I have questions related to power consumption of BGM13S, i referred the following link to reduce the power consumption during advertising and connected state ( Attached snap shows the result that i got )
The soc-empty example application is running, no other external components are connected to board.
I have connected the 500mAh battery to BGM13S ,which was 3.95V yesterday evening at 6:20pm, today evening at 6:20pm it is 3.85V [ Battery life will be nearly 20days ].
My questions are :
1. How can i improve the battery life to at least 6 months......................................?
2. Energy profiler is enough to see the power consumption of entire custom board (or) we need any hardware component like DC analyzer.......................?
I have connected the 500mAh battery to BGM13S ,which was 3.95V yesterday evening at 6:20pm, today evening at 6:20pm it is 3.85V [ Battery life will be nearly 20days ].
How did you come up with the 20 day estimate?
Advertising current is 25uA. Battery capacity is 500mAh. Assuming you can use all that energy optimally, the runtime is 500mAh / 25uA = 20000 hours. That's 28 months. Looks like there's plenty of margin if your target is 6 months.
Since you mentioned the measured battery voltages, I hope you don't assume that the battery voltage will behave linearly throughout the lifetime of the product?
0
Thank you for your reply.
I made an excel sheet regarding power consumption of BGM13S using energy profiler tool and calculated the battery life as you mentioned, but theoretical calculations are not matching practical scenario's.
Battery voltage is reducing gradually, but the rate of reducing is more i guess, why because I connected fully charged battery ( 500mAh , 3.95V ) to custom board but it lasted only for 20 days and the Battery voltage was 1.7V .
whats might be the problem, how can i resolve it, Is there anything i'm missing.........?
Is Energy profiler is enough to see the power consumption of entire custom board
Thanks,
vijay
0
Ok, makes sense. So you have actually tested that the battery will run out in 20 days. Something is definitely wrong then.
By the way, you are exceeding the absolute maximum ratings of the BGM13S. I just checked the datasheet:
if the full battery voltage is 3.95V then you are clearly exceeding the spec. It might not explain the increased current consumption, but something to be checked anyway.
Is Energy profiler is enough to see the power consumption of entire custom board
I am using Energy Profiler to measure the current of the products that I am working with and it think it is a reliable tool to show the actual power consumption. I use WSTK and hook my custom PCB to the expansion header (VMCU and GND, pins 2 and 1 of the EXP header).
Can you share the schematics of your custom board? If you can't upload it here then I suggest you open a private support case and ask for a review of the design. There must be lots of current leaking somewhere because the battery should last much longer than 20 days. And your energy profiler graphs look OK to me.
1
Hi, the Energy Profiler specifications from the Radio Board User's Guide:
7.3 AEM Accuracy and Performance
The AEM (The Advanced Energy Monitor) is capable of measuring currents in the range of 0.1 μA to 95 mA. For currents above 250 μA, the AEM is accurate within 0.1 mA. When measuring currents below 250 μA, the accuracy increases to 1 μA. Even though the absolute accuracy is 1 μA in the sub 250 μA range, the AEM is able to detect changes in the current consumption as small as 100 nA.
The AEM current sampling rate is 10 kHz.
Note: The AEM circuitry only works when the kit is powered and the power switch is in the AEM position.
And Tyler already told how to connect a custom design into the EXP-header. It is still good to remember that it can provide only 3.3 Volts so if the real system voltage is different, the current consumption might also be a bit different (most likely max tens of percents, not many times).
Reviewing the hardware design might be a good idea as suggested.
BR, Juha
Correct Answer
0
Sorry, the battery is connected to voltage regulator, which outputs the voltage <=3.3V.
I'm connecting 10 pins from the WSTK to custom board as described in following link
Ok, I will create a private support case and ask for a review of the design
Thanks & regards
vijay
0
Sorry, the battery is connected to voltage regulator, which outputs the voltage <=3.3V.
I kind of guessed this might be the case. Go ahead with the support case and make sure you add details about the regulator used.
Next thing I would measure is powering the custom board from an adjustable lab power supply and check the current consumption at different input levels (3.3V, 3.4V, 3.5V etc). If you connect a multimeter in series you will probably see the current go up the higher the input voltage. (Just a theory, would be nice to hear what's the result)
Based on how fast you drain the battery I guess the current is high enough that you can measure it with some cheap digital multimeter.
0
Hi,
i got a good results out of it.( Practical output : coin cell battery( 225mAh) is connected to custom board.,Its almost 16 days still the voltage is same ).
But the issue is with connection, it is giving some gatt service conn time out error.
Thanks & regards
vijay
0
But the issue is with connection, it is giving some gatt service conn time out error.
It's better to create a new topic/thread for this connection issue. Also, you need to provide more details about the error (does it happen every time you try to connect etc..)
0
Hi,
Yes, It happens every time.
I found that, while creating a soc-empty project we need to specify part and boards details in device configuration as described in below snaps.
But if i specify only part details and generate the code, connection with custom board(Only BGM13S ) is possible, but if i specify both part and boards details, connection with custom board(Only BGM13S ) is not possible.
Please find the attached snaps and let me know if you have any solutions
I found that there is no external oscillator in our custom board, hence we are using LFRCO oscillator which is giving connection issue .
If we generate a source code by giving device configuration as specified in the above Boards_Inc.png the power consumption is high ( around 3.5mA ), Hence we need to use PLFRCO oscillator.
I followed the below steps to use PLFRCO, but BGM13S is not advertising
battery life validation
Hi,
I have questions related to power consumption of BGM13S, i referred the following link to reduce the power consumption during advertising and connected state ( Attached snap shows the result that i got )
https://www.silabs.com/community/wireless/bluetooth/knowledge-base.entry.html/2017/08/25/optimizing_currentc-C4uw
Current consumption :
1. Advertising = around 25μA
2. Connected = around 15μA
The soc-empty example application is running, no other external components are connected to board.
I have connected the 500mAh battery to BGM13S ,which was 3.95V yesterday evening at 6:20pm, today evening at 6:20pm it is 3.85V [ Battery life will be nearly 20days ].
My questions are :
1. How can i improve the battery life to at least 6 months......................................?
2. Energy profiler is enough to see the power consumption of entire custom board (or) we need any hardware component like DC analyzer.......................?
Thanks & regards
vijay
What's the logic here?
How did you come up with the 20 day estimate?
Advertising current is 25uA. Battery capacity is 500mAh. Assuming you can use all that energy optimally, the runtime is 500mAh / 25uA = 20000 hours. That's 28 months. Looks like there's plenty of margin if your target is 6 months.
Since you mentioned the measured battery voltages, I hope you don't assume that the battery voltage will behave linearly throughout the lifetime of the product?
Thank you for your reply.
I made an excel sheet regarding power consumption of BGM13S using energy profiler tool and calculated the battery life as you mentioned, but theoretical calculations are not matching practical scenario's.
Battery voltage is reducing gradually, but the rate of reducing is more i guess, why because I connected fully charged battery ( 500mAh , 3.95V ) to custom board but it lasted only for 20 days and the Battery voltage was 1.7V .
whats might be the problem, how can i resolve it, Is there anything i'm missing.........?
Is Energy profiler is enough to see the power consumption of entire custom board
Thanks,
vijay
Ok, makes sense. So you have actually tested that the battery will run out in 20 days. Something is definitely wrong then.
By the way, you are exceeding the absolute maximum ratings of the BGM13S. I just checked the datasheet:
if the full battery voltage is 3.95V then you are clearly exceeding the spec. It might not explain the increased current consumption, but something to be checked anyway.
I am using Energy Profiler to measure the current of the products that I am working with and it think it is a reliable tool to show the actual power consumption. I use WSTK and hook my custom PCB to the expansion header (VMCU and GND, pins 2 and 1 of the EXP header).
Can you share the schematics of your custom board? If you can't upload it here then I suggest you open a private support case and ask for a review of the design. There must be lots of current leaking somewhere because the battery should last much longer than 20 days. And your energy profiler graphs look OK to me.
Hi, the Energy Profiler specifications from the Radio Board User's Guide:
7.3 AEM Accuracy and Performance
The AEM (The Advanced Energy Monitor) is capable of measuring currents in the range of 0.1 μA to 95 mA. For currents above 250 μA, the AEM is accurate within 0.1 mA. When measuring currents below 250 μA, the accuracy increases to 1 μA. Even though the absolute accuracy is 1 μA in the sub 250 μA range, the AEM is able to detect changes in the current consumption as small as 100 nA.
The AEM current sampling rate is 10 kHz.
Note: The AEM circuitry only works when the kit is powered and the power switch is in the AEM position.
And Tyler already told how to connect a custom design into the EXP-header. It is still good to remember that it can provide only 3.3 Volts so if the real system voltage is different, the current consumption might also be a bit different (most likely max tens of percents, not many times).
Reviewing the hardware design might be a good idea as suggested.
BR, Juha
Sorry, the battery is connected to voltage regulator, which outputs the voltage <=3.3V.
AN958: Debugging and Programming Interfaces for Custom Designs(https://www.silabs.com/documents/public/application-notes/an958-mcu-stk-wstk-guide.pdf)
Ok, I will create a private support case and ask for a review of the design
Thanks & regards
vijay
I kind of guessed this might be the case. Go ahead with the support case and make sure you add details about the regulator used.
Next thing I would measure is powering the custom board from an adjustable lab power supply and check the current consumption at different input levels (3.3V, 3.4V, 3.5V etc). If you connect a multimeter in series you will probably see the current go up the higher the input voltage. (Just a theory, would be nice to hear what's the result)
Based on how fast you drain the battery I guess the current is high enough that you can measure it with some cheap digital multimeter.
Hi,
i got a good results out of it.( Practical output : coin cell battery( 225mAh) is connected to custom board.,Its almost 16 days still the voltage is same ).
But the issue is with connection, it is giving some gatt service conn time out error.
Thanks & regards
vijay
It's better to create a new topic/thread for this connection issue. Also, you need to provide more details about the error (does it happen every time you try to connect etc..)
Hi,
Yes, It happens every time.
I found that, while creating a soc-empty project we need to specify part and boards details in device configuration as described in below snaps.
But if i specify only part details and generate the code, connection with custom board(Only BGM13S ) is possible, but if i specify both part and boards details, connection with custom board(Only BGM13S ) is not possible.
Please find the attached snaps and let me know if you have any solutions
Thanks & regards
vijay
Hi,
I found that there is no external oscillator in our custom board, hence we are using LFRCO oscillator which is giving connection issue .
If we generate a source code by giving device configuration as specified in the above Boards_Inc.png the power consumption is high ( around 3.5mA ), Hence we need to use PLFRCO oscillator.
I followed the below steps to use PLFRCO, but BGM13S is not advertising
https://www.silabs.com/community/wireless/bluetooth/knowledge-base.entry.html/2019/06/18/kba_bt_0409_usingtheplfrcoaslow-frequencyclo-vPoh
Like to know the advice from you all.
Thanks & regards
vijay
Hi,
PLFRCO is introduced in BGM13S22F512GA-V3 as specified in below document.
https://media.digikey.com/pdf/PCNs/Silicon%20Labs/191018649.pdf
I tried it with v3 boards. It worked for me, now there is no issue with connection.
Thanks & Regards
vijay