Our full technical support staff does not monitor this forum. If you need assistance from a member of our staff, please submit your question from the Ask a Question page.

Log in or register to post/reply in the forum.

CS475A radar level sensor - why doesn't the integration require more than 1 second?

Bo 2 Nov 1, 2018 04:57 PM

In Normal Power Mode and with the default 10-second integration time, the 0M! command returns "00014", indicating data will be ready in 1 second.

Does this mean the sensor is always measuring and integrating, even between M! commands, and a valid distance-to-water is always available within a second?



Bo 2 Nov 10, 2018 09:43 AM

To test this theory, I left the sensor defaulted to Normal Mode with 10-sec integration time.  My logger program was set to sample the distance every 5 seconds.  When I moved the sensor a few inches further from its target (file cabinet) to simulate a step change, I was indeed able to see the reported distance slowly ramp up to the new distance.  It settled on the new distance in about a minute.

So I believe the sensor, in Normal Mode, measures the distance every second and calculates a new result every second, damped according to the integration time

I think this does mean that the logger can get the latest calculation by sending the M! command at any time.


jeff Nov 21, 2018 11:57 AM

In normal power mode the CS475A is always measuring, and it will respond back to a datalogger requesting a measurement in a second.  Where the default integration time for the sensor is 10 seconds, it would need at least 10 seconds worth of data to integrate over before it could start responding with data, and at that point you could request new data from the sensor every second if you wanted.  It would just basically be integrating over the last 10 seconds worth of data, and it would just be a moving 10 second window, where a datalogger could request a new reading from the sensor anytime, up to a rate of once per second.  The way the CS475A currently operates it does take about a minute to settle in on the reading when there has been a step change, even if you do reduce the integration time.

Bo 2 Nov 23, 2018 06:07 PM

I'm used to SDI12 devices being comatose except when they receive a M! command, so it took me a while to realize the CS475A was sampling at 1HZ all the time, continuously refining its estimate of the distance to the water.

That I can get all that computing power in 1 second, with one simple command, is amazing.

Thanks Jeff.


Log in or register to post/reply in the forum.