Early Ice on the western shore of Lake Erie 11/24/14

A few years ago after the purchase of my first sailboat, an Ericson 29, I soon realized that there was a limit to how much involvement family and friends desired in my "new to me" sailboat.   Not many were willing or wanting to spend an entire day on the boat.    A few hours..  that would be fun.   A day.. uh maybe..    A couple of days...ugh... Thanks..I'll keep that offer in mind.   Also the people I know are busy with their lives, and oftentimes our schedules do not coincide.   On top of that my first mate strongly dislikes sailboats.  And I'm being kind in saying it like that.   So as it turns out the boat that can sleep 4 or 6 usually has one on board.   That means single handing.    Anyone who has sailed a decent sized sailboat knows that the rudder requires a lot of attention especially in a an IOR designed hull like the Ericson 29.  Tying off the rudder is not an option on most IOR type boats.   They need to be actively steered.   

My Ericson 29 did not have an Autopilot and the expense of buying an autopilot was prohibitive.   However as a controls engineer I had access to controls.  Hence my Homemade Autopilot project started in 2011.

Here is a picture of the control box which is a 8x8x4" plastic water tight box.  About $15 at Menards. 



What is in the picture? 
The PLC mounted on a din rail that is screwed through the back of the box
To the right is the perf metal housing for the 12 volt to 24 volt DC inverter power supply.   I think it can output about 1 amp at 24 volts which is plenty to power the PLC and the two relays and drive the PLC inputs.
The little perf board wire tied to the 12 volt to 24 VDC power supply is a 5 volt regulator chip since I needed 5 volts to power a RS232 to RS485 converter which is tucked under the terminal strip.   I think I wrapped the converter with
electrical tape to prevent shorts.  The converter is used to adapt the flux gate compass to the RS485 serial ports on the bottom of the PLC.   The bottom serial port on the PLC is used to communicate with the LCD display that serves as a tuning/setup/display device for the autopilot.
The two relays are used to drive the electric actuator.   They are easily replaceable since they are socketed.

 The current incarnation uses an Automation Direct PLC which is very inexpensive as PLCs go and the software to program it is free.   

The PLC used in this project was this one:


This plc has 4 - 24 VDC inputs and 4 - 24 VDC outputs.    It also has two analog inputs and two analog outputs.   The project was started before analog I/O cards were availabe for the Click family of PLCs.  I also used a add on card for the PLC to gain additional inputs. 

The additional inputs were used on a cabled waterproof pendant which has a joystick and two buttons on it.   A resume button and a tack button.  
The joystick is nice since if you are in Man mode you can steer the boat with the joystick.   Pretty nice if you have to steer but want to plant your butt in the corner on a pillow for comfort.

Here is a picture of the pendant:


I also used a display unit to show the compass heading, deviation from desired heading, and tuning parameters, and mode settings.
The display is a standard Automation Direct display which communicates with the PLC via a cable that looks like a phone cable, but is not.

This is the screen that was used.   


 Here is a picture of the actuator I used.   This is a 12" stroke model and I used the available end fittings.  This was purchased from Surplus Center in Nebraska.  http://www.surpluscenter.com/

The actuator has a built in potentiometer that is used as a rudder position feedback device to the autopilot control.


Here is a link to the PLC program.


Here is a link to the Screen/display program.


 Here are wiring diagrams.  I just realized that these diagrams are incomplete.   I will update them.

PLC Inputs

PLC Outputs

Relay and compass wiring

The compass I used was a military surplus fluxgate compass purchased off Ebay for about $100.  It is in a nice cast aluminum watertight box and it works very well.
But since then newer compass chips have come out for less money and they work just as well.  The CMPS10 chip is a good compass and I have tested it.  


However the CMPS10 has been discontinued and the replacement is the CMPS11 which I have not yet tested but it looks good despite the price increase from the CMPS10.


What would I do differently?

The Automation Direct screen is nice in that it is viewable in daylight.   And it is nice to be able to tune the autopilot with the little screen.
However the screen is really not required.   Newer PLCs are now being sold that have built in web servers.   If I am going to use a Tablet for navigation at the helm, then pulling up a webpage on the tablet to tweak the PLC autopilot controller makes a lot of sense.   It also cuts about $150 out of the cost of the system.

When I first selected this PLC, Automation Direct did not have analog I/O cards, they only had PLCs with onboard Analog I/O.   Also, I thought it would be important to be able to tie the navigation computer to the autopilot so the navigation computer can control the autopilot.     With a Nav computer link, a compass link, and a screen link to the PLC, three serial ports are required.    In hindsight, I think a Navigation computer link is really not required and may not even be desirable.  Sailboats do not move fast.   There is plenty of time to think about the desired heading for the autopilot.  

When sailing close to the wind, having a wind indicator so the autopilot can sail to the wind direction would be a great advantage.   So a second analog input would be desirable for a analog based wind indicator.

With these things in mind, a different PLC could be used.   I'm thinking that the Siemens S7-1212 or 1214 would be a very good fit.  That PLC has a built in webserver.  Tie the PLC to a small Wireless AP for communication to a Tablet and that simplifies the hardware further and gets rid of some wires and it eliminates the need for a dedicated screen/MMI for the PLC.


If you make any improvement to this design please let me know and I will post them on this blog.

Happy Autopiloting ! :-)


Licensing stuff:

I grant this software/design a GPL3 open source license.

PLC based Autopilot design Copyright (C) 2015 Dave Cole
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 3 of the License. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program. If not, see <http://www.gnu.org/licenses/>






The marina where my boat is at does not like boat stands and they charge a premium (about $300 per winter) to have a boat stored on stands.     I paid the premium last winter but this fall I decided to make a cradle for the boat.  The boat's basic specs are 10 meters long - about 33 feet, 11 ft wide, draft of about 6 feet and a empty weight of about 12,000 lbs.  The boat is stored with the mast up.   The boat yard transports the boat on the  cradle with a hydraulic yard cradle trailer.  So the cradle frame must be robust.  A lightweight cradle frame will not work. 

People have asked me why some other marinas prefer cradles and the only reason I can come up with is that for some marinas using cradles improves their efficiency.   They can place the boats closer together in storage since narrow hydraulic trailers are used to set the cradle/boat.    Also using cradles can use less labor.  Once a cradle is made for a particular boat, placing the boat into it's cradle with a travelift is fairly simple, whereas moving stands around, placing the stands and then chaining them together can be some work.  

I watched Craigslist and Ebay for cradles that might fit this boat and found nothing that was even close, so I started looking for basic cradles that I could rework.
I found one in Sandusky for a lot less than I could purchase the steel.    So I drove their with my trailer and picked it up just before the weather really turned cold (highs of about 10 degrees in early November!)

The cradle base I purchased is about 17 feet long and 7 feet wide.    I cut off all of the uprights with a plasma cutter and then ground them close to flush with the top of the channel steel framework.  The cradle frame is made of 8 inch channel on the outside with a mix of 8 and 6 inch cross members.   The existing frame welds were adequate.   

But where to place the uprights?  At what angle? And how tall?  

 Before I launched the boat in 2014 I used a plumb bob, an inclinometer (Harbor Freight), a few long tape measures and some boards (for use as straight edges)  to determine the location and angles of the existing stands which were holding the boat securely.  The boat was supported by 7 stands.  Three on each side plus a bow Vee stand.   I referenced the port/starboard measurements off the centerline of the boat, and for a fore/aft location reference, I used the point that is formed between the centerline of the boat and the waterline of the boat at the bow.      From these references I measured to the center of contact point for each stand support pad.   (You have to visualize where that point is and make an educated guess since the reference point for the pad is actually at the pad/hull surface.)   I also measured from the ground up to these pad contact points.    In this situation, the boat's keel was sitting on several 6x6 wood blocks that actually measured 6" square.  So putting all of this together resulted in 7 support locations in 3D space over the framework (which was already welded).  Keep in mind that you need to be able to get travelift or crane straps under the boat while the supports are still in place.

Here you can see my hand written notes when I sketched the locations of the support pads in relation to the centerline of the boat and the front reference point on the bow. 


With this used cradle base that I purchased, I got lucky as I had support steel in the base where I needed it so I could attach the stands uprights.  The rear two pads were shown in the drawing as being placed tilted in from the side but when doing the welding I also decided to tip them forward slightly.    While that didn't hurt anything, I tipped them towards the front further than I should have by a few degrees. 

Here is a shot of the working drawing I used to place the uprights.



For the new vertical support tubes, I purchased some seamless structural 1 1/2" OD tubing from a local steel surplus yard.   It is surprising how much steel tubing is used for the uprights and then the diagonal supports for the uprights.   I believe I used about 60 feet of steel tubing.   I reused the pads and screws that came with the used frame. 

For the bow I decided not to use a Vee type stand top.    I decided to go with a ladder type support and use two laminated 2x6 southern yellow pine boards as cross members that was carpeted to give the bow better support.    With the swept keel, there is a lot of weight on the forward boat support stands and I was concerned that a V type stand there would have to much point contact pressure and possibly damage the boat.   The flat 2x6 (times 2) beam covered with carpet distributes the load better.    The ladder support moves up and down via two scaffold jacks.  Scaffold jack feet are bolted to the 2x6's which are laminated together with epoxy and the scaffold jack screws fit into the vertical "ladder" members.   A cross member stabilizes the "ladder" and fore-aft support is via a diagonal steel pipe that clips over the ladder cross member and attaches to the crossbeam in the base.    The ladder support sits on short steel pins that are welded to the top of the 8" channel.  In this way, the entire ladder support can be lifted off the base so the boat keel can be slid in between the uprights with the travelift without lifting the boat over the front bow support.   The travelift operators seems to really appreciate this feature. 

 Here is a picture of the boat in the cradle.   As you can see it still needs a paint job which will have to wait for warmer weather.