Wednesday, June 28, 2023

Speed and Heading — Basic, But Not Always Simple

Speed and heading are two fundamental parameters in marine navigation, but as instruments and navigation programs become more sophisticated, we can face challenges in keeping these basic parameters in order.  I do not refer to speed over ground (SOG) or course over ground (COG), these one-time elusive parameters are now the simplest to understand and to measure. They come from the GPS instrumentation; meaning their abbreviations and place in NMEA sentences are all very clear.

The challenge comes with boat speed through the water (knotmeter speed) and more often with the heading of the boat, being the angle between the centerline and either true north or magnetic north. These are sensor measurements that get transmitted to the nav program via NMEA sentences or N2k parameters. 

The issue is that these parameters can be transmitted in more than one NMEA sentence, and these days it is not uncommon to have more than one heading sensor, and if these might be from different companies they might be telling you the heading in two different sentences. At that point we need to learn if the sensor software will let us choose which sentence to use, some do, some don't, or if our navigation program lets us filter out the ones we do not want—again, some do, some don't. 

That is in large part the issue. Seems simple enough, but I know from recent experience this can eat up hours of time sorting out, so these notes are an attempt to alleviate that issue—and to document the findings we needed to resolve a related case.

Heading is the more common issue, so we look at that first. Here are the primary NMEA sentences that involve the heading of the boat, keeping in mind that electronics can create proprietary sentences, plus some electronics still export old sentences that have been deprecated—a NMEA term meaning do not use these anymore. 



Old equipment still using HDT use it in the form: 

$--HDT,x.x,T*hh


Likewise you will also sometimes see:

$--HDM,x.x,M*hh.  Both of these are old.


When using N2k (See Introduction to NMEA 2000—with a Review of NMEA 0183) there will be a conversion between PGN parameters and NMEA sentences. The correlation is shown below as presented in an Actisense gateway manualSee a similar list at RosePoint Navigation






Note that the vessel heading (PGN 127250) can be exported to several NMEA sentences, which can potentially cause a conflict if other devices are also exporting the heading.  


Abbreviations in ECS


Now taking a look at how these parameters are presented in various ECS, with a note that electronic charting system (ECS) is the official name of any navigation program that is not a type-approved ECDIS. In short, any "nav program," "charting software," or console "chart plotter" interface to a GPS sensor. We cover this distinction in our text Introduction to Electronic Chart Navigation.



It is not easy to find official sources for recommended abbreviations. Bowditch, for example, 2021 edition, page 437, Glossary of Abbreviations, gives S for speed and Hdg for heading. STW and CTW are not listed. S is also used for set, south, slow, sand, and the sea-air temperature difference correction!


A more useful source is the International Electrotechnical Commission (IEC) Glossary online  where we find these definitions:



According to this important source, CTW and STW are well recognized abbreviations. Also it appears that CTW and "heading" could be used interchangeably. In principle we have a heading at the dock but not a CTW. ECS that use both (HDG and CTW), such as qtVlm and Expedition (identifies CTW as "course"), are more general.  Some consider that CTW should be HDG + leeway, but this is just the sort of thing that has to be sorted out within the ECS in use.


Looking at the speed parameter, that can be made complex as well. Generally we think of STW as the speed in the direction we are pointed, our HDG, but if we have much leeway it will reduce the proper knotmeter reading, as shown below and discussed in Leeway effect of knotmeter speed.




New age instruments

There is every reason to be sure to have a traditional paddle wheel knotmeter as well as both a magnetic compass and a magnetic heading sensor on the boat, but with those in place we can improve on both with new GPS based sensors, often called "satellite compass" or "satellite speed log" and indeed some provide both data and more.

A single GPS sensor can measure SOG and COG but it has no way to know your heading, but two GPS sensors in line with the centerline, say one at the bow and one at the stern, does know your heading, which is just the bearing from your stern location to your bow location.  Two such sensors make up a "satellite compass" that can measure your true heading at any time, in addition to COG and SOG, being how the pair is moving relative to the fixed earth. The heading can also be exported in magnetic units using the USGS software program called GeoMag. It computes accurate variation based on your Lat and Lon and the World Magnetic Model, which is updated every five years. (You can access GeoMag on your phone with the USGC app CrowdMag.)

The technology for this type of instrument is well advanced; the two GPS units do not have to be on the bow and stern, they can be just a foot or so apart in a single unit. There are several models, in the $1,000 price range.  There is a lot of math involved and they do require at least 5 satellites in view for best performance. Once the heading is known, the SOG in the direction of the COG can be projected onto the direction of the heading, called the longitudinal velocity, which is the speed you are moving in the direction you are headed, and projected onto the perpendicular direction, called the transverse velocity. The latter is the speed you are moving sideways. 





Here we see that one GPS tells COG the course over ground, and we can imagine the computation of SOG as the distance between the locations at time T1 and T2 divided by the time interval (T2-T1).  With two GPS sensors we can compute HDG and then project the SOG vector onto that heading and the transverse direction.

Note that we are not measuring leeway. The instrument does not even know we are in water. When this data gets reported to us in a NMEA sentence it would likely be VBW.

We would be getting ground speeds, not water speeds. This does not help analyze leeway but could be helpful docking and it could help us interpret current sets and our thinking on the derived parameter course made good (VMC).

Furuno has an attractive advancement of this concept in their SCX 20/21 Satellite compass. They use four GPS sensors so they get more reference lines to compute around, six in principle, two fore and aft, two abeam, and two diagonal.  This way, with still more math, they can measure and account for all vessel motions (roll, yawl, pitch) and correct for them, in addition to reporting heel angle (roll).  This allows them to get good results with fewer satellites.  It is a relatively small box, with list price of about $1,400, and taking only about 2 watt of power.  We see these on several high-tech race boats.


On the left is the SCX 21 for NMEA 0183 and the one sitting loose on the deck is the SCX 20 for N2k. This image and the one below are from a 2020  Panbo review of the instrument. The latest instrument includes 2023 improvements.


The displays from the two units above. Note that the headings differ as neither one has been carefully aligned with the center line in the demo pictured here.

I do not have direct experience with this unit, but the specs and output seem impressive:

— heading to ± 1º
— accurate, stable heading leads to clean radar trails
— very high GPS accuracy of ± 5m
— SOG to ± 0.02 kt with 5 satellites
— VBW speeds to ± 0.02 kts
— rate of turn (ROT) output for AIS broadcasts
— barometer ± 1mb with calibration offset option            
— air temp ± 2ºC 
— heel output for leeway computations and anemometer corrections
— data smoothing and offsets available on all outputs
— DR mode for lost GPS signals

_______________

I have been reminded in a comment by navigator Mark D'Arcy of the related issue that ECS often have access to multiple GPS sensors, which in turn can present similar conflicts that might lead to the vessel position jumping around a bit on the screen. Most ECS and indeed all ECDIS have a means of selecting what will be the primary source of GPS and setting this will solve that problem.