Dear Sirs,
I regard myself as an experienced TD7200 user, now working on my third design using this part, so far with excellent results. However I have now got a problem. The TDC7200 apparently gets into a state where it no longer makes measurements:
The device is used in measurement mode 1, measuring times between 100ns and 200ns with a 10MHz clock. The only change to the default setup I make is to write 00h to CONFIG2, to change calibration2 to 2 clock periods.
The sequence is use to measure is : a) Write 01h to CONFIG1 to start a measurement
b) After an interrupt caused by the stop pulse, I read TIME1, CAL1,and CAL2
I then repeat step a) The maximum measurement rate is one measurement per second.
Note that I do not use the TRIG output or the INTB output.
The current design works for many hours successfully, however sometimes the TDC7200 stops measuring. The behavior of the stuck state is :
a) The TDC7200 apparently responds correctly to the write to CONFIG1, as it sets the TRIG output high.
b) It definitely gets a start and stop pulse, in the correct order.
c) It then apparently starts the read cycles correctly, as it controls the DOUT pin. ( I have checked this by adding a weak pullup to the pin)
d) The output of TIME1,CAL1, and CAL2 are always zeros.
It is possible that sometimes the stop pulse is missing, caused by a software problem, although I have not yet definitely detected this. Can this initiate the stuck state? I assume that with no stop pulse the counters would overflow. The data sheet says that this will reset the measurement, in which case the next write to CONFIG1 should start a new measurement. If the write to CONFIG1 occurs before the counters overflow, can this be a problem? Is it possible to deliberately terminate the measurement by writing 00h to CONFIG1?
regards
Cosmo Little
