Gas Thruster RCS Drone - Update
Added 2019-06-02 11:23:12 +0000 UTCAs you can probably tell from the title, I've been working on a drone which is controlled using compressed air!
It's main source of lift is produced by the two propellers in the central blue duct, which only provides yaw control. The pitch and roll are controlled by exhausting compressed air from 3 nozzles positioned at the tips of each arm.
I was going to make an update video of the setup, but as I haven't tested it yet, I thought I'd leave it till the final video, which I'm aiming to finish this week (still lots to do!).
I've spoken with Joe Barnard (BPS.Space) regarding his recent RCS endeavours (His video: https://youtu.be/10Pdl0Omet8 ) and he's given me a lot of information regarding the valve triggering control system.
I'm aiming for some test flights (hopefully not crashes) over the next day or so and have the video ready for next Friday (or Wednesday/Thursday for early access).
Apologies for the lack of information as I'm rushing to get this tested with the shockingly ok weather here in the UK. Let me know if you have any questions!
Comments
Critical pressure ratio for air is 1.89. So 60 psi gauge = 413 kpa gauge = 514 kpa abs. 514/101 >> 1.89 so you should be able to use a de Laval nozzle successfully. The problem is that maximum allowable area of the choke point is determined by the mass flow rate provided by the rest of your plumbing (likely the valves are the culprit here). If you don't achieve sonically choked flow in the throat, then increasing the cross sectional area of the nozzle will decelerate the flow. At 60 PSI gauge pressure, and room temperature, the maximum throat area, in meters squared is Q / 1210 where Q is the mass flow rate in kg/s and 1210 has units of s / (kg m m). This was computed from the red equation here: https://www.grc.nasa.gov/www/k-12/airplane/mflchk.html
Zaak Beekman
2019-06-07 20:17:02 +0000 UTCExactly, I've setup the flight controller to output pitch and roll values, then I've done the 120 degree mixing and valve triggering in an arduino. They're are off the shelf valves, the same cheap valves Joe is using from amazon. I probably should have looked for lighter valves, but these are so cheap and can actuate in about 10ms.
Tom Stanton
2019-06-03 12:34:37 +0000 UTCAre you using the thrusters in a configuration that mimics a tricopter? Also, are you using your own valves or something off-the-shelf? I experimented a few years ago with valves based on pneumatic landing gear actuator valves for R/C planes, and those were around 70g with a servo.
U.S. Water Rockets
2019-06-03 12:20:44 +0000 UTCThe main reason for 3 nozzles is to save weight, each vavle is 120g! Since there's no torque like on a drone, three points of thrust should be fine. Also with 4 nozzles, theres still the possibility that one nozzle has to produce all the correction thrust, if it tilts on the diagonal axis between pitch and roll.
Tom Stanton
2019-06-03 05:40:32 +0000 UTCWouldn't the setup be easier if you had 4 nozzles?
U.S. Water Rockets
2019-06-03 01:30:59 +0000 UTCThanks for the update.
Doeke Zanstra
2019-06-02 17:00:59 +0000 UTC
