Flying Cherubs

[clivee1]

After 3 years of sailing UK Cherubs we have decided to fit foils to our boat.

The UK Cherub class split from our Australian cousins in the early 90s when in response to dwindling fleet numbers and perceived threat from new UK SMOD asymmetric classes the class adopted asymmetric kites before the fleet down under were ready to do so.

In 1997 the asymmetric got bigger and since 2005 we have had twin wires 15.5m^2 of up wind sail, 23m^2 kites on 12' boats that weigh 70kg fully rigged. We have a small but vibrant fleet with recent nationals being won by boats designed by Kev Ellway, Paul Beiker and a derivation of the Nuplex 12' skiff from David Lee. T foil rudders are universal.

Crew weights are typically 130 - 150kg. Within the fleet there is a mix of young light adults / couples and ageing skiff sailors sailing with their offspring.

It is one of the few active open development classes remaining in the UK.



Foils are fully legal.

Many will ask why they are not already standard. the boats are very similar to Kiwi R Class where they have been shown to work.

I think that the broad range of conditions we see in the UK and a the daunting prospect of trying to stay on the side of a short trapeze dingy during crashes have put most off.

Last year Graham and Eddie Bridle converted their boat having won every thing possible with Eddie rapidly outgrowing the restricted space at the front of the cockpit. They showed that the boats can fly but have not managed to reliably get around the race track.



Compared to the Moth our sail area to all up weight is similar upwind and potentially much better downwind.

(15.5m^2 / 210kg Vs 8.25m^2 / 110kg upwind).

With twin wires and 1.8m beam our sail carrying ability is better.

However our hulls are only one foot longer, nearly 3x the water line width and carrying twice the weight so getting to take off speed is going to be a challenge. We do not have the aero dynamic efficiency of the Moths high aspect una rig and minimal hull. I have no expectations of getting close to their performance.

As an un athletic 50 year old sailing with my 13 year old son I have no desire to do 30kts on the wire. Others can try that.

We typically sail at 8-10 kts upwind. It would be nice to see this increase. Downwind we routinely do 15-18 and occasionally hit 20.

The biggest challenges are going to be take off at low wind speeds and having good enough control to avoid the crashes that would eject us off the front.

Our solution is a twin centreboard Pi foil configuration with a large main foil.

I was going to make the foil with a 1.6m span but have reduced it to 1.45m so that it still loads onto the trailer and so that we can roll it for rigging with the foil retracted against the hull. The foil area is 0.22m^2, giving 30% more area to weight than the more recent moth foils. This gives a 10:1 AR similar to the Blade Rider or first Mk2 foil but lower than the more recent ones.

The Pi foil configuration allows the wand to control the pitch of the entire foil which should give better height control than just the flap.

It also allows a centreline wand to be used.

It is was much easier to build as there are no structural T joints or complex flap assemblies.

We have used a DAE41 foil section that was developed for low Reynolds human powered flight.

It has low Cd for Cl near zero to Cl ~1.5 which is higher than we would ever want to use because at that level the induced drag would be higher than the hull drag if we were low riding. The section has quite a high pitch moment. To counter this we have added a tail plane to the foil.



This can be adjusted to tune the load on the push rod.



We sailed for the first time 2 weeks ago. The boat would take off readily but height control was very poor, and we had some bruising crashes.



We had our second sail last weekend after tweaking the tail plane. Control was much better but we were both nervous after the previous crashes. Youtube link below - how do you embed a youtube vid?.

We have no commercial aspirations for the project. No expectations of breaking records but do hope to have some anarchic fun.



Come and see us at http://www.uk-cherub.org/forum/index.php/topic,1946.0.html

or http://www.uk-cherub.org/doku.php

or better still come and join us on the water or in the air.

 

[kprice]

Where is the electronic control?

 

[Doug Lord]

Clive, that is very cool-best of luck! I'll post it on boatdesign.....

 

[atg]

Cool idea Clive. I would like to see what you have going on up on deck, i.e. compensation for the shifting fore/aft center of lift. Perhaps the tailplane handles things in a way I do not appreciate. Regards, Karl

 

[clivee1]

Hi Karl,

Simplistically this is how the control works. I will try and take a picture this weekend.



It is geared so that the movement in the control line is large and the loads low.

90 degrees of wand rotation (1m of height) has 70mm of control line movement.

The tail plane is there to compensate for the pitch moment to keep the position of the centre of lift in the right place and the loads on the push rod sensible. I think that there will be some gain to be had in allowing the push-rod forces to increase with speed.

Any elasticity in the system will mean that the boat will not rise and may even drop slightly at higher speeds. We need to be careful with this. a combination of elasticity and friction will lead to a jerky control.

During the first sail it became clear that there was not enough load on the push rod and the control was very poor.

I had not made allowance for quite how much the angle of the flow over the tail plane is effected by the main foil.

For the second sail we made a slightly bigger tail plane.

Its angle can be adjusted whilst capsized but not with any real precision.

Initially during the second sail it would not take off. The wand was not able to push the push rod down.

After a couple of adjustments Alex was able to push on the top of the push-rod with the control disconnected and manually make us take off. The load is probably still higher than ideal but the boat then started to fly pretty nicely.

I am working on a cam system so that the ride height control does not change the take off AoA.

Ideally I would like to be able to adjust the ride height, Take off AoA and control gearing independently.

It seems that the controls the Moths have evolved all effect all 3 to different degrees.

Clive

 

[Boatshed]

Thanks Clive. I think you've caused me enough trouble trying to get the RS300 round the course without you showing me that Cherub.

Congrats on your flights !

 

[makerofthings]

Lovely work clive, glad the bigger tailplane is helping, you look rock solid in the video... More please!

Cheers. D.

 

[slip knot]

I'm amazed at how smooth that is flying on only your second try.

Great work.

 

[RobG]

In regard to ride height systems, Moths have found mechanical linkages to be superior to tensioned line systems.

There's no indication in the diagram of how return tension is applied to reduce the AoA. There needs to be a lot of tension in the system to make it work well. I don't know a precise number, but maybe 20 to 30kg depending on the gearing of the wand and return system.

Interesting that you can use a manual override to get foiling, I guess that's the advantage of having a second pair of hands. ;-)

 

[couchsurfer]

.

...flying cherubs??.......old news :mellow: <_<

 

[clivee1]

In regard to ride height systems, Moths have found mechanical linkages to be superior to tensioned line systems.

There's no indication in the diagram of how return tension is applied to reduce the AoA. There needs to be a lot of tension in the system to make it work well. I don't know a precise number, but maybe 20 to 30kg depending on the gearing of the wand and return system.

Interesting that you can use a manual override to get foiling, I guess that's the advantage of having a second pair of hands. ;-)

Hi Rob,

I suspect that mechanical linkages are superior, however the simplicity of the tensioned line is very attractive.

I hope to minimise the disadvantages by gearing it so that there is lots of movement and low tension in the line.

Over the range of movement of the wand, the line moves 70mm. I think that this is about 5x as much as the movement in a Moth linkage. The effect of any stretch is correspondingly reduced.

I learnt to drive in a car with cable steering it would never be used now however the cable shift on my road bike is pretty crisp.

The return force is generated by having the pivot on the foil forward of the centre of lift. The idea is that the tailplane can be tuned to control the position of the centre of lift and how it changes with speed.

During the little bit of flying we have done so far I have been trying to judge the reaction of the wand tip on the water and from that had estimated that we are currently running with about 20kg of load on the centreboard push rod.

I had felt that this was perhaps a bit high. Interesting that you numbers are even higher.

I am very aware that there is now thousands of man years of Moth flying experience out there and that we have been doing it for 2 weeks, we have a lot to learn.

The manual take off was a useful experiment to try and gauge the load on the push rod. I don't see it being routine.

Hopefully we can find the AoA for the foil that gives the sweetest take off.

I would like controls that independently adjust the take off AoA, the ride height and the gearing.

The forecast is looking good for tomorrow with luck we will get some more airtime before it gets too cold for us.

Clive

 

[atg]

On the pivoting DB it is all about keeping the loads down. I am curious about how the tailplane is trimmed; that is a cool idea.

Ideally the wand would trim the tailplane and not the foil itself. That would keep loads down.

You will want to take a hard look at friction and put bearings everywhere.

I used an old magic box with an alu tang to mount the turning block aft of the head of the foil. So I could easily adjust the linkage length with that, even loaded. I did however manage to rip it right off the deck with the deck paint and plexus though, so you may wish to sand down to the carbon and rough the surface of the aluminum up a bit before gluing one on. Nothing like trying to sail dead down through chop to the channel entrance with a daggerboard flopping around willy nilly trying to pitchpole you because the linkage went away. Might be good to have an emergency lock on it somewhere in fact.

Fundamentally the wand is fighting a sort of losing battle when the pivot is at the hull exit. I ran a counterspring for a long time on the moth made from several strands of mandrel-dipped rubber UAV launcher tubing hose clamped to some nylon plugs. Then I added a motorcycle steering damper, which survived salt water immersion remarkably well for several years. Let me know if you need one; I'll give you a good price on it. Looks pretty much new actually.

One problem is that the spring rate needs to change with the fore/aft position of the crew in the boat. As you vary the loading between the rudder and mainfoil, the moment about the pivot changes.

Another problem is that during takeoff, with a spring, the wand load is high because the foil is not generating lift, so the spring needs to be manually engaged as you take off, which is pretty much impossible on a Moth. If you have no spring then great, but your operating load on the wand is going to be higher, leading to more wand drag. Maybe the crew can trim the spring. Two people is a luxury.

I could go upwind just fine but the corners in marginal air were very tough due to takeoff, and downwind the settings were hard to dial in. Basically, it was just too much manual adjustment all the time to be competitive around a course for a singlehander.

The new system is much better, though it is hard to get the wife to sign up for RIB driving and childcare duties together. Maybe a stainless T top will help...how hard is it to bend 1.25" stainless stanchion stock?

K

K

In regard to ride height systems, Moths have found mechanical linkages to be superior to tensioned line systems.

There's no indication in the diagram of how return tension is applied to reduce the AoA. There needs to be a lot of tension in the system to make it work well. I don't know a precise number, but maybe 20 to 30kg depending on the gearing of the wand and return system.

Interesting that you can use a manual override to get foiling, I guess that's the advantage of having a second pair of hands. ;-)

Hi Rob,

I suspect that mechanical linkages are superior, however the simplicity of the tensioned line is very attractive.

I hope to minimise the disadvantages by gearing it so that there is lots of movement and low tension in the line.

Over the range of movement of the wand, the line moves 70mm. I think that this is about 5x as much as the movement in a Moth linkage. The effect of any stretch is correspondingly reduced.

I learnt to drive in a car with cable steering it would never be used now however the cable shift on my road bike is pretty crisp.

The return force is generated by having the pivot on the foil forward of the centre of lift. The idea is that the tailplane can be tuned to control the position of the centre of lift and how it changes with speed.

During the little bit of flying we have done so far I have been trying to judge the reaction of the wand tip on the water and from that had estimated that we are currently running with about 20kg of load on the centreboard push rod.

I had felt that this was perhaps a bit high. Interesting that you numbers are even higher.

I am very aware that there is now thousands of man years of Moth flying experience out there and that we have been doing it for 2 weeks, we have a lot to learn.

The manual take off was a useful experiment to try and gauge the load on the push rod. I don't see it being routine.

Hopefully we can find the AoA for the foil that gives the sweetest take off.

I would like controls that independently adjust the take off AoA, the ride height and the gearing.

The forecast is looking good for tomorrow with luck we will get some more airtime before it gets too cold for us.

Clive

 

[Phil S]

Flap less foils? Full tilt boogie?

Karl has been doing this for quite a few years and maybe has more hours than anyone else so his experience, problems and solutions outlined above may well be the best available. Ian Ward did it first, way back, and has done some refinements with his Glide Free Laser foil kit. But no one has done a competitive moth foiler with this system.

Apart from the control and power issues Karl has outlined I see a few other fairly significant disadvantages.

Firstly the foil section has to be compromised between an ideal take off section and a high speed section, not easy on a moth when the range is from 7 to 30 kts. Maybe easier if your target speeds are more conservative which is maybe where Clive and Glide Free are aiming. Basically more camber is needed or it will not take off.

Secondly the leading edge of the tilt foil needs to be different to a flap foil so it can still work at a wider range of angle of attack, basically it needs to be blunter, and have more camber forward which will both add drag at high speed, so that when it is tilted LE up for take off it does not just stall.

The only advantage I see is the lack of drag from the hinge line, against the added drag from the open Tee hinge with associated large gaps and local thickenning. I would think it would balance out.

The flapped sections now used by moths are relaitvely thin with maximum camber well aft, close to the hinge line, and with quite sharp almost symetrical leading edges. Max thickness is not far infront of 50% chord. Quite like laminar flow and some early supersonic aircraft sections.

Having the max camber near the hinge line also means the flap position does not hugely distort the camber line, so with flap up at high speed the section is almost symetric.

The flap down position at take off adds camber but also adds angle of attack of the total flapped section as the trailing edge drops, but without tilting the sharp leading edge outside its relaively narrow low drag range.

The moth system may look like it just happenned by accident but it really is a very clever solution to a complex issue, and based on the number of boats now using the original Prowler system with great success it should not be ignored if you are planning to foil a different type of boat. If moth sailors have not invented a better system in over 10 years it can not have much wrong with it.

 

[clivee1]

Hi Karl,

it is just the main horizontal that rotates.

The verticals are fixed. Yes it would be nice for the wand to control the pitch of the tail plane.

The tail plane is clamped between the 2 booms. It is tapped on its 1/4 cord line and has a symmetrical section so that it generates no pitch load that could disturb its setting.

We are finding that it is very effective at controlling the load on the push rod.

Too much and the wand cannot generate enough AOA on the main foil to take off, too little and there is no effective control.

We sailed briefly last Sunday and could not work out why it would not fly as well as it had the previous week.

On coming ashore we found that the tail plane had been disturbed and that we did not have as much load on the push rod.

I have some printed parts on order to allow it to be set with greater precision and to reduce the chances of it being knocked.

Best solution I came up with for the child care was to take him sailing with me.

Phil,

The main reason I went away from the single T foil is that the cube square law is working against us. We are carrying twice the weight of A Moth and would need a foil 50% wider. I did not think that I could build a reliable T joint at the first attempt.

With this system all the parts are easy to build and there is no high risk T joint.

We could go to much thinner foils without the risks. I am not sure why thin foils are perceived to be so much better. Glider wings are not thin and they are looking for high performance.

The DAE 41 section operates at low drag at these reynolds numbers from Cl near zero to ~1.4.

I do not believe that the Moths load the foil any harder than this. The induced drag would just be too high.

Playing with Java foil (no substitute for the real world) I cannot see the advantage that the flapped foil gets.

The DAE41 would appear to have lower Cd than the 63-312 laminar section at all Cl at these reynolds numbers.

It would be a very different story if we were working in the same region as aircraft that laminar flow sections were developed for.

I would love to see a polar for a Moth foil.



I certainly take you point that the Moths are very refined and are working very well but we are not sailing a Moth and have some different problems to solve.

Clive

 

[atg]

Yes many smart people have looked at the problem. And some dumber ones like me.

On the flapped vs flapless front Phil has made a good articulation of the pros and cons. Nonetheless if one actually runs the numbers, the flap does not come out ahead of a flapless foil. Flapless allows pushing the structural limits harder, getting more span for a given thickness, and also playing with geometry e.g. anhedral. So the drag you might pick up at takeoff due to a thin nosed 2D foil section is offset, because the foil area can be smaller and the span larger, thereby cutting induced drag considerably. So takeoff drag can remain the same, while high speed drag goes down because the foil area is smaller. Flapless cleans up the T. This comes at the expense of wand load; the paddle does generate a fair bit of resistance unless the entire system is optimized to eliminate both wand loads and friction loads.

I think my current system is pretty close in performance actually to a flapped foil. If there is ever another local regatta I will test the theory. It could be a bit lighter, but for a prototype it is OK. After that I should really fire up the router and make an optimized mainfoil mold, as my current foil is bigger than it has to be.

I think Clive's boat demonstrates that the world of ideas is far from exhausted when it comes to wand-driven foiling solutions. They may not be easy to dream up, but neither was the flap system. In fact the more I did with my stuff the more respect I had for John Ilett and company. Lots of people have tweaked control systems on the water, but not many have developed entirely new ones.

So hat's off Clive. Keep us posted. I agree it would be much more fun with a kid in the boat to share the experience, but right now we are working on swimming lessons, so it is a bit premature! Perhaps my next tilting foil will go into a Cherub hull.

K

Flap less foils? Full tilt boogie?

Karl has been doing this for quite a few years and maybe has more hours than anyone else so his experience, problems and solutions outlined above may well be the best available. Ian Ward did it first, way back, and has done some refinements with his Glide Free Laser foil kit. But no one has done a competitive moth foiler with this system.

Apart from the control and power issues Karl has outlined I see a few other fairly significant disadvantages.

Firstly the foil section has to be compromised between an ideal take off section and a high speed section, not easy on a moth when the range is from 7 to 30 kts. Maybe easier if your target speeds are more conservative which is maybe where Clive and Glide Free are aiming. Basically more camber is needed or it will not take off.

Secondly the leading edge of the tilt foil needs to be different to a flap foil so it can still work at a wider range of angle of attack, basically it needs to be blunter, and have more camber forward which will both add drag at high speed, so that when it is tilted LE up for take off it does not just stall.

The only advantage I see is the lack of drag from the hinge line, against the added drag from the open Tee hinge with associated large gaps and local thickenning. I would think it would balance out.

The flapped sections now used by moths are relaitvely thin with maximum camber well aft, close to the hinge line, and with quite sharp almost symetrical leading edges. Max thickness is not far infront of 50% chord. Quite like laminar flow and some early supersonic aircraft sections.

Having the max camber near the hinge line also means the flap position does not hugely distort the camber line, so with flap up at high speed the section is almost symetric.

The flap down position at take off adds camber but also adds angle of attack of the total flapped section as the trailing edge drops, but without tilting the sharp leading edge outside its relaively narrow low drag range.

The moth system may look like it just happenned by accident but it really is a very clever solution to a complex issue, and based on the number of boats now using the original Prowler system with great success it should not be ignored if you are planning to foil a different type of boat. If moth sailors have not invented a better system in over 10 years it can not have much wrong with it.

 

[atg]

Oh I just saw the photos you posted Clive. For some reason they would not appear before.

Interesting that with totally different systems we both had problems with too much wand loading at takeoff! With an extra set of hands you can do a lot about that though. I thought of having a bicycle brake handle on the tiller to pull and get a bit more daggerboard rake for takeoff, then release once flying. Hydraulic or cable. You could do the same to somehow push harder on the rod at takeoff. It has to be convenient wherever you are in the boat while taking off; moving around to adjust it is slow out of tacks and the competition will be long gone if it takes any time to sort out at all.

Seems like the joint between uprights and the foil could actually be rather fair when optimized.

Decavitator used two uprights on the main foil for the reasons you describe, so you are in good company there I think.

All this has me motivated to try to get out more often. My wife works weekends so it isn't always an option, but that should change for the better shortly.

Good Sailing,

Karl

 

[GinPalace]

Just wanted to congratulate you on your progress!

As an out of shape ex i14 sailor with a 7year old I'm inspired by what you've done and hope you'll continue to share your progress! I can only think of one thing more fun than skiff sailing with my son, and that is foiling with him!

GinPalace

 

[clivee1]

Thanks Gin Palace.

I too sailed 14s in the 90s - Before Children.

I started sailing with Alex in an RS200 when he was 8 and we got our first Cherub when he was 10.

There are other dads in the fleet that I sailed against years ago and had not seen since. It has been a great 3 years for sailing.

We are being very selective about the conditions for trying the foils and have not been out again but the forecast for tomorrow is promising.

Clive

 

[RobG]

Clive Everest, on 11 Oct 2014 - 01:49, said:

[…]

During the little bit of flying we have done so far I have been trying to judge the reaction of the wand tip on the water and from that had estimated that we are currently running with about 20kg of load on the centreboard push rod.

I had felt that this was perhaps a bit high. Interesting that you numbers are even higher.

The load I noted is what is required in the return system of a tensioned line. It needs to overcome the maximum pushrod load plus any loses in the system and needs to respond quickly, so likely its way more than is in the pushrod.
I think your approach is very interesting, have you considered a single-boom tailplane? It would allow for a single connection to a control system to adjust the tailplane (elevator) AoA and hence main foil AoA. The loads on the adjustment mechanism should be significantly lower than adjusting the main foil or flap directly.

However, response might be muted and not sufficiently direct.