Hi.
When threshold level at ultrasonic burst and decay is set as 255 (248 + 7),
it should ignore the burst and decay time period as shown in Fig.32 (and FAQ 5.2).
Is there any exceptions?
Since in my test case, IC responds too short time(equivalent to ultrasonic decay time) although threshold level is set to 255.
Thank you.
Hello Kondo-san,
Thanks for the question and for posting to the sensing forum! I have not come across the scenario where the burst and decay are not ignored (triggering a UMR) when the threshold is set to 255 (248 +7).
Make sure there is enough clearance when between your echo decay as well as your threshold settings. This will help avoid triggering a result during the time the threshold is also decaying to the next point. You may also parse the UMR result for validity, by running a system diagnostics command (command 8) you can obtain the decay time of the sensor. If your measurement result occurs before the decay time indicated by the diagnostic and if the UMR result indicates that the peak amplitude is 255, then it is a safe indication that you are capturing the burst.
Below is an example of a measurement I took, I set a lot of the threshold markers near the burst to see if I could cause any of them to trip and generate a result but they seem to operate fine until I got the return:
Another thing to check, make sure that you are using the correct threshold preset for your measurement.
Best,
Isaac
Thank you for your reply.
I couldn't clearly understand " enough clearance when between your echo decay as well as your threshold settings". Is this means threshold value setting needs enough clearlance from echo decay though the threshold is set to 255? If so, is there any guidelines?
Below image is an example of what happens im my test case. Threshold value 255 period of Red threshold line is100μsec longer than Green threshold line. In case of Green threshold line, IC responds 0.4msec in case of Green threshold line (which is not suitable), while IC responds 1.6msec in case of Red threshold line (desired value). Is this caused by not enough clearance from echo decay?
Hello Kondo-san,
Thanks for the additional details and the waveform. This helped me understand the problem and it is something we have come across before. It seems as the data comparator ramps up during the burst and listen period and the data is not saturated which can be falsely triggered at around 0.067m (~0.4msec).
One way to overcome this is by saturating your signal initially to eliminate the initial signal ramp such that it will not falsely trigger the threshold, we cover other ways how you can overcome this issue in section 5.6 of this FAQ: https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/579804/pga460-q1-pga460-frequently-asked-questions-faq-evm-troubleshooting-guide?tisearch=e2e-sitesearch&keymatch=pga460%25252525252520deglitch#
To clarify what I was talking about before in my previous post is ensuring that you have enough margin horizontally, if the threshold falls too close to the line it may accidentally trigger your threshold during your ringing period. As seen in the images below if I do not give the threshold line enough margin horizontally (in the time domain) you can see it was trigger and measured incorrectly. But when given some margin it reacts correctly.
Let me know if these suggestions help!
Best,
Isaac
