Linking Fly-by-wire systems
 

Hi everyone!
Im completely new to PPRuNe, so i apologize in advanced if im in the wrong area.

Basically, as a college project i chose to link the sidesticks in the A320 to make it more of a safety feature. Just like how the 777 yoke moves simultaneously, my creation was to do the same with the airbus. I would like to ask any fellow airbus engineers out there or anyone who has the knowlegde on how i can come across making the side sticks move at the same time? What tools or motors etc i could use to make this work? I know it sounds ridiculous, however i need to get it done by December. The lecturer wants to know rough costs, what i would need to make it happen, and if it will be realistic or not. Any help would greatly be appreciated! :)

Not realistic!

Do you know any sort of creations i can add to any sort of aircraft for my college project? I initially planned to insert force feedback motors within the sidestick to link the two together. I know none of this is realistic, however i need all the help possible. Unless anyone can think of a different idea which i could do? Thanks

Since this is a school project you have a major advantage of not having to worry about real world issues such as certification which would make this a very unrealistic endeavor indeed.

BTW: I am not a pilot or expert on anything to do with A320 so take the following with appropriate grains of salt.
I see two possibilities, in either case a Raspberry PI or Arduino is a good choice for controller and logic.

A: If you really want to have the sidesticks move together look into Radio Controlled (RC) components, there is a lot of available servos and such for a moderate price.

B: I recall though that the sidesticks are more of a 'force' rather than position device so tracking the two may not help.

You could consider adding a tactile (aka haptic) "disagree" warning that would be much more likely to be noticed during a stressful situation than "yet another alarm" either audible or visual. See AF447 thread/report for more details on this.

The disagree logic is already implemented so the project would be reduced to determining the best stimulus and activator to use. Possibly indicating the direction of mismatch could be useful.

Good luck and don't forget to list your professor's name first on any resulting paper or press release :)

How about a modification to fit an ‘Engine Cowl Unlocked’ warning system that could also trigger a ‘Take-off Configuration Warning’ on existing aircraft.

Relevant info here:
https://www.flightglobal.com/news/ar...-a320n-415047/

The "relevant info"s are from july 2015!

It is installed on all A320NEO with PW1100G engines.
If more than 1 of the 3 fan cowl door latches are open you get a ENG1(2) FAN COWL NOT CLSD warning message on the upper ECAM as soon as you turn the ENG/MODE selector switch to the CRANK or IGN/START position.

That's not a bad shout! I know a fair amount of the A320 as my brother flies them. However with him being away a lot, I don't really get time to ask him question's such as these. The Pi is pretty much a small computer right? If so, then how can this help with linking the systems? Sorry for the delayed response! Had a lot of college work to sort out.

I suggested the Pi or Arduino since they are low cost and have a lot of available interface boards of various types available.
Either should be more than capable, which one probably comes down to what is in use at your place.

Although possible to electronically link two controls using old school analog technology using a computer for the logic greatly simplifies thing, especially when adding features and dealing with unforeseen issues.

1: Thought experiment: Left side stick moves, your system moves the right side stick, how does the system know who moved it and whether to in turn move the left one. If not properly tuned and designed very easy to get positive feedback, aka an unwanted oscillator.

2: From my limited knowledge at least some linked yokes are designed with a break away feature to handle potential jams. What is the equivalent in your system?
A lot easier to implement in software than pure HW.

Final question: Is the class goal primarily to demonstrate adding a feature to an existing system or is it to improve safety, either one is totally valid but the approach is different for each.

Improving safety will require finding an optimal solution, not just replicating an existing system.
This will require understanding the reasons side sticks were introduced and how best to improve situational awareness in that context.

MurphyWasRight,


Your an absolute god! That's helped me out so much! I know this project is quick wacky, however you've given me the ideas to continue the initial idea. One quick question, would I link some sort of servo to the side-stick along with the Pi/ Arduino in order for it to work?

Basic requirements would be to have a position (or force) sensor on each side stick and along with an actuator each connected to the controller.

The sensor is already in place as part of existing device so you would need to add the actuator.
As mentioned earlier research RC (model aircraft) servos as a possible actuators since they are a reasonable priced and the right size.

Do you have access to a pair of real Airbus side stick for this project?
If not ask your brother to 'borrow' a couple, one at a time though :E

Not forgetting of course that airbus sidesticks use summing, 20deg on left stick 20deg input to flt controls, right stick moves 20deg then another 20 deg input to flt controls, total 40deg input!

Yeah that's a good idea! However my brother only flies them, doesn't engineer on them, :( so I guess finding a pair or at least a side stick might be a little hard for me (unless I buy a second hand one which will cost an arm and a leg). So basically connect the Pi to the servo, which is then connected to the force sensors?

Without a pair of real side sticks I would look into RC controllers that typically have a pair of joysticks for a demo.

Again this partly depends on the point of the assignment, to demonstrate ability to modify HW or improve safety.

Yo aer correct on the basic architecture, not sure of your constraints but might be worth finding a partner who has embedded system and/or servo control loop knowledge.

Thanks for all your help MurphyWasRight. I don't think I would have gotten far without you in this project! How about you fly down from the US and pretend to be a student in my class? Then we can work together ��. None the less I believe this servo and pi theory will be a success. Thanks again! :)

Simplicity is best.

Nothing prevents you from mechanically linking side stick controllers to translate control feedback from one side to another.

You can still have your Airbus style resolvers/input potentiometers on each stick, with the added redundancy that if the electrics in one side stick assy fails, it can still be flown normal, as the inputs are mechanically translated to the other side.

If you're set on approaching this from a purely computing viewpoint, I think your best bet would be to grab a pair of PC force-feedback joysticks and write some custom code to translate inputs from one into feedback into another (there's lots of Open Source force-feedback APIs out there).

However, such an approach would be miles away from the redundancy, accuracy, and practically required for serious use in aerospace applications.

Also, Google up some images of a complete Airbus side stick assembly - it's very much a mechanical marvel just as much as it is an electronic one.

Interesting thought and depending on the nature of the class assignment might be easiest path. Would still be good to demonstrate some form of breakaway or override as is done for yokes.

Sometimes the simple way is best.

Reminded me of a hack many years ago for a high school play where the requirement was to release a leaf the middle of Charlie Browns soliloquy to said brave last leaf.

I came up with a string and clothespin, the other techie used a solenoid to release the leaf.
In the test to decide which to use mine worked silently and well, the solenoid made a terrible 'clack/buzz' noise.

Engineering truths: F=MA and you can't push a string (unless it is wet and frozen,in engineering there is always an out :)

Sorry for the late reply guys! - Yeah that's true! Simplicity can be the solution, it's never hurt. I was initially thinking about linking them with some sort of Rob behind where the motors are. Then feed it behind the MIP and down to the second side stick.