Electronics -- System Architecture

[Dockhead]

After reading this: Panbo: The Marine Electronics Weblog: Adventures in NMEA 2000 Wiring - Part I

I think about splitting up the network into two networks with a bridge, like this guy did.

Seems eminently sensible: (a) you can shut one network down altogether at anchor; (b) network will be more robust and easier to troubleshoot; and (c) better power supply -- less voltage drop.

So it would seem that I could give each one its own 24v/12v regulated power supply, which will supply a fixed 12 volts output with input between 20v and 32v. That will isolate the networks from voltage fluctuations in the main 24 volt DC power system on board.

One network, call it Blue, will have:

WSO-100 at the masthead
Depth, speed, temperature near the base of the mast
DSM-250 master network monitor at the nav table
2x T8 Zeus MFD's at the nav table (but not powered from the network)
NMEA2000 to USB bridge at the nav table
GPS

Then on the other network, call it Gray:

Engine interface harness
Fuel flow monitor
Fuel tank level monitor
2x Triton displays on the scuttle
Zeus Z8 MFD at the helm
Triton display and pilot control at the helm
AIS black box
Autopilot
Rudder sensor

Gray network backbone becomes much shorter (about 20 feet versus 120 feet) since it does not extend forward of the nav table.

Blue network becomes much straighter -- doesn't have to veer into the engine room or up into the cockpit.

Both networks become much simpler with fewer nodes.

And I can leave Blue Network switched on most of the time -- I'll have weather, GPS, wind data 24/7, which I constantly need. But a whole bunch of other stuff can be shut down at anchor -- saving a lot of power.

I like it. Thanks, Panbo.

[Dockhead]

Originally Posted by xymotic

It would sure be nice if you could take one of these new super wide screens and orient the damned things UP. Course up with the position bottom third would be the way to go.

For radar, I agree -- I always use Head Up.

For plotter, I do not -- I always use North Up.

But the new MFD's allow you to change the "point of view" for the radar display to "Look Forward", so that the center point is shifted down, giving you more room ahead. I think it will be ok.

Besides that, the whole issue may be more or less irrelevant if you have another window open, leaving the radar or chart display effectively square -- I think that's the real point of the superwide displays in any case.

[xymotic]

Originally Posted by Dockhead

For radar, I agree -- I always use Head Up.

For plotter, I do not -- I always use North Up.

But the new MFD's allow you to change the "point of view" for the radar display to "Look Forward", so that the center point is shifted down, giving you more room ahead. I think it will be ok.

Besides that, the whole issue may be more or less irrelevant if you have another window open, leaving the radar or chart display effectively square -- I think that's the real point of the superwide displays in any case.

I usually use radar overlay.

In any case it would be what? 5 minutes of programming?

[Dockhead]

Originally Posted by xymotic

I usually use radar overlay.

In any case it would be what? 5 minutes of programming?

My present network doesn't do radar overlay very well, because there is a glitch with processing heading data which I have had since I owned the boat. So I don't really have any experience with this mode.

For collision avoidance, heretofore 90% of all radar use in my case, I would definitely not want any kind of overlay and definitely head up.

[goboatingnow]

Dockhead, given the extensive setup you propose, I would fully agree with splitting the 2K network

I would re examine your split though , Id say firstly split(a) nav items that are close physically together, (b) key funtionality you want to keep sailing/navigating,

Dave

[xymotic]

Originally Posted by Dockhead

For collision avoidance, heretofore 90% of all radar use in my case, I would definitely not want any kind of overlay and definitely head up.

It's easier to avoid colliding with 100% of the objects when 100% are on screen

And I agree, head up, and I want as much screen real estate in front of the boat as possible. To my knowledge none do it, maybe the simrad beaming to an ipad wil orient vertically?

[K_V_B]

Originally Posted by Dockhead

After reading this: Panbo: The Marine Electronics Weblog: Adventures in NMEA 2000 Wiring - Part I

I think about splitting up the network into two networks with a bridge, like this guy did.

I wonder if there exists a NMEA2000 network isolator, that would allow to simply isolate part of the backbone. A simple switch that in one position would just transparently pass power and data through, but in the other position uncouples one part of the backbone and puts a terminator on the other. Dos such a gadget exist?

Mmmm. Would probably be easy to build something like that yourself.

[Sandero]

Very useful discussion here. I think it would be useful to readers to include the rough cost of the items including the cable runs. It all sounds terribly spendy. My system is quite old and functional and so I am facing an upgrade. Smaller yacht not as extensive. I suppose the information developed can be used but *scaled* down...

The stuff is coming out fast and furious and it's almost impossible to keep up on the developments. I wonder out loud what sort of support the mfg's will provide as they race ag head with new product offerings.

On a theortical note... are we approaching the limit of the data that a yachtsman/navigator can deal with? It's all very interesting, can be very useful (prudent and safety wise) but it can be distracting, complex and turn the experience of sailing into a huge video game... almost. I've witnessed my own seduction of high tech instrumentation limited only by the fact that my gear has not failed and my budget for such things is limited (to necessities).

As Dockhead has been heading in the direction of a complete state of the art digi nav package I would like to hear his thoughts on the above.

Note that as the N2K thing has happened... unless you are in that market, you tend to not focus on the details, costs, benefits of N2K and it all seems like a steep learning curve including the fast changing range of product offerings. The mind boggles.

[colemj]

Originally Posted by defjef

Note that as the N2K thing has happened... unless you are in that market, you tend to not focus on the details, costs, benefits of N2K and it all seems like a steep learning curve including the fast changing range of product offerings. The mind boggles.

The N2K part of things is just the network and communication protocol. For that, it is SO much easier than 0183 with little to no learning curve at all. All of the other stuff you ask about would be the same for NMEA0183 equipment, since it is the advances in electronics, displays, etc that have changed the most and the fastest.

Mark

[Dockhead]

Originally Posted by K_V_B

I wonder if there exists a NMEA2000 network isolator, that would allow to simply isolate part of the backbone. A simple switch that in one position would just transparently pass power and data through, but in the other position uncouples one part of the backbone and puts a terminator on the other. Dos such a gadget exist?

Mmmm. Would probably be easy to build something like that yourself.

Well, you can just unplug stuff. If you have a faulty T connector, you can just replace it, or replace it with a connector, or a termination resistor. It seems to me this is extremely straightforward, and an advantage of NMEA2000. Note to self: Have an extremely clear, detailed map of the network and a bag of spare connectors, drop cables, termination resistors, etc. on board.

[Dockhead]

Originally Posted by defjef

Very useful discussion here. I think it would be useful to readers to include the rough cost of the items including the cable runs. It all sounds terribly spendy. My system is quite old and functional and so I am facing an upgrade. Smaller yacht not as extensive. I suppose the information developed can be used but *scaled* down...

The stuff is coming out fast and furious and it's almost impossible to keep up on the developments. I wonder out loud what sort of support the mfg's will provide as they race ag head with new product offerings.

On a theortical note... are we approaching the limit of the data that a yachtsman/navigator can deal with? It's all very interesting, can be very useful (prudent and safety wise) but it can be distracting, complex and turn the experience of sailing into a huge video game... almost. I've witnessed my own seduction of high tech instrumentation limited only by the fact that my gear has not failed and my budget for such things is limited (to necessities).

As Dockhead has been heading in the direction of a complete state of the art digi nav package I would like to hear his thoughts on the above.

Note that as the N2K thing has happened... unless you are in that market, you tend to not focus on the details, costs, benefits of N2K and it all seems like a steep learning curve including the fast changing range of product offerings. The mind boggles.

It's looking like roughly $25,000 without any labor (I plan to do it all myself). So far. And without the SSB, and without a computer (now plannign to just use my laptop).

I really don't think it's all that bad. Those are 2012 dollars. A vastly cruder system like what I have now, with perhaps 10% of the functionality, cost about the same, I think, in 2000 dollars. I remember the RL80CRC+ was something like $5,000 just by itself.

Now it's not really state of the art -- I am passing on a lot of things. No FLIR (ha, ha, in my dreams only), no 12" MFD (but I'll have two 8" ones or an 8" plus a 6.4" at the nav table, plus an 8" one at the helm), no water tank monitoring, no AC power monitoring, no DC power monitoring, no multiplex switching (which seems insane for an ocean-going boat), one relatively cheap radar only (B&G 4G; passing on the 4' open array , no millimeter-accurate GPS heading sensor, only two out of four scuttle-mounted instruments replaced, no turbo boost sensor, no bilge alarms (doing those freestanding), no cooling water flow sensors. There are a jillion things you can do with N2K and I'm passing on most of them; mine will be a modest system relative to the full range of what's available these days.

As to too much data -- I really don't think there is any change in the amount of data. Before N2K, we still needed to know how much fuel we have and how much is being used, what's the oil pressure and water temp of the engine, we still needed to know the weather, we still needed to know whether there's another vessel on a collision course with us, etc., etc., etc. It's just that with the new systems, we have all this data highly organized and available at our fingertips wherever we are on the boat. It's going to be really great; I'm totally psyched about my new electronics.

[goboatingnow]

There are a couple of 2K bus isolators like Maretron : Marine Electronic Instruments
http://www.adfweb.com/download/filef...04-E7x_ENG.pdf
Mystic Valley Communications

The big disadvantage is trying to find potential cable fault in a boat wide 2K network, This requires you to locate the T connectors and try and find the damaged section, by removing it and re-terminating. Its a good idea to make up a long-ish length of NMEA2K cable so that you can route around the suspect section ( its great if its short, if its the one up the mast , well!!!. The other issue is when you disconnect sections to determine where the fault if often you strand sensors and displays on each of the seperated sections, and you cant tell if the damm network is actually OK.

Were going to see far more issues like this as NMEA 2K expands to cover more total boat automation.

NMEA2K should be star and hub , like Ethernet is now, way more resilience.

Dave

[Dockhead]

Originally Posted by goboatingnow

There are a couple of 2K bus isolators like Maretron : Marine Electronic Instruments
http://www.adfweb.com/download/filef...04-E7x_ENG.pdf
Mystic Valley Communications

The big disadvantage is trying to find potential cable fault in a boat wide 2K network, This requires you to locate the T connectors and try and find the damaged section, by removing it and re-terminating. Its a good idea to make up a long-ish length of NMEA2K cable so that you can route around the suspect section ( its great if its short, if its the one up the mast , well!!!. The other issue is when you disconnect sections to determine where the fault if often you strand sensors and displays on each of the seperated sections, and you cant tell if the damm network is actually OK.

Were going to see far more issues like this as NMEA 2K expands to cover more total boat automation.

NMEA2K should be star and hub , like Ethernet is now, way more resilience.

Dave

I've heard other engineers with the same complaint about N2K needing to be "star and hub".

As to rerouting around the mast, no need, at least in my case. The only N2K device I will have up there will be the weather station. If I suspect a problem in the mast cabling, then I just disconnect it and pop on a terminator. Et voila.

I will leave my old wind instruments up there for backup. I have just figured out that for $120, I can buy a little box from Raymarine which bridges SeaTalk 1, which will be output from the old ST60+ wind instrument which I will be leaving in place, to NMEA0183. All my plotters have NMEA0183 connections, and will bridge NMEA0183 to N2K. Cool! So I'll have an alternate source of wind data already on the network, in case the weather station or mast cabling goes tits up for some reason. Also depth (can't have too many sources of depth data, can you?).

I could have actually left my entire instrument system in place and got the data into the new network like that, but now I'm already well into the new Triton displays, ultrasonic wind instrument, etc., etc., so I'm not going back to that plan. And who knows when I'll have the mast down again.

[goboatingnow]

I will leave my old wind instruments up there for backup. I have just figured out that for $120, I can buy a little box from Raymarine which bridges SeaTalk 1, which will be output from the old ST60+ wind instrument which I will be leaving in place, to NMEA0183. All my plotters have NMEA0183 connections, and will bridge NMEA0183 to N2K. Cool! So I'll have an alternate source of wind data already on the network, in case the weather station or mast cabling goes tits up for some reason. Also depth (can't have too many sources of depth data, can you?).

why bother with 0183, use the ray convertor to go directly to Seatalk NG, which is 2K. ( all you then need is a Device net convertor cable. for the Simrads
Raymarine SeaTalk1 to SeaTalk NG Converter Kit

When you are bridging like you suggested , you always want to confirm that for example , while the MFD might have 0183 and 2K and it will display data received on both , but it may or may not bridge the two protocols. ( a common gotcha on some RayMarine gear).

[Dockhead]

Originally Posted by goboatingnow

why bother with 0183, use the ray convertor to go directly to Seatalk NG, which is 2K. ( all you then need is a Device net convertor cable. for the Simrads
Raymarine SeaTalk1 to SeaTalk NG Converter Kit

When you are bridging like you suggested , you always want to confirm that for example , while the MFD might have 0183 and 2K and it will display data received on both , but it may or may not bridge the two protocols. ( a common gotcha on some RayMarine gear).

For $89 -- good plan! I didn't realize how easy it is to make these old instruments play with N2K.