The tilt rotor story began when one fine day Joe Saroli walked in with a book titled ‘VTOL MILITARY RESEARCH AIRCRAFT’ by Mike Rogers. The cover of this book portrayed the XV-15 tilt rotor in various phases of flight. Joe always likes unusual aircraft and commented ‘let us build one of these with a couple of 10’s’ (10’s meaning 1.6 cc engines). My reaction was that 10’s would not carry the required load and that with 2 engines the weight would be mostly concentrated at the end of the wings. The other point I made was that with such an aircraft, the speeds of the engines would have to be very closely matched, an almost impossible task without a cross coupling shaft.
The idea of the tilt rotor played on my mind and I started giving it a bit more serious thought. After some preliminary calculations, I came to the conclusion that the minimum power requirement would be a ‘60’ (10 cc) engine. The engine would have to be in the middle of the wing with shafts running to the end of the wing, where power to drive the rotors would be geared down with a bevelled gear box. Over the next few weeks the write up in Mike Rogers’ book was studied to see if the idea was really feasible. It was concluded that the idea was feasible, but it would cover a lot of ground which had not been covered in our modelling experience. Joe was at this time an airplane flyer, and I had flown both planes and helicopters. It was decided to have a go at making a model of the V-22, a military version of the tilt rotor. More over an Airfix model of the V-22 could be purchased. Various ideas were discussed with a few of our flying friends and we reached the following conclusions:
Since the success of the project (by this time it had become one) depended so much on the control system, I set about designing an analog mixer for the rotors. After months of work, I managed to get a mixer working. It worked fine as far as the pulse widths for the servos was concerned (observed on a scope), but went absolutely berserk when the servos were connected. In short it suffered from noise. At this time the ‘PIC’ micro-controller appeared in some of the electronic magazines. A design of a servo mixer appeared in RCM&E. My son, Sheil, was at university reading electronics. I abandoned the analog mixer and Sheil was assigned the task of developing the mixer for the project using the PIC micro-controller.
In the mean time design work on the project made slow but steady progress. There were no doubts about the helicopter part of the aircraft, as it would just be another ‘Chinook’, but there was a bit of uncertainty about the aerodynamic flight. At this stage it was decided to build a half scale model of the final model with 2 contra rotating propellers driven from a single ‘30’ (5 cc) engine. Such an aircraft was built and flew very well. Some of our flying friends were a bit sceptical about it flying at all, as it looked very big for a ‘30’ sized airplane. The aircraft crashed on a windy day. It was later analysed that problem was lack of elevator authority due to the swept forward wing. Overall it was thought that the test proved that airplane mode flight should not really be a problem. The lesson learned from the crash was that for a model with swept forward wings, one needs large control surfaces!
During the summer of 1995 the Paris airshow came up. It almost coincided with my birthday, and as a birthday present my son bought a channel tunnel ticket to cross the channel and visit the Paris airshow. The reason for the visit to the Paris airshow? They were displaying the V-22 Osprey at the Paris airshow. (the first time it was being flown outside America). I wouldn’t miss the opportunity, it would be my first time in the tunnel and of course the first time to see the V-22. We set of for Paris at 5 am. The first disappointment was on arriving at the airshow. My opinion was that Sandown is better organised. The second disappointment was that they were not flying the V-22. But all was not lost as there was some stunning flying to witness and we could at least see the V-22 and the XV-15 in flesh & blood. This was also a turning point in our project. We were very fortunate to have met Colby O. Nicks at the Bell tilt rotor stand. Colby was the research engineer on the XV-15 program. He seemed genuinely interested in what we were doing and spent over an hour of his valuable time to discuss the various aspects of the tilt rotor. He listened very patiently to what we had to say and eventually he commented that most of what we had said was applicable to the tilt rotor. One important aspect, he pointed out, where our thinking was flawed, was the fact that we were trying to balance the rotors around the tilt axis. He told us that the rotors should not be balanced around the tilt axis, as the C of G needs to move forward in the airplane mode. Up until now the idea had been to build a model of the V-22. Colby told me he was in a position to release information on the XV-15, but he could not do so for the V-22 as it was a military aircraft. Since the XV-15 was a nicer looking aircraft, and had slightly better proportions for modelling, we were delighted at his offer to send information on the XV-15. We returned from the Paris airshow at 2 am the next day - very tired but at the same time very happy having met Colby O. Nicks.
While we were waiting for information to arrive from Colby, it was decided to concentrate our efforts on aspects of the project which were not dependant on having this information. One of the major tasks was to develop the mixer for the rotor control. Considerable effort was made by Sheil, to not only implement the mixing, but to enhance the mixer by having a degree of error checking on the radio signal to provide a fail-safe system. It did lead to a lot of heated discussion (not argument) between father and son, but in the end it led to a system without which the development of the tilt rotor would not have been successful. The final design incorporates a dual receiver / transmitter fail safe system which goes a long way to guarding against radio interference and failure. I feel without Sheil’s efforts this task would have been almost impossible.
A couple of months after the Paris trip a large brown envelope arrived in the post, crammed with information on the XV-15. Before the Paris trip all effort that had been made to balance the rotors was set aside and a new approach was made. one scheme to balance the rotors around the tilt axis involved mounting the engine on the tilt tube to counter balance the head. The idea of the engine tilting in the wing was not very appealing in any case. The revised scheme was to have the engine mounted towards the front of the fuselage, hoping that this would result in a more favourable centre of gravity. This eventually evolved into the present design, with the engine towards the front of the fuselage and a belt reduction drive to the centre of the wing, aligned with the drive shafts. With the rotor now not balanced around the tilt axis, tilting the rotors presented a problem using normal servos. This problem was solved by using screw jacks to tilt the rotors. From the information received, the drawings were scaled to achieve the final design of the model at 1/5th scale. The 1/5th scale was chosen based on our initial tests with the airplane version of the V-22. At 1/5th scale the model XV-15 ended up with a fuselage length of over 8 feet and a wing span of 5 feet! I was very fortunate to have access to an industrial camera which made the task of scaling was made relatively easy. Fuselage formers, ribs for the wing, tail plane, vertical stabiliser etc. were copied from the drawings supplied.
Construction of the fuselage was the first major assembly to commence. It was felt that with the drive system to the wing having been designed, we could commence construction of the fuselage since most of the complex mechanics would be in the wing. The fuselage was constructed using the traditional balsa frame work and 1/16th balsa skinning. Carbon fibre re-enforcement was used in high stress areas like the engine assembly and the under-carriage position. The fuselage has been more or less as off the drawing board, and it has proved to be adequate in strength and at the same time very light weight. The fuselage along with the tail plane and vertical stabiliser were finished a long time before the rest of the parts were constructed. The rotor used Hirobo 3 bladed rotor parts, with a modified swash plate from Hirobo.
Due to pressures of normal life the project took second priority and was effectively shelved for a very long period. After a lot of ‘pushing’ from David Bell (one of the many friends who have been instrumental in the success of the project) the project was revived. Major assemblies included the wing drive shafts, rotor nacelles and the tilting mechanism. These were manufactured, with weight and adequate strength being the major factors in their design. The drive shafts in the wing were made from carbon fibre tubes. The drive shafts and the end coupling were tested to destruction and failed at about 10 times the calculated torque. The assembly for the rotors was also designed around carbon fibre tubes and aluminium bearing holders. After months on the lathe and hours of head scratching the final design evolved. The result was a very light weight but a strong and rigid transmission / rotor system. Although all the parts to start construction of the wing were in hand, it was felt a more prudent approach would be to bench test the drive system. Accordingly a test rig was set up evaluate the performance of the various parts. No major problems were encountered and after extensive tests, it was decided to begin construction of the wing. Crudely measuring the thrust from the rotors left no doubt in my mind about the ability of the ‘60’ to be adequate to power the then estimated 8kg all up weight. The construction of the wing was relatively straight forward. It was decided to install all the major assemblies in the wing, but to leave the wings without the balsa sheeting.
Towards the end of 1998, the model seemed almost ready for its first test hover. For various reasons the project was put on the back burner and hence it was decided aim for a specific day to aim for the first hover. The 1st January 1999 was chosen as the day for the very first hover. It may have been premature, but an attempt was made on the 1st of January 1999. After the engine was fired up, for the very first time both the rotors were spinning together. There were doubts in my mind and at the same time a fair bit of confidence, as lot of time had been spent in setting up the rotors. Although the rotors had been tracked individually, on the first attempt at application of power, the aircraft had a tendency to roll over to the right. Pitch adjustments were made to the left rotor to balance the lift from the two rotors. After being satisfied that everything was OK, it was decided to roll the model out for its first hover. During the roll out one of the under-carriage legs collapsed (bad design). The model was run up again for its first hover. Hover it did - only for a few seconds and came down rather hard, parting the nose from the rest of the fuselage. On inspection of the radio gear I noticed that the error lights were lit on the error indicator (this indicated the presence of interference). Since the fail-safe had been set to go to neutral on the servos, the model had come down rather hard. The damage was not substantial and was fixed the next day. The task now was to find out the cause of the interference. The model was fired up a few times with changes of the receiver position etc. to try and cure the interference problem - but in vain. I had been pretty sure that there was no metal to metal noise and mind was set on some other cause for the problem, possibly long servo leads. Eventually a friend gave me a telephone number to get in contact with Chris Golds, who had built and flown some large models. I had a very pleasant conversation with Chris. He seemed very keen on my project and was very encouraging. As for the radio problem, he assured me that he had used much longer servo leads without any problems and hence the problem must be either with the radio gear or some metal to metal or metal to carbon fibre noise. On this reassurance I decided to have a closer look at all the possible areas of metal to metal noise. Indeed Chris was absolutely right - I found a couple of carbon fibre rods that could vibrate against some aluminium parts. The carbon rods were insulated and the fail-safe made to hold the last command. The model was fired up again. Yes you have guessed it, no more interference. An attempt was again made to hover. The aircraft seemed very positive in roll control, but very spongy in pitch and yaw. Eventually after several attempts to hover the XV-15, I came to the conclusion that the spongy pitch and yaw control was due to the lack of tortional rigidity of the wing. (Again there were some people who discounted this theory! I wonder who that could have been). The only way to prove this was to sheet the wing (the wing had been not been sheeted to enable us to observe all the mechanical parts). The wing was sheeted, and an attempt was made to hover. All the problems with spongy controls disappeared. But this was not the end of all the problems. In calm conditions it was possible to hover the XV-15, but the work load to achieve a sustained hover seemed very high. However, it did hone my hovering skills! After several sustained hovers at the factory, it was decided to take the XV-15 down to my local flying field (Wormwood Scrubs). It was quite blustery at the scrubs, and after spending a total of about an hour at attempting to hover the XV-15, I only succeeded in hovering it for 2-3 seconds a few times. It was time to give up. Back to base and it was concluded that the XV-15 could not be hovered in any wind without recourse to gyro stabilisation. The model had felt very stable in yaw, but the opposite was true for roll and pitch. This brought the project at a crucial point in its development. Right at the outset it was felt that gyros would probably be needed. The problem was that although the control system had been designed very well, no provision had been made to allow the use of gyros. By this time Sheil had a job and the task was left to me. Since my software skills were not very good, it took me a long time to work out a way to incorporate the use of gyros. Even so the software did not work quite right. Fortunately Sheil managed to sort out the problems. With my experience at hovering the XV-15 I was of the opinion that pitch control was the main control that needed stabilisation. Hence a gyro was installed to cater for pitch. This transformed the model. A second gyro was installed for roll and now the model was much more stable. It was again taken out to the scrubs. The result was a very satisfactory hover in fairly wind conditions. The third time out to the scrubs resulted in the very first forward flight in helicopter mode. A very satisfying experience! After a few sessions, the first major failure occurred. One of the gear boxes failed while in forward flight. Fortunately I was able to shut the engine and do an engine off landing, I hasten to add without any problems. I had never been sure about the gears (from the Morley MXA helicopter). I looked up some data on the gears and realised that the gears were grossly under rated. It was decided to change the gears and do a bit of tilting before up rating the gears. After fitting new gears, the XV-15 was taken out several times and gradually put in forward flight by tilting in 2-3 degree steps. Once it is in forward flight, the XV-15 becomes remarkably stable. All this time the flying had been done using NHP carbon fibre blades. Although I was aware of the fact that the rotors could not be tilted fully forward without using twisted blades, my computer modelling had suggested that I could possibly tilt by about 25-30 degrees. The XV-15 was flown with the rotors tilted by about 35 degrees. It appeared to be flying forward at a fairly high speed. It was impossible to tilt the rotors any more with the present set up without sever loss of thrust. The model was been grounded after a mishap which resulted in rotor damage. The mishap was caused by loss of power due to a gear-box failure.
No attempt was made to repair the relatively small amount of damage. As much flight testing as could be done with untwisted rotors had been done. Efforts were being made to have moulds machined to enable twisted blades to be made. Although a lot of companies were approached for this task, not many companies were even willing to quote for the job. One or two companies, who did, quoted prices, which seemed beyond the scope of this project, which right from its conception was embarked upon as challenge rather than for commercial gain. However like always perseverance pays off. This time perseverance by the secretary of one of my customers in Coventry, June Fairbrothers paid of. She managed to find an engineering company in Coventry, Jet Blades & Engineering. She arranged for a meeting with Mike Kirby of Jet Blades. I had a very pleasant meeting with Mark, who spent a considerable amount of his valuable time to discuss my project and show me the manufacturing facilities they had. I was convinced that if anyone could make the moulds for the twisted blades, Jet Blades could. Mark was very enthusiastic about the project and was willing to commit to make the moulds at a price I thought must hardly cover his costs. While the model was grounded new gearboxes were made with up rated gears. It was decided to rebuild the wing as a few areas of weakness had shown up. Although the wing had adequate strength for flight loads, there was no margin for slightly messy ‘arrivals’. The redesigned wing should cater for such ‘arrivals’. Several incidence of interference were encountered at the scrubs. Every time the model was returned to the factory and hovered the interference was not present. There were several other reported incidence of interference at the scrubs from other flyers. I was of the opinion that for the sake of safety, it was best to incorporate a dual receiver / transmitter to enhance the already advanced fail-safe. The dual system could be justified in view of the complexity of the project and from the value of the XV-15 in terms of time and money.
FLYING THE XV-15
Flying the XV-15 has been a mixture of anxiety and pleasure. During the initial test hovers there were a few anxious moments. Initially when the wing was not sheeted, it was almost impossible to hover. After the wing was sheeted, it had a more positive response, however it was still hard work hovering it. The situation was improved by adding weight at the bottom. However it was accidentally discovered if the weight was too far out in the lateral direction, it adversely affected the stability. This instability was put down to the overall lightweight model and a very large model. The stability was further impaired in windy conditions. Although the extra weight did not present a problem in hover, it was felt to be too a high a penalty for airplane mode flying. (It would have to fly much faster). It was felt at this stage that the best solution to the problem was to have gyro stabilisation. Fitting the 2 gyros (pitch & roll) overcame the hover stability problems. However once the XV-15 got into forward flight, the situation changed dramatically. It became the most stable aircraft I have ever flown. It has an amazing amount of control power and yet is extremely stable. At one stage, with a small amount of tilt, I let go the controls and it carried on in a perfectly straight line. It had no tendency to roll, pitch up or down, or yaw! When the project was commenced, it was felt that the initial transition would be a difficult stage to go through. This was an anticlimax as I can honestly say that I have never thought of the aircraft as being either in helicopter mode or airplane mode. It just goes from one to the other in a very natural way. This has fulfilled one of the design criteria laid down - it should be possible for anyone with a reasonable ability to fly both planes and helicopters to fly the model. So far it has proved to be very much so. The other criteria were the requirement of a standard engine without recourse to fancy pipes and having to tweak the needle for ultimate power. This has also proved to be true. Until the twisted blades are available and the rotors are tilted fully forward, the full flight envelope cannot be explored.
As it stands the model has been flown with about 35 degrees of tilt. Why can the rotors not be tilted any more with the normal NHP blades? A lot of flyers (some very experienced ones) have asked me this question. The answer very simply is because of the effective airflow over the blades in forward flight, resulting in negative thrust being produced by part of the blades near the root. Computers modelling of the rotor performance indicated sever loss of thrust at about 40-degree tilt. Just like a propeller, the blades would need a twist. This would of course make the blades inefficient in the hover. Aerodynamics is a lot about compromises, and the blades have to be designed to compromise between hover performance and forward flight performance. On the real XV-15 the blades have been optimised for forward flight. This is due to the fact that the whole concept of the tilt rotor is to have the vertical take off and landing capability of a helicopter and the fast and efficient performance of an aeroplane. Without this ability the tilt rotor would have no application in real life. The compromise on the real XV-15 has been in the hover mode. The model XV-15 is not constrained by these requirements and hence the blades have been designed to have a favourable performance in both hover and full forward flight, with each mode at a disadvantage because of the blade design.
The greatest advantage with a model this size is that you never get disoriented! The model XV-15 dwarfs normal models, including those powered by 30-45 cc petrol engines! During the last flight the model weighed in at 9.1 kg. It is estimated that it will weigh between 10 and 11 kg when fully finished. At best, the power with the present engine (an OS 60 SF) could be very marginal. One choice would be to replace the 60 SF with the 60 SX. This would make about 25% more power being available. The other would be to move up to a ‘90’, providing perhaps 50% more power as compared with the present set up. Perhaps the best approach may be to evaluate the performance of the XV-15 with the OS 60 SF using the twisted blades. With twisted blades, the worst case would be in the hover. If the XV-15 can be hovered with twisted blades in the current configuration, its performance should improve in forward flight with the rotors tilted forward.