IP38 N2K Upgrade

Robert, I did much the same thing on my Express. I did however rationalize spending quite a bit on Airmar N2K transducers:

Airmar PB200 ultrasonic weather+compass+gps
Airmar DT800 depth/temperature
Airmar CS4500 + ST60 Speed Pod ultrasonic speedo

I also have the Maretron battery and engine monitors.

Olsonist,

Yeah, I'm excited most about the engine and electrical monitoring. That's really breaking new ground, way more fun than replacing outdated instruments.

I think I prefer fewer sensors on top of the mast, for now. At least until it's a little more clear to me how well my different monitors behave with redundant data on the network... selecting, failover, etc..

It would be really handy to have an 'N2K matrix' somewhere.. (Panbo?)

Instruments (input) on the X axis, Displays (output) on the Y axis.

Listing information about compatibility, data display options, failover, conflicts, calibration requirements, etc.

It would be amazing if one of these manufacturers did it themselves, but I'm not going to hold my breath..

robert

Hi Robert,

I can't get your "idisk" link to open; it wants a user name and password. If you send me the image, I can put it right into your entry. (It would be great if forum members could that too, but not possible yet.)

Ben,

Here's a direct link, the iDisk Public folder apparently doesn't like direct linking to files..

http://islandless.net/home/docs/KAII_Network.pdf

Thanks, and keep up the good work, this project would have been much more difficult to approach without the information I learned here!

robert

That's a heck of an N2K system, Robert; I think you're wise to plan for the max, and in phases. You've obviously thought about N2K power issues, but I think you could save some money and hassle by switching from a Mini-size backbone to the new "Mid" size. I've been meaning to write a Panbo entry, or two, about this stuff for a while, and now I will.

Yes, I agree with you, the MINI backbone might be a little difficult in spots.. I'd love to hear what your thoughts are on the Mid cables.

I haven't settled on my exact cable runs yet, but I will be in the next week or 2.

Could you clear up a question I have?

I've been assuming that with the multiple Power circuits available in maretron's MINI power distributor,(8amp/leg for a mid powered, MINI cable system)

I'm able to:

1) have separate breakers for each leg (allowing for 'under power' and 'under sail' instrument-breaker distinction)

2) with one leg off, the rest of the network is properly terminated, and able to pass data to and from any device on the remaining powered network leg.

3) as a leg is given power, the already functioning leg will see the new devices as they begin operation.

robert

Robert, what are the Micro Dist boxes in your diagram? On your 15m mast run to the wind instrument, doesn't that have to be part of the trunk? Or is it and I'm just not reading it right? The Maretron installation guide says

The maximum cable distance from any device on a branching drop line to the trunk line is 6 meters (20 feet).

Robert, I think the answers to 1,2, and 3 are all Yes. I have even seen some NMEA literature that described how you could put more than 2 isolated power feeds into a backbone, if needed. But there are definite rules about common grounding, etc.

Olsonist, Robert is referring to what Maretron calls a Multiport Box, which essentially lets one backbone T and drop become 4 drops.

Ben, Olsonist,

Ben: good news! It seemed like it was part of the design (after reading Maretron's "giude to n2k installation" repeatedly). Roger that on the Grounding.

Everything I can glean about Maretron's claim of 'each leg of the backbone can handle cable max voltage' seems to point to 2 breakers being part of the equation...

Olsonist: again from gleaning info from Maretron's guide, and a conversation with, um, Rich? at navstore.com (who really helped gather a pile of n2k stuff on short notice, to the point of helping with some of the shipping of things that should have been in stock but weren't) explained that if you're going >6m (aka: up the mast) then an 'inline' terminator would sit on the end of that, the longest leg.

Ben, am I correct in this understanding?

As for weight, and installation, I think I'm going to have to hack a cable to something shorter than .5m, place the inline terminator right at the masthead, send a loop to the wind sensor, and use the terminator to secure the line running down the wire run.

There wouldn't appear to be any other way to send a single strand >6m (and remain compliant with any recommendation) if not..

also, looking 1st hand at the SimNet connectors, I'm impressed. A mostly waterproof 'no thread' connection isn't the easiest thing to come up with.. My only question on SimNet is do they make 90deg connectors?

many thanks,

robert

Robert, I'm wrestling with the N2K masthead weather station issues right now. Normally the N2K rule is no drops over 6m. But if you want to run an N2K cable up the mast then you have to find a workaround.

The Airmar PB200 Installation doc (p16) says:

IMPORTANT: When using a cable that is longer than 6m (20'), remove the termination resistor at the last node/tee on the NMEA 2000 network. Insert the male-to-male pin into socket 5 of the WeatherStation connector to activate the termination resistor located inside the WeatherStation instrument.

Dan has actually done this and so he can comment. I assume it means that you have a backbone with a tee that doesn't have a termination resistor. The drop goes up from the tee to the PB200 with the special pin inserted.

Another thing about the PB200 is that it is both N2K and 0183 compatible at the same time. Of course, this means that it has a *very* non-N2K connector needing *another* $50 adapter cable.

Right, an N2K cable run longer than 6 meters, like up a mast, has to be made into one end of the backbone and terminated with a resistor (or also run back down the mast, and terminated elsewhere, which is awkward). The Furuno SC30 diagrams I put up yesterday show numerous ways to terminate backbones.

https://www.panbo.com/archives/2009/03/furuno_sc30_nmea_2000_every_which_way.html

Furuno and Simrad offer long cables with a terminator built in, Maretron has a sleek inline terminator that can screw between its Weather Station and cable, Airmar's Weather Station has a built in terminator switchable with a pin, and Garmin has its own, somewhat clunky, inline:

https://www.panbo.com/archives/2009/02/garmin_gws_10_hands_on_1.html

a bit of advice that the guy @ navstore.com gave me on the phone:

the micro Field-Attachable connectors don't fit into the mounting tube for the maretron wind sensor (standard 1" SS tube). this necessitates hacking a premade cable (its connector will fit) and placing the inline terminator somewhere else at or about the masthead..

hope this helps,

robert

"Dan has actually done this and so he can comment. I assume it means that you have a backbone with a tee that doesn't have a termination resistor. The drop goes up from the tee to the PB200 with the special pin inserted."

--> I have not connected via N2K yet, but this is what I understand from reading the docs.
--- last line ---

Hi folks,

So I'm down in Lauderdale, boat's on the hard getting new bottom paint etc, and due to the fact that I've had to remove a dead watermaker (6 modules, prolly 150lbs total, thru-hull, tap to the main water system, weee!) I called an authorized maretron dealer down here to secure the cable install and help speed things along..

imagine my surprise when they called back quite impressed, and throughly confused about a few things.. "we're not prepared to do the math, i think, but you're probably wrong, we're calling maretron"

I'm not going to name em, as they're nice guys and I did kinda lay the largest (planned) n2k system they've ever seen on em with a cold call..

Apparently they were not totally up to speed on the 'inline terminator on the >6m length despite not being the "end of the branch" thing"...

anyway, they need to do the math and talk to maretron before we start, so I'll ask their main sticking point here:

Ben, Panbots, can you have a backbone that is MINI sized, feeding MINI to 4x micro distribs if said distribution boxes are in the middle of a larger MINI network ie: there are more downstream?

I figured it was obvious since maretron makes a mini UP, micro DOWN distribution box, but maybe it's not so obvioius if one of the leading ft lauderdale shops, fully versed (they say) in N2K installation balks at the mere thought..

ideas? I'm putting her back in the water wednesday, not that that matters, the thru-hull will be installed and waiting for the N2K cable..

haulout and watermaker removal picts below (very little 'marine electronics interest' but wtf, its the interwebs, enjoy) :

http://gallery.me.com/rcrecine#100121&bgcolor=black&view=grid
http://gallery.me.com/rcrecine#100128&bgcolor=black&view=grid

thanks in advance

Robert

"Apparently they were not totally up to speed on the 'inline terminator on the >6m length despite not being the "end of the branch" thing"..."

Not sure what you mean, Robert. My understanding is that no one section of N2K cable can be longer than 6 meters unless it is part of the backbone (aka trunk).

"anyway, they need to do the math and talk to maretron before we start"

Right. There's a simple formula for determining if there is a voltage drop problem in a given N2K network design (found in the Maretron install guide), then much more complicated forumlas for solving such issues. That's how you determine which cable --- mini, mid, or micro --- for a backbone and/or drops, and how to arrange power taps.

"Ben, Panbots, can you have a backbone that is MINI sized, feeding MINI to 4x micro distribs if said distribution boxes are in the middle of a larger MINI network ie: there are more downstream?"

I don't understand the question.

At the risk of getting more confusing, here's a detailed manual on the DeviceNet cable/power system that N2K is very much related to. You'll see here some of the techniques like device daisy chaining that NMEA itself doesn't like but manufacturers like Furuno and Simrad have decided is reasonable:

http://www.odva.org/portals/0/library/Publications_Numbered/PUB00027R1_Cable_Guide_Print_Copy.pdf

Ben,

Let me try to explain more clearly.. They seem to think that one cannot have more than one multi distributor box per network backbone leg. Their installation manager claimed that the multi box "acted like part of the backbone" and not in a manner that would allow, for example , each mini to micro T to feed 4 devices via a multi box, assuming >1 T per network side.

The difference is minimal, functionally, it simply means more t's, and drops for the install, but I swear they acted like I was from mars by trying to simply keep the t's and connections to a minimum.

Also it's clear that they've never done an N2k install that wasn't a small 4-5 device string (which this is, albeit with room to grow later.

We'll see what maretron says next week,

Thanks again for yer help, I'll post all lessons learned.

Robert

That's pretty silly as one of the main illustrations in the Maretron installation guide shows two multiport distribution boxes hung off a backbone. They are simply an alternate way to doing Tees and Drops.

It's worth noting for readers that the no-device-more-than-6-meters from the backbone rule still applies. You can have a 2 meter cable to the distribution box and 4 meters to any or all of the attached devices, or 3 and 3, etc.

I believe that you can also incorporate multiport boxes into a backbone simply by routing both sides through it (like you can use a Furuno break-out box as an entire backbone).

N2K network diagrams tend to be quite linear, and we tend to think they should actually look that way. But I don't think the data itself cares how many turns it takes, or how many connections like tees and multiport boxes it passes through.

The essential factor is that the data can successfully travel quite a ways (100 meters if you're using Mini or Mid size cables) along a network's backbone -- i.e. a continuous electrical path terminated at each end with a resistor -- but only 6 meters when branching from the backbone and terminating at a device.

What a difference a weekend makes..

I met again with the install folks I'd been talking to and it would seem that the entirety of all miscommunications and concerns stem from a busy friday causing not quite enough time to look at my diagram's without me on hand to decipher..

heh..

In the end it's the simplest of all solutions, their install manager spent time over the weekend and it made more sense to him quickly. We sat down this morning and things went pretty smoothly after we looked at my diagram together.

we should start in to it on thursday.

Thanks again Ben for the replies and panbo, in general. I'd be up a creek without the insight of your testing, and the great variety of equipment reviews!

robert

Robert
I read your reply to my question re an AP for an IP 38 and liked it. Really, it's made think alot about a rotary drive. I plan to buy a raymarine AP in Chicago at the boat show in January.
I'm still a little confused on the install location. Any chance you can sent a sketch? Thanks: mike