Fullscreen HTML5Med Res (??MB)HTML4

RADIO CONTROL BY BOB YOUNG The philosophy of R/C transmitter programming; Pt.2 This month, we will look at some of the broader issues governing the design and programming of comput­er transmitters. Last month we covered some of the fundamental aspects in regard to model design and their influence on successful program­ ming. Perhaps I should point out that this series of articles is not intended to be a stepby-step programming guide. There are far too many different brands of transmitters, each using a dif­ferent programming technique, for that approach to be successful. Instead, I want to establish the fundamental principles upon which programming is based and show how an understanding of these principles can help simplify the programming process and improve safety. Transmitter design This is a fine example of a modern computer controlled 6-channel R/C transmitter. It has memory to cater for up to four different models and host of programming options. 74  Silicon Chip As noted last month, modern transmitter design has been driven largely by the requirements of the international class competitor allied with the need for mass production and market­ing. The smaller the number of models any one manufacturer can produce to capture the largest market share, the more efficient the operation. This development has crept up slowly and probably began with Phil Kraft when he introduced his Signature Series. This radio had settings for throw in different directions and dual rates. I recall one of the Australian Kraft factory team crashing on one occasion during an aerobatic contest. He hit the ground inverted at the bottom of an outside loop. When I questioned him about what caused the crash, he informed me that he had the dual rate switch set to low instead of high and the loop diameter was too great for the height available. Fig.1: this is the configuration for a “flaperon” wing, showing the direction of servo rotation to achieve (a) flaps or (b) ailerons. I had been flirting with dual rate at the time and after that I dropped it as I had also found that learning two complete sets of control responses detracted from the purely instinctive response so necessary for high level performance. I am not a great fan of gadgetry for this reason. As a manufacturer I have to play the game and provide these gadgets but personally I like simplicity and prefer to rely on my own physical dexterity. Soon after, Futaba introduced their “J” series with FM transmission and a few mixers for elevator to flaps and rudder coupling. Both these radios used potentiometer adjustments, no channel allocation and no model memory. Flying a different model meant readjusting the pots if the models were not correctly set up, a situation we dealt with last month. R/C system designers eventually found a better way and that was the computer encoder. Now models of all types could be flown, the sky literally being the limit. This has lead to the overly complex computer transmitter, designed to be all things to all people, which in many instances has so many features that it just simply overwhelms the beginner and sports flyer. Basic requirements Let’s look at what a modeller really needs from a transmit­ter. To begin, it is essential that you have a clear understand­ing of what your requirements are. At the most fundamental lev­ e l, this may involve deciding that the transmitter is to be used for cars, boats, aircraft or helicopters. This may involve choos­ing the ideal physical layout such as wheel or stick transmitter, or the ideal program configuration such as the helicopter specif­ic transmitters now available. This may seem pretty obvious but what is not so obvious is the next step. That is to decide what is the best system for your branch of the hobby. Fixed wing flyers, for example, fall into broad categories such as beginner, sport, glider, scale, aerobat­ ic, pylon, ducted fan, etc and each cate- gory places different demands on the R/C system. Sport flyers have the minimum requirements in regard to auxiliary features. Scale may call for a relatively large number of channels with few mixing features. F3B gliders place the most stringent demands on the R/C system in regards to complexity of programming. It is in trying to produce a radio that will cover all of these requirements that has lead the R/C manufacturers to produce the very complex transmitters we now see in the model shops and the computer makes it all possible. Manufacturers claim that the key to this flexibility is model memory. Some transmitters now offer up to one hundred model memories, a mind bending figure. In view of the fact that many modellers have taken off with the wrong program loaded, such a large number of memories certainly ups the odds in this area. Last month, we looked at model memory and decided that for modelOctober 1997  75 Fig.2: this is a glider in “crow” landing configuration, sometimes referred to as “butterfly” mode. Flaps are down and the ailerons are up but still providing aileron function. Elevator trim compensation is sometimes applied. lers flying fixed wing sport models, model memory presented more of a danger than an advantage. If the models are basically the same type and correctly set up, then model memory is a rela­tively unimportant feature. Having said that, there are several aspects of fixed wing aircraft operations whereby model memory may become very import­ant. Such is the case of a modeller who specialises in F3B (multi-task gliders) for example. F3B models use variable geometry with each configuration stored in a separate model memory. These memories are switched in flight so that with the flick of a single switch, the entire aircraft geometry may be reconfig­ ured. One F3B model may use as many as six or seven memories. Under these circumstances one hundred memories suddenly shrinks to about fifteen models. However, here we are talking about the most specialised, highest level competition flying that exists in this sport. The average club flyer has no need for anything remotely like the sort of R/C system called for in F3B. Futaba, for example, in their Super 7 transmitters originally had all three model types, powered fixed-wing aircraft, helicopters and sailplanes, however the sailplane features were lacking. They then released (1993) a sailplane specific system, the 7UGFS, which had to relinquish the helicopter features to make way for the complicated F3B program­ming. So the crux of the matter comes back to choosing the cor­rect system for your needs, remembering not to get too ambitious with your first radio. Since 1993, computers and memory chips have made enormous strides and the very latest transmitters on 76  Silicon Chip offer have covered all of the gaps, albeit at a mighty price. Choosing a system So what sort of radio should you choose? Can the potential dangers of model memory be minimised while still holding on to the advantages? I believe there is a way to have the best of both worlds but it involves thinking outside the square. So let us proceed with a more detailed examination. For example, take a modeller who regularly flies sport, aerobatic (F3A) and glider (F3B) models. This would be most unusual modeller I might add, for most modellers specialise in only one or two branches as a rule. For this particular examina­tion we will use the programming manual for the Futaba Super 7 system (7UAPS and 7UAFS), as written in good English by Don Edberg and published by Dynamic Modelling, Irvine CA. This is as excellent publication which not only gives the programming steps but also the aerodynamic theory behind why such steps are necessary. There are other manuals rewritten for other systems and if they are not available in your area, then as a last resort you should fall back to the factory manual. Unless that is, you have an American system such as the Ace Radio Micro­pro 8000 system which has an excellent factory manual. Before we start, it is necessary to look at some of the innovations that have crept into modern transmitters. First is channel allocation, which is the ability to assign each front panel control to a particular channel number in the data trans­ mission sequence. This is a very useful feature but should be used with the greatest of care. Not all transmitters have this function but when it is used it should be used with as much consistency as possible from model to model. If the wrong memory is loaded accidentally, coping with a reversed control is one thing but if ailerons became elevator, for example, then all hell would break loose. This becomes increasingly difficult as we move into the more complex programming systems which are made possible by another trend in model design and that is two independent servos for each control which are electronically but not mechanically cou­pled – see Fig.1. This trend has been accelerated by the increasing size of models, the need for some degree of fail-safe servos on these very large models, the falling cost of servos and finally the ability to mix two different functions into one control. The more obvious configurations such as elevons and V-tail have always called for two servos with mixing on each servo. However, such configurations as the airbraking system desig­nated CROW (both ailerons UP and flaps DOWN), ailerons mixed into flaps, and the ultra weird “ailevator” configuration where ailerons are mixed into elevators on a standard fixed wing model, all demand two independent servos for their functioning. So the very first thing we notice in the Futaba manual is the attempted consistency applied to channel allocation through­out the programming descriptions. In the first instance, the channel allocation given in the Super 7 manual for setting up a sports model is ch1 – aileron; ch2 – elevator; ch3 – throttle; and ch4 - rudder. This is the traditional Futaba channel allocation which is still used on their non-programmable transmitters. The F3A model will follow similar lines with perhaps ch5 allocated to retracts and ch6 allocated to flaps, if these are used. There would be no problem running these two types of models from the one 6-channel transmitter without model memory, using the techniques discussed last month. However, when we change to F3B mode the problems begin. Channel allocation suddenly becomes a very different matter. In the F3B model we are dealing with the multi-servo wing as a mandatory item. The F3B model The F3B model is a multi-task model which calls for a very high level of aerodynamic sophistication. Variable camber wings are a must for this type of model in order that the launch, speed, cruise, endurance and landing tasks are all carried out in the most efficient aerodynamic configuration. Thus, the ailerons and flaps are called upon to perform multiple roles, with the ailerons performing the functions of flaps, ailerons or speed brakes in the one flight, often with any two simultaneously engaged. Flaps likewise may be called upon to perform as flaps, reflexed trailing edges to increase speed or even ailerons in some models. In the CROW (landing) con­figura­tion, the ailerons are both moved UP to provide airbrakes (whilst still performing as ailerons) and the flaps are at maximum droop. Elevator trim is mixed in to compensate for the trim shift caused by the flaps and ailerons and coupled aileron/rudder may be engaged to compensate for the loss of aileron efficiency in the UP position – see Fig.2. To add to the programming confusion, glider wings may be two, three or four servo types, depending on the complexity of the design. Added to this are additional problems of complex mixing of elevators with flaps, flaps with elevators and rudder with ailerons. The F3B glider is the most complex program of all model types and I believe the prime driving force shaping devel­opment of the modern computer transmitter. Yet when I had completed the F3B module for the Mk.22 TX and I needed to run the final testing, I looked around for someone with a four servo wing and could not find one in an easily accessible location. The best I could find at short notice was a two-servo wing. In all of Syd- ney there is not a handful of these complex models yet they dominate transmitter development the world over. As I stated previously, in trying to cater for the handful, the transmitter designers have made life really tedious for the average flyer. The channel allocation for the F3B model called for in the Futaba UAFS/ PS system is ch1 – right aileron, connected to the aileron stick; ch7 – left aileron, not connected to any front panel control but slaved through an inverting mixer from ch1 to provide the equal and opposite drive signal. Ch3 is right flap and ch6 is left flap, both connected in parallel with ch3 con­nected to the throttle stick to provide flaps. Ch2 is elevator and ch4 is rudder. Ch5 is left unused. From the above it would appear at first glance that there is no possibility of flying sport and F3A models from this trans­mitter configuration without model memory, which is oddly enough quite wrong. We still have a throttle stick, aileron stick, elevator stick and rudder stick all in the correct locations. So long as all models are fitted with seven channel receiv­ers and absolute consistency is adhered to in regard to channel allocation and servo directions, there is still no danger of model memory causing a catastrophic result if the wrong memory is accidentally loaded. Even if the sport program is loaded for the F3B model, at least one half of each control will work in the correct sense and direction, although one half of the flaps working would certainly raise the adrenalin levels for a while. However, problems could arise if the channel allocation was changed to squeeze in a 6-channel receiver in one or more models in your fleet. So as a general rule, the larger the number of channels available in your system the easier and safer programming becomes. One final point: many manufacturers make a big deal about their trim memory function. This function stores the trim offsets for each model. Here is one function that you could well do without. Make sure that all servos are correctly neutralised and that all trims are in neutral for all models, when the final trimming of each model is complete. That is the only approach if you want to avoid a nasty surprise one day. SC The answers! to 260,000 questions, ALL in one book! The largest range of replacement semiconductors in the industry! Call now to get your new NTE cross reference book for just $25. Stewart Electronic Components P/L P.O. Box 281 Oakleigh 3166 phone (03)9543-3733 fax (03)9543-7238 P.C.B. Makers ! If you need: •  P.C.B. High Speed Drill •  P.C.B. Guillotine •  P.C.B. Material – Negative or Positive acting •  Light Box – Single or Double Sided – Large or Small •  Etch Tank – Bubble or Circulating – Large or Small •  U.V. Sensitive film for Negatives •  Electronic Components and •  •  Equipment for TAFEs, Colleges and Schools FREE ADVICE ON ANY OF OUR PRODUCTS FROM DEDICATED PEOPLE WITH HANDS-ON EXPERIENCE Prompt and Economical Delivery KALEX 40 Wallis Ave E. Ivanhoe 3079 Ph (03) 9497 3422 FAX (03) 9499 2381 •  ALL MAJOR CREDIT CARDS ACCEPTED October 1997  77