After a four week period of time, I have not received any feedback from B&G USA or Navico Corp. about this problem.

However, running an "end sweep" thanks to some professional and interested ex-B&G software experts, they were able to confirm the error in B&G's N2k network sub-routine deriving magnetic heading.  The explanation we received is that the heading compass that was used to prototype the system apparently outputted magnetic heading in the field allocated for "heading sensor reading".

Also these now independent experts contacted and discussed the bug with B&G/Navico and learned that the flawed N2k network sub-routine code for deriving magnetic heading is operating in many B&G and Navico systems.  An anonymous B&G software developer promised to correct the sub-routine, but also advised that, "due to the extensive platforms where this code is operating, and due to B&G detailed software quality assurance" it will take B&G/Navico a "long" time to issue firmware that corrects the bug.

In the meanwhile the ex-B&G experts and current B&G employees verified that the "old" 0183 sub-routine that derived magnetic heading works properly!!!  This leads to a "workaround" which they inform is wire the compass's 0183 heading output (10Hz rate) to the B&G H5000 cpu processor 0183 input and in Sources page, assign the heading input to the 0183 port that the heading sensor is wired to.  I will do this and report if the work around actually works or not.

One has to have a sense of humor operating a boat, the irony in this particular case is that the justification to spend $15k upgrading navigation and sailing instruments was due in no small part to eliminate two proprietary compasses, one 0183 compass, and one N2k compass from the boat; and replace these heading sensors with one compass writing the same heading data to all instruments/systems on the boat!  Ha. 

Which heading sensor?  The Precision 9?

Hello Dan:

I don't know what heading sensor B&G used to develop their N2K software running in instruments/systems deriving magnetic heading.

According to NMEA PGN 127250, the heading sensor should output (a) "heading sensor reading", (b) heading sensor deviation, and (c) variation if true heading is enabled.  The instrument/system reading PGN 127250 is supposed to derive magnetic heading to be "heading sensor reading" plus "deviation".  If the instrument/system is operating in true heading mode, then true heading is magnetic heading plus variation.

I hope this information is what you are interested in.

Oh, you mean the defective software is in the H5000 receiving PGN 127250 and that the make/model of the heading sensor sending PGN127250 does not at all matter? 

Hmm. So when I calibrate my heading sensor, I ended up positioning it in the wrong direction and that cancels out?   So I wouldn't notice the anomaly as my autopilot computer and head are on a different (ancient proprietary) network, and positioned low so that I don't even look at the AP heads's magnetic course number.

Hi Dan:

The defective software resides in the H5000 CPU which does not derive magnetic heading properly according to NMEA PGN 127250 standard. 

I am advised that B&G's H5000 software derives magnetic heading properly when reading NMEA 0183 heading data.

I hope this helps.

Would you know if the Zeus 3 has the same problems. e.g. if I do this 0183 fix for the H5000 processor, will the Z3 follow along or conflict?

Hello Dan:

I have been informed by ex B&G software Experts that the "buggy" subroutine B&G uses to derive magnetic heading using N2K formatted data affects all Navico products deriving magnetic heading using N2k data.  This would then include the Zeus product family.

There is a simple test to determine if your Zeus product is calculating magnetic heading properly, use a software tool like NMEA reader and a NMEA 2000 to USB converter and observe PGN 127250 broadcast on your N2k network.  Find "heading compass reading" and "deviation". (assuming heading compass is properly compensated).  Add heading compass reading plus deviation (keeping track of deviation sign).  The result is magnetic heading.  Compare this result with the magnetic heading reported by Zeus.  This test is irrelevant in the case where heading compass deviation is zero.

I hope this is useful.

Thank you