[Bit of a Rant] Is the NMEA Corrupt or Incompetent?

[StuartWeibel]

I led an international standards initiative for 15 year, and there is much in this thread that rings true.

Still, as a DIY neophyte marine electronics installer, I successfully linked my own collection of devices without buying any expensive connectors or backbones using the NMEA 0183 standard, and it has worked quite nicely for some years.

Neatness counts. Do it once, then do it a second time, but with obsessive attention to neatness, so you can sort it out later when you have a problem.

Buy an expensive wire stripping tool. Buy and expensive ratcheting crimping tool. use high quality, sealant-shrinkwrap terminals and high quality terminal blocks.

In the end you'll know what you're doing and your stuff will work, and you'll understand why.

Here is my simple tutorial on how to do it for those so inclined:

Installing and Networking a Class B AIS Transceiver with a Chart Plotter and VHF Radio using NMEA 0183

[Lake-Effect]

Originally Posted by CarinaPDX

Agreed that "corrupt" is far too harsh of an accusation. And I see less of an issue with the NMEA standards than the individual manufacturers' choices to subvert them by using proprietary sentences, PGNs, connectors and extensions (e.g. RM's original SeaTalk is based on an early 0183 standard but wasn't upgraded AFAIK). By offering new features of interconnection within a product line with proprietary sentences or PGNs (e.g. selecting a target on the AIS or MFD that then makes a DSC VHF call) a manufacturer can get a lead on its competitors without giving away its idea early by having to jointly agree to them. And the high cost serves to discourage smaller companies (and even some larger ones) from getting NMEA-certified product status - hence NMEA-compliant products.

^ ^ ^
This.

With their standards, NMEA is trying to hit the sweet spot of reliability, interconnectability, and exclusivity (protection). By choosing to be even more exclusive, it's easy to see that many manufacturers very much fear a market where the Far East can cheaply produce compatible gear; because that will eat into their fat "marine" markup, as is already happening.

I don't envy the NMEA member companies' predicament. I don't know what would be their best way forward. Currently I've been helping on one yacht, trying to integrate some good pan-tilt-zoom cameras into a big name-brand multi MFD setup.... and it's been misery. Integration into the marine system completely thwarts any other means of viewing and controlling the cameras. And the MFD/system maker has been zero help.

When someone finally gets around to producing a reliable, marine-quality sunlight-readable Linux "appliance" at a good price-point, there will be a jump in the adoption of more open marine solutions.

[genaea]

I agree with the poster above who said why use cabling at all. New systems could be totally PAN/WLAN driven and allow one to not use any cabling for communications.

The good news is your boat can be driven by open source hardware/software IF one is technically competent to do so. While you might not have all the same features located exactly where the manufacturers devices place the features, you could replace most/all the systems with microcomputers/microcontrollers and open source software. And, if one were really tech savvy, you could have a totally integrated system that far surpasses what the big companies are doing (in the recreational space anyway) for pennies on the dollar to what is available from marine manufacturers. You would just have to know how to flex your coding muscles to perform this, but the sky is the limit.

Not to say this is for everyone, but it is available for those who are savvy enough in this aspect to benefit from it. While the rest of folks may be relegated to using the NMEA off the shelf components and the associated prices.

It is obvious most people will chose to pay the NMEA piper and be off sailing rather than configuring their own onboard (OSI or PLC) based system.

[Lake-Effect]

Originally Posted by genaea

I agree with the poster above who said why use cabling at all. New systems could be totally PAN/WLAN driven and allow one to not use any cabling for communications.

If you have to run power wires anyway, wireless sensors and instruments don't make sense to me. Power on the backbone is one of the good features of NMEA2000.

[genaea]

Power on the backbone is one of the good features of NMEA2000.

Agreed, and also one of the great features of PoE ethernet as well, if you chose to go the non-wireless route.

They are updating PoE specs all the time for ever increasing power requirements of the devices connecting to them.

We have a lot of Multi-CCD Pan/tilt/zoom video cameras powered by PoE. The latest spec (802.3bt) delivers up to 100watts of power.

[TBear60]

Addition to this rant:
Granted Electronics manufacturers are a little "prima-donna"-ish!

But how about the VHF radio manufacturers? They have many models with many features. However, most of what goes out the door are entry level radios that only have NMEA 0183 which is essentially useless in Marine applications (coastal and Oceanic). The government mandated DCS functionality which needs a GPS (lat/lon) source to be effective is unavailable to the radio from any chartplotter except Garmin! They are the only charplotter builder still offering NMEA0183 as standard equipment.

Granted you can get a Nmea 0183 to 2000 converter pigtail ($100?) to tap into a NMEA 2000 backbone (another $100), but why? Current manufacturers are offering GPS equipped VHF radios for a nominal $50 upch arge at the retail level.
And you know if there is only $50 upcharge to the retailer, it's only costing the builder about $5-10.
My gripe is why aren't all the manufacturers offering the GPS equipped machines as their entry-level offerings? Simply eliminate NMEA0183 from the VHF radio!
For that matter, why not also include NMEA 2000 on all VHF radios. Sell the combined VHF/GPS/NMEA2000 radio for $200. It would sell like hotcakes and they would make a killing (fewer skus, parts bins, etc) due to lower manufacturing costs!

[rslifkin]

Originally Posted by TBear60

For that matter, why not also include NMEA 2000 on all VHF radios. Sell the combined VHF/GPS/NMEA2000 radio for $200. It would sell like hotcakes and they would make a killing (fewer skus, parts bins, etc) due to lower manufacturing costs!

They do. The Standard Horizon GX1850G has NMEA2000 and internal GPS for about $225. Some of the lower end models have GPS but not NMEA2000. The cheapest units that have neither are well under $200.

[TBear60]

rslifkin, thanks for the reply. Yes, you are correct. I forgot about the GX1850G. Good choice! The cheapest units sell for $149.99 in the US. Still, sticking with my original rant: why? You're required by law (USA) to have DSC, but it's useless on the entry level model! (that is, unless you have Garmin!)

[CarinaPDX]

Originally Posted by TBear60

The government mandated DCS functionality which needs a GPS (lat/lon) source to be effective is unavailable to the radio from any chartplotter except Garmin! They are the only charplotter builder still offering NMEA0183 as standard equipment.

Not true. The latest and greatest (as well as older) Furuno MFDs include an NMEA 0183 output port which could be used to provide GPS data to the VHF. They also have internal GPS receivers so an NMEA 0183 input is less necessary. Otherwise it would have to be added through an external 2000>0183 converter if needed. I do think it would be better if they had one 0183 input port for an external GPS receiver, as N2K GPS receivers are scarce and expensive.

Buying anything new I would insist on N2K functionality as it really cleans up interconnections.

Greg

[wholybee]

Originally Posted by genaea

I agree with the poster above who said why use cabling at all. New systems could be totally PAN/WLAN driven and allow one to not use any cabling for communications.
snip

Wireless works great, except where it won't. Wireless equipment falls in one of several unlicensed bands with limited bandwidth that must be shared with other unlicensed users. I don't want my critical navigation equipment to depend on how many other (often non-marine) users are sharing the band.

Case in point, the marina I am currently at, Bluetooth and wifi are both basically non-workable do to the number of users in a small area. Anything on my boat that uses the 2.4Ghz or 5Ghz bands is almost useless.

NMEA 0183 is extremely simple and extremely reliable. NMEA 2000 a bit less simple and a bit less reliable, but still more simple and reliable than wireless.

[Panama Red]

NMEA is the standard until it is not.

SeaTalk was the Raymaine standard until it wasn’t.

Lighthouse was the Raymarine standard until it wasn’t (now you need Lighthouse 3).

On my boat no part is standard. A Johnson impeller pump is not standard with a Johnson Impeller Pump (same part number).

Unlike the aviation industry there is no regulating authority to enforce standardization in the marine industry.

Example when I need a 3/8 stainless steel bolt, I get one from West Marine, it doesn’t fit, order from Defender it doesn’t fit. Go to Home Depot and get one...and it fits.

I don’t think Raymarine Axiom is compatible with Raymarine Axiom.

Marine standards means that it is not compatible...

[motu42]

As mentioned earlier, check out https://signalk.org


Disclosure: I am one of the original founders of the project, although I am not involved much anymore.


The reason we started Signal k was the difficulty in building a modern integrated marine system on our own vessels. As a long term geek and IT professional I worked daily with distributed enterprise systems, where global connectivity (internet) is assumed, and sharing data is a must. Its the 21st century!. We live in a connected world, but I on my chartplotter I still cant drag and drop my position, routes, or waypoints on my mates vessel in a simple standard way - I can share data in my car, my lounge, anywhere but my boat.



My observation was that the marine industry was still in the 1970's mini-computer era, where proprietary systems of limited function existed stand-alone, and any sort of connectivity and sharing of data was difficult and expensive. Mostly it only existed on the brochures :-)



NMEA was not a solution. They had tied physical wires to data transport, how exactly do you send NMEA0183 or NMEA 2000 over the internet in a standard way? How do you send it by Wifi, ethernet, cellphone? Well there is no standard, you invent your own - which means every-one has yet another incompatible method.


Then you face the problem that the vessel id is not in the data sent by either NMEA standard!. So if you share NMEA data between vessels its impossible to tell which data is yours! Again the solution is to wrap it up in yet more incompatible one-off solutions.


And then there is the licensing issues.



The solution was an open source data protocol, that could be used, improved, and shared without cost - a communal resource. Thats Signal K.


More, its designed to be simple text (JSON), so a human can read it for dev and debug. Text can be sent over almost any transport, like serial wires, Mod/Canbus, ethernet, wifi, cellphone, HF radio, satellite phone, the whole internet. So now there is a clean data protocol and you pick your choice of physical network.


OneNet is really no-OneNet. Same problem, data and physical tangled, licenced, it will be obsolete before adoption. They should have created a modern internet friendly data protocol, and a separate 'marine ethernet' spec so the cables would plug together.



Signal K takes us into the 21st century, everyone can contribute and use it.

[Beausoleil]

Some further minor corrections:

• NMEA 0183 was published in 1983, whereas IEEE 10BASE2 (aka ThinNet or CheaperNet) was published in 1985. So 0183 was before ThinNet.
• NMEA 0183 is derivative of EIA-422 / RS-422, which uses differential signalling. It happens to work with RS-232, but isn't as robust. It was based on 8 year-old tech (RS-422 was published in 1975). RS-232 dates back to 1960, so it was 23 years old by then.
• ThinNet used BNC connectors, not CATV-style (properly known as F-connectors)
• NMEA OneNet uses gigabit ethernet (IEEE 802.3ab or 1000BASE-T), which was spec'd in 1999 - so 22 years, not 38.

That being said, the thing about NMEA is they're extremely conservative, and their aim is to protect their members - the manufacturers first, and the boating public is always second...

And yes, I've been doing networking since before Ma Bell was split up, ThinNet was new, and Windows was a glint in Bill Gates' eye...