Swan 57 Ketch Autopilot

As owner of S&S Swan57 CB Ketch Lianda Hull #40, we noticed that when under sail at medium conditions of sea & wind from reaching to downwind, the autopilot is not coping, disconnecting very quickly. We wonder if this is normal & this system was made to work only in very light conditions, motoring.

 Is it possible to adjust or reinforce this original system?

Thanks for your advise,

Merry Christmas to you all

Marc Busschots

Hi Marc,
in order to help I need more information on your pilot and a precise description of the problem. First of all: is it a Neco?
Whatever kind or brand, it is very unlikely that a sensible person deliberately installed in your beautiful boat an autopilot which is only good for light wind conditions or for motoring. The most likely case (in particular if it is a Neco) is that there is a calibration problem or, in the worst case, a malfunctioning. Both cases can be fixed.

Daniel, 411/004

Hi Marc,
in order to help I need more information on your pilot and a precise description of the problem. First of all: is it a Neco?
Whatever kind or brand, it is very unlikely that a sensible person deliberately installed in your beautiful boat an autopilot which is only good for light wind conditions or for motoring. The most likely case (in particular if it is a Neco) is that there is a calibration problem or, in the worst case, a malfunctioning. Both cases can be fixed.

Daniel, 411/004

Dear Daniel,

Thanks for your quick reaction.

We have on Lianda a Neco autopilot with B&G Hydra pilot control.

We sailed her all summer using the autopilot without problem in the protected waters & light winds of new England.Just noticed that it will take some time like 30 sec to accept a new course; For instance if you motor  on a course of 360° then you alter your course to let's say 90° , you'll have to wait for a minute or so prior to switch the autopilot on steering the new course, if not it will bring her back to her previous course, but once on the course the Pilot will hold the course very precisely.

 Sailing down the US coast towards Florida, we encountered more sea & wind, mainly beamreaching ( a joy!). But the autopilot will disconnect after 30 seconds,we hear cleary the sound of what we believe is the relay switch. We always trim the sails to have the lightest helm possible, the steering gear & autopilot motor are perfectly clean & dry & look in very good order.The helm is light;in fact the Autopilot is not coping with any sea more than a calm sea. do you need more infos ?

Thanks for your time & help

Marc

Lianda 57/40

Marc,
I am afraid that we will enter in some very technical discussion regarding a particular problem and it might not be of interest in this forum. I suggest we exchange e-mail addresses by private message.
I get that your system is made of a Neco Drive Motor and a BG Central Unit (Controller) with its compass and rudder reference.
The slow response (30 sec for changing course) is by itself a sign of malfunction and should be fixed. It is likely but not sure that the same agent that causes the slow response might cause the other problem.
I can surely help you in finding if the problem is in the Neco Drive and, if so, I will suggest how to fix it. I will not be able to help you to fix the BG itself but I might give you some indications.
One point I need to know is: what happens to the rudder wheel when you notice the slow response or the `not coping` situation. Does the autopilot disengage the clutch so that you can actually manually steer?

Daniel, 411/004

Marc and Daniel
The maximum rudder torque to expect on a Swan 57 is about 4 kNm or 2950 ftlbs. The autopilot is normally not sized to cope with this, but to handle about 2/3 of this torque. Hope this is helpful for your investigation
Best regards
Lars

Thanks for your help, professor,

Merry christmas from Lianda, capt & crew

Marc

Lars,
thank you for your input.
I browsed the characteristics of the Neco autoplilot and found the following.
The torque at the output shaft of the Neco Drive Unit 17DR8 (the most powerful Neco servo motor at 24V) is 20 lbf ft while the recommended HO-to-HO turns are 4.5.
Assuming that the HO-to-HO angle at the rudder is 90 degree, the resulting overall torque ratio between the Drive shaft and the rudder shaft is 1:18.
According to this simple calculations the Neco Drive would be able to provide a torque at the rudder of approximately 360 lbf ft (488 Nm) which appears to be much lower than 2/3 of the one you cite.
Is there something I overlooked?

Daniel, 411/004

Daniel
Pls note that there is also a reduction between the autopilot drive and the steering wheel shaft. I have the information that there are 11 teeth on the drive sprocket, and 48 teeth on the wheel shaft sprocket, it would be good to confirm this.
This affects your calculation.
Best regards
Lars

Dear Lars,
I appreciate very much your interest and help.
I know about the ratio you mention and it was indeed taken care in my calculations. I followed the indication specified in the original Neco Marine Autopilots leaflet
http://styret.no-ip.org/docs/Neco%20Marine%20Autopilots.pdf
which gives simply the recommended HO-to-HO turns of the Neco Drive sprocket and then I assumed a 90 degree swing of the rudder blade. A swing of 80 degree is probably more likely but this does not change much because I am still missing a factor of ten or so!
I would appreciate if you could look at the Neco leaflet and comment. You will indeed find there a more powerful motor but I understand that it is not the most common and surely it is not like mine.

Daniel, 411/004

Daniel
It appears we have a mystery to solve here.
Suggest that we first look at the steering system numbers.

For the Swan 57, the wheel turns from hard over to hard over are 2.47, and with the sprockets 48/11 the autopilot drive would then need to turn 10.78 turns.
The ratio between wheel and rudder angle is 11.68, and the autopilot would produce 20 x 48/11*11.68 = 1019 lbsft in the rudder, i.e. about 1/3 of the maximum.
Marc, if you turn the wheel from hard over to hard over does this mean 2.47 turns, and how many turns does this produce on the autopilot drive?

It would also be good to check the Swan 411 in this connection, Wheel turns are 2.33, the sprockets as far as I know 33/11, and drive turns then become 6.99. The ratio wheel/rudder angle is 11.03, and the autopilot produces 20 x 33/11*11.03 = 661 lbsft in the rudder.
The maximum rudder torque is in the order of 1100 lbsft, and the autopilot produces about 2/3 of the maximum.
Daniel, if you turn the wheel from H.O. to H.O. does this mean 2.33 turns, and how many turns does this produce on the autopilot drive?
Maybe we should settle down and also think about other things for the next few days.
Merry Christmas !
Lars

Lars,
you are right, it is time for other thoughts...
...still I just received an answer from our friends of Lianda, 57/040, and I like to share it.
Neco Drive type: 17DR8, 24 volt. (This means 20 lbf ft at the sprocket)
Wheel turns hard-to-hard: 2.35
Wheel sprocket to rudder: 15 teeth.
Wheel sprocket to autopilot: 25 teeth.
Autopilot sprocket: 11 teeth.
Autopilot sprocket turns hard to hard: 5.25
Ratio autopilot-wheel: 11:25

According to the Neco leaflet, if the autopilot turns hard-to-hard were 4.5 and the total rudder swing were 70 degree, the torque applied to the rudder shaft would be 390 lbf ft; for the same swing and 5.25 turns, the torque becomes 455 lbf ft.
If the hard-to-hard swing angle were smaller, the computed torque would increase.

Regarding my boat I will share the data after I see her during the holidays!

Happy Holidays to everybody!

Daniel, 411/004

This is probably something you all are well aware of, but I had to learn it: To effectively steer the boat in all conditions the pilot has to have a fast response. Ten seconds hard over to hard over or a little less. This came from data obtained from the BOC where they sail single handed and the pilot is vital. On PLEIADES, a Buchan 37 that I built I ended up with a Wood Freeman pilot. We sailed her from Seattle to SF. We had NW winds about 20 knots most of the time. During the day we put up a 3/4 oz chute. The pilot steered all the way. A person would tire out very quickly. We found that a person could steer well for about 20 minutes, After that he would become sloppy. I sailed ATALANTA, a 74 ft. ketch down the coast. It had a Robertson pilot. It was powerful but way too slow. It would not handle the boat in other than very mild conditions, like maybe eight knots over the deck on a reach. Any pilot will handle the boat to weather; it's downwind where you need the fast response. On my present boat CYGNUS, S&S 40, 1971 I have a B&G Network pilot driving a home built hydraulic ram. Hard over is 8 sec. It has handled the boat in all conditions I have encountered so far in the twenty years I have owned her. The reason for building my own power unit was that the stock drive is not fast or powerful enough. To test for speed you punch in a 100 degree course change and time the wheel movement.

Pete, Marc and Daniel
Thank you for the information.
As far as I know the Neco drive on the S57 uses a 17.5 rpm motor, and on Lianda this would mean 18 seconds hard over to hard over. On a 57-footer more power would be required to reduce this to 8 seconds .
I would like to ask if there are other Swan 57:s out there still with Neco autopilot? If so, how many turns do they have on the autopilot sprocket hard over to hard over? I would expect it to be about 10 turns, and this makes the pilot stronger but slower. Comments about the performance would be appreciated.
Best regards and Happy New Year!
Lars

Pete, Marc, Daniel and Lars,
Swan Lake (57/027 Ketch) has two independent autopilots (Raymarine 7000 and Robertson AP 300 HLD 2000)constantly used in very trying conditions. Please advise if I can provide any helpfull information.
Alberto

Dear Alberto
Thank you for your message.
Of interest here for each autopilot is the rudder torque provided, and the time from hard over to hard over.
If you do not have the torque maybe you have some other numbers from which the torque can be worked out.
Best regards
Lars

Dear all,
I have the numbers regarding my 411 and, as usual, they are the same given by Lars!

Autopilot-to-rudder-wheel sprocket ratio: 11:33
Hardover-to-hardover turns of autopilot sprocket: 7.5

Assuming (sorry but I forgot to check also this) a 17DR8 Drive at 12V the numbers lead to a torque at the rudder of approximately 16x3x11.03=529 lbf ft and a ho-to-ho time of 13 s.
If it were a 35DR8 there would be 12x3x11.03=397 lbf ft and 6.5 s.
Indeed I already know that the autopilot in my boat performs perfectly in most situations.

I have a couple of other questions for Lars.
I understand that the rudder of the 411 is not compensated. Making it very clear that I like it as it is, I wonder if there is a fundamental reason for such a choice made by the designer.
Moreover I wonder if it is possible to have the formula used for rudder maximum torque and the physical values (area, equivalent lever arm, speed, etc) for our boats.

Daniel, 411/004

Dear Daniel
Thank you for the confirmation of the numbers.
Afraid your conclusion about ho-to-ho time is not correct, with 7.5 turns and a 17.5 rpm drive I believe the time should be 26 seconds.
Suggest you confirm this in practice.

Referring to your questions:
At the time your yacht was built it was common practice to use skeg rudders, in fact this arrangement was used on all S&S Swans. As far as I can remember it was never proposed to balance these rudders, and I think Rod was against it, because he knew of several cases where a line was caught and jammed in the horisontal slot created above the balanced forward part.
A skeg rudder was considered to be safer due to the support and protection from the skeg, and a spade rudder deemed unsuitable for ocean crossing. There is a rather big difference in the loads in the stock, for a skeg rudder there is only torque, while for a spade rudder also important bending loads must be considered. The stock therefore has to be much stronger.
Rudder loads for yachts were not well known at that time, early work on the subject was Rudder Design Data for Small Craft by A.F. Molland 1978.
Molland has published a book with the title Marine Rudders and Control Surfaces in 2007, there is much ship information but also a section on yacht rudders.
Referring to the maximum rudder torque there are various formulas, but they use assumptions for rudder angle, boat speed, and lift coefficient. Let us take an example - if a small boat has a big rudder, the formulas will produce a big torque. In practice, however, a small boat turns easily, and the actual torque is much less. Here empirical knowledge related to yacht size comes in, and some steering system manufacturers provide such information on their website. I can show you where to find this if you are interested.
Best regards
Lars

Dear Professor, Alberto, Daniele & Peter,

Wish you all a lovely year 2012 & fair winds.

Thanks again for your support.

We tested the autopilot on flat sea, for Lianda to alter course of 100 ° it takes 35 sec.

Thinking about Swanake hydraulic autopilot system, it is probably a more accurate & powerful device than the old Neco & B&G.I followed part of Swan Lake South American cruise, very interesting, we plan to follow the same way next year.

Still, if possible, I would like to rely to the original system which proofs by time that it is a  reliable one too, but, basicaly, if possible, let's try to keep things originaL.Also  I am not so fond to play with the rudder system.

If we do not succeed, then help of Alberto as how, where, by who.. the hydraulic system was installed will be most appreciated.

Best wishes,

Marc

Hi Marc & gang,
Concerto, 57-022, has now sailed with 2 autopilots, for several years and good range of conditions. I had the same resevations about keeping original, but added to the Neco, a B&G ACP2 and type3 hydraulic ram as an entirely separate, spare backup autopilot. This has been entirely a success, and steers so much better that the Neco, plus having the remote vision capability of steering from anywhere on the boat, or steering the boat back to yourself if you go overboard. So far we have sailed the entire North & South Pacific including crossing the Tasman three times, with the hydraulic autopilot steering the whole way. The Neco never coped with fast broad reaching and running conditions in above moderate seas, and the new setup does so effortlessly at half the power consumption. If you need any details of how we did it, please let me know.
smooth seas to all,
Robert