24th MAY 2016
It's been a very long time since my last blog entry about my attempt to build a full size x-wing cockpit, and I've received a few messages asking me if the project has died.... In fact, no, it has come on quite a long way, its just been that my time has been incredibly tight over the last 2 years working on various films as well as other projects. Much of my spare time has been put towards building and programming for the x-wing, rather than blogging about it. However, all of the positive messages and interest I have received has encouraged me to get back on it...
Incidentally, as a side note, one of the jobs I did in 2014, was working on VFX for Disney's live action Cinderella for 6 months at Pinewood studios, the first "big film" I had worked on, which was exciting. Whilst I was there, security was insanely tight because of a certain film that was also being filmed at the studios, I'm sure most of you can guess what film. I didn't get to see that much, as the studio doors were usually shut, but the highlight was seeing the full size new style x-wing cockpit on a motion rig that was being filmed outdoors on the back lot. I noticed straight away that the dimensions were off, which gave a clue to the redesign. I won't get into what I thought about episode 7, but I don't mind the new x-wing design, although I think the original looks better overall.. Other cool things I saw at Pinewood was the top gun section of the Millennium Falcon, some moisture vaperators, a half covered up crashed tie fighter, and Harrison Fords air ambulance when he broke his leg....
Targeting Computer
Anyway, back to my cockpit...... A lot has happened since my last blog, so I need to think back over 2 years. Not long after my last post, I got the targeting computer working.
Original Targeting Computer
My Targeting Computer
I put quite a bit off effort into matching the original, even down to the slight trail on the image. For my first version, the image was being generated by a separate program that I wrote and sent to the screen using a Blackmagic video output card. The targeting data could be sent from the main game using a local network connection.
I've actually been working on the game (I mean simulator) for a few years now and its coming together. I'll do a separate blog post about it, as there's lots to show and talk about, but here's a cheeky preview:
My X Wing Game
Pedal Controls
I knew early on that I wanted full controls including a joystick, throttle and foot pedals. I already had a joystick with built in throttle and a twist for pedals, but always wanted the full experience using foot pedals. I looked into commercially available foot pedal game controllers, but wasn't too impressed with the options. Without going high end and spending a fortune, they all seem to be quite plasticy and not very solid. At the same time I noticed my sister had an exercise stepper machine lying around and It caught my eye. It was pretty solid and I realised that upside down, it could maybe work quite well as flight pedals. So my sister wasn't best please when she found out I had taken it all apart, but when I asked her if she had used it at any time in the last 5 years she said nothing, ha.
Stepper Machine...Or Flight Pedals?
In order to make it work, I needed to fit a potentiometer (or pot) to measure the movement. To read the pot and make a USB game device I used a Teensy 3.0 like I mentioned in my previous blog post. Originally I had quite a neat solution using a rotation pot fitted inside the stepper to read the motion of the bar that takes the weight of both pedals. In the end I decided that the accuracy of the data was too limited and jittery as there was only a few degrees of rotation being used to read the full range of movement. So I decided in the end to use a linear pot mounted externally to read the data, more ugly, but when correctly positioned, can read the full range of motion giving accurate clean data...Nice.
Pedals (Obviously still need painting)
Pedals + Feet
Linear potentiometer to read the position of the pedals.
Ugly but it works.
Joystick and Throttle
Now the main joystick for the simulator was at one point going to be a blog post all on its own. What with the time and effort put in, and the amount different things I tried. In the end, it was a mostly wasted effort as I took a different solution in the end.
I had decided that I was going to try and make my own joystick from scratch, as mounting a PC joystick into the simulator was going to look awful, what with the big bulky base they all have. There was a big bit of metal with curves in it that came off the stepper machine, and at the right angle, kind of looked like it could be the base of a joystick. So I drew up some plans and got to work.
Rough joystick design
It was going to be mounted to the base of the cockpit, which would be fine for forwards and backwards (pitch) motion, but would hit your legs for left and right (roll )motion. So in the end I went for a Spitfire style design, where by the whole stick goes forwards and backwards, but only the top part goes left and right.
Spitfire Control Stick
My Design
The base was fairly easy, I made a block out of delrin and used 2 pots as the bearings with which to mount the stick to the block. The block could then be bolted down. The center springs could be external so was fairly simple to mount, although still needed some experimenting to get the right tension.
Base section of joystick
The tricky bit that drove me mad was the top section with the left right (roll). I needed to measure the rotation with a sensor, plus also have a spring or several springs to center the stick. I got very close several times by making my own springs by bending rods of metal, but getting the right tension in the spring and solidly fixing it proved too problematic. In the end, I went for a more advanced solution, involving a geared down motor with a built in rotation encoder. This solved the problem in one as the sensor picked up the rotation and the motor itself could provide a centering force depending on the position of the stick. Again I used a teensy 3.0 to read the data as well as control the motor using an H-Bridge chip and external 12V power supply. The other neat thing is that the motor would have enabled force feedback on the stick. I got as far as experimenting with a slight jolt every time the guns were fired.
Original custom spring approach
Motor with encoder solution
The working joystick
I mounted the trigger at the back of the stick, also inspired by the Spitfire stick. There was also going to be some other controls on top.
The design of the stick was meant to represent the stick seen in A New Hope (hence the boxy block on top of the stick), but I wasn't trying to do a perfect replica. In the behind the scenes photos the stick looks quite impractical, and for me ergonomics and practicality are taking precedence over screen accuracy, especially as the stick is seen for under a second in the film, and I think not at all in the Empire and Jedi.
You can just about see the shape of the original joystick in this behind the scenes still from ANH.
Stick in use. Notice the control box bottom right where all the wires end up.
My stick actually worked ok, and I had many hours of flying time in flight simulators such as Cliffs of Dover, and Rise of Flight without the motor ever burning out (it got quite hot after extended use). However, in the end I made the tough decision to scrap the stick, and use a commercially available option after all.
I was looking into building a throttle, and was realising that the chances of making it look great were rather slim, as well as getting the right tension and smoothness. So I looked at the best options that I could buy and came across the Thrustmaster HOTAS Warthog, a replica of the A10 Warthog joystick and throttle. At £300 for joystick and throttle it's one of the most expensive consumer setups you can buy. In the end I couldn't resist due to the solid construction and sheer number of switches. I mainly bought it for the throttle unit, but I was also toying with the idea of swapping out my joystick also. The base of the stick unlike any other I have seen was fairly compact and is bolted to a heavy metal plate which can be removed. The stick is made of metal and the smoothness and accuracy made my mind up in the end.
Thrustmaster HOTAS Warthog
Throttle and Joystick
The stick is temporarily fixed with a lot of black electrical tape.
I roughly mounted the Thrustmaster stick to the original metal base of my joystick, and realised that I could use the original forward/backward motion as a way of moving the joystick out of the way when getting in and out of the cockpit. I was originally going to have a spring and locking mechanism to keep it in place, but I now think I'm actually going to try and automate the movement with a linear actuator which would be far cooler...
Side Control Buttons
One other input I have, in addition to the Pedals, Joystick, Throttle and Front Panel, is two button panels for the right hand side opposite the Throttle. I found these in a skip outside a film lab in London where I was doing some work. They would have been used for video switching between different inputs and would have been rack mounted. With a fair amount of trial and error I was able to work out the wiring of the buttons in order to make them individually light up as well as work as inputs. With a Teensy device in each box they both act as a separate game controller.
2 video switchers for side panel buttons. I think they look suitably 70s.
Anyway, hope you've enjoyed this update. This project is most certainly not dead!
More to come very soon....
