Welcome to the third of our SenixVIEW Video Tutorial series: Basic Sensor Configuration.
We return to the main Workspace window. And, in this case, we are going to click on the “Measure” icon on the tool bar. This brings up the measurements window where we set basic sensor configuration. Measurement activation has a number of choices. “Continuous” is where it’s generally left although “Start on poll” is sometimes used. The suite of synchronization choices we will revisit in special section. Let’s leave it on “continuous”. Power save is disabled. And that is the normal mode of operation. The sensor don’t require a great deal of power. And only in battery operated, power starved situations would we consider the enabling power save. Measurement interval is at default or at 50 milliseconds. That’s 20 readings a second. For illustration, let’s change that and slow it down to a hundred milliseconds. That’s 10 readings a second. Temperature compensation is none by default, we will make that be internal and now we are going to use an internal temperature sensor within the body of our distance sensor to track changes in the measurement environment.
To the right, in the sensitivity box, you will see gain profile and this can be set to any one of six basic settings. Free air, low medium and high, and stilling tube, low, medium and high. We will leave it on the free air medium for now which is the default. But in a stilling tube environment, it is still necessary because the sound is contained within the tube, to reduce the gain and that enables operation in that case. In extreme cases, we may need to go to “Advanced” and in the advanced gain settings we are able to see the change in sensitivity with time or distance for each of these standard settings and make some adjustments to them. I don’t recommend you do that unless there is a specific reason and we’ll revisit this in certain application sections. So let’s leave that for now.
I’d like to call your attention to the “Filter Options”. This is the on-board signal processing capability. We have the ability to put in “Input Rejection” filter in place. It can be one of three. X of Y samples, which I’ll explain in a moment. The “Farthest of Samples” or the “Closest of Samples”. Closest of Samples might have any one of the number readings in there. If you put in ten, it would return to you the shortest reading among the last ten that were made. In the case of Farthest, that would be the same. You might be able to see that those two could be used to pick the peak of a wave or trough of wave.
The “X of Y Samples” is a little bit more complicated. The default is typically 3 out of 4 but I like to use 2 out of 2. What it does is to apply a 12 percent arbitrary filter. It falls out of the digital math, but when you set 2 of 2, what it says is only output a measurement if two measurements in a row are within 12 percent of each other. The 12 percent being arbitrary. And that eliminates any strange reflections or hiccups you might have with the sensor, and sometimes, ultrasonic transducers do tend to hiccup, not often! But, this will filter out any of that. I like the use of 2 of 2 and usually it has no impact on what you are doing. Except, in some cases in tanks where it may.
Signal averaging is of particular interest. Here, you can choose “none”, or “Running average” or a “Box Car average”. Running average gives you the mean of a number of samples, default actually is two here. You get a signal to noise improvement that is proportional to the square root of the number of samples. So the sample of 10. The square root is little over three, it will be about a three time improvement to signal to noise. If I put in a 100, it will be a ten time improvement in signal to noise. But it obviously takes more readings to get there. So, if we were, for example, measuring 10 times a second and we said give me the average of 10 samples, we would have the mean of 10 every second.
In the case of a “Running Average”, it will bring in the latest reading, give you the average of 10, throw out the oldest. Bringing the latest reading, throw out the oldest give you the average of 10. And that would be a one second mean output, 10 times a second. If you did it as a Box Car average, it will not output until it is taken in 10 samples and it will give you the mean. It will take in 10, and then will give you the mean. So the Running Average is a little bit more responsive to change in the measurement environment but it does impose something of a low pass filter on your data. I am going to leave that at Running Average. Often, by the way, I will set this to 1 second interval and samples to 60. So, I have a minutes worth of data every second. Distance limiting a has couple of possibilities. “Slow/Fast” is one filter and “Rate of Change” is another. They are both generally related to machine control applications and we will deal with those a bit later. They are not used very frequently. So we will leave that at “none”. Let’s okay out of this window.
And now, you will notice that we have a not equal sign between the Workspace and the Sensor. What we have to do is download to the firmware the changes we have made in the Workspace. Two ways to do that: Sensor menu-Come down to “Move Workspace to Sensor” and you will notice that the pencil will write to the right. Alternately, you can right click on the Workspace icon. See the other pencil come up and we can just drag it to the sensor and release it. It achieves the same thing.
So those are some of the basics in sensor configuration. In our next session, we will discuss the configuration of our Analog Outputs and show you some of the test facilities that are present in the program for verifying your house wiring in terms of Analog Outputs.
If you have any questions, please contact Senix customer service. We’re here to help!