The simulator is now complete!
The 3D printing is complete, so we installed all the electronics and tightened the last cable ties before putting the glider’s seats and flooring back in place. Bolting the flooring and seats back onto the glider’s frame really gave us a sense of accomplishment: we had always said that this would be the last step before our first proper flight.
We put the We calibrated tested the rear simulator. After some troubleshooting of 1 of the momentary buttons on the control panel, everything was ready for a first flight.
We put the 2 parachute harnesses donated to the unit by a gliding club member (thank you very much Adrian!) and took our seats.
There is something inherently wrong when 2 pilots are flying different flights yet sit in the same aircraft… (and fly without a canopy).
We still have some cosmetic work to do when face-to-face scouting is allowed again but we have reached our goal: Endeavour ESU’s simulator is really complete and working now.
Thank you to our 3 sponsors for the opportunities you gave us:
The red PLA has arrived! It’s not quite the same colour as the first roll we used but it’s close enough. The 3D printer is going to be busy for a few days!
The big day.
This morning we made some final adjustments to the wire ropes and cabling before checking all cables and tightening the remaining cable ties.
When all was in place, we connected everything together and calibrated our software, running our first test flight.
After a bit more calibration work and software tuning, we were ready for our first flight using the front simulator (G-DDYR-1):
SUCCESS!!!! We had an 18 minutes first flight, exercising all controls successfully. From winch launch to soaring to a safe landing, all worked well.
Once the red PLA we ordered finally arrives, we’ll be printing the display panels for the rear simulator.
Cabling time! Very early on through our design process, we decided to use Cat5e or Cat6 network cables for the long runs required to connect the sensors, buttons, switches and displays to their respective controller enclosures. Now that these are all in place, it was time to install and tie cables securely in the glider after crimping the RJ45 plugs at both ends of the cables (and testing them before installation).
We deliberately did not tighten our cable ties initially to allow us to reposition cables if needed. Unfortunately, we weren’t able to use re-usabel (velcro) cable ties everywhere but we did where we could to reduce the amount of non-recyclable materials we used.
The last remaining “big item” on our list was to install the LEDs in place to provide lighting inside the glider. We identified that the initial plan to use a contiguous 7m string of LEDs wasn’t going to work. So we looked at options and decided to split the string into 4: a long, 5m (150 LEDs) strip running along both sides of the glider, a 1m strip along the top of the wing box and 2 shorter, 50cm long strips along either side of the wing box.
We made the modifications to the lighting controller, adding 3 new connectors, and to the LED string. We installed 5V DC cables and data wires for the 4 LED strings before gluing the LED strips in place.
We added a connector to the cable of the rear simulator for the LSM6DS3 sensor.
We started tuning our software and modified the lighting controller software to allow for the 3 new LED strings.
We made final adjustments to the enclosures before screwing their lids and tightening cables ties to secure the cables in place.
The time to install the potentiometers in the simulator and to connect them to the rudder and trim cables as well as the air brake tubes has come!
We spent a lot of time looking at various designs before deciding on our design. This also required a fair bit of adjustment, creating longer or larger parts as needed. This demonstrated the power of using 3D printed parts for both prototyping and final assembly.
We installed and tweaked the trim potentiometers.
The air brakes potentiometers And the rudder potentiometers:
We also greased, re-routed and adjusted the rudder cables a final time to reduce friction against the aircraft frame. We made some small adjustments to the stops on the front air brakes tube and the trim cables.
We assembled all our components together and realised that we had a problem with our 6 dimensions of freedom sensor (6DOF) that we had seen once or twice before but not paid enough attention to. The sensor had a habit of locking up randomly: sometimes after a few seconds, sometimes after a few minutes. There was no obvious pattern or triggering event. The sensor was simply preventing us from controlling the glider for no obvious reason. We spent several hours investigating. We tried several MPU6050 sensors, different libraries and even used an I2C protocol analyser to try and identify a problem before deciding that we needed to find an alternative solution. Fortunately, we had access to a different 6DOF sensor (LSM6DS3 breakout boards) that we could easily adapt our boards for. Modifying the code was harder as we had to abandon a lot of our earlier code. We decided to base our readings mostly based on the gyroscope rather than the accelerometer and wrote code to convert angles into a 10 bit value that could be passed reliably to the simulator software.
After resolving these 6DOF sensor issues, we gave the airframe a good vacuum cleaning and got on we more angle grinding! We finished removing parts of the frame that protruded or would prevent us from installing plexiglass panels to close holes in the sides.
We created the bottom and top rear shelves, covering the top shelf with insulating cork before securing the power supplies and enclosures to shelf. We also installed the bolts to secure the rear enclosures to the bottom shelf.
We fixed back panel to frame and adjusted rear air brake spring before adjusting the trim cables and rear pivoting plates.
We spend a lot of time finding the optimal place for the release cable sensors. We had initially thought of placing the sensors in the nose of the glider, attached to the remaining release cable, but decided on a different solution today. When we took over the airframe, the nose hook (aerotow hook) of the aircraft was still there but the belly hook (winch hook) had been removed and, presumably, sold or re-used elsewhere. The cable connecting the nose and belly hooks was lying on the floor of the glider. When cleaning the airframe and removing unnecessary cables and fittings, we had decided to leave this connecting cable “just in case” it would be useful later. This proved to be a very wise decision and allowed us to clean and re-use the cable for the release sensors. We connected the connecting cable to a spring attached to the front plate where the front springs holding the stick are attached (albeit facing towards the nose of the airframe). This allowed us to put a little bit of tension in the release cable and provided a convenient place to attach a magnet and the 2 release sensors. Once the front floor of the glider is put back in place, the cable, magnet and sensors will disappear from view and should be well protected.
We installed the wiring on the top rear shelf: black cables for 230V AC and white cables for 5V DC. We will need to add the cabling for the LEDs when these are fitted, most likely fitting a connection box behind the lighting controller enclosure.
We started installing the green trim cables and the cable supports we designed. The cables were easier to work with than the yellow rudder cable, which was a welcome relief.
We had initially planned on using a custom made steel plate pivoting along an axis at the back of the simulator. Unfortunately, the lockdown was imposed as we were discussing the specification with a company who had volunteered to help us. Rather than wait for the lifting of the lockdown, we decided to create temporary plates from PLA using a high fill percentage.
We created a wooden support for the front simulator control panel
and we created the wooden support for the rear simulator control panel.
We cleaned remains of the tubes we cut when we separated the tail of the glider.
We installed new, blue, purpose-made guides in the swivel clips for the front air brake tube to allow the tube to move more freely and effortlessly. Our previous attempt at creating a connecting piece didn’t quite work out when it broke inside the tubes. We re-printed the connecting piece in PLA using a 100% fill rate. We tried very hard to break it but failed, so this seems to have solved our problem.
We also fitted a spring to the rear air brakes lever to force the air brakes closed by default.
We installed the connecting piece between the old and new front air brake tubes. Together with the guides in the swivel clips, this made the front air brake slide correctly and smoothly along the desired path.
We created the wooden back panel that will host the control panels for the lighting and both simulators.
We soldered wires on all the buttons, switches, potentiometers and LCD screens for the control panels (front and rear), ready to fit them to the panels.
We continued our work on the rudder cables today, trying to find the optimal route and placement for the cables. We made slow progress trying to find a solution to the positioning of the pulleys. The cable draws the pulleys towards the centre and away from the sides, thereby causing the cables to rub against the frame’s tube. The cables are already showing initial signs of wear.
We worked on the rudder cables today. We re-routed them and installed 3D printed cable supports to bring the front rudder cables below the top of the rear rudder pedals.
COVID-19 Still running with a reduced team, unfortunately.
With both stick spring anchor rings printed, we started by fitting the rings and attaching the springs. After spending some time readjusting the front rudder glider floor anchor points, both sticks were setup and ready for use!
We finished the afternoon by checking the attachements for our 6DOF sensors.
Note: this photo is more recent but it shows the rear stick ring in place, connected to its 4 springs.
COVID-19… Face-to-face scouting activities have been suspended for the foreseeable future, like many other activities (and all schools). We’ll be running with a very reduced team for a while.
Before the country went into lockdown, we purchased enough wood to create the external supports for the simulator: some form of cradle or braces to keep it upright and horizontal while in use or storage.
So today, we set about constructing 2 supports: 1 at the front of the glider, between the 2 seats, and 1 at the back.
We also took another look at the rudder cables and decided that we needed to reroute and swap the existing cables. There was a high risk that the cables from the front rudder would catch on the rear rudder pedals. In addition, the pulleys needed to be pulled closer to the sides of the glider to reduce the friction on the cables.
We had a good day on the airfield on March 7th and finally managed to get some flights. What a nice change!
We reconvened on March 15th after our leaders attended their 1st aid refresher course.
We installed the rear rudder pedal springs (there is no need for new springs on the front rudder pedals as the original springs are sufficient). After tightening the rudder cables, we proved that the springs we had selected were just right. Success!
We also created the anchor points on the floor of the glider for the stick springs, ensuring that the springs were set at 90 degrees from each other and equidistant from the centre of the stick. We attached the springs to their anchor points on the floor of the glider. We also agreed on the design for the anchor points on the stick. Once these are modelled and printed we’ll be able to test our springs.