I am going to upgrade the instruments and sensors on my Island Packet 37, and from what I can tell after reading many posts and manufacturers manuals etc, it will be best to run an N2K network trunk and then drops off that.
My plan is to run an Airmar PB200 at the masthead, and make use of the transducer's ability to output in both N2K as well as NMEA 0183. First question that comes up is "Do I need to run the large, 9-conductor cable all the way down the mast, or could I get the NMEA 0183 sentences somewhere downstream by translation from the N2K PGM's that the transducer is sending out, and by doing so, just run an N2K cable down the mast?"
I currently have all of my NMEA 0183 sentences going into a ShipModul multiplexer which has small screw terminals for each input and output channel...sentences from Garmin GPS Map162 on one channel, and speed, depth and wind sentences from my B&G instruments on another channel. I would think that I would input the NMEA 0183 data from the PB200 on an unused channel (that would come directly from a NMEA cable as you show in your blog review coming out of the junction box if I run the "big" combination cable down the mast), but how would that data get into the multiplexer if I just run the N2K cable down the mast? And how is the data from the Simrad IS20 Wind and Combi instruments going to get into the multiplexer. Since those instruments will be connected to the N2K trunk, will their data be included with the data from the PB200?
Currently the multiplexer outputs NMEA 0183 sentences (generated by Nobeltec VNS) which control the autopilot for following a route. These come out of a multiplexer output and currently that cable feeds the autopilot display directly. With the new network, how will the NMEA 0183 sentences generated by Nobeltec be introduced into the N2K network so that the Simrad AP28 will follow them? On a Simnet network, they use an AT10 to inject NMEA 0183 into the Simnet network, but I need something to go from NMEA 0183 to N2K.
Jon, I think there are many ways to approach this, none of them perfect quite yet!
The Simrad AT10 does translate many standard NMEA 0183 messages to standard NMEA 2000 messages, and vice versa, and there are patch cables able to adapt SimNet connectors to regular N2K. An AT10 should be able to get your VNS autopilot commands to the AP28 and your IS20/PB200 depth, wind, etc. data to VNS. But it might be slow on the VNS side, and does not handle some messages like GPS satellite status.
A Maretron USB100, which would drop right from the backbone to your PC, translates 2000 to 0183 a lot quicker, and more completely. But you still might need a way to connect the PB200's 0183 line to your PC if you want to calibrate it or update its firmware.
Far better, I think, would be a direct NMEA 2000 to PC connection. Airmar has a gateway, and it's WeatherCaster software will now run over N2K. Then you could just run its thinner N2K cable up the mast. VNS doesn't understand N2K yet, but hopefully that will happen, or maybe some one will develop a software translator. In the meantime you could use the AT10 to your MUX of direct to a serial port.
This entry might help:
https://www.panbo.com/archives/2009/06/wireless_n2k_wind_more_tacktick_style.html
"In the meantime you could use the AT10 to your MUX..." I can't see from the images if the cable coming out of the AT-10 contains both Tx and Rx conductors? I need to be able to direct data from a MUX output to the Pilot (on the Tx conductor[s]), and also direct all the NMEA 0183 sentences from the N2K network into one of the MUX inputs with the Rx conductor[s] .
"Far better, I think, would be a direct NMEA 2000 to PC connection." I would rather convert the N2K data to 0183 and then feed it through the MUX to the PC, so that I can use the capabilities of the MUX for filtering, etc. So if the AT-10 is capable of both sending and receiving, will it be sending out all of the PB200 data in NMEA 0183 as well?
What are the two yellow connectors for on the AT-10? Are they for Simnet in and out? If connecting to an N2K network, would you just use one?
If I did decide to run the 0183 cable from the masthead, could the 0183 sentences for the Pilot be transmitted over that same cable, or would I still need the AT-10?
Thanks,
Jon
Yes, the AT10 has RX/TX wires, and it will try to translate all N2K messages it sees on the backbone that it understands, like PB200 wind, heading, etc. But the translations are not perfect, and the data rate has to be slowed down to fit on NMEA 0183. Speed issues, and data overflow, is why your MUX has those filtering abilities. If the data stayed in N2K format, you wouldn't have to mess with filtering.
Also, O183 data sent to the PB200 will not go to the autopilot. After all, 0183 was not designed to be a networking protocol; we've just forced that role on it with MUXs and so forth.
OK, am I correct that
1. feed my autopilot data from the MUX output to Simrad via AT-10.
2. feed PB200 data plus speed, depth & temp from the AT-10 back to input side of MUX.
3. there seems to be no need for the Airmar 0183 cable all the way up the mast.
4. N2K backbone will continue up mast and terminate at PB200.
Since MUX passes all NMEA 0183 data to PC, will that satisfy Weathercaster?
What would you think of the Actisense gateway vs the AT-10?
To consider additional recommendations, could you tell us what wiring supplier your using, if there are additional displays, how you predominantly use your IP37 (ocean cruising ?), is power ever an issue, are there other displays, are their other N2K components your thinking of down the road ?
... and Thank You for responding to the survey!
jon,
I just did a similar re-instrumentation of my IP38.
to answer your questions:
1) yes, i'm accomplishing this from MacENC to the AT-10
2) yes, I have full (sailing) instrumentation from n2k through the AT-10
3) correct, no need.
4) correct, you technically need the inline terminator, but it will work without it, as I found out while installing my system.
have fun.
Dan, here are the answers to the questions you posed:
I'm planning to get my N2K wiring components from Maretron. The diagram above cut off the IS20 Wind display, which I plan as part of the system. The only other display is not part of the N2K network...it is a Panasonic Wireless Display at the helm, linked to the laptop below. I do not plan ocean cruising; we typically sail around Lake Michigan and Lake Huron and are usually connected to shore power at night, with anchoring out occasionally, so power is not a concern. The only other component I foresee is possibly a radar unit at sometime in the future.
I'll look forward to your comments.
Jon
If you said you anchored often, then I have some ideas that will appear on Panbo's front page shortly.
For transducers, you might want to consider adding an CS4500 ultrasonic speed sensor to go with that PB200 to enable a reliable calculation of true wind angle.
"Since MUX passes all NMEA 0183 data to PC, will that satisfy Weathercaster? What would you think of the Actisense gateway vs the AT-10?"
Jon, I'm concerned that you're headed for disappointment here. The 0183 data you are going to get from the AT10 will only be subset of the 2000 data on your backbone, and it may come in slowly. Regarding the PB200 and WeatherCaster, you should get the standard stuff -- wind, position, COG/SOG, heading -- but you won't get satellite status (not translated by the AT10) and you will not be able to calibrate the PB200 or manage its data output.
https://www.panbo.com/archives/2008/05/simrad_at10_handy_but_limited.html
The Actisense gateway, and other Third Party Gateways about to come out, are nothing like the AT10. They pass all NMEA 2000 PGNs to a PC at full speed, and they can put PGNs back on the backbone (checking that they're valid).
The PB200 is a good example of all modes possible:
* Using 0183 and Airmar's Combiner Box with USB to PC, WeatherCaster works well, though it only sees data coming from the PB200 (and possibly a Smart Triducer attached to the Combiner).
* Using 0183 via an AT10, WeatherCaster is apt to be slow, will not see some data like satellite status, and won't do any calibrating, let alone firmware updating. (I haven't actually tried it, but am fairly confident of those predictions.)
* Using an Actisense NMEA 2000 Gateway, or Airmar's own Gateway, WeatherCaster FLYS. Plus it sees other sensors on the backbone, let's the user choose among redundant PGNs, supports data logging...and that's just Airmar's first rev of WeatherCaster N2K.
Ben,
Does WeatherCaster N2K work with the Actisense NGT-1 gateway? I have an NGT-1 and have been experimenting with it but I don't find anywhere a list of software that supports it. Of course, very few PC plotting apps support N2K at all.
I think it would be extremely useful to have an app that would translate N2K messages that come in via the NGT-1 (or USB-100, or whatever) to 0183 and put them on a virtual COM port. This would allow for backwards compatibility with existing apps without the limitations of a hardware component like the AT10 (such as cost, speed, limited and non-expandable translation set, additional equipment in the chain, etc.).
Adam, The NGT-1 seems to work well with WeatherCaster N2K. There's a good screen shot here:
https://www.panbo.com/archives/2009/06/wireless_n2k_wind_more_tacktick_style.html
I tried a simple calibration of the PB200, which worked, but haven't tried uploading firmware to the PB.
As far as N2K to 0183, the Maretron USB100 does that right in the box. It can be a translator or a straight N2K gateway, as seen with N2KAnalyzer and N2KView.
Ben, does the USB100 put N2K data on a comm port in a "standard" way? Indeed, is there a standard for serial presentation of N2K data?
Will the USB100 work as a straight N2K gateway for non-Maretron software?
Adam, There is no "standard" way to access N2K data on a PC...yet. The NMEA is working on a standard called the Third Party Gateway but it is not yet published. As I understand it a developer will work with an SDK provided by a Gateway manufacturer like Maretron or Activision and the resulting program will be certified to work properly with that particular gateway.
In fact, Rose Point's Coastal Explorer can already address N2K directly through the USB100, using a patch program, but they don't recommend it because they don't yet support any data that isn't already translated into NMEA 0183 by the USB100. It's going to happen, though, I'm pretty sure, and it's going to be good.
http://beta.coastalexplorer.net/wiki/Maretron_N2K_Gateway.txt
Ben, that's interesting. So at the moment we have the USB100, which normally operates in 0183 conversion mode but can just dump the N2K stream onto the comm port, and the NGT-1, which more or less just dumps the N2K stream onto the comm port and has no 0183 support.
(I did some research on writing a virtual comm port that presents the N2K stream from an NGT-1 as 0183. It looks manageable and I might consider doing some work in that direction.)
The big question is of course whether the N2K stream from these two devices (and similar forthcoming ones) is the same or different. I'm going to guess the latter. (Are you able to test whether the Maretron N2K plugin for CE works with the NGT-1?)
Presumably this forthcoming TPG protocol will address the situation but I have to say that in apparently typical NMEA fashion this effort is woefully behind. The idea that PC plotting is now mainstream -- CE, MaxSea TZ, Nobeltec -- but that you can't get a N2K stream onto your PC in any standard way seems nuts, but "nuts" appears to be par for the course in marine electronics.
Frankly I'm surprised vendors like Actisense have even bothered to enter the market with a N2K-only product before the TPG is defined. You'd expect they would want to implement it in firmware, though I guess that updated drivers could handle any necessary translation.
On a related note, I've been interviewing marine electronics installers for my upcoming project and 2/3 have said to me either "we haven't done N2K" or "we don't do PC networking". Amazing.