A month ago I saw a news article about the re-discovered remains of an experimental aircraft catapult that was unearthed just outside oxford. As I read I thought it would be an interesting combination of realms of simulation that we could easily model and set myself a challenge to create the system within Dymola.
Lots of catapult methods were developed and experimented with in the early days of aircraft catapults, including using weights, flywheels and (my favourite) rockets. Rocket assisted catapults were implemented during World War 2 to launch fighter aircraft off the bows of Royal Naval merchant (CAM) ships.
It was only after the end of World War 2 where pressure driven catapults became the standard design on aircraft carriers, still widely implemented today. But with the introduction of electromagnetic catapults, first on the US Navy Gerald R. Ford-class aircraft carriers, pressure driven catapults may become of a thing of the past.
One of the first things that I wanted to get right was the catapult mechanism. The article didn’t mention much about the specifications of the system, other than using a “2000 psi” “pneumatic ram”. So I took specifications from the much more widely deployed steam based mechanism, implemented on US Naval aircraft carriers. There is a lot more information about them, especially from the two following sources Thermodynamic analysis of the C-13-1 steam catapult for aircraft launching from an aircraft carrier and Steam Powered Catapults.
While there is enough information within those documents to develop a lot more detailed system, I just wanted to create the dynamics of the system. So I focused exclusively on the pressurised side of the system, without modelling the charging or support elements, which significantly simplified the system.
To that end, consisted of 4 different fluid components, the accumulator, supply pipe, valve and piston. The accumulator was only acting as a pressure vessel, with the appropriate volume of steam within. The pipe before the valve acted as a pressure loss dependent on the length of the pipe. The valve is a version of hydraulically actuated ball valve, with no flow until the valve starts to rotate. The piston is a simple expanding volume piston. The system is assumed to include no leaks and as there is little detail about the connections, no nozzles or orifices have been included.
Below is the diagram showing the system, including all the above components. Also included are the 1D translational components that are representing the shuttle. From top to bottom, shuttleStop is the spring damper that stops the shuttle at the end of the 82m long travel; shuttleMass is the mass of the shuttle, including the pistons and aircraft latch; shuttleHold is the mechanism to hold the shuttle in place before launch.
Figure1 : Launcher catapult fluid system with logic and 1D translational shuttle
Also included in the system is the launch sequence logic, which pressurises the piston one second before releasing the shuttleHold and allowing the shuttleMass and shuttlePosition output flange to move. The mass will continue to move until it hits the shuttleStop spring and damper that brings the shuttle to a stop at the end of its travel. When the shuttle reaches 0.25m before the shuttleStop the valve is closed and the holdPlane Boolean turns false to cause the release of the aircraft.
I may have taken a little bit of artistic licence with this bit…… The article didn’t state the aircraft intended for use in Oxford, but the image depicts an Avro Manchester. Now I don’t know about you, but it’s the Avro Lancaster that stuck in my mind over the Manchester, if not for the original Airfix box art alone.
Figure 2: The original artwork Airfix-72 scale Lancaster B1 kit, returning to land with 1 engine on fire and wing shot to bits
So that’s what I made…
Figure 3: Dymola Model of the Avro Lancaster built using the UAVDynamics library
The model was built using components from the UAVDynamics library, almost all were just re-parameterised versions of existing models. The only sub-systems that needed specialist creation was the twin tail model, which just required adding a second tail aerofoil model to an existing, conventional T-Tail design.
While principally designed to model electric vehicles, the modelling of ICE aircraft is not a difficult inclusion, but for this application I used speed actuators rather than torque or an engine model to represent the 4 Merlin V12 27-litre engines.
Once parameterised (and adding the visuals) some testing was required to tune the flight controller and a couple of the control elements to be able to stably control the bomber in non-assisted take off and flight.
To make sure the system worked as expected, initially I recreated the “Dead Load” experiment, which uses a wheeled cart with the appropriate weight instead of a plane. The easiest way to do that was to use one of the Suspensions vehicles with no suspension but with Pacejka Tyres and the mass bumped up to the same as the Lancaster.
Video 1: This resulted in a rather familiar looking sight, and reminded me of a classic Citroen advert and, of course, the loss of the first Stig in Top Gear.
Video 2: Classic, (rather eccentric) Citroen Advert
Video 3: The final moments of the original Stig on Top Gear
With that system working as expected, it was just a matter of swapping dead load for the Lancaster and launching it, and after telling the controller to throttle up and then pull up at the appropriate time.
One final bit of tuning involved adjusting the pressure of the tank to ensure that the Lancaster was leaving the launcher at an appropriate speed. This involved reducing the pressure from the dead launch as the Lancaster was producing a large amount of thrust itself. After that element was complete the result was very pleasing:
Video 4: Launch of the Lancaster using a steam powered catapult
Now I need to think of other things I want to launch…..
Written by: David Briant – Senior Project Engineer
Please get in touch if you have any questions or have got a topic in mind that you would like us to write about. You can submit your questions / topics via: Tech Blog Questions / Topic Suggestion.