
Channel Modes
Understanding the eTPU Channel Hardware, Rev. 0
Freescale Semiconductor
25
blocking time could be a fixed delay or a variable driven by the measured
speed of the shaft. The second match could then be setup to request service
if the following transition does not occur within a selected time, indicating
possible stalling of the shaft.
If this technique is used to reject a noisy edge, it may be necessary to process
both the rising and the falling edges of the input signal, as noise pulses on a
falling edge may also be detected by a
DetectARisingEdge()
transition.
3.10 Match B, Double Transition
In the
Match2DoubleTransition()
(m2_dt) mode, Transition A does not block matches nor issue a service
request. These actions occur on Transition B. Otherwise the action is identical to the previous mode. If
Match B is set up to occur after Match A, a pulse detection on the channel will be detected, as shown in
Figure 6
. Note that if Match B occurs before Match A, then only MatchB_TransA will request service, and
only MRLB will be set.
This mode has an obvious application in qualifying an input pulse with a lower and upper limit. A less
obvious application would be on a single transition time-out, where the service needs to be delayed until
some time after the transition. If DetectBDisable() is selected, the Match A could set the blanking time for
an input transition, and the transition time could be captured without a service request. The Match B would
then be setup to request a service thread during which the software could test TDLA to find out if the
transition happened at all. This might be used where the software only needs to confirm that something
expected actually did happen.