May 11, 2010

Special! Datalog report for the new sensor from Dexter Ind. : dPressure

Hello everyone!  This is a datalog report of the new Dexter industries sensor coming: end of May.  Here it is!

 Coming in late May, I present the new "dPressure" from Dexter Industries!  The dPressure is a very interesting sensor.  It has two nozzles: the first on can sense pressure in both PSI (Pounds per Square Inch), and kPa (Kilopascals).  The second nozzle can be used as a vacuum sensor (sense negative gauge PSI or kPa).  It's also very easy to use too!  All you have to do is connect an NXT wire to the sensor and to the NXT and there you go!  It, so far, can only be programed in NXT-G 1.X, and NXT-G 2.0.

                   As you can see in the pictures, it can connect to any one of the NXT's sensor ports.  It also has four data hubs:
1. Port
2. PSI
3. Raw data
4. kPa
The programing block works quite well.  It's very simple; therefore, it's easy to use!  I made a simple program that displays the current PSI on the NXT's screen:
The program works perfectly!

I have made a test rig to make some graphs of the sensor gaining up till the pair of Power Functions XL-Motors won't go any longer:
In the picture, there is the NXT; the dPressure sensor is connected to the side of it (NXT).  There is also two LEGO large Pneumatic pumps, a water bottle to hold the compressed air, and some other Power Functions accessories (including an IR Receiver, and remote).  How it works: when I start the datalog program, I start the pumps (using the Power Functions Remote) that fill up the bottle.
  You can see the results here:
 Here, you can see the 60 second test (the horizontal line is in 5 second increments [for all the graphs]) of the water bottle filling up with air.  That flat spot is the motors slowing down because of lack of strength.
 You can see the same thing here, it rises up, then levels out.
There are some ideas that come to mind when I see these results.
1. Have the bottle as one big touch sensor.  When I press down on the bottle, the pressure will rise, naturally.
2. A compressor that will fill up with air, but when there's enough air, it will stop pumping.
3. A silly scale to weigh objects (same idea as #1).
4. Measure depth below sea level, and as it goes lower, the pressure changes (idea by Brian Davis)
5. Detect automobile traffic by placing a long tube across the road (or walkway), and when a car (or person) drives (or walks) over it, the dPressure can sense that (idea by Brian Davis)
6. Sense when someone tosses and turns at night by connecting the dPressure to an air mattress (idea by Brian Davis)

In my experience with LEGO products, I think, overall, this is a great sensor to purchase!  It's easy, fun, and can be used for various reasons!  Play Well!


  1. Nice write up. There are some things it would be interesting to learn a little bit more about as well. For instance, you say the sensor can measure "negative pressure"... by that, I assume you mean negative *gauge* pressure (pressure lower than ambient). It would seem this is rather sharply limited to a pressure about 1.8 psi below ambient however. Making some assumptions from your graphs, I'm guessing the sensor makes out ("off scale high") around 35 psi, or a shade over 2 Atm - is this the case?

    It's also unclear to me what you mean by "datalog program": does the DI block function within the NXT-G Edu datalogging environment, or did you write your own small program to log sensor values to a standard text file? If they second, what are the units on the horizontal axis? Seconds? Something else?

    I'm also guessing that each of those runs above was from a *different* experiment, correct?

    As to your ideas, there are some others you could use. For instance, measuring depth below the water surface (as the depth increase, so does the pressure, in a very reliable way, the rate of air loss from a working pneumatic system, detecting automobile traffic (yep, there's a way), measuring how someone tosses and turns at night (hook it up to an air mattress, and log the pressure), etc. In fact the one use I've never quite understood for sensors like this to "stop a compressor", as there are very reliable, completely mechanical methods of doing this (LEGO polarity switch activated by a small pneumatic cylinder opposed by a rubber band), but it will certainly work that way (why choose the mechanical solution? See my recent blog post on the subject at theNXTstep blog, and realize that the mechanical version frees up one input and one output port that you'd otherwise need).

    It does look like a nice, useful sensor - thanks for writing it up!

  2. Thanks Brian! I have added changes to it.