Hi Antonio, Gerat to hear that it works. The driver can lock up if it does not receive the right interrupts. Branoislav Kusy has some fix to make it more resilient (in case for some reason an interrupt is missing), but those interrupts should be received on an undisturbed platform.
Best, Miklos On Tue, Dec 21, 2010 at 8:20 AM, antonio rosa <[email protected]> wrote: > Hi Miklos, > > I have been working on it with my partner Francisco J. Luque. Finally, we > checked that the driver works properly. The radio goes to sleep state and > the oscillator is turned off. > > I would like to apologize about the mistake and explain why we came into it. > We tried to check the RF230 sleep state using TinyOS2, by monitoring the > oscillator pin, as we did some time ago using TOS1x. In the past, we got > signals like shown in Xmesh.JPG. Channel 1 shows the SPL_TR signal and > channel 2 shows the oscillator signal (We took the metal case off from the > IRIS module to access the pins). You can see how the oscillator is turned > off when SPL_TR is high. Using the time scale, the 16 MHz sinus can be seen > in the channel 2 when SPL_TR is low. The mote uses XMesh to manage the radio > implementing a LPL of 125ms wake up cycle . > In this way, we checked that the power off was properly executed on > TinyOS1x. We tried to do the same thing now, but using TinyOS-2.1.1. This > time, the oscillator was allways on. That is because we thought that the > RF230 did not entered into sleep state. We have check today that the > behaviour of the mote is modified when we connect the probe to the > oscillator pin. After a HW reset, the SPL_TR was commuting propperly but, > once the probe had touch the oscillator pin, it stopts. It seems that the > RF230 do something wrong and the driver on the microcontroller stops > working. This does not happen in TinyOs1 but happens in TinyOs2, we did not > find the reason. > We finally found a way to make our test without perturbing the RF230. We > used a 100 kOhm resistor in series with the probe and we saved the signals > you can find at TO2.jpg. This is the proof of the LPL is properly executed. > > We apologize again because of the waste of your time and hope that the > signal pictures can be useful for you in the future. > > Thanks and best regards. _______________________________________________ Tinyos-help mailing list [email protected] https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
