Tuesday, April 8, 2008

Outfitting - Autopilot

We actually installed this some months ago, but a spate of recent posts on HunterOwners.com has convinced me there aren't enough step by step install guides for either this autopilot or these boats, so I'm putting this up as a public service. It's not exactly a step by step guide, either, since we did it long enough ago I don't remember it all very neatly, but hopefully it fills in some gaps for someone.

Both based on pricing and in order to stay consistent across our electronics suite, we elected to purchase a Raymarine S1 Wheelpilot (I see that this has about to be superseded by the new MX-5... so goes the marine electronics industry). We wanted something that would interface seamlessly with our chartplotter and not require a terribly complicated installation and the S1 fit the bill. The Raymarine wheel pilots have received mixed reviews going back several generations, with some reports of wheel slippage, inaccuracy, excessive wear and breakage, but they are quite widespread so we decided to give it a try. Then, too, there is relatively little selection available for wheel pilots on the market. The alternatives beyond that are complex below-deck ram hydraulic units, which are overkill on a boat our size, and which are just as prone to breakage but more complicated to repair.

The Raymarine installation instructions left something to be desired. On the face of them, they look extremely comprehensive and accurate, but as you go through them you find that they seem to be missing the bits of guidance which would be most helpful (or at least helpful to someone who has never installed an autopilot before). So I started with the easy part; the Smartpilot ST 6002 module installation.

The brain of the system, this has to be positioned somewhere without excessive electro-magnetic interference (EMI), which can be difficult on a small boat. It also isn't very amenable to liquid, so it pretty much has to be somewhere in the cabin proper. As it happened I had a suitable patch of open bulkhead over the chart table, which had the added advantage of being convenient to the battery panel and fairly close to the cockpit. I slapped it up and ran the spaghetti mess of wiring right down through the table into the engine compartment and wired it up. (Ignore the wiring on the overhead there; it goes somewhere else and just hasn't been secured yet.) The little snap connectors took a little practice to jam the wires into just right; I'm still not sure I have a lot of faith in their holding power but there shouldn't be a lot of strain, either. I used an awl or screwdriver to hold them open, much easier on the fingers.

After I got the Smartpilot installed, I had to put up the fluxgate compass. This was also pretty straightforward; they want it as close as possible to midships and centerline (you know, every piece of electronic equipment seems to want to be there--boat builders should just center a massive server rack or something in the middle of the cabin for all this junk [except then that would violate the other ubiquitous stricture to not place any of it anywhere near any other of it... easy enough to follow if you are installing things on the Queen Mary, I suppose]). I found a nice spot forward of the head and snaked the wire down and around to the corepack without any issue.

The actual wheel and motor weren't terrible to install, either. The wheel clamps on to the spokes of your existing wheel and the motor fits into a slot on the back of it. The only iffy bit is mounting the motor bracket properly on the binnacle. You can see it in the photo at right (covered in a plastic baggy, starboard side of the binnacle just where it curves inward). I let Mandy handle that part--she measure more often and more accurately. Also, I loathe the binnacle and anything to do with it.


Part of the reason I hate it is also the reason we didn't do the tidy thing and run the power cable up through it for the motor: it's jammed full as it is and absolutely inaccessible. The binnacle is an Edson and they have a good reputation but I'm not thrilled with this one. The interior layout is bizarre and unwieldy, and we had enough trouble snaking a tiny new transmission cable down it a couple years ago to warn me off even attempting to route the fat power cable through it. Also, the bottom is glassed in pretty good and clearance for the cable to come out would have been minimal (without drilling--but I like the thick glass there and don't want to go punching holes in it). Anyway, our less elegant solution is to run the power cable out the engine control panel and under the cockpit grating, and secure it on the outside of the binnacle and into the motor.

We mounted the control panel next to the engine control panel for convenience to the helm. We had to put a hole in the boat for this (horrors!) but it is a tightly sealed one and looks good (Mandy did it).

That left the rudder position sensor. This doohickey measures the rudder angle directly and allows the pilot to steer more efficiently rather than simply yanking the wheel back and forth and waiting to see what the compass winds up at. This was also the most complicated bit of the installation. The sensor is spring-loaded, so it's difficult to keep it in a neutral position while working on it. It has to be attached with a metal rod directly to the steering quadrant, so your install locations are limited to those within reach of the rod. The quadrant itself is not horizontal, but the sensor has to be in a plane with the movement range, so it has to be intentionally mounted at an angle as well. Finally, you are doing the whole thing jammed up in the narrow bit of the stern, upside down.

It's possible to mount the sensor itself upside down and calibrate for it, but that seemed complicated and fraught with potential problems. Our solution was to build a platform for the thing and attach it to an L-shaped bracket mounted to the underside of the deck. Here are the results:

The angles are all correct and everything is hooked up right, but only after a lot of measuring (by Mandy) and swearing (by me). Recently, I ran across a number of other photos of sensor installations and by far the most common way of doing it is to glass a platform to the hull and build a post or something up to rest the sensor on. This allows a much simpler platform for fiddling with angles, attaching the rod, etc, etc, and is probably the way to go. You'll notice from the photo that we simply used wood rather than starboard or metal for the mount, and this is why: I'm always considering how to install things in a way which will leave my options open to reinstalling things differently. None of the wiring got cut, we didn't thru-bolt or glue in anything, it's all sturdy enough but still easily disassembled. I figure I can go back and glass it up, rebuild it with starboard, whatever; but I want to give it a trial run before I make it permanent. So we may redo it with the hull-mounted platform. Then again, maybe not--I like having the space beneath it available to shove stuff in. We'll see how it goes.

After all that, we took the boat out on the Sound to commission it (not a lot of swinging room on Lake Union, certainly not uninterrupted). For some reason, perhaps as a private joke, Raymarine insists you circle the boat in 360s for long periods of time as part of the commissioning process. You do this until the autopilot tells you to stop. My impression is that it does so only after it has had a good laugh at your expense and senses that you are about to give up, because it gave me the nod bare seconds before I was going to pack it in anyway.

After all that, I incorrectly specified that it was calibrated on a true heading rather than magnetic. As it happened, this hasn't made much difference since I just plug in the adjusted value, and the pilot shows a "T" after it and the chartplotter an "M" but as long as the number is the same they are both happy. I imagine that happy state of affairs won't continue once we start to change declinations, however, so I'll have to re-calibrate it again before we go.

So far it has worked great, but we have yet to have it out and on while under sail, which is probably the real test. We'll have to be much better about balancing the boat properly. Under power, however, it's terrific. Holds a good course, doesn't use much juice or make excessive corrections, and has an excellent interface with the chartplotter. Indeed, we can simply set up camp under the dodger and give commands to the autopilot from there, without ever having to touch the pilot control panel. It follows waypoints or headings as you request.

No comments: