I'm using the CR2032 to power up my nrf51822 with LDO setup, it works well in the begining. But when the voltage drops below 2.6V, my phone cannot scan the BLE. Is it little bit weird because the voltage range of nrf51822 is 1.8V-3.6V, isn't it?
I'm using the CR2032 to power up my nrf51822 with LDO setup, it works well in the begining. But when the voltage drops below 2.6V, my phone cannot scan the BLE. Is it little bit weird because the voltage range of nrf51822 is 1.8V-3.6V, isn't it?
You need to be a little more specific about which type of board. ie, custom, devkit, etc.
However independent of this, 2.6v is getting near the functional end of life for a CR2032. So, unless your design has a lot of local capacitance, vbat is probably dropping significantly during the TX cycle, which could only be verified by monitoring vbat synchronous with TX activity on an o-scope.
To improve the situation, you should modify your code to make sure LED use is minimized and also advertisements should not be broadcast more than 1 per second when running on a lithium button cell. Also, local capacitance should be at least 22uF for button cell operation. Look in the devzone and people recently discussed the TI white paper on BLE and button cells, though you can arrive at the same conclusions with a little math and some battery specs.
Also, use of the LDO would not be recommended as it increases by 35% the required current during a TX burst. Always use the onboard switcher.
For your reference, here is the Panasonic voltage vs. discharge time curve for a CR2032. You will notice that 2.6v is just before it crashes to nothing. This is with a tiny 15kohm load. This is much smaller than the normal 11.8mA (4dBm)TX burst or 16mA using the LDO.
