View RSS Feed


Connections and catharsis

Rate this Entry
I'm an idiot.

Let's just get that out of the way from the start.

I used to think I was fairly bright. Recent evidence to the contrary makes it fairly hard to cling to that belief.

Case in point.... I thought it would be fun to provide my autopilot with navigation info. Makana has a perfectly functioning GPS/chartplotter at the helm. She also has a perfectly functioning wheelpilot at the helm. Both have NMEA-0183 capabilities. So I thought... why not get them talking to each other?

At best, with all the cross-currents and tide-rips in Puget Sound it would be nice to be able to auto-steer towards a place, rather than just maintain a compass heading. It'd be a good option to have. And, at the very least, connecting things like this has a very high Gadget Quotient. I'm a geek, and things like this intrigue me.

Plus, how hard could it be? I mean, an NMEA channel only involves two wires, right?


So I did my research. Came to the conclusion fairly rapidly that the Garmin chartplotter has one notion of NMEA wiring nomenclature, and the Raymarine wheelpilot has a slightly different naming standard.

On the Raymarine, there are two connectors for an inbound NMEA connection
-- in "+" and
-- in "-"

Seems pretty straightforward.

On the Garmin Echomap, there are 2 connections for each of two ports
-- Tx (out) and
-- Rx (in)


I thought, hey, I'm a bright guy, each side has two wires, "out" on the Garmin side must mean "voltage out", which would mean it connects to "-" on the Raymarine side, right? And then, obviously, it makes sense that "in" on the Garmin side would connect to "+" on the Raymarine.

So I tried that. Didn't work.

How do I know?

Well, I went up to the wheel and pressed the button on the wheelpilot and got a polite message that said "No navigation data". Along with a rude beep. Seems pretty unambiguous.

Actually tried a couple of different permutations. As a once-upon-a-time math-major, I can calculate that with 4 wires on the Garmin side and two terminals on the Raymarine side, there are something on the order of 1.6 gazillion possible combinations.

I think I tried them all. At least once. Each followed by a trip to the wheel. And a beep.

In hindsight, I wish I'd had my "fitbit" on. Pretty sure I would have hit my 10,000 steps just between the panel and the wheel.

But I digress.

In flagrant violation of several different Man Rules, I called Raymarine for help. Waited on hold for about 30 minutes, and when I got through the nice fellow on the other end told me - in about 30 seconds - what I needed to know. Apparently it isn't just that Garmin have a different idea about what to name the wires... they have a different wiring protocol entirely.

As it turns out, the "+" and "-" on the Raymarine are two sides of the same circuit, and the "-" side needs to go to boat-ground. But on the Garmin side, each of the NMEA wires is half of a separate circuit. The Garmin uses a common ground approach, so the "-" side of each connection is already covered when the device itself is connected to boat-ground.

OK. That kind of makes sense.

So I went back to Makana and connected Garmin "Tx" to RayMarine "in (+)"
And connected Raymarine "in (-)" to boat ground.

Tried it.
No navigation data.

Well, the thing has to be wired right... right? I mean, I have GPS data "TXing" into the "in" side of the wheelpilot, and everything now has a proper ground.

So it must be a configuration problem. Or a software problem. I'm a software guy, this is good.

After a few paths through the menus, I discovered an configuration option that seemed promising. "Search for data sources". That seemed like a winner, after all there must be GPS data coming in, I just need to let the wheelpilot find it.

So I pressed it. And got a somewhat disconcerting message on the screen.

At this point I'm not entirely sure what that means, but it seems to be saying something like "hey, dude, you're trying to do things that won't work."

In fact, despite all the information in the Raymarine manual that show how to use the "track" function to steer to a GPS-provided waypoint... this message seems to be saying that the SPX5 wheelpilot I have will not even do what I want it to do. Ever.

So I called Raymarine. Again.

Another 30 minute hold-time, another nice fellow on the phone, who tells me that the message I saw has nothing to do with what I'm trying to accomplish. Apparently that function is for finding Raymarine data sources on the Raymarine-proprietary "SeaTalk" network, and NMEA-0183 connections are not part of that network.

And then he tells me how I can "test" my wiring. He said that if I place a multimeter across the Garmin "Tx" wire and the boat ground, I should see voltage fluctuations. Nominally close to zero, but periodic spikes of 3-5 volts would indicate that the GPS is successfully transmitting NMEA-0183 sentences across the wire.

Emboldened with Great Knowledge, I go back to the boat. Pull out my trusty multimeter, put the leads across the terminals of the handy terminal strip Christian thoughfully put behind the panel to make these connections cleaner and easier. Sure enough, multimeter shows periodic spikes of just about 3 volts. So the GPS is sending.

Then I put the meter across the terminals where I have connected those wires to the Raymarine. Same spikes there. Which means... hmmmm... it means those sentences are successfully making it out of the GPS, through the wires, across the connections and all the way to the Raymarine.

So why isn't it working?


Another call to Raymarine. The third, if you're keeping score. Another 30 minute wait. Another sad description of the symptoms, albeit with the proud ability to tell the tech that I have tested the wiring and verified it is valid.

The tech asks me some questions about the GPS, I assure him it is all functioning and has valid position data while I'm doing this.

Then he asks ".... and an active waypoint?"


Yeah, no, I tell him, the boat is sitting in the slip, I'm not going anywhere right now, I just want to make it so navigation data is available to the wheelpilot. Once I know that is working, I'll go out on the water and put a waypoint in to test it.

He laughs. And patiently - like he's talking to a moron (which, apparently, he kinda is) he tells me that I need to have an "active waypoint" on the GPS before it will send the right navigation data to the wheelpilot.


Apparently getting the wheelpilot to steer to a place requires that you... uh... actually specify a place you want it to go.

Yeah, that made perfect sense. As soon as he said it.

Back to the boat. Select a waypoint on the GPS. Press the button on the wheelpilot.

And magic happened. All of a sudden, the autopilot knew where I wanted to go.

No beep!

Yup, "EV-ENTR" (the Everett Marina Entrance) bears about 293T, which would involve a turn of about 25 degrees to starboard from my west-facing slip.

Total victory!


Turns out I had the connections "right" since shortly after the first call to Raymarine.

But this isn't really about the connections. It's about catharsis. Writing the story is much cheaper than therapy.

There is, though, some possible value for the reader who makes it this far.

Yeah, sure, if you want help sorting out connections between a Garmin GPS and a Raymarine wheelpilot, I might be able to help. Two wires. How hard can it be...right?

But if you want to feel smarter? I can absolutely help with that. Just knowing what an idiot I am automatically raises your IQ.

You're welcome.

Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	IMG_2013.jpg 
Views:	474 
Size:	137.2 KB 
ID:	19364   Click image for larger version. 

Name:	IMG_2014.jpg 
Views:	451 
Size:	89.6 KB 
ID:	19365  

Submit "Connections and catharsis" to Digg Submit "Connections and catharsis" to Submit "Connections and catharsis" to StumbleUpon Submit "Connections and catharsis" to Google Submit "Connections and catharsis" to Face Book

Updated 06-18-2016 at 03:05 PM by bgary

Tags: None Add / Edit Tags


  1. Loren Beach's Avatar
    Not sure if this helps in the slightest, but when we installed our new Lowrance plotter/HD radar setup, our electronics shop guys said that it could be connected to our existing Raymarine AP. An interface device would be required. I opted to keep the devices/systems separate.
    Being able to correct for cross track error would be nice, but we are so used to course plotting and inputting any needed change to the AP heading that the potential result did not seem worth the hassle.

    Updated 06-20-2016 at 09:57 AM by Loren Beach
  2. toddster's Avatar
    Raymarine does seem to have pretty knowledgable and responsive technical support. There are a lot of 0183 problems caused by different manufacturers kludging their existing RS232 serial ports into "sorta" an 0183 port. There is a fairly detailed discussion of the different flavors in the Shipmodul mux manual and a couple of others.
  3. bgary's Avatar
    Quote Originally Posted by Loren Beach
    Being able to correct for cross track error would be nice, but we are so used to course plotting
    Yeah, I doubt I'll use it. I'm sorta nuts about constantly running bearings and keeping a good DR plot anyway.

    Plus, very mindful of the Aegean incident a few years back - there is no electronic that I would trust to "be in control".

    But it was fun to work through it. And I learned a lot about my systems along the way.
  4. EGregerson's Avatar
    "Writing the story is much cheaper than therapy. "

    Reading the story is therapy! therapy.

    One great thing about sailing is the lessons are so expensive... it becomes the stuff you never forget.
  5. olsenjohn's Avatar
    I absolutely loved the piece. Great writing that we can all relate to.