oscilloscope connected to first pin at P1 header. Please find attached 'screenshots' from oscilloscope. Basically output is similar in both cases, which makes me wonder, whether resistor that was mounted on my board is really 10 Ohm ( I was told so). Could you tell me what are the real values of current consumption in specific points of data transmition for ble_app_proximity (peak values that can be easily seen on plots that I have attached)? Additionaly I made screenshots after 5, 14, 17.5 hours of continous operation of the system. After 17.5 hours voltage droped from 3.21 V to 3.05V. Could you verify if this is reasonable value for app that is advertising some data with frequency < 1Hz (battery was brand new at the beginning of test)? I know that discharge curve for CR2032 battery has slope close to -90 degrees at beginning and because of that I would say that it is correct behaviour, but I am not 100% sure of that. Now when I have access to oscilloscope I will be able to check whether turning off INTENSET will influence current consumption. I will run current test for 2 more days and then I will check this.

    Thank you in advance for your help, Best regards, Rafał Urbański

    Apps comparison.zip

    Battery discharge.zip

  • 0 in reply to Rafal

    You can find detailed values for graphs like these in the S110 SoftDevice Specification, that I previously linked to: https://www.nordicsemi.com/eng/content/download/13238/213065/file/S110_SoftDevice_Specification_v1.2.pdf

    The numeric values for the symbols can then be found in the PS: https://www.nordicsemi.com/eng/content/download/13358/214991/file/nRF51822_PS%20v2.0.pdf

Related