Gas station without pumps

2012 October 16

Pressure sensor assembly

In Pressure sensing lab possibilities  and PC board for pressure sensor, I outlined a lab based on modeling water in a tube using electrical circuit techniques:

Two reservoirs of water connected by a thin hose, with a pressure sensor near the bottom of each reservoir.

Capacitor and screw-terminal side of the breakout board.

Sensor side of the breakout board

I ordered some “biomedical” pressure sensors (Freescale’s MPX2300DT1) which have a differential output of 5µV/V/mmHg, and will need an instrumentation amp to get usable signals.  I also designed a breakout board to mount the pressure sensor, but had not had the spare time to solder and glue the board to test the whole lab.

I finally got the board soldered last night. The surface mount soldering was not to difficult, nor was supergluing the board in place (which I did after soldering). I think I managed to do the gluing without fouling the back of the sensor, but testing will be needed to confirm that.

Sensor side of the breakout board (before adding glued-on standoffs).

Connector side of breakout board.

Side view of breakout board (before adding standoffs).

I now need to make the standoffs (2.4 mm thick), and drill a 1/8″ hole in a ¾” PVC male threaded plug and glue everything together (again without plugging the port).  Luckily I’m not planning to use this pressure gauge at high pressure, so the weakness of a superglue joint as a seal is not too important.

The breakout board with two layers of cable-tie material as standoffs. The cable tie material ends up being about 0.2mm too thick, but I think that may be good enough.

The standoffs should be 2.4 mm (including glue) to match the height of the flat part of the sensor. The thickness matching has to be pretty good, since we want the standoffs to relieve any stress on the glue joint for the sensor.  I think we could fill in a shortfall of 0.1–0.2mm with extra glue, but we don’t want the standoffs to be too thick, as that would require extra glue at the port, and the port only extends 1.3mm above the flat part of the sensor.

I going to try using material from plastic cable ties as the standoff, since cutting a uniform 2.4mm thick piece of plastic would challenge my shop skills. The cable tie material is easy to cut to length, but is it a suitable thickness?  At 1.3mm, one layer is definitely too thin, but 2 layers may be a bit too fat.

The finished sensor assembly with the sensor glued to a 3/4″ PVC plug. The double thickness of cable ties does not seem to have interfered with the gluing, and I see no trace of glue inside the hole for the port.

After the superglue has had a chance to fully harden, I’ll try some tests of the pressure sensor. I can think of several I want to try: leak test, static pressure calibration, vibration test without water, and vibration test with water.

The leak test is the simplest: just put the plug into a pipe and put some water in it (only a few inches of water pressure).  Does it leak?

The static pressure calibration will require providing power to the sensor.  At a nominal 5µV/V/mmHg, I should get 9.34 µV/V/in H2O.  With a 5V power supply, I should get 46.7µV/in H2O, which is too small to measure with my multimeter or oscilloscope, so I’ll need wire up an instrumentation amp to see anything.  I’ll want a gain of at least 1000.

Once I can detect pressure, I’ll want to see if shaking the sensor on my homemade shaker table produces a signal when there is no water present, and when there is water present.  If the sensor is too sensitive to vibration, I may need to rethink this lab—perhaps going back to the idea of recording the step-response from a bursting membrane.


  1. […] Pressure sensor assembly yesterday, I talked about putting together the pressure sensor circuit board. What I didn’t […]

    Pingback by Pressure sensor miswired « Gas station without pumps — 2012 October 18 @ 00:06 | Reply

  2. […] Pressure sensor assembly […]

    Pingback by Rethinking the pressure sensor lab « Gas station without pumps — 2012 October 23 @ 17:27 | Reply

  3. […] showed them the tiny medical-grade pressure sensor that I had considered using, but ultimately rejected as too fragile, switching to a sturdier sensor with an easier-to-assemble […]

    Pingback by Twenty-first day of circuits class | Gas station without pumps — 2013 February 28 @ 20:52 | Reply

  4. Hello, I’m trying to use one of the MPX2300 sensors, but am struggling with the op amp circuit design (I’m a lowly ME). Would you be willing the share/sell the op amp circuit design you used?

    Comment by Rob Beane — 2016 February 10 @ 05:56 | Reply

    • Because I have students design an amplifier for the pressure sensor as one of the lab exercises in my Applied Electronics class, I’m not willing to publish a solution. The key is to use an instrumentation amplifier, not just a single op amp. You can either buy an instrumentation amp (the INA126PA is the one we include in the parts kit for the lab) or you can make one out of 2 or 3 op amps. I’ve found that a 2-op-amp design, an op amp used as a unity-gain buffer for a voltage reference midway between the power rails, and another op amp for second-stage gain works well. The 2-op-amp instrumentation-amp design can be found on the INA126PA data sheet. The advantage of buying an instrumentation amp chip is that the resistors are very carefully matched, so common-mode rejection is excellent. The advantage of building your own out of op amps is that it is cheaper and you can get better op amps for your application (the INA126PA instrumentation amp does not have rail-to-rail outputs, for example, which limits its usefulness at low voltages).

      Comment by gasstationwithoutpumps — 2016 February 10 @ 08:14 | Reply

    • What is the application? Do you really need a tiny, disposable pressure sensor? That’s about the only advantage of the MPX2300, as the packaging is not protective of the chip and there are no integrated electronics. If you have the space for a larger, slightly more expensive (about $16) sensor, then something like the MPX4250 series which include the amplifier in the sensor may be a better bet for you, as you can hook them up directly to the A-D converter input of a microprocessor.

      Comment by gasstationwithoutpumps — 2016 February 10 @ 08:34 | Reply

      • This is great, and I’ll put some work into it. I want to measure the pressure within a medical device I’m working on, so the MPX2300’s medical compliance is a major sell. For the initial proof of concept, I may just use the MPX4250 like you mentioned.

        Thanks again – It’s very appreciated.

        Comment by Rob Beane — 2016 February 10 @ 12:28 | Reply

        • If you need something that will be in an IV line or other thing that needs to be sterile, then the ability to sterilize the MPX2300 and its compatibility with high-salt solutions like blood is indeed a selling point (as is disposability). If you’re looking at something like a blood pressure cuff (the application in the textbook), then the MPX2300 is not such a good idea—a bit too fragile. In either case, for quick prototyping, I’d go with an integrated-amplifier sensor, and switch to the final sensor with external electronics on the next iteration.

          Comment by gasstationwithoutpumps — 2016 February 10 @ 13:08 | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: