Please clarify the radio state diagram for the nRF52: it shows a black dot between TXIDLE and RXIDLE states, having arrows from TXIDLE or RXIDLE leading apparently back to TXRU or RXRU states (triggered by TXEN or RXEN tasks) without going through the DISABLED state.
(Note the diagram is different for the nRF51. The new diagram seems to imply a functional change, but no functional change is discussed in “Migrating from nRF51 Series to the nRF52 Series”.)
Does that mean you can go from the RXIDLE state to the TXRU state by triggering the TXEN task? Does that reduce the ramp up time (since you don’t go through the DISABLED state?)
Context: I am testing both the nRF51 and nRF52 chips, not using any existing protocol (writing my own, raw prototocol.) The protocol does enter the DISABLED state after any transmit or receive, and does occassionally switch from receive to transmit. I am wondering if there is any benefit to not enter the DISABLED state?
The Nordic blog “Intro to SB/ESB” devzone.nordicsemi.com/.../ seems to imply that ramp up is always incurred (130us on the nRF51, 40us on the nRF52 in fast ramp up mode.) If that blog is correct, then I would guess that the answer to my question is: yes you can make that transition but it is no faster than going through the DISABLED state.
The nRF52 Product Specification>RADIO >Electrical Specifications >Radio Timing says that Tswitch is at least 20us, and the fast ramp up is at most 40us (if the frequency is not changed.) So I suppose that a protocol where you did not enter the DISABLED state could at best be only a little faster (40-20=20uS faster.) So maybe the question is moot. Or the best advice is, use nrf52 in fast ramp up mode if you are concerned about ramp up.
