A crash course in stability issues for SailRocket
by Paul Larsen 4 May 2007 09:53 BST
Paul Larsen looses steerage of SailRocket at 35 knots © SailRocket
On Day 12 and Run 7 with 18 knots of true wind blowing across Walvis Bay’s incredible speed sailing strip in Namibia, SAILROCKET’s pilot Paul Larsen lost steerage at 35 knots. His craft spun in towards the beach. The round-up was aggressive and caused the solid wing sail to backwind violently and slam down into the cross beam. Sailrocket stopped abruptly. The beam was shattered.
It was virtually a repeat of their first run at ‘speed-spot’ two months ago.
Yes, there was some swearing.
‘We had been very cautious about choosing our wind strengths. The last two runs had gone well in slightly lighter winds and we had run up to 30 knots both times. There was a hint of instability at the end of Run 5 which I mentioned. We thought that by going to the smaller ‘high-speed’ rudder at 30 knots that this would be resolved. Well this run proved that it made little difference. I did a nice and gentle approach to the course and built speed gradually (for this boat anyway). At a little over 30 knots the helm felt light so I flicked up the big rudder. With just the little rudder down there was some weather helm but I could easily hold it. Sailrocket kept accelerating. I was well positioned on the course and still a conservative distance off the shore. If all went well and I had good control then I planned to squeeze in closer. The helm increased and I felt that if it got any worse as the speed increased that I could be in serious trouble. We were expecting neutral or even the beginnings of the much sought lee helm at these speeds. I sheeted in a little to help relieve it. Sailrocket began to yaw to windward. I stomped on the left C-tech rudder pedal but I knew I didn’t have the steering power or room to correct it like on the wild Run 4. All of the team in their different positions along the course knew what was about to happen. I fought the round up for awhile but then it was no use. If I had halted the turn half way I would be up the beach in an instant. The rig compressed the COMPOTECH strut into the beam which fractured the beam and shattered the strut. It all rained down. The fact that the beam was already broken could have softened the impact of the wing. I was in waste deep water by this stage and could pull all the carnage ashore.
This was not a pretty sight. We were all gutted. My mind reeled at what had just happened. What had just happened? What is wrong here? Why does this keep happening despite all our modifications and tuning. Whilst everything was fresh in my head I tried to evaluate what I had seen and felt for later reference. It was hard to explain.’
Now, a few days later, I can say that what lay before us on that windy strip was merely a hard lesson accompanied by a slap in the face. Sometimes progress is destructive. We lashed the wing down to a log and limped home with the wounded platform. The boat had done what we had worked hard to eliminate. I hadn’t used the brake that we had built to prevent it but that is secondary. The way it happens is just so quick and the window of opportunity for the brake is small. I missed it this time but that wasn’t what was bugging me. The boat is fundamentally unstable and I can’t control it.
I called Malcolm and followed this up with a quick report into sponsors and supporters. Positive offers of support came back immediately. It really helps to turn the tide, ‘Oh no, that sucks, well…chin up, you guys can come back. What do you need?’
We pulled SAILROCKET apart on the windiest day since we arrived and went back to the apartment… to think hard. Something was fundamentally wrong. Our beliefs had been shaken. In times like this I find that a healthy diet and hard exercise is simply no substitute for a well stacked fridge!
The following morning we went back to speed-spot and recovered the AEROTROPE wing. It was amazingly intact. I turned it over expecting to see a huge area of mashed up composite… but there was nothing other than some scratches and easily repaired cracks. The central COMPOTECH spar had stood up to the abuse, the FIBREFUSION ribs had absorbed an alarming amount of shock and suffered some small local delaminations but overall…she had all survived. I can’t believe how strong this thing is. Watch the video and you will understand. The beam had obviously failed before the wing hit it. The flaps had copped a bit of damage. The repairs there are always fiddly but that’s ok. We built it, we’ll fix it.
I began to come to grips with the damage and the fact that my lovely ‘first build’ was not going to stay perfect forever. These initial crashes were like the first scuffs on your first new set of footy boots. Get over it and go play football! The psychological separation is almost a relief. This boat is a tool, the goal here is to build and sail the world’s fastest boat… not the prettiest. No doubt there will be more composite work to come.
We did a fully set-up weigh in the other day. In full sailing mode Sailrocket weighs in at 192.5 kg (no pilot… 82kg in case you were wondering). She is light. We can afford to ‘spend’ a few kilos on repairs and mods. That’s why we took the advice of SP way back at the start and went the pre-preg carbon / Nomex way. This decision is paying dividends now. So now that the damage is just resigned to a boat building exercise, the big issue is what is causing it and how can it be cured?
Chris Hornzee-Jones who designed the wing looked at it from my perspective and worked backwards into his understanding of Malc’s calculations (Malcolm’s performance simulation). He deduced that whilst the simulation showed ‘freeze frames’ of control scenarios it didn’t make it obvious how this would relate to the dynamics that would be required by the pilot.
It appeared that Sailrocket is fundamentally unstable and even when going straight, it is teetering on the edge of instability. It is almost divergent. Like trying to fly an arrow backwards. The faster you go, the worse it gets.
Now you can technically fly an arrow backwards. Most modern fighter aircraft do exactly that. They are unstable way beyond what a pilot can control and require a computer to maintain stability. The pilot simply directs it. If the computer fails then the next button you push is to eject. No point hanging around except to grab the car keys!
So Chris suggested fitting a fixed skeg at the back to give Sailrocket some idea of direction and an ability to track in a straight line. The rudder can do this but it needs constant super fine adjustment.
Malcolm factored this skeg into his simulation and it virtually transformed the handling of the boat and dampened down all of its unruly characteristics. Malcolm’s simulation did predict what was happening all along but you had to know where to look. Chris looked at the pilot’s perspective first and resolved that the response amounts and reaction times needed were unrealistic… especially when you consider the result of a mis-match.
We had been focusing on making the boat have more of a tendency to turn away from the beach i.e. more lee helm, but all the while this had distracted us away from a far bigger issue. Well, all I can say is that it has our full attention now.
We are not about to jump onto the first wagon offering hope here but we feel that we have discovered the fundamental issue. Now we have to resolve it. The model worked perfectly so what is the difference? One big difference is that there is a human attached to the rudder and not a strong mechanical device. The rudder was also inclined on the model. The thing about the model was that once it was up and running you virtually didn’t have to steer it. It tracked straight and true. This is what I liked most about it. Replicating it in full scale is not proving to be that easy.
Once I slipped off the ‘straight and narrow’ with the little rudder, it didn’t have the power or advantage to correct the craft. In fact we wonder now about the wild ride experienced in Run 4. The fact that the aft planing surface had broken off and that the back end was riding low in the water may have helped with the tracking and assisted our very narrow escape. Maybe.
So this is where we find ourselves.
The brains back in the UK (and some in Valencia) are swarming over the problem and the support network is in action. Helena and I have already begun the repair. We want to be back out there in around two weeks. With the amount of support we have been offered from HOMETRACK, SP, COMPOTECH and a host of others, I’m sure we can do this. We must however be armed with a thorough understanding of the problem and a very solid solution. This will define the timeline for our return.
We will forget 50 knots for now and focus on regular 40s.
We receive, consider and reply to all offers of advice from all directions (the mention of a skeg had been made before but we were looking elsewhere...cheers Rik).
We share all of this with you so at the very least this project can share its understanding of a new concept in a constructive way and pave the way for the more adventurous of you who might like to follow some day (also for the enjoyment of those who have trail-blazed before).
As usual, it is our ability as a team to come back from these lows will define our end result.
As a potential world record breaker SailRocket is still a diamond in the rough… but a diamond nonetheless!
Cheers Paul
More Information: