Self steering - the progress, Part 2

[Chrisc]

Not a lot of progress in the new pics from my last post, but a lot of thinking and experimenting with full scale mock-ups given that I am now departing from accepted windvane practice. I thought a lot about this post because if my ideas are totally wrong then I am going to appear a first class idiot, but, better to try and fail and share than not to try at all...
Being old, retired and liking my comfort, my self steering requirements are perhaps slighty different. Since I have all the time in the world, I now only sail in nice weather, and if where I want to go is into the wind, then I go the other way. As my wife likes tease me, 'gentlemen don't sail to windward.'
In this regard I need light weather, eased sheet performance from my gear and as such have always had reservations as to the air vane - trim tab linkage. Linkages are offset to one side of the airvane and represent a degree of inertia to be overcome, particularly in light weather. In past constructions I have tried small cantilevered weights on the opposite side of the linkage to overcome the inertia, but without any appreciable success.
The new gear air vane to trim tab is activated by 2mm spectra line cables in teflon sheaths. I know this arrangement doesn't find a lot of favour, friction being the stated reason, but I have found friction in my set up to be almost non-existant and believe that the poor steering that these systems are known for is due to the absence of feedback between the trim tab and rudder. If I may explain : If the airvane is deflected 45 degrees from the vertical then the trim tab should rotate 15 degrees thereby turning the rudder. When the rudder reaches 15 degrees of rotation then it and the trim tab should then be in line so that oversteering does not occur. As far as I am able to ascertain, every cable operated system has the cable sheaths terminating on the rudder head. Whilst there are many benefits to this it does mean that that this arrangement cannot provide the necessary feedback ie the trim tab will stay at 15 degrees regardless of the rudder angle until the airvane returns to the vertical position, and this results in huge 'S' curve steering.
I have made a lot of mock-ups and have a system devised whereby I gain feedback. In the experimental stage it works flawlessly, the down side being that the cable/sheath termination, trim tab tiller and geometry requuired at the rudder head, whilst simple, needs to be accurately made, and each system needs to be tailor made to each individual boat. Perhaps this is why manufacturers have not gone down this line.
But, as we all know, many a slip between theory and practice...
More as it happens.

[s/v Beth]

Chris there is a incredible amount of knowledge out there on this subject, including a number of us that like to tinker and make our own windvanes and self steering devices. Don't reinvent the wheel, and google "sheet to tiller steering" and DIY windvanes.

[Jim Cate]

Chris, the vane that I built, and the AutoHElm that inspired it use a big counterweight on the air vane to provide what the inventor called "synthetic feedback" if IRC. In these designs the vane assumes some initial position as the boat goes off course, the tab is turned and the rudder follows shortly thereafter. The boat begins to return to course and the angle of attack on the vane reduces, and the vane, under the influence of the counter weight begins to move back towards the vertical position. This reduces the angle of the trim tab, the force on the rudder is reduced and it too returns to the neutral position. Sheesh, that's kinda hard to verbalize, but it really does work! I believe that all trim tab vanes tend to wander a bit, but can say from thousands of miles of usage that it isn't too bad.

So, don't over think this problem. It sounds like the dyneema/teflon combination works as well as I anticipated, and should provide trouble free lifetimes.

Keep on truckin', and let us know how it gets on!

Cheers,

Jim

[Chrisc]

Yes, there are a lot of well understood principles in wind vane steering - size relationship of trim tab to rudder and airvane, degrees of rotation, feedback and damping - none of which I would wish to change, although I do have a few doubts about the hard and fast 20 degree off the horizontal for the airvane axis, given that it operates in a dynamic environment. But my doubts are not enough to alter this one, either.
My efforts relate purely in trying to streamline the system's operation within accepted parameters, and the two bugbears I identify to light weather performance are friction and inertia. Jim, thank you for your succinct reply about a heavier counterweight providing a type of feedback. But, if I could build in mechanical feedback to my cable operated system then I could use a lighter weight, which would mean less friction and inertia...
I know, chasing rainbows and all that, but the whole self steering thing is a life long interest, and I like to tinker.

P.S. Have you seen the new micro hydraulic rams developed for the robotics industry? Imagine the beautifully smooth connection they could provide airvane to trim tab. Perhaps on Wndvane construction No.8.

[GILow]

Hello Chris,

I know you have a more recent thread on progress, but after seeing your most recent post and rather excellent construction, I was reading back over your posts and this thread touched on something I had been fiddling around with.

To cut the story short, after reading half a dozen books on the subject I am going for an auxiliary rudder with trim tab, for lots of reasons this would be best for our needs.

I was also looking at the problem of feedback (alluded to in only two of the books) and had come up with a spring. Yes, a simple blade of spring stainless steel mounted in a stationary position in relation to the boat, which presses against the horn that controls the trim tab as the auxiliary rudder swings across to its full travel, thus countering the force of the vane. Downside I can see is that it will exacerbate any friction in the system by resisting the vane, but since that would only happen at the end of the travel I feel this might not matter. Upside is that if got right, spares would be easy to fabricate and carry to deal with loss of spring force over time.

You are well ahead of me on this, my mockups are in still mostly in Lego and plywood scraps at the moment (I am working on a design for the rudder to be able to swing up out of the water when not in use, as it seems one of the most problematic aspects of auxiliary rudders are their effects on manoeuvring, and we have a bit of a wiggle to get into our marina pen) but I would be interested in your thoughts on this, and if you are still concerned about the feedback aspect at all. As only two of the books mentioned what you are discussing, I was not sure if I was worrying too much (as Jim suggests), but both of the books were better than the others so (with deference to Jim's experience), I am still interested in the issue.

Matt

[Chrisc]

hello Matt,
I appreciate your spring idea but wonder is it will provide consistent feedback given that it has to operate in varying wind strengths - would the action of the spring not work back through the linkage to the air vane?
All the cable systems I have seen have the cables and sheaths terminating on the rudder head so no feedback is possible given that the termination points turn with the rudder. Attached a crude drawing of my solution to the problem. If you make a mock-up of this, you will see that by adjusting the cable connection point to the trim tab tiller you can set whatever feedback you like. In my case, 15 degrees of trim tab tiller deviation will turn the rudder 15 degrees, at which point the rudder and trim tab re-align. Jim is probably right in that there is a bit of nit-picking going on here, but I think it worthwhile to try and construct a system as good as I possibly can.
Would be very interested to how your spring develops.

[Capt Rottnest]

Chris, the vane/trim tab system I'm building I'm sticking with the 80% ratio, (trimtab tiller distance/trimtab axis to rudder axis distance). The idea is the trimtab moves 15˚, causing the rudder to move 15˚ but because the cables terminate a small distance behind the rudders axis, they pull back the trimtab to be aligned with the rudder.

As you noted, cables terminating at the rudder axis causes oversteering. Anyway Bill Belcher recommends 80% so I'll see how that goes.

[GILow]

Originally Posted by Chrisc

hello Matt,
I appreciate your spring idea but wonder is it will provide consistent feedback given that it has to operate in varying wind strengths - would the action of the spring not work back through the linkage to the air vane?
All the cable systems I have seen have the cables and sheaths terminating on the rudder head so no feedback is possible given that the termination points turn with the rudder. Attached a crude drawing of my solution to the problem. If you make a mock-up of this, you will see that by adjusting the cable connection point to the trim tab tiller you can set whatever feedback you like. In my case, 15 degrees of trim tab tiller deviation will turn the rudder 15 degrees, at which point the rudder and trim tab re-align. Jim is probably right in that there is a bit of nit-picking going on here, but I think it worthwhile to try and construct a system as good as I possibly can.
Would be very interested to how your spring develops.

Chris,

Yes to all your points, you are way ahead of me on this.

There is a significant difference with my system I realise, and that is that I have based the control arms of my trim tab on the control horns used on aircraft trim tabs, so they stick out sideways from each side of the trim tab (I have considered an option with it sticking out only on one side, and using a push pull arrangement, but I suspect there may be friction problems, that one is still sitting on the "to test" list). Therefore, as the main rudder rotates, there is a degree of "forward - backward" motion available to utilise. However, your point about varied feedback now has me worried.

I realise now that such a setup as I propose is not nearly as visually elegant as yours either (actually, your whole build looks extraordinarily tidy) but being an auxiliary rudder rather than a main rudder, I can get away with a bit more of the ugly stuff. I will sketch what I have and scan it when I next have access to a scanner during the week.

HOWEVER, don't let Jim talk you out of wind steering perfection. We shall blast past him in perfectly straight lines as he meanders along with his elegant "S"'s, and pinch the best anchorages in Tassie before he gets there.

Matt

[Chrisc]

Its taken longer than I anticipated to update the build progress but now the gear is finished and we are a little over a month into a three month cruise around NZ.
The windvane works amazingly well. Mostly the winds have been heavy but I have used it down wind in 8 - 10 knots of breeze with a moderate swell running with excellent results.
The aim of the exercise was to produce a self steering system that would steer the boat as well as an autopilot, ie without the 'S' bends that wind gear tends to produce. There is no real reason to do this other than to see if it could be done.
In this, I have succeeded, but,(and there's always a 'but') the gear works too blooming well!!
I steer an arrow-straight course on all points of sail but the rudder is constantly moving, making tiny corrections to the course. I can see that I'm probably going to chew out the rudder bearings fairly quickly if I let it go on like this.
Better I think to de-sensitize the gear and save the wear and tear. I will post some pictures of the finished installation when the opportunity allows.

[GILow]

A three month cruise around NZ sounds like a serious test. Look forward to seeing the finished result.

Matt

[Steve Pope]

Chrisc, what size teflon outer did you use and where did you source it. I made a similar setup based roughly on the auto-helm using 1.2mm SS wire push-pull system, unfortunately the wire wasn't up to the job. I have been thinking along your lines with dyneema, but haven't been able to find a suitable teflon outer. A problem I had with the wire was keeping an even tension on the wire, the non working wire would flex and after several days would break at the upper connection.

[Chrisc]

Hello Steve, I started off using spectra line in a nylon sleeve for the control lines but was a bit concerned about the chafe and friction. Now I am using bog-standard Shimano bicycle brake cables and sheaths. The rationale for this is that this stuff is really cheap so if it should prove to be an expendable item then you can carry spares - brake cables come in a variety of standard lengths, each with a small ferrule soldered to one end and the sheathing you can buy by the mile. In reality, I have not yet had to replace any of the actuating cables but I did take a lot of care over their placement. The H28 is a well mannered old lady and very light on the helm. When self steering I adjust my main so that I have slight weather helm and then only use the windward cable connected to the trimtab.

[Steve Pope]

Hi Chrisc, havn't been down to Whangamata for a year or three, I did sit out a 3 day blow a few years back on a mooring just west of the the ferry crossing, my passengers got sea sick from the swell coming in though.
I had thought about using bike cables and tried to find if there was a ss outer and inner available, no luck. I did get 2mm id. bike inner teflon tubing from H S White & son to run the 1.2 ss wire through, which I then ran the teflon through a heavy wall plastic outer. This lead to the problem that the outer could / would compress as the load came on altering the length of slack 1.2 wire, this therefore had a limited life and after repairing it twice on the way to Tonga / Fiji meant it wasn't good enough for serious short handed cruising.
I would like a set up that will last a lot longer than I have experienced so far.
Do the inner and outer brake cabling that you are using show any signs of rusting or deterioration so far?

[GILow]

Steve, Campagnolo (bike components manufacturer from Italy) make a very high quality brake cable that will do what you ask. It is what I have used for similar projects in the past and it is bullet proof. I have yet to wear one out.

Matt

[Steve Pope]

Thanks GILow, I will see what is available here in NZ I would imagine Campagnolo stuff would be available.

Originally Posted by GILow

Steve, Campagnolo (bike components manufacturer from Italy) make a very high quality brake cable that will do what you ask. It is what I have used for similar projects in the past and it is bullet proof. I have yet to wear one out.

Matt