NMEA Network

[CarinaPDX]

bstreep- we may have strayed a bit .

It would be interesting for the OP to come back in and see if this is of any value to him (i.e. would he even be considering an N2K-based system). Personally I think it makes sense to consider one for all but the tightest budgets, for the reasons that have come out from several posters. If that is not of interest to the OP, then agreed let's move on to other solutions.

Greg

[downunder]

Originally Posted by bstreep

Would it be possible...

For us to just address the OP's question? Rather than argue merits of systems?

Great discussion. and informative.

[colemj]

Originally Posted by bstreep

Would it be possible...

For us to just address the OP's question? Rather than argue merits of systems?

Let's look at the OP's question: "It's time for an upgrade to a NMEA network with new equipment; log/depth/wind etc. I'd like to keep it as simple as possible and as cheap as possible. Looking also for iPad integration (MID WIFI app).
Any advice regarding the hardware and general network issues would be very welcome."

Where exactly have we strayed?

Mark

[goboatingnow]

well it may be thread drift, but its a useful drift. The OP has enough to go on, there's not much to be said.

To dig a little deeper heres some of the issues, likely to dog the standard

* Bus system, faults replicate over the while bus, making location difficult

* Bus system faulty nodes cannot be automatically isolated, ( jabber,etc)

* No separation of high speed and low speed traffic as is common now in cars ( my car has three separate can networks).

* No central error reporting or collating system , like OBD on cars etc. Difficult to see bus errors or node misbehaviour. MFD have poor functionality, usually because you need a specialised CAN stack. ( ie promiscuous node)

* The standard does not encompass internetworking so bridges are difficult

* The issues of (A) auto terminating resistors (b) rise time control , needed on SIMnet are showing that the spec isnt detailed enough

* Lack of reasonably priced or portable diagnostic tools

* delays in extending the spec, 2 years to sort out AIS, distributed swicthing PGNs still not signed off. where do chargers, invertors, etc fit in.

* We not yet seen very large or stressed systems ( ie with full electronic engine and gearbox control systems etc)

* lack of user knowledge, how do you diagnose faults at sea, its not simple to understand.

* growing number of private PGNs, that destroy the open nature of the spec.

etc etc

What NMEA should be doing

* Introduce a star based option, with intelligent switches to handle isolation and control bus misbehaviour

* Spec a common error reporting system and an error repository that can be easily accessed by simple tools

* Promote a diagnostic tool spec.

* Go open source and lets other contribute, allowing standard to grow quickly and in advance of actual product innovations



Dave

[colemj]

Originally Posted by goboatingnow

well it may be thread drift, but its a useful drift. The OP has enough to go on, there's not much to be said.

To dig a little deeper heres some of the issues, likely to dog the standard

* Bus system, faults replicate over the while bus, making location difficult

* Bus system faulty nodes cannot be automatically isolated, ( jabber,etc)

* No separation of high speed and low speed traffic as is common now in cars ( my car has three separate can networks).

* No central error reporting or collating system , like OBD on cars etc. Difficult to see bus errors or node misbehaviour. MFD have poor functionality, usually because you need a specialised CAN stack. ( ie promiscuous node)

* The standard does not encompass internetworking so bridges are difficult

* The issues of (A) auto terminating resistors (b) rise time control , needed on SIMnet are showing that the spec isnt detailed enough

* Lack of reasonably priced or portable diagnostic tools

* delays in extending the spec, 2 years to sort out AIS, distributed swicthing PGNs still not signed off. where do chargers, invertors, etc fit in.

* We not yet seen very large or stressed systems ( ie with full electronic engine and gearbox control systems etc)

* lack of user knowledge, how do you diagnose faults at sea, its not simple to understand.

* growing number of private PGNs, that destroy the open nature of the spec.

etc etc


Dave

Are you describing NMEA2000 or NMEA0183. I can't tell the difference from your description.

Oh, wait, you mentioned needing terminator resisters. Yup, 0183 has the edge there.

Mark

[Shoalcove]

Don't stop now! I'm in the process of planning to shift to a NMEA2000 network (precipitated by a want for an autopilot) and am following this closely. Feel free to mention more good vendors -especially if you know any in Canada!

[goboatingnow]

Originally Posted by colemj

Are you describing NMEA2000 or NMEA0183. I can't tell the difference from your description.

Oh, wait, you mentioned needing terminator resisters. Yup, 0183 has the edge there.

Mark

well yes, isnt it scary that NMEA actually didnt deal with all the issues in 0183, when they invented 2K,at least with 0183, you could stick it into a PC and actually see what was happening.

[CarinaPDX]

Apparently, the (imaginary) better is the enemy of the good (enough). Your list of shortcomings is useful IFF we are a committee to design a better standard. We are not. There is no available alternative that comes close to N2K - certainly not 0183.

As someone who has installed an N2K system, I can say that in practice, for a system one would find on a cruising boat, it works and does not suffer from the lack of professional tools, etc. There have been some amazingly complex networks described on Panbo, and some of them have had a few problems, but again pretty modest given the complexity and what one would expect from any alternative.

Greg

[goboatingnow]

Yes but thats not the point Greg, what we are seeing currently is small systems, usually around the sensor/display pair paradigm. its only now as 2K is getting a head of steam, with lots of players entering the market that issues will arise. This is a replay of 0183, it worked fine for the simple stuff, then it got horribly complicated and bastardised.

Sure its all we have, its could be fixed though, maybe it will in time, Im not holding my breath. It doesnt bother me, I have a some good diagnostic tools, I can actually see whats going wrong

PS ( its was interesting a a connect fest some time ago to see the sheer number of bus errors and protocol violations as all sorts of stuff got connected up!).

anyway all out of typing juice now

[Ziggy]

Originally Posted by Askopelos

Hello to everybody - newbie here...
The electronic systems on board our Colvic Atlanta are almost as old as the boat herself. It's time for an upgrade to a NMEA network with new equipment; log/depth/wind etc. I'd like to keep it as simple as possible and as cheap as possible. Looking also for iPad integration (MID WIFI app).
Any advice regarding the hardware and general network issues would be very welcome.
Cheers...

Consider the Garmin GMI-10/GWS-10/Smart Tri-ducer bundle (see here: Garmin GMI 10 GWS 10 and Thru Hull Smart Triducer Bundle).

1. Includes Wind, depth, speed, and a color LCD graphics customizable display (analog/digital/time history, etc)
2. N2K network, with a provision to also display NMEA 0183
3. $1,073 for the whole package (I think this might be as cheap as possible)
4. You can add a N2K MFD, compass, gps, radar, redundant transducers, etc later
5. Pretty straightforward to install
6. Wifi bridge available from http://www.digitalmarinegauges.com/seasmart.htm

[CarinaPDX]

Dave: Actually, I can plug my Actisense bridge into my USB port (how many laptops still have a serial port? And RS-422, not RS-232?). I can see everything except the radar and chart images with free software, and could get all them if I wanted to buy the software. Try that with half a dozen individual instruments each with a 0183 output. Perhaps use a multiplexer? Not for the novice...

Perhaps I am the one now without a sense of vision: I don't see that a cruising boat needs a lot more complex system than the one I have. I might add a bridge to the engine bus, but that seems pretty unlikely to result in problems. But a cruising sailboat is unlikely to get even as complicated as powerboat systems that exist today.

I appreciate what you say about the many technical shortcomings - all true. They just don't get in the way of today's systems working. I think the larger point is that the NMEA standards are done by committee, whose members have a record of dragging feet and conflicting interests which has resulted in something less than perfect. Too bad, but this is what we've got. It works. It will be a long time before we have another widely accepted standard that is better (a looong time if NMEA does it again!).

I'm heading out too, so others can get in.

Greg

[colemj]

Originally Posted by goboatingnow

well yes, isnt it scary that NMEA actually didnt deal with all the issues in 0183, when they invented 2K,at least with 0183, you could stick it into a PC and actually see what was happening.

I have two ways of doing this with the N2K system - a Maretron 2000-USB gateway and Actisense 2000-0183 bridge.

It was actually more difficult to do it with 0183 to PC

Mark

[DotDun]

This has been discussed ad-nauseam in other threads.

NMEA2000 is only a partial step forward compared to 0183. Dave is correct, NMEA fell short of doing their job of delivering a true next-generation network. Other than the ability to have multiples of the same sensor connected plus the 'ease' of running a single cable, N2K really doesn't deliver any more functionality than 0183 and a multiplexer. If you've ever had to troubleshoot a bus network, you would have a different opinion of them.

In fact, I'd venture to say the N2K certified Furuno NavNet 3D running the current 2.07 firmware would bring down a N2K network that is fully loaded. It spews out frivolous data querying different nodes utilizing ~8-10% of the 250k bandwidth. Put that on a net with 49 other devices and you're sure to have problems that won't be obvious by simply unplugging various devices/segments.

Unfortunately, as NMEA members planned, the consumer is stuck. It doesn't make long term sense to put a bunch of money into 0183 gear.

[CarinaPDX]

What I have tried to convey, apparently to little effect, is that we are discussing two quite different issues here: 1) an engineering one, of the imperfections of the current NMEA 2000 standard and its suppliers/products, and 2) a practical one, of which instruments to install on a cruising boat now, say this year. None of us with electrical engineering backgrounds and an awareness of the standard disagree about its flaws. But that is not what this thread is about, at least going back to the OP. Of the options available (0183 and 2000) N2K wins hands down in turns of capability. And it works. Yes, we can use diagnostic tools and see a lot of bad stuff going on but again, not of use to the practical application - at the user level it works. And most if not all modules can be field-upgraded to fix the various problems. Furuno has been shipping new system upgrades for NavNet 3D products on an annual basis. If there is 10% extra traffic, that is concerning from an engineering POV, but not a practical one. 49 devices? Perhaps on a gear-mad power boat (ref: Panbo) or a megayacht. Very unlikely on a cruising sailboat, which is the real question here.

Saying that "N2K really doesn't deliver any more functionality than 0183" is like saying a Lexus doesn't do much different than a Model-T. Which at one level is true, but not for most of us. And troubleshooting a Lexus is far more complex - not really relevant for most of us, particularly given the reliability. For practical cruising boat applications N2K works spectacularly. [Edit: And will work better in the future as the software/firmware bugs get fixed.] If there is a desire to tear apart the engineering specs for N2K, and the lack of adherence to the spec, and any other technical issue of interest to EE's, then start a new thread.

'Nuff said.

Greg

[Askopelos]

Good morning all and thank you to everybody that has posted a reply to this thread. It will take me a while to digest and process all this information. For the first time I feel that I'm moving in the right direction. In the future I'll be rather more careful with the terminology that I use; 'Cheap', however cheerful, was probably misleading! I wanted only to suggest that Im working with a limited budget... Value for money, reliability and future-proof are my legitimate concerns.
Back soon... Thanks again...