It seems like only last weekend that Pat Mackenzie and I sent aloft a small electric powered foam model attached to a set of strings and sent the modelling world into turmoil over the use of processors aboard those globally significant, universally critical and politically sensitive aerial devices known as Control Line Aerobatic models.
As a famous Canadian once said “we do these things not because they are easy but because they are hard”. What?? Stop yelling!,.. he wasn’t Canadian???... I thought he renounced his citizenship so he could run for Prime Minister?
So while the ner-do- wells immediately took up the chant to burn these (me in particular) heretics at the stake and make promises to each other to fight to the last drop each other’s blood, we quietly kept flying and (no one thought ask us about this) crashing. Yes, we hit the gym floor many many times and raised more questions about control systems than we answered. (this was surely a system to be in fear of).
Since most models occasionally get a little slack in the lines ours would be no different. But “regular” models have essentially free moving control systems and things like gravity, inertia and centripetal force tend to move the control surfaces when not under the control of the lines such that the model may at some point move to the outside of the circle restoring control. Here was lesson number one. The control surfaces in a fly-by-wire model are not free moving but rather held in position by the servo (small servos are ridiculously powerful). If our model got loose on the lines there was nothing to move the bell crank and since the servos would hold the last commanded position till death do us part you would find yourself in a loop that would not let go. Here comes the gym floor. (good thing we were only using foam models)
Part of the learning curve was to guess or approximate how control of flat plate models weighing four ounces would transfer to full size aerobatic models. Would there be enough “feel” to permit accurate control inputs? Would the system be fast enough to keep up with fairly rapid hand movements? Could servos be found that were small enough, powerful enough and yet light enough? We built several models and made changes to the code that controls the model. We still had crashes but we began to understand why things did what they did and how to avoid the really bad stuff.
The biggest problem was that there would be no way to really know how this system would behave with a heavier and faster model. My feeling was that the greater mass and higher speed would make life easier.
Sometimes life has a way getting in your way when all you want to do is play with your toys. Work was hard to come by and then my own government tried to force me out of business for its own political gain. Models cost money. Complex models cost more and consume more resources. I had made the decision the day we first flew in the gym that I would take a fly-by-wire model to the W/C’s and it was now time to get down to it.
I really did not have the money to go or to build a new model. Then life intervened and I got a small economic boost which let me commit to going. Oh ya, I still need a model. Just any model would not work. It had to be full take apart, light, have room enough to house the components, be easy on the purse strings and be able to be built and tested in a couple of months. In retrospect I should have built from scratch but being essentially cheap I decided to use an unfinished SkyWriter fuselage, a previously flown SkyWriter elevator and stabilizer and a super light one piece Saturn wing I had built but not used.
If all you wanted to do was to build this up as a non take apart, IC model you would have been able to proceed fairly quickly. I needed to gut the nose of the fuselage, rebuild a completely new structure, make a one piece wing a three piece wing (center section and two panels) and find a way to align a fully assembled tail with new take apart hardware. Then we would have to find a way to mount the servos, run the electrical connections (you would not believe the number of wires!), mount the motor, cool the whole thing, make a take apart cowl and get some sleep.
We were leaving in just over a week. There was no paint on the model. The tail was not mounted, the servos were not installed, the control electronics had not been completed let alone tested. I mounted the stab and checked five ways from Sunday that it was straight. It took me about five hours. It was 11:00 p.m. I was looking at the front of the model in the alignment jig. The incidence angle looked too big. I must be right because I do not have time to do it over. I did it over twice more before I was happy with it.
Thankfully the Saturn wing came from a lost foam jig and I was able to use the jig to maintain alignment while I removed all center section sheeting, the regular bell crank, installed the new take apart hardware and the fly-by-wire bell crank, replaced the sheeting and cut the wing into three pieces. Did I mention the bell crank? The bell crank itself is complex, the need to mount the Hal Effect sensor accurately would almost prove our undoing. The sensor is mounted to the tip of a pair of wires with heat shrink to protect everything. It had to mounted such that when the bell crank was in the level position the sensor was in the middle of the magnetic field created by the two magnets on the square piece of the bell crank. This was not so easy to do as even a small movement would bias the controls. After several hours it was safely secured with GOOP.
Wing covered, servos mounted two days prior to blast off. Color paint to be sprayed tonight on my balcony. Keep it simple and don’t look back. The control electronics are almost done. Its Wednesday night. We leave for Bulgaria Friday at 7:00p.m. Box, what box? The one I need to pack the model in. Fortunately I have one and just need to rearrange the dividers. Time to hook up the servos to the control surfaces. Model not yet flown. Controls not yet moved.
Over to Pat’s house Friday morning to collect the control electronics and test the system. All hooked up and we have nowhere near enough movement on the elevator. It was biased way up. More tests. 12:00 noon. We decide the sensor has not been installed properly. I needs to be done over. That’s it I’m done for. Not enough time and too much money at stake. If I don’t go I’ll save about a thousand dollars. I tell Pat that I can’t go and go to my car. I call my wife to tell her the news. I’m basically crying.
My wife tells me to try to find a way to solve the problem. She asks why we can’t move the magnet block. I have looked at the block and determined that we would destroy the wing center section if we tried to use tools to free it (it was glued). I think some more and decide that we have come too far to quit and go back into Pat’s shop to see if we can move the block. Then out of the blue of the western sky comes..... a wrench that just happens to be small enough to fit the block exactly and another set of channel locks that also fit. By compressing them towards each other the block should break loose and we can reset the mid point. SNAP!! (oh how I hate that sound) The block let go. Nothing was harmed. 1:00 p.m. Pat has not yet finished packing. We reset the block and fix it in place. The system is powered up and.....EVERYTHING WORKS!!!!!
Rush home, pack, call the limo get to airport and check in. Almost miss the connecting flight at Heathrow, last to board. It’s a bright RED box. Where is my box? “Your box did not make the flight but will be delivered tomorrow”. You can pick it up here. (hour and a half drive) I needed to have the model so we could install the control system electronics and get some much needed practice. Practice time at a World Champs disappears exponentially as people arrive from around the world. A spectator for Friday I drive to Sofia on Saturday to pick up the model. Model safely on board but am asked to wait to give a ride to another modeler who is awaiting the arrival of his girlfriend in twenty minutes. We leave two hours later. Day wasted.
Everything spread out over my bead I start to assemble the model. Wires, wires, connectors, fish lines pre-installed in model. Only one chance to get the wiring in place. Most of the system is installed by Sunday night. Completed early Monday morning (Technical inspection at 1:00 p.m.) After some discussion we pass inspection and I head to the practice field. Motor fires up and we get a needle setting (my timer processor has a “needle”) Off we go.
The first flight of a full F2B fly-by-wire stunter. Not bad, need to adjust some trims. Plug in the programming box and dial in some expo and some down elevator. Second flight ( I realize I will have to go from zero to 100 in just a couple more flights.) I’m fairly confident. (This is not my first rodeo) Some single rounds, single squares (trim not quite right), inverted, (need some more down trim) land. More trimming (I am thinking that I just might pull this off as we are progressing fairly well).
Pat and Ivan were across the field flying combat. Pat said it did not look right from the first loop. I took off and decided it was now or never to tackle the complete pattern. Too soon??? Probably but I would not be able to get practice time so easily the next day. I left out the wing over just to be safe and went straight to the round loops. First loop was not precise. I had problems getting the bottom in the right place. Second loop was bigger and came lower than I thought it should. The third loop started ok but as I got to the backside I was no longer in control. I was giving full up and the model was not going to miss the ground. BLAMMMMM!!!!!. The model bellied into the ground ripping the nose to pieces but leaving the wings and tail mostly untouched.
I could only surmise that we (I) had not set something up correctly or simply had a brain fart and lost control. For most of the last year the model sat in its box untouched since arriving home. I beat myself up pretty good thinking I had missed something. A few weeks ago I decided to strip the plane of its hardware and see if the wings and tail were able to be reused. I was removing the elevator servo and noticed that the horn was “lose” but still firmly bolted to the servo. Upon further examination it was found that one of the gears had stripped. It was missing two cogs. Enough to account for about twenty degrees of movement. It now began to make sense – None of the manoeuvres had seemed very sharp except the first couple. The gear was failing with each flight.
I was a spectator for the balance of the 2012 W/C’s.
I WILL be back.
Kim.
Topic: Fly-By-Wire the story unfolds (Read 1132 times)