Comment by Teque5 on May 27, 2009 at 10:59pm

Is there a guide somewhere on how to hook up the pressure sensor? Does one tube go in front and one in back - how does it work?

Comment by Simon Kerr on May 28, 2009 at 1:23am

The pressure sensor is basically a MEMS version of a Pitot/Static system used on airliners, etc. The airspeed indicator on these has a diaphragm that receives Pitot pressure on one side of it, and Static pressure on the other, and converts the difference into a gauge reading.

My interpretation of the datasheet is that port P1 is the "pitot" port, and P2 is the "static" port, since during operation, P1>P2. You could use some thin tubing that opens to the front of the aircraft to feed P1, and another tube to the side of the aircraft to measure static pressure.

I haven't got one of these yet, but I work on the real deal on the commercial airliners, and a few issues come to mind that may or may not matter at this scale:

1. For accurate pitot readings of ram air, you need good clean air. Commercial jets have their pitot tubes pretty much outside the windows of both captain and 1st officer, to get ram air before the airframe and wings etc can disrupt that airflow. The pitot tubes also stick out far enough from the skin to get air from the "free stream flow" instead of the boundary layer air next to the skin. Propellers also have an effect, since they obviously disrupt the airflow, which is why the Spitfire had its pitot tube halfway along the wing, just below and forward of the leading edge. Perhaps that approach would be best for a front prop-driven UAV.

2. The tubing you use to transfer the air pressure from the environment to the sensor could play a part too. Avoid sharp turns in the tubing and definitely avoid kinks or squashing the tube, particularly as it meets the sensor, or you'll get all sorts of venturi effects which will stuff up the accuracy of your readings.

3. Consider the effect of sideslip or skidding, particularly if you're aiming for aerobatic manoeuvres, where the nose of the aircraft and therefore any pitot tubes and static ports aren't necessarily facing the oncoming flow. Crossfeeds from dual, oppositely placed pitot tubes and static ports should counter this effect, but may be overkill depending on your application. A simple T-junction in your tubing could work though.

4. Finally, the static port could be fed via a tube to the outside of the airframe, with the opening perpendicular to the airflow, or it may be sufficient to leave the tubing off the P2 port on the sensor if sensor is going to be at the static pressure inside the UAV anyway. It depends on how well you seal up your $32.50 Ardupilot Shield!

Hope this explanation and the links provided help peoples understanding of pitot/static systems, airspeed indicators, etc.

Simon.

Comment by Marty on May 28, 2009 at 1:29am

My first shield hasn't even arrived yet and it's already superseded

Comment by Bryan Cuervo on May 28, 2009 at 4:28am

EagleTree
sells a nice little Pitot tube for $10.00 that looks ideal for this settup. They also provide instructions on placement away from propwash, etc.

Moderator Comment by Chris Anderson on May 28, 2009 at 4:28am

Marty,

Just to be clear, it's the same shield--just $5 cheaper.

Moderator Comment by Chris Anderson on May 28, 2009 at 4:33am

I think P1 is the top port, so that's the one you connect to the Pitot tube. Double checking...

Moderator Comment by Chris Anderson on May 28, 2009 at 4:36am

I don't think you need a proper Pitot tube (twin tube). It should be enough to just connect the top nipple to a single aluminum tube in the airstream and leave the lower nipple unconnected to record ambient pressure in the cockpit.

Comment by Marty on May 28, 2009 at 6:37am

Looks as if the pressure sensor has changed Chris or is it just the angle ? I should have added added a :P to my previous post. It's the only downside to trying to stay at the front of development I'm afraid :)

Moderator Comment by Chris Anderson on May 28, 2009 at 6:41am

Marty, it's identical. No changes.

Comment by Marty on May 28, 2009 at 7:21am

No worries Chris thanks for the info. Well done Jordi

Comment

• ‹ Previous
• 1
• 2
• 3
• Next ›
• Page