2025-01-15 05:39 AM - edited 2025-01-15 05:54 AM
Hello,
I would like to go a bit deep in the inductive wake up function. Unfortunatly seems that i will need to work with delta 1 at least for the moment, but due that it is a battery powered device i need to avoid as much as fake awake when there is no RFID card in front.
1) Im playing with your 3916B Discovery kit connected to my antena instead ST antena board, and im seeing for example that on phase awaken, mostly of the “fakes” are values under the reference, like i set as reference 120 and fake shows a value of 119, never 121. Is that realistic to say that fakes will follow that logic? Or it is just coincidence…
2) Has ST any statistics about behaviour with Delta 1?
3) In addition, my feeling is that Just rebooting sometimes the reference is taken for example as 120 and sometimes in a range of 140. i think im not modifying anything. Any idea about why this could be? I mean, is normal to read such different values as reference? Im using Discovery firmware, with no modification.
4) Is is possible to set awaken with delta 1 but only +1 not –1?
5) is it possible to configure the transceiver to trigger the interrupt not at the first delta 1 detection and to do it for example after 3 times in a specific time range?
6) Is it ok to use Amplitud and phase awaken at same time? Or this could trigger me other problems
7) any idea of how to minimizate this fake awakes?
Any help will be higly apreciated, thanks!!
Solved! Go to Solution.
2025-01-16 12:21 AM
Hello Ignacio,
ad 1)
Likely you are looking into some cooling/settling effects in your system (inductors, power supply, caps). I presume there is no general guaranteed direction of this settling. You can do statistics on your system and verify they all have the same direction.
ad 2)
delta=1 is inherently unstable. Actual value can can always be in the middle between two values (analog). When digitizing you will have either the lower or the higher value - even if your noise floor is at minimum.
ad 3)
Could be the regulator gets adjusted to a different value sometimes.
ad 4)
Not with ST25R3916, ST25R200 and newer lets you select the direction.
ad 5)
Not possible, but you are free to do such on MCU side. You can enable wto and I_wut interrupt and let your software look at the values and decide there.
ad 6)
Yes, this is possible and not problematic. Power consumption will increase as pulse duration will increase to ~59us.
ad 7)
Auto averaging could be an option. It would help if the internal reference (at higher precision) is already between two bits. So you need to perform some tricks here as well. E.g. set wut=10ms and let wakeup run with auto averaging for 100ms, then switch only wut to 100ms and push the MCU to low power.
Regards, Ulysses
2025-01-16 12:21 AM
Hello Ignacio,
ad 1)
Likely you are looking into some cooling/settling effects in your system (inductors, power supply, caps). I presume there is no general guaranteed direction of this settling. You can do statistics on your system and verify they all have the same direction.
ad 2)
delta=1 is inherently unstable. Actual value can can always be in the middle between two values (analog). When digitizing you will have either the lower or the higher value - even if your noise floor is at minimum.
ad 3)
Could be the regulator gets adjusted to a different value sometimes.
ad 4)
Not with ST25R3916, ST25R200 and newer lets you select the direction.
ad 5)
Not possible, but you are free to do such on MCU side. You can enable wto and I_wut interrupt and let your software look at the values and decide there.
ad 6)
Yes, this is possible and not problematic. Power consumption will increase as pulse duration will increase to ~59us.
ad 7)
Auto averaging could be an option. It would help if the internal reference (at higher precision) is already between two bits. So you need to perform some tricks here as well. E.g. set wut=10ms and let wakeup run with auto averaging for 100ms, then switch only wut to 100ms and push the MCU to low power.
Regards, Ulysses
2025-01-20 01:25 PM
Dear Ulysses,
thanks for your quick answer.
Soon we will start here the real test and try to find the best management for this delta 1. Your ad7) idea is so welcome.
Im sure this is temporary, i dont like to work with this delta, but until we have a better option we eill need to manage it.
If during the implementation i have doubts i will come to you
thanks!