Imux NMEA communication error

[Tabaqui]

Had a similar issue. Also a Raymarine/iMux/OpenCPN system. Position would for a second or so swap south/north and the back again (Live at N70deg, suddenly position is shown as S70deg).Annoying as it trips the anchoralarm. Only noticed this on OpenCPN tho, not on the chartplotter or any of the iPads/phones that use the NMEA stream.


Not really sorted it, more a workaround. Set up OpenCPN to ignore the RMC sentence from the network and use a GPS puck instead.


Edit: I have tried enabling and disabling NMEA cheksum. Did not do any difference. Both the VHF and the chartplotter have separate GPSantennas, error remains regardless of wich is turned on.

[rgleason]

Originally Posted by Tabaqui

Not really sorted it, more a workaround. Set up OpenCPN to ignore the RMC sentence from the network and use a GPS puck instead.

Edit: I have tried enabling and disabling NMEA cheksum. Did not do any difference. Both the VHF and the chartplotter have separate GPSantennas, error remains regardless of which is turned on.

Agreed, with a puck GPS all is good for me too.

[mkgsail]

I am back aboard our boat finally and have resumed working on this issue. I now realize that, at power up of the system, the first RMC sentences are coming from the AIS system as I intend (GPRMC). But shortly thereafter, the sentences change to IIRMC. I assume this means the RMC sentence is being generated on the seatalk bus which should only happen if I had a GPS on the Seatalk bus. The iMux shares the AIS RMC data if there is no GPS data on the seatalk bus (my intended operating mode) but if it detects GPS data on seatalk it makes that the GPS source.
I have no idea where the IIRMC would be coming from. This only happens when I route the iMux NMEA 0183 out to the autopilot computer. But per the Raymarine manual it does not produce an RMC sentence. The only lat/long output it is supposed to produce is GLL and I do not see that via opencpn. I intend to write Brookhouse and ask them if the imux might create a RMC from data available.
Side note: the IIRMC data time field is populated with the time of the first IIRMC sentence after powerup and remains static thereafter.
And the core problem remains: After 1-2 minutes the latitude of the IIRMC is wrong and varies considerably.

[Tehani]

Hi all. I am new here, but I have seen for years that your problem is quite frequent, and it has some variants.
In general, this problem occurs when a device receives the same data from two or more different sources, or as it is in this case, when a data loop occurs.
The mkgsail installation is supplying GPS data to the imux via NMEA0183 from the AIS. When the RS422 output of the imux is connected also to the pilot it happens that the pilot translates the GPS NMEA0183 data to seatalk. The imux translates again from seatalk to NMEA, and resends. We have a data loop.
It is difficult to solve this problem with an imux Brookhouse, because it is not able to translate from NMEA0183 to Seatalk and thus avoid the loop. It is also very likely that this imux will only translate the low resolution seatalk datagrams for the coordinates (0x50 and 0x51) and ignore the high resolution datagram (0x58).
I understand your goal: you want the pilot takes GPS and also navigation (RMB, XTE) data, but this can only be achieved by substituting the imux for a bidirectional gateway in all channels (WiFi included). With this device you will simplify the installation and everything will work. In this case you should not connect anything to the pilot's NMEA0183 input. My advice is that you buy one that also has NMEA2000 to be able to adapt new equipment in the future.

[mkgsail]

Tehani,
Thank you your observations on the situation. I do believe it is something along the lines of what you describe. However, the Smartpilot is not supposed to create an RMC sentence so it is not clear to me how what is apparently a seatalk RMC sentence is being generated.
Also, I have the iMux-ST which does translate between Seatalk and NMEA-0183.
It is also not clear to me how a dataloop as you describe it could cause a latitude data error. Even if various NMEA-0183 instruments are repeating the data, that is not a reason for the data to become corrupted unless NMEA protocols are being violated.
I have an email into Brookhouse for their advise on this subject. But while they were quick to respond last year, I have yet to hear back from them this year. I hope they have not gone out of business. The email is not bouncing so ??

[Tehani]

Mkgsail,
When a data loop exists, a very negative effect can occur for all the devices. That is, the data that originates the loop occupies all the available time without leaving the communication channel free at any time.
This may be the origin of the erroneous data. Collapses occur in NMEA and / or collisions in the case of Seatalk.
SmartPilot does not create an RMC statement, but creates its equivalent in Seatalk (Datagrams 0x50, 0x51 and 0x58 = Coordinates) when you connect the GPS data to the pilot input.
These Seatalk datagrams are sent to the imux, and translated back to NMEA0183. Here is the problem.