Whipstaff, my way

Footy Class General Discussion (Footy)
1

“Niel, I saw the whipstaff arrangement in your boat, and liked it. I don’t recall, though, what material you made the various bits from…nor did I see how you attached the tiller to the rudder post. Any advise for those of us who might like to make one?”

Thanks…Bill H

Bill and all, following are the specs and some pics of the whipstaff rudder linkage that I use on “Tanto”. This system worked better than I had expected. Steering is very smooth and I have not had any difficulty turning the boat in adverse conditions even though Tanto sports a rather large rudder.

The geometry of a steering system is important to understand, so a comparison of the ubiquitous push/pull linkage and a whipstaff one is the place to start. First, whipstaff steering involves a tiller arm extending from the rudder shaft that engages with a pin and slot arrangement to an arm extending from controlling mechanism. Whipstaff, or tiller steering, was the common linkage in vane sailing. When model yachting changed over to radio control the push pull linkage became the norm for two reasons; first, the availability of off the shelf parts from r/c aircraft, and second, the large distance on most model yacht classes from centrally mounted electronics and the rudder well aft.

The push/pull system converts rotary motion (at the servo) to linear motion (connecting rod) and then back to rotary motion (rudder arm). Because the neutral position of the rudder servo is at 90 degrees to the direction of travel of the connecting rod the rotary motion of the servo moves the rod in an arc, with the linear, fore or aft travel of the rod diminishing the further the servo moves from neutral. What this means for you and I is that with the push/pull linkage most of the steering leverage occurs with little movement of the rudder servo and as one calls for more steering at the transmitter less leverage becomes available because of the diminished rod movement at the end of the rudder servo’s arc. This is why so many sailors, particularly new ones, over steer.

Whipstaff steering is a rotary motion to rotary motion linkage. As such, the rudder input from the servo is output directly to rudder making the steering effect more even across the turning arc. Because the engaging pin from one arm (in Tanto’s case the servo) slides in a slot in the other there is linear motion, only this motion is across the hull at the intersection of the turning arcs of the two arms.

My version of whipstaff steering uses the following:

Servo Arm - 2-1/2" long balsa core/carbon sandwich arm w/ pin (5/8ths, 2-56 bolt w/ section of 11.5GA S/S hypo tubing as sleeve), 2-3/16ths from servo pivot to center of pin.

Rudder Arm - 1/16th Delrin, 5/16ths wide by 3-1/16th long, slot is 1" long by shy 1/8th wide, starting 1-5/8ths inches from rudder shaft, ending 2-58ths from rudder shaft, w/ 7/32nd diam. brass rod drilled for 3/32nd diam. rudder shaft and drilled crosswise and tapped for 4-40 set screw, silver soldered to a 1/64th thick brass plate w/ clearance holes for 2/56 bolts tapped into mounting holes in the Delrin arm. Delrin was used because it is a bearing material. It is important that there be very tight tolerance in the slot to pin interface. The pin must slide easily in the slot but not be a sloppy fit or the rudder might vibrate. Expect to replace the rudder arm when it gets warn.

The distance from the rudder shaft to the rudder servo pivot is just over 4 inches. At neutral the center to center distance from the pin on the servo arm and the rudder shaft is 1.800 inches or just under 1-13/16ths.

I find that the even steering of my whipstaff system superior to the other systems I’ve tried. It is ideally suited to the Footy because the distances between the r/c guts and the rudder shaft are so compressed. It was used in the vane sailing days because it was easy to set up and understand. Push/pull systems would have worked with a vane but were not used. I wonder why?

If any of you decide to try a whipstaff linkage I hope that this has been of help. I credit Angus with re-introducing me to this method. If one searches past posts there are several other photos and explanations of whipstaff systems.

Good steering - Niel

2

Thanks very much for the complement and the superb explanation of why whipstaffs are good. I personally can’t see that there is a lot of excuse for any other arrangement in a Footy, so long as you have the gear available to make the rather fiddly slide mechanism.

Being ever a pernickity bugger, I would point out that (given a suitable slide mechanism), you can have any angle you like between the pivot axis of the rudder and the servo shaft. In vane systems the two axes were more or less parallel. Most of mine have had them more or less at right angles with the servo mounted horizontally under the tiller with its ‘back’ to the transom. This makes the system incredibly compact.

Startt whipping some staffs, people!

:devil3::graduate::graduate::zbeer::zbeer:

3

Very clean servo set up. I likee.

The whipstaff steering looks the business. I will have to try it out.

:zbeer:

4

Nigel, thanks for the compliment on the removable r/c board. The pin and slot connection of my whipstaff linkage is the quickest way to disconnect the rudder servo from rudder arm, making swapping of the electronics out and back in a snap. You can refer to my post #18 in the “Servo Question” thread for my rational for being able to change r/c boards quickly.

5

Neil - thank you for going into such useful detail. May I ask a couple of simple questions

Where does one source Delrin, and come to that what is it?

And how did you go about cutting that beautifully even slot?

Can do with wood and carbon fibre but am only gradually learning other materials

Many thanks

Cheers

Charles

Five Footys (one sunk!)

6

I have attached some photos of a whipstaff fitted to a 507 by Keven Jackson for me that i thought might be of interest

7

It’s very simple but two things worry me:

  • the servo arm tracks an arc - does this bend the tiller wire at it’s extremes or is the tip of the servo arm trying to slide on a shaft?
  • what adjustments are possible in the rudder’s throw, apart from moving the servo?
8

Hi Charles,

Delrin or Acetal is a plastic bearing material, somewhat harder that Teflon. I had some 1/6th thick sheet left over from a project many moons ago. You may be able to source some from an industrial supply company. They would probably be willing to send you a sample. Another option would be an after market odd lot outlet.

Cutting an accurate slot is very important. It must match the pin in your system with minimum clearance. The method that I use is simple but requires patience because you sort of sneak up on the final dimension. First, drill pilot holes a tad smaller than the slot width at both ends of the slot you want. With Delrin or Teflon it is a good idea to create a “centerpunch” type location to start your drill hole using an awl or the point of an x-acto knife rotated about. This helps to keep the drill from wandering while it cuts through the material.

Once you have drilled the holes then cut a groove with the x-acto knife from the edge of one hole to the edge of the other hole. Then do the same for the other side of the slot. Then you can either keep cutting the groove deeper with the x-acto or, if you have some experience with a jeweler’s saw cut the slot sides with it. The grooves will act as guides and help the saw track, cutting the thinner material in the grooves. Once you have the rough slot cut out it can be refined to final dimension with jeweler’s files.

One last thing, I would start by cutting the slot in the sheet before cutting out the outline of the tiller arm. This way, if you screw up the slot (and if you duplicate my experience then you probably will the first time) then you haven’t invested your time in ruining a beautiful arm. It is also easier to hold a sheet while cutting the slot than a narrow strip of plastic.

Good luck, and please post your linkage when you have it finished.

9

Rusty Nail,

I believe that in Andy’s version the post connected to the servo both rotates and telescopes. It is an interesting solution that eliminates the slot in the linkage. But, my interpretation is a function of being able to remove the r/c components quickly in case of a breakdown. Andy’s rudder servo is mounted securely in the boat and I think would take some time to remove, but it is more compact. I am also not a big fan of having a slot in the transom, but his system would work just as well for an under-the-boat mounted rudder.

In any whipstaff system the ratio of the two arms determine the rudder throw. In most push/pull systems folks opt for a 1-to-1 linkage between the rudder arm and the tiller arm. That seems to be standard although there are some push/pull set ups I’ve seen where the servo arm is longer than than the rudder arm which gives the rudder more throw. In a whipstaff system the opposite is true, a longer servo arm will give you more throw.

The average servo’s throw is about 60 degrees, half on either side of neutral, or the centered position of the rudder. Most digital radios allow you to adjust this throw so getting the amount of throw that you want mechanically is no longer as critical. If you have an older r/c system the amount of rudder throw in a whipstaff linkage can be worked out on paper fairly easily. You need the distance between the servo’s pivot and the rudder’s pivot as starting points. You should draw a straight line between these points. Then from the servo’s side you draw a line from the pivot point 30 degrees off the line between the two pivot points. Then, from the rudder’s pivot point draw a line corresponding to the desired rudder throw. The intersection of the two lines is the location of the pin (assuming that the pin is on the servo arm) and the outer end of the slot on the rudder arm at the desired rudder throw. The back end of the slot (closer to the rudder) can be found by measuring the distance from the servo’s pivot point to the intersection of the two angled lines and transferring that distance to the straight line between the two pivot points.

One caveat that needs pointing out, make sure that the linkage you design will fit inside your boat. If there is a great distance between the servo and the rudder then the arcs the linkage must take to provide the rudder throw will also be large. After all, you wouldn’t want to get the system installed only to find that the arms swing out and punch a hole in the side of your boat!

10

Interesting thread:D

Flavio uses a mechanism (at least in Folgore) similar to Neil’s in the first post
This arrangement of servo and tiller requires quite a long slot in the tiller arm to accomodate the sliding motion.

Hence, no doubt, History Man’s question about parallelism of the slot.

While thinking about the inherent geometry of this setup I imagined a version which has very much less relative motion, and hence requires a shorter slot

If I had to produce such a slot which would give no play or binding, and since I don’t have access to machine tools or routers I would “assemble” it from two “sides” with nice straight edges and bond them to the basic tiller so that the slot is parallel, smooth and the right width.

Material: Delrin and acetal are good, I would be happy also with plywood, PCB material or nylon (for the sliding sides) how about an offcut from a Tywrap?

andrew
The staff were not happy about being whipped, even tho I introduced it as an incentive regime:mad:

11

Due to the fact that my own version of whipstaff ( on board Folgore ) has been so kindly shown as example , I wish to add more detail about it as well the results of my own “scientific” investigation.

As first approach I have done several sketches aimed to understand the effect of geometry modifications ( tiller length and max angle, as well servo horm deflection and length too )

I have built a preliminary arrangement to make a “dry test” in order to check the clearance between hull and sticks.
To be able to investigate - quickly - geometry modifications on first wooden servo horn I have made several holes in order to be able to move the pin at different distances from the axis

As third step, I have done a simple excel spreadsheet to have a better understanding of the system

and these are the most intersting results :

1 - There is a remarkable "gain effect"In other words, when the rudder is moving port and stbd from the center, rudder angle increase is almost double of servo angle.
Amplification is dropping to about 1.5 at large angles of rudder ( around 30°)

2 - As side effect due to the fact that servo torque is ( or it should be ) constant at various angles, torque on rudder blade is increasing from small to large angles.

Conclusions

Whipstaff is so sensitive that , it is possible to use only trim buttons for steering, using sticks only for tacks and jibes .

On models bigger than footys, the incrasing torque effect at large rudder angles could be a method to use a smaller and lighter servo

Last but not least, it is an easy removable system

Flavio

12

When Angus introduced this subject back in 2006, one of the features that he was very keen on was the ability for a Whipstaff to be relatively insensitive to servo movement close to the neutral position, progressively giving a faster response as the Tx stick was moved further.

Flavio’s chart of servo against rudder rotation illustrates perfectly that this doesn’t actually happen with the set-ups shown - there is a gradual falling off the the rudder response with increased servo rotation. The fact that on Flavios setup there is initially more rudder rotation than servo rotation is simply a function of the relative position of the actuating pin to the distance between the rudder stock and the servo spindle. If the pin is on the servo arm and is closer to the rudder than the servo then the magnification effect will be large. The opposite will happen if the pin is closer to the servo.

My tests have just used bits of card and drawing pins but show quite clearly that the best sytem is where the tiller and servo arms are parallel and square to the rudder blade and the link rod between servo and rudder is roughly square to both arms. The amount of rudder rotation is tweeked by adjusting the relative lengths of the two arms. If the arms are the same length then rudder is roughly proportional to servo up to about 40 degrees, and if the servo arm is 75% of the rudder tiller then 40 degrees of servo will give about 30 degrees of rudder. This is ideal for a yacht where the received wisdom is that rudder movement should be no more than 30 degrees and a Hitek HS55 servo gives +/- 35 degrees.

What Neil and Flavio’s Whipstaff does give is the ability to mount the rudder stock internally and very close to the transom which is not possible with the conventional systems.

I’m sorry if all this is a bit wordy, but I’m not too good at including attachments to the Forum messages - if you would like to see the spreadsheet data and plots just forward your E-Mail address, I might even be able to photograph my bits of cardboard.

Cheers,

firstfooty

13

I hope these attachments reproduce O.K. The spreadsheet text didn’t look too clear on the preview - not up to Flavio’s standard I’m afraid, but . . . .

However unclear the images they illustrate what I mean by In Line and Offset more clearly.

The Whipstaff arrangement most closely represents Neil and Flavio’s layouts. In practise it is effectively the same as Angus’ as shown in AndyT’s post, just rotated into the same plane.

The data in the spreadsheet is just the measured results from the rig test. Although I could get Excel to calculate the Whipstaff numbers I can’t do the same for the other two systems. I think my little grey cells are wearing out.

One of the more interesting points to note about the conventional systems is the way that you can get different results on Port and Starboard. I think this must be put down to the effects of the varying connecting link angles.

You can also see the falling off of response shown by the Whipstaff design as noticed by Flavio. This doesn’t seem to happen with the other designs until much larger angles, by which time you get a stiffening of the links due to over-centre effects.

So - unless you particularly want to reproduce the effect of sitting in a dingy and swinging your arm about then I wouldn’t think the Whipstaff is the way to go. Even low cost 2 Channel radios have servo reversing switches these days, so if you feel the need to move the Tx stick to the right when you actually want to turn left then it is a trivial correction to make. In the final analysis at least both the traditional layouts give roughly proportional responses up to the sort of angles we are looking for, but the In Line system must win out by having equal responses on both Port and Starboard tacks.

Cheers,

firstfooty

14

Flavio’s system, while beautifully executed, has much different proportions than mine. In my setup the neutral position the pin is roughly centered between the two pivots so the arcs that they move through are about the same. The distance between the rudder pivot and the servo pivot is 4 inches and the slot length that I use is only one inch long.

When the servo arm/pin location is moved closer to the rudder pivot the result is more radical turning closer to neutral and a proportionately larger rudder throw. You would also need a longer slot like Flavio is using. The opposite would take place if the servo arm/pin location were to be set up closer to the servo, a dampening effect and smaller overall rudder throw. I can see how FirstFooty would characterize Flavio’s setup as a dingy setup, but that may be more familiar to new guys who come from sit-on sailboats.

By the way, the “conventional” push/pull system that FirstFooty prefers is actually an adaption of “Norwegian steering” that was used on boats with mizzen masts that were too small to justify a wheel and cable steering system. In these boats the tiller arm that the skipper used could slide side to side (for use on different tacks) in a sleeve mounted on a pivot attached to the deck just aft of his sitting position. Attached to the pivot’s small arm was a push/pull connector shaft that passed around the mizzen to an arm on the rudder. Sounds kind of “old school” doesn’t it?

Anyway, for practical purposes (that is steering feel in real life situations) I like my set up a lot. It “feels” more even where the push/pull systems I used to use felt twitchy when you just move the transmitter stick a little and soft when you need steering the most, in tacking or strong winds. As a matter of fact, I gave up on the push/pull system a long time ago. For all my larger class boats I use drum and cable steering which gives the most proportional control of any steering system (it also doesn’t require that the servo and rudder shaft be in any way aligned). My whipstaff steering feels very much the same even though my Tanto is only 1/3 the length of my 36/600.

15

If you are fortunate enough to own 2.4ghz then you can use +/- proportional movement as well as dual rate. So with the advent of the TurboRix £31.17 TX&RX and its PC programmability its as cheap as a normal 27Mhz kit bit of a no brainer

16

Hi
This thread is providing lots of new ideas on how to set up an internal control system for a rudder which is located as far back inside the hull as one can achieve. I’ve managed to follow all of the ideas up to the last post, however, I’ve never heard of ‘FlySky’. Please can you give me some more information or links to where I can find out more?
thanks
Phil

17

Here you are http://www.r2hobbies.com/products.php?cat=71 unfortunately the price has just gone up to £31.17 for the 2.4Ghz transmitter and receiver ~ Amazing value for a 6 channel set that includes the programming lead and software Its actually called TurboRix (sorry got the name wrong) additional receivers http://www.r2hobbies.com/products.php?cat=23 are £12.12 and you will have to add the p&P

18

Why not drive the rudder shaft directly off a rotary actuator??

The web link below shows some tiny lightweight linear and rotary actuators based on muscle wires.

At 7 quid these are so cheap that I may have a go with them. They do not accept a servo input from a R/C receiver but instead just simple switches.

http://www.active-robots.com/products/motorsandwheels/linear-actuator.shtml

(Thanks to Lawrence for the link)

19

Somewhere (I can’t remember where, but somewhere on this forum) there are some drawings of the last generation Jackson/Richardson whipstaff system with servo and rudder axes at 90 degrees. This uses a telescoping mechanism on both tiller and whipstaff - easily achieved by using pultruded (for exact tolerances) carbon rods and tubes for the two slide mechanisms. Keven Jackson’s great contribution was to realise that, instead of the rather elaborate designs that I had been faffing about with , the whipstaff-tiller coupling could easily be made out of a piece of fuel tube. Only special skill - a reasonably steady hand, a pin chuck and a 0.8 mm twist drill.

This probably makes it easier to make without excessive play than a pushrod system, it is much more compact, lighter and does not have the not uncommon system of pushrod (as opposed to yoke) systems with long pushrods of the rudder over-centring with consequent loss of control and burnt-out servos.

Of course neither you nor I would ever launch a boat, or even test her dry, with the steering gear in such a lamentable state. I also believe in flying pigs, werewolves and the Eternal Goodness of Tony Blair.

Cheers.

:zbeer::zbeer::devil3::graduate::graduate:

20

Angus,

Is this what you were referring to http://s111.photobucket.com/albums/n138/angusrichardson/whipstaff/

This was part of the thread you started back in 2006 with http://www.rcsailing.net/forum1/showthread.php?t=3521

The trouble is, as Flavio’s and my own work has shown, your main aim in using a whipsatff to get “a relatively coarse input on the stick to have only a small affect on the rudder so that we do not start to zigzag all over the place.” isn’t supplied by the systems discussed here - neither Neil’s nor Andrew Jackson’s. They both demonstrate an noticable tendancy for diminishing rudder response with increasing servo application. Kevin and Neil’s systems are effectively identical except Neil’s rudder and servo spindles are parallel whereas Kevin’s are perpendicular to one another - but the end results are the same.

The simplest - if not the cheapest - way would be to use the method mentioned by Andy Trewin with a computer controlled transmitter which would allow you to dial in virtually any form of response your heart desires, whether you yearn to turn to the left when you actually move to the right or the possibly more useful sensitivity adjustments.

Cheers,

firstfooty