Sheet to tiller self steering, windvanes, and autopilots

<< First  < Prev   1   2   3   4   5   ...   Next >  Last >> 
  • 03 Feb 2017 09:19
    Reply # 4585018 on 1493031
    So... with wind and sea up, we did the best approximation of Arne's idea without bungee -  bias the vane and servo off-centre to give the vane a constant force to act against, and not let it have to deal with a 'sign change'. That helped quite a bit. David's suggestion of gearing down the servo/tiller linkage helped, but there was quite a compromise between damping oscillations and getting sufficient tiller motion. I geared up the vane/servo linkage to compensate. I also shimmed the rudder stock, which had a little slop. Net result: oscillations of up to 25 degrees, in fair sized quartering seas and wind just shy of taking a first reef. 

    I'll head into Whangarei tomorrow for sheaves/bearings/new blocks etc. and see if that helps...

  • 02 Feb 2017 19:44
    Reply # 4583951 on 1493031
    ...and in perfect conditions of steady, moderate winds and totally flat seas, with the vane buttoned up as well as I can manage and connected to the nose of the tiller, we're seeing the same rhythmic oversteer. The amplitude is reduced, but the period is the same. :-( 

    There is no sign of damping as the boat yaws through the vane vertical/ servo vertical/ tiller centered position.

    Last modified: 02 Feb 2017 19:47 | Anonymous member
  • 02 Feb 2017 00:39
    Reply # 4581739 on 4581459
    Annie Hill wrote:
    Bryan Tuffnell wrote:Fantail is one of those boats that I think a beginner would tend to oversteer on a run. The helm does load and she naturally yaws when running, and a steady grasp of the tiller resists this and holds her straight.

    I'll take your word for that - I have to confess that I always found her easy to steer regardless of the point of sailing and hadn't noticed any tendency to yaw excessively.  But then, a dainty fin-and-spade boat, with junk rig is something of a contrast to a hulking great steel brute, with a long keel!  BTW, you do reef early, don't you?  So many people who haven't sailed junk before hang on to sail downwind because of the novelty of being able to go fast in the direction that they want without having to add extra sails.  But a reef or two often makes things easier without noticeably reducing speed.

    Annie, has the vane ever operated successfully on a run?
    Only in a flat sea and/or light conditions.  It was pretty good in Tasman Bay, but on this coast struggles a lot more.  I guess this would confirm the boat's tendency to yaw, with the then concomitant over-correction by the wind vane.  I did, once or twice, slack off the lines on the chain which sometimes helped and is a crude way of achieving what David suggests: moving the lines further forward.
    Annie, I agree that Fantail is easy to steer on any point of sailing, with one proviso. I think you, and anyone with a bit of experience behind them, would know that when running you turn the tiller only to keep the heading constant. In my limited experience, with a flat sea and steady breeze, the helm can be left briefly even when running; there is no great tendency to yaw. However if either sea or wind fluctuates and the heading is maintained, a gentle but definite weather helm develops. I think (and have recently seen) that some people have a tendency to turn the tiller in response to rudder pressure, rather than heading change, and I feel that it's chasing the constant tiller pressure that causes oversteer. Pure speculation on my part, but I have a feeling that this is the key point that makes both a beginner and wind vane more challenged when running when conditions are anything other than very good (both need to ignore feedback from rudder pressure). I'm wondering if Fantail's short coupling and large rudder develops that little tiller pressure when she wants to yaw; the after edge of the rudder traveling in a greater arc than the leading edge. Maybe?

    We've got strong, gusty, choppy conditions today and it's a bear to try Arne and David's suggestions... particularly as I can't find any bungee aboard. Hopefully things will steady and I'll have something meaningful to report soon.

    Gak. The nearest automated weather station is report 38 knots. Perhaps tomorrow?

    Last modified: 02 Feb 2017 06:57 | Anonymous member
  • 01 Feb 2017 20:14
    Reply # 4581459 on 4578219
    Bryan Tuffnell wrote:Fantail is one of those boats that I think a beginner would tend to oversteer on a run. The helm does load and she naturally yaws when running, and a steady grasp of the tiller resists this and holds her straight.

    I'll take your word for that - I have to confess that I always found her easy to steer regardless of the point of sailing and hadn't noticed any tendency to yaw excessively.  But then, a dainty fin-and-spade boat, with junk rig is something of a contrast to a hulking great steel brute, with a long keel!  BTW, you do reef early, don't you?  So many people who haven't sailed junk before hang on to sail downwind because of the novelty of being able to go fast in the direction that they want without having to add extra sails.  But a reef or two often makes things easier without noticeably reducing speed.

    Annie, has the vane ever operated successfully on a run?
    Only in a flat sea and/or light conditions.  It was pretty good in Tasman Bay, but on this coast struggles a lot more.  I guess this would confirm the boat's tendency to yaw, with the then concomitant over-correction by the wind vane.  I did, once or twice, slack off the lines on the chain which sometimes helped and is a crude way of achieving what David suggests: moving the lines further forward.
  • 31 Jan 2017 23:41
    Reply # 4578560 on 4578392
    David Tyler wrote:

    Bryan, you're 180 degrees wrong about the geometry. This servo worked on Tystie, too well. It is one of the cornerstones of servo pendulum design that when the servo tiller is pulled out one way, the servo blade follows it out the same way, thus putting in the necessary negative feedback. So, if the servo tiller is facing forwards, it must be below the power axis. 

    Darn, that's twice now! David, I understand that the servo blade and its tiller must operate in the same sense to damp its response and limit it's actions. 

    Where I'm confused is when the boat yaws. Say the stern yaws to port. The servo would sense this, swinging to starboard and correctly turning the tiller to starboard if left to itself. However, because the rudder isn't fully balanced it senses the yaw, and turns the tiller and by extension swings the servo. But, because the vane is doing its job it has also sensed the yaw to to port and has tumbled to starboard, holding the servo tiller, so the angle of attack on the servo is now acting against the desired starboard swing of the servo, trying to turn it to port instead, positively coupling to the original yaw to port. If this happens, this action would compete with the correct function of the unit as a whole. In other words, does feedback from the rudder moving due to yaw, which Fantail's does, damp the servo's yaw response (out of existence, possibly)? Does this make sense? This is what I really want to be wrong about, because the solution to that would be to put the servo tiller above the pendulum axis (or reverse its direction), which would compete with the servo's need for damping.

    I'll try both shifting the tiller attachment point and the bungee - thanks guys.

    Here's that photo:

    http://www.junkrigassociation.org/Sys/PublicProfile/30361922/Photo/

    61733681/61733710/0?dh=0&cppr=0



    Last modified: 01 Feb 2017 02:02 | Anonymous member
  • 31 Jan 2017 22:00
    Reply # 4578392 on 1493031

    Bryan, you're 180 degrees wrong about the geometry. This servo worked on Tystie, too well. It is one of the cornerstones of servo pendulum design that when the servo tiller is pulled out one way, the servo blade follows it out the same way, thus putting in the necessary negative feedback. So, if the servo tiller is facing forwards, it must be below the power axis. The amount that it below the power axis versus the distance of the end from the servo vertical axis determines how much negative feedback there is. The servo tiller can also be above the power axis, in which case it must face aft. 

    Have a look at this photo of the current servos on Tystie. The same basic geometry, except that the power axis angle is 30, not 45. The mass distribution is better, and the friction is better, but these mark 2 servos work in exactly the same way as the mark 1 servo that you have.

    The problem you have is to do with simple harmonic motion. The period of the vane gear is the same as the period of the boat, so they are oscillating in sync. Try what I suggested: move the lines out to a greater radius on the tiller. This will get them out of sync.

    I'm waiting on that side-on photo to see whether the gear is in any way different from how I made it. I'm almost convinced that it used to work on a run, but my memory ain't what it used to be.

  • 31 Jan 2017 21:25
    Reply # 4578338 on 1493031
    Anonymous member (Administrator)

    When reading about this, I understand that Fantail has the luxury problem of too powerful vindvane gear  -  quite the opposite of the problems I had when steering my 23’ Malena with a direct drive Belcher gear:

    (see JRA Magazine 51 p.16 – 18.)

    To get the straightest course upwind in uneven wind strength, Belcher ‘s advise was to set the counterweight low. In that case the weight itself would help to pull the tiller “up” to bear away in case a wind gust made the boat heel over (and increase weather helm).

    When running before, Belcher’s advice was to move the counterweight up to make the vane top heavy. This was to help keeping a straight course with the boat rolling. The vane thus tried to make the boat turn to the side it rolled (..or at least to try to hinder the opposite from happening...), even before it had started to round up. This worked surprisingly well, as long as I kept the sailplan-to-hull balance right.

    Another armchair idea for Fantail:
    How would it be to use bungees on both sides of the tiller to make it want more to stay in centre?

    Arne

    PS: ..and yes, I repeat my suggestion of making artificial weather helm with a bungee. Nothing could be simpler than trying that...

     

    Last modified: 01 Feb 2017 21:29 | Anonymous member (Administrator)
  • 31 Jan 2017 20:48
    Reply # 4578228 on 1493031
    Ah - David, your reply came in while I was typing mine. Thanks, and yes, you're right about the blocks, and there's certainly scope to remove some of the friction in the system. It's a bit of a trick to balance friction and backlash in the setup, but there's opportunities to improve both.

    I do see the servo blade operates in the wrong sense though and still wonder if ultimately a redesign is required. Any thoughts?

    Last modified: 31 Jan 2017 21:05 | Anonymous member
  • 31 Jan 2017 20:41
    Reply # 4578219 on 1493031
    Annie and Arne, i have a feeling that you're both right about the vane. I think there's a bit that could be done to improve the vane's performance but I'm wondering if its design doesn't suit the boat. Fantail is one of those boats that I think a beginner would tend to oversteer on a run. The helm does load and she naturally yaws when running, and a steady grasp of the tiller resists this and holds her straight. 

    I'm inclined to think that the scalloping downwind is due to a combination of Fantail's behaviour and the design of the servo. Once Fantail begins yawing, the servo blade swings to the wrong side to correct this. I'm guessing that the reason is because the angle of attack of the servo blade, which is set by the combination of the vane angle and feedback from the rudder, overpowers the servo blade's tendency to swing due to the yawing action of the hull. My guess is that the motion of the tiller feeds back through servo, which is resisted by the vane, resulting in the pitch of the servo blade being set in the wrong direction. All this happening in concert with the correct action of the system makes it hard to analyse.

    Annie, has the vane ever operated successfully on a run? To my untrained eye it looks as if the servo tiller needs to be above the rotating axis of the servo in order to prevent the tiller feeding the servo blade pitch incorrectly when running. I'd like to be wrong about this as it would be a major redesign to correct. 

    I'd really like to hear from David and anyone else who has ideas...


  • 31 Jan 2017 20:28
    Reply # 4578195 on 1493031

    OK, the video is showing the classic rhythmic oversteer that is normally due to lack of negative feedback, but can also arise due to other causes:

    1. Friction - check all bearings, main rudder included, and get rid of it.
    2. Backlash/lost motion: Arne has a good point, on other points of sail but a dead run, the gear is generally only applying helm in one direction, but on a dead run, it is having to work in both directions. There were  two blocks in the primary linkage, turning the lines through 90 degrees, that tend to nod to and fro causing lost motion. These were meant to be temporary, and to be replaced by fixed sheaves, but have maybe become "temporarily permanent". Biassing the gear with a bungee is worth trying.
    3. Inertia and mass balance: we can't do much about inertia, but we can make sure that everything is working towards a mild self-centring effect. Vane bottom-heavy, servo blade front-heavy.
    Annie is right: a balanced spade rudder is the worst type to try to drive with a pendulum, as it is so easy to overpower. This was the case on Blondie's boat Pilmer, a Kingfisher 20. We could not get a Hasler SP pendulum gear to work unless we fixed the steering lines to the tiller at three times the normal radius (another thing worth trying), though of course, this only gave one third of the normal deflection, and the vane gear could not be used to tack the boat. This was partly due to inertia, as well as the balanced rudder - the heavy plywood vane was slow to keep up with the speed of turning of the boat. We invented what Blondie called a "gate gear" (because it swung around a vertical axis), and this did the trick, but it never went into commercial production. The 45 degree power axis of the gear that is now on Fantail is headed in this direction, but is not as aggressive in its self-centring effect. Nevertheless, these were too aggressive for use on Tystie, and were replaced by servos with an angle of 30 degrees - still very effective at yaw damping, but with a wider range of helm angles than the 45 degree servos.

    There's a whole chapter in John Letcher's book Self Steering for Sailing Craft that discusses all of this in detail.

<< First  < Prev   1   2   3   4   5   ...   Next >  Last >> 
       " ...there is nothing - absolutely nothing - half so much worth doing as simply messing about in junk-rigged boats" 
                                                               - the Chinese Water Rat

                                                              Site contents © the Junk Rig Association and/or individual authors

Powered by Wild Apricot Membership Software