I do not know how many people here could have the problem I will talk
about, but I have a 36 ft sailboat, several Garmin Gps and an Autohelm
Autopilot, and I have used the NMEA interface since 1997, not without
problems.

The fact that worries me more is that the problem is recurring every
time Garmin launches a new gps model, appearing that, every time, they
forget the need of maintain the compatibility with older units. Now,
the problem reappeared again, and this is the third time I will have
to fight with their technical support people because this problem.

Please, read the following problem report, and perhaps, if you have
similar hardware and application, do the tests I have done and report
the results here, and if possible, join me in the fight with Garmin,
so that they correct the problem in a more definitive way, like I am
proposing, and to do so, open yourself a new problem with Garmin
technical support, based in your own reasons.

Thanks

Pascal Goncalves

S13 00/W 038 27

GpsMap 276C NMEA Interface Problem

Background

I am a Garmin customer since 1992, and I had owned more than 10 Garmin
gps models since that time. My use of gps is mainly in marine mode,
since I have an 36 feet sailboat witch have an Autohelm ST4000
autopilot working interfaced with my 1997 Garmin GpsMap 130. At
that time, I have found a problem with this interface, as I observed
that the bearing to next Wpt (BRG) showed by autopilot , was 23
degrees less than the waypoint BRG showed on the gps, what caused the
initial error of equal magnitude (23 degrees) to the left, on the
course of boat.

As the local magnetic declination here in Salvador – Bahia – Brazil
(S13 /W 038) is 23 degrees West, I concluded that the autopilot was
incorrectly applying the Magnetic deviation correction to the True
Bearing, thinking it was Magnetic Bearing. One day, I have changed
the NMEA version used by GpsMap 130 from 2.0 to 1.5 and the problem
disappeared.

Four years ago, in 2001, I have purchased a new GpsMap 76, and one
year later, when Blue Chart finally supported the South America East
Coast, I decide to use it interfaced with the same old autopilot, and
them, the problem of 23 degrees was found again, and we had reported
this problem along others to Garmin.

Because a problem with the limitation on the length of waypoint names
supported by the old autopilots (max 4 characters) witch not allowed
the use of trackback of the GpsMap 76 with old autopilots, Garmin
make an improvement on the autopilot compatibility, introducing in
version 2.90, the so called "Advanced NMEA Setup" and after upgrading
the software of my Map76 the problem of the 23 degrees disappeared
again. I ended thinking that Garmin finally changed something that
was in error with the GPRMB NMEA message sent by the gps, the same
time it introduced the ANS.

Last year (2004), I had purchased a new Garmin GpsMap 276C and a few
weeks ago (28/jan/2005), I have finally interfaced the new
Map276C with the AH autopilot, and for my surprise, the old 23
degrees problem reappeared again.




Test Results

I had spent the last 2 weeks analyzing the problem, making several
tests, using several autopilots and gps combinations, and displaying
the NMEA messages with Windows HyperTerminal on my laptop, in order to
determine what is causing the problem.

• Tests witch resulted in the 23 degrees problem:
o GpsMap 60C with AH/ST 2000+
o GpsMap 276C software 2.7 with AH/ST 2000 Plus
o GpsMap 130 NMEA v 2.0 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 4000 (*)
o GpsMap 276C software 2.7 with AH/ST 7000
o Gps V software 2.5 with AH/ST 4000
o Gps III Plus with AH/ST 4000

• Tests witch NOT showed the 23 degrees problem:
o GpsMap 130 NMEA v 1.5 with AH/ST 4000(*)
o GpsMap 176C software 4.0 with AH/ST 4000
o GpsMap 176C software 4.0 with AH/ST 6000
o GpsMap 76 software 3.8 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 5000 Plus (software #14)
o Gps III Plus with AH/ST 5000 Plus (software #14)
o Gps Magellan with AH/ST 2000+
o Gps NavMan with AH/ST 6001

(*) My owned equipment

I have obtained the NMEA OUTPUT messages sent by some of the gps
involved in the tests, using the HyperTerminal software, and the
result was:

• Gps witch SEND the GPBWC message: GpsMap 76, GpsMap 176C,
GpsMap130 with NMEA v 1.5
• Gps which NOT send GPBWC message: GpsMap276C, GpsMap 60C, GpsMap130
with NMEA v 2.0, Gps V, Gps III Plus.


As a sample, see below the messages sent by GpsMap 76 and GpsMap 276C:

GpsMap 76 NMEA output:

$GPRMC,002040,V,1248.678,S,03827.627,W,10.0,344.7, 210205,23.0,W,S*0E
$GPRMB,V,0.00,L,AIC,BA01,1248.413,S,03827.698,W,0. 275,345.4,10.0,V,S*2F
$GPGGA,002040,1248.678,S,03827.627,W,8,10,2.0,18.6 ,M,-10.5,M,,*5C
$GPGLL,1248.678,S,03827.627,W,002040,V,S*40
$GPBOD,345.6,T,8.6,M,BA01,AIC*04
$GPBWC,002040,1248.413,S,03827.698,W,345.4,T,8.4,M ,0.275,N,BA01,S*7D
$GPVTG,344.7,T,7.7,M,10.0,N,18.5,K*47
$GPXTE,V,V,0.00,L,N,S*11

GpsMap 276C NMEA output:

$GPRMC,124752,A,1248.704,S,03827.581,W,0.0,103.9,2 40205,23.0,W,A*34
$GPRMB,A,0.22,L,,G1,1247.472,S,03827.840,W,1.260,3 48.4,,V,A*01
$GPGGA,124752,1248.704,S,03827.581,W,1,11,0.9,2.7, M,-10.5,M,,*61
$GPGLL,1248.704,S,03827.581,W,124752,A,A*41
$GPBOD,358.3,T,21.4,M,G1,*0B

Notes:

1. As I have only the PC/USB cable that not allows the use of NMEA, I
had made an adaptation in the original power/data cable witch came
with the unit to connect it to a serial DB9 port in my laptop in order
to obtain via HyperTerminal, the NMEA data sent by the gps.
2. The manual of the GpsMap 76 do not says that it sends the BWC
message, but my tests showed that it sends it, at last in the 3.8
version, witch works OK with the autopilot.
3. I became very surprised when I have taken a close look at the
GpsMap 276C manual, as it says, in page 90 that the unit ALWAYS
sends the following: RMC, GGA, GLL, BWC, VTG, XTE, and RMB, but In
page 107, the manual says it sends RMC, GGA, GSA, GSV, GLL, GPBOD,
RTE, and WPL. (nothing about RMB and XTE and BWC)


Considerations


1) Looking at the tests results, we can see that the GpsMap 276C (as
the GpsMap 60C, GpsV, Gps III Plus, and the GpsMap 130 NMEA 2.0) does
not send the GPBWC, witch I think, is the message that works well with
the old autopilots.
2) The manual of the AH/ST 4000 Plus states that the autopilot gets
the Waypoint name and the Bearing to Waypoint from messages APA,
APB, BWR, BWC and RMB, in this order, and I think that it uses the
RMB only if the BWC is not received.
3) We had made some tests with the AH/ST2000+ with at last two
Magellan gps receivers, witch sends APA, APB etc, not only RMB, and
the interface worked OK.
4) I believe that there is an incompatibility or software error in
these old AH autopilots software, which interprets the wrong way the
RMB message, mistaking the True Bearing as if it was Magnetic Bearing,
and so, applying the magnetic deviation correction (W23 = - 23) witch
is causing the problem.
5) It is impossible to correct the software of these old units,
because their design does not allow this and because all these units
are out of production.
6) The only unit in the test that does not showed the 23 degrees
problem was the AH/ST 5000 Plus which is in reality, a brand new unit,
with last year updated software to version 14, made by RayMarine , not
an old Autohelm unit, so probably it has the software corrected for
the problem with RMB message. I had tested it with GpsMap 276C and
with a Gps III Plus.

Final Conclusions

1) As Garmin GpsMap 130 NMEA V 1.5, GpsMap 76 V3.8, and GpsMap 176C
4.0 all works OK with the old AH autopilots listed, and these units
also sends the BWC message, I think that the problem with the RMB
message could be easily bypassed if the gps sends the BWC.

2) So, I am asking that Garmin analyses this and make ASAP, in a next
software update, a change in the GpsMap 276C software so that the BWC
NMEA message is included in the batch of messages always sent by
GpsMap 276C, as stated well clear in page 90 of the manual.

3) As a better idea, why the "Advanced NMEA Setup" of the GpsMap 276C
could not be similar of the GpsMap 3006/3010, witch allows the
selection of each sentence that the gps unit will send? This is the
obvious solution for the problem of compatibility in the communication
with other NMEA devices, as stated in this Garmin Manual. See pages
73-74 of the 3010 manual where it says: "check any other equipment's
owner manual to see what sentences are required to be enabled for
functionality".


============ The END =============

Nice analysis Pascal!
A few suggestions:
First of all, to verify if your conclusions are correct, download a (free)
trial-copy of a program called "Nematalker". You'l find the site with a
Google search. Under "Custom" you can define your own GPS sentences. By
generating a set of sentences to control your autopilot you can experiment
and see how your autopilot reacts if you leave the RMB sentence out and use
WBC instead or if you change/delete parameters in the RMB.
An important question is wether the Magnetic Variation used by your
autopilot is the value as received from the GPS in the RMC sentence, or if
it is the internally stored value for variation, defined with setup.
If the autopilot uses the variation from the GPS, one solution would be to
set var in the RMC to 0.
If you purchase a NMEA multiplexer from Brookhouse (standard model:
US$149.00) it will include a number of features that allow you to adapt or
modify the NMEA data from the GPS. The "editing on the fly" feature can be
used to set magnetic variation in the RMC to 0 or, if that does not work,
you could "convert" the RMB to a BWC sentence. Some of the parameters for
BWC are not available in RMB, but the autopilot probably only looks at
lat/long, distance and bearing.
You can try this all out in advance by simulation using Nematalker. Once you
know what changes are necessary in your Garmin output to make your autopilot
work correctly, we can see what mux "edit on the fly" directives are
necessary to achieve this.
If you can convince Garmin to add the BWC sentence, that would be great, but
I doubt it.
Regards,
Wout

"Pascal Goncalves" wrote in message
om...
I do not know how many people here could have the problem I will talk
about, but I have a 36 ft sailboat, several Garmin Gps and an Autohelm
Autopilot, and I have used the NMEA interface since 1997, not without
problems.

The fact that worries me more is that the problem is recurring every
time Garmin launches a new gps model, appearing that, every time, they
forget the need of maintain the compatibility with older units. Now,
the problem reappeared again, and this is the third time I will have
to fight with their technical support people because this problem.

Please, read the following problem report, and perhaps, if you have
similar hardware and application, do the tests I have done and report
the results here, and if possible, join me in the fight with Garmin,
so that they correct the problem in a more definitive way, like I am
proposing, and to do so, open yourself a new problem with Garmin
technical support, based in your own reasons.

Thanks

Pascal Goncalves

S13 00/W 038 27

GpsMap 276C NMEA Interface Problem

Background

I am a Garmin customer since 1992, and I had owned more than 10 Garmin
gps models since that time. My use of gps is mainly in marine mode,
since I have an 36 feet sailboat witch have an Autohelm ST4000
autopilot working interfaced with my 1997 Garmin GpsMap 130. At
that time, I have found a problem with this interface, as I observed
that the bearing to next Wpt (BRG) showed by autopilot , was 23
degrees less than the waypoint BRG showed on the gps, what caused the
initial error of equal magnitude (23 degrees) to the left, on the
course of boat.

As the local magnetic declination here in Salvador - Bahia - Brazil
(S13 /W 038) is 23 degrees West, I concluded that the autopilot was
incorrectly applying the Magnetic deviation correction to the True
Bearing, thinking it was Magnetic Bearing. One day, I have changed
the NMEA version used by GpsMap 130 from 2.0 to 1.5 and the problem
disappeared.

Four years ago, in 2001, I have purchased a new GpsMap 76, and one
year later, when Blue Chart finally supported the South America East
Coast, I decide to use it interfaced with the same old autopilot, and
them, the problem of 23 degrees was found again, and we had reported
this problem along others to Garmin.

Because a problem with the limitation on the length of waypoint names
supported by the old autopilots (max 4 characters) witch not allowed
the use of trackback of the GpsMap 76 with old autopilots, Garmin
make an improvement on the autopilot compatibility, introducing in
version 2.90, the so called "Advanced NMEA Setup" and after upgrading
the software of my Map76 the problem of the 23 degrees disappeared
again. I ended thinking that Garmin finally changed something that
was in error with the GPRMB NMEA message sent by the gps, the same
time it introduced the ANS.

Last year (2004), I had purchased a new Garmin GpsMap 276C and a few
weeks ago (28/jan/2005), I have finally interfaced the new
Map276C with the AH autopilot, and for my surprise, the old 23
degrees problem reappeared again.




Test Results

I had spent the last 2 weeks analyzing the problem, making several
tests, using several autopilots and gps combinations, and displaying
the NMEA messages with Windows HyperTerminal on my laptop, in order to
determine what is causing the problem.

. Tests witch resulted in the 23 degrees problem:
o GpsMap 60C with AH/ST 2000+
o GpsMap 276C software 2.7 with AH/ST 2000 Plus
o GpsMap 130 NMEA v 2.0 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 4000 (*)
o GpsMap 276C software 2.7 with AH/ST 7000
o Gps V software 2.5 with AH/ST 4000
o Gps III Plus with AH/ST 4000

. Tests witch NOT showed the 23 degrees problem:
o GpsMap 130 NMEA v 1.5 with AH/ST 4000(*)
o GpsMap 176C software 4.0 with AH/ST 4000
o GpsMap 176C software 4.0 with AH/ST 6000
o GpsMap 76 software 3.8 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 5000 Plus (software #14)
o Gps III Plus with AH/ST 5000 Plus (software #14)
o Gps Magellan with AH/ST 2000+
o Gps NavMan with AH/ST 6001

(*) My owned equipment

I have obtained the NMEA OUTPUT messages sent by some of the gps
involved in the tests, using the HyperTerminal software, and the
result was:

. Gps witch SEND the GPBWC message: GpsMap 76, GpsMap 176C,
GpsMap130 with NMEA v 1.5
. Gps which NOT send GPBWC message: GpsMap276C, GpsMap 60C, GpsMap130
with NMEA v 2.0, Gps V, Gps III Plus.


As a sample, see below the messages sent by GpsMap 76 and GpsMap 276C:

GpsMap 76 NMEA output:

$GPRMC,002040,V,1248.678,S,03827.627,W,10.0,344.7, 210205,23.0,W,S*0E
$GPRMB,V,0.00,L,AIC,BA01,1248.413,S,03827.698,W,0. 275,345.4,10.0,V,S*2F
$GPGGA,002040,1248.678,S,03827.627,W,8,10,2.0,18.6 ,M,-10.5,M,,*5C
$GPGLL,1248.678,S,03827.627,W,002040,V,S*40
$GPBOD,345.6,T,8.6,M,BA01,AIC*04
$GPBWC,002040,1248.413,S,03827.698,W,345.4,T,8.4,M ,0.275,N,BA01,S*7D
$GPVTG,344.7,T,7.7,M,10.0,N,18.5,K*47
$GPXTE,V,V,0.00,L,N,S*11

GpsMap 276C NMEA output:

$GPRMC,124752,A,1248.704,S,03827.581,W,0.0,103.9,2 40205,23.0,W,A*34
$GPRMB,A,0.22,L,,G1,1247.472,S,03827.840,W,1.260,3 48.4,,V,A*01
$GPGGA,124752,1248.704,S,03827.581,W,1,11,0.9,2.7, M,-10.5,M,,*61
$GPGLL,1248.704,S,03827.581,W,124752,A,A*41
$GPBOD,358.3,T,21.4,M,G1,*0B

Notes:

1. As I have only the PC/USB cable that not allows the use of NMEA, I
had made an adaptation in the original power/data cable witch came
with the unit to connect it to a serial DB9 port in my laptop in order
to obtain via HyperTerminal, the NMEA data sent by the gps.
2. The manual of the GpsMap 76 do not says that it sends the BWC
message, but my tests showed that it sends it, at last in the 3.8
version, witch works OK with the autopilot.
3. I became very surprised when I have taken a close look at the
GpsMap 276C manual, as it says, in page 90 that the unit ALWAYS
sends the following: RMC, GGA, GLL, BWC, VTG, XTE, and RMB, but In
page 107, the manual says it sends RMC, GGA, GSA, GSV, GLL, GPBOD,
RTE, and WPL. (nothing about RMB and XTE and BWC)


Considerations


1) Looking at the tests results, we can see that the GpsMap 276C (as
the GpsMap 60C, GpsV, Gps III Plus, and the GpsMap 130 NMEA 2.0) does
not send the GPBWC, witch I think, is the message that works well with
the old autopilots.
2) The manual of the AH/ST 4000 Plus states that the autopilot gets
the Waypoint name and the Bearing to Waypoint from messages APA,
APB, BWR, BWC and RMB, in this order, and I think that it uses the
RMB only if the BWC is not received.
3) We had made some tests with the AH/ST2000+ with at last two
Magellan gps receivers, witch sends APA, APB etc, not only RMB, and
the interface worked OK.
4) I believe that there is an incompatibility or software error in
these old AH autopilots software, which interprets the wrong way the
RMB message, mistaking the True Bearing as if it was Magnetic Bearing,
and so, applying the magnetic deviation correction (W23 = - 23) witch
is causing the problem.
5) It is impossible to correct the software of these old units,
because their design does not allow this and because all these units
are out of production.
6) The only unit in the test that does not showed the 23 degrees
problem was the AH/ST 5000 Plus which is in reality, a brand new unit,
with last year updated software to version 14, made by RayMarine , not
an old Autohelm unit, so probably it has the software corrected for
the problem with RMB message. I had tested it with GpsMap 276C and
with a Gps III Plus.

Final Conclusions

1) As Garmin GpsMap 130 NMEA V 1.5, GpsMap 76 V3.8, and GpsMap 176C
4.0 all works OK with the old AH autopilots listed, and these units
also sends the BWC message, I think that the problem with the RMB
message could be easily bypassed if the gps sends the BWC.

2) So, I am asking that Garmin analyses this and make ASAP, in a next
software update, a change in the GpsMap 276C software so that the BWC
NMEA message is included in the batch of messages always sent by
GpsMap 276C, as stated well clear in page 90 of the manual.

3) As a better idea, why the "Advanced NMEA Setup" of the GpsMap 276C
could not be similar of the GpsMap 3006/3010, witch allows the
selection of each sentence that the gps unit will send? This is the
obvious solution for the problem of compatibility in the communication
with other NMEA devices, as stated in this Garmin Manual. See pages
73-74 of the 3010 manual where it says: "check any other equipment's
owner manual to see what sentences are required to be enabled for
functionality".


============ The END =============

Have a look he http://brookhouseonline.com/nmeamux.htm or he
http://brookhouseonline.com/pdf%20fi...%20Editing.pdf for
details about the on the fly editing.
Wout


"Wout B" wrote in message
...
Nice analysis Pascal!
A few suggestions:
First of all, to verify if your conclusions are correct, download a (free)
trial-copy of a program called "Nematalker". You'l find the site with a
Google search. Under "Custom" you can define your own GPS sentences. By
generating a set of sentences to control your autopilot you can experiment
and see how your autopilot reacts if you leave the RMB sentence out and
use
WBC instead or if you change/delete parameters in the RMB.
An important question is wether the Magnetic Variation used by your
autopilot is the value as received from the GPS in the RMC sentence, or if
it is the internally stored value for variation, defined with setup.
If the autopilot uses the variation from the GPS, one solution would be to
set var in the RMC to 0.
If you purchase a NMEA multiplexer from Brookhouse (standard model:
US$149.00) it will include a number of features that allow you to adapt
or
modify the NMEA data from the GPS. The "editing on the fly" feature can be
used to set magnetic variation in the RMC to 0 or, if that does not work,
you could "convert" the RMB to a BWC sentence. Some of the parameters for
BWC are not available in RMB, but the autopilot probably only looks at
lat/long, distance and bearing.
You can try this all out in advance by simulation using Nematalker. Once
you
know what changes are necessary in your Garmin output to make your
autopilot
work correctly, we can see what mux "edit on the fly" directives are
necessary to achieve this.
If you can convince Garmin to add the BWC sentence, that would be great,
but
I doubt it.
Regards,
Wout

"Pascal Goncalves" wrote in message
om...
I do not know how many people here could have the problem I will talk
about, but I have a 36 ft sailboat, several Garmin Gps and an Autohelm
Autopilot, and I have used the NMEA interface since 1997, not without
problems.

The fact that worries me more is that the problem is recurring every
time Garmin launches a new gps model, appearing that, every time, they
forget the need of maintain the compatibility with older units. Now,
the problem reappeared again, and this is the third time I will have
to fight with their technical support people because this problem.

Please, read the following problem report, and perhaps, if you have
similar hardware and application, do the tests I have done and report
the results here, and if possible, join me in the fight with Garmin,
so that they correct the problem in a more definitive way, like I am
proposing, and to do so, open yourself a new problem with Garmin
technical support, based in your own reasons.

Thanks

Pascal Goncalves

S13 00/W 038 27

GpsMap 276C NMEA Interface Problem

Background

I am a Garmin customer since 1992, and I had owned more than 10 Garmin
gps models since that time. My use of gps is mainly in marine mode,
since I have an 36 feet sailboat witch have an Autohelm ST4000
autopilot working interfaced with my 1997 Garmin GpsMap 130. At
that time, I have found a problem with this interface, as I observed
that the bearing to next Wpt (BRG) showed by autopilot , was 23
degrees less than the waypoint BRG showed on the gps, what caused the
initial error of equal magnitude (23 degrees) to the left, on the
course of boat.

As the local magnetic declination here in Salvador - Bahia - Brazil
(S13 /W 038) is 23 degrees West, I concluded that the autopilot was
incorrectly applying the Magnetic deviation correction to the True
Bearing, thinking it was Magnetic Bearing. One day, I have changed
the NMEA version used by GpsMap 130 from 2.0 to 1.5 and the problem
disappeared.

Four years ago, in 2001, I have purchased a new GpsMap 76, and one
year later, when Blue Chart finally supported the South America East
Coast, I decide to use it interfaced with the same old autopilot, and
them, the problem of 23 degrees was found again, and we had reported
this problem along others to Garmin.

Because a problem with the limitation on the length of waypoint names
supported by the old autopilots (max 4 characters) witch not allowed
the use of trackback of the GpsMap 76 with old autopilots, Garmin
make an improvement on the autopilot compatibility, introducing in
version 2.90, the so called "Advanced NMEA Setup" and after upgrading
the software of my Map76 the problem of the 23 degrees disappeared
again. I ended thinking that Garmin finally changed something that
was in error with the GPRMB NMEA message sent by the gps, the same
time it introduced the ANS.

Last year (2004), I had purchased a new Garmin GpsMap 276C and a few
weeks ago (28/jan/2005), I have finally interfaced the new
Map276C with the AH autopilot, and for my surprise, the old 23
degrees problem reappeared again.




Test Results

I had spent the last 2 weeks analyzing the problem, making several
tests, using several autopilots and gps combinations, and displaying
the NMEA messages with Windows HyperTerminal on my laptop, in order to
determine what is causing the problem.

. Tests witch resulted in the 23 degrees problem:
o GpsMap 60C with AH/ST 2000+
o GpsMap 276C software 2.7 with AH/ST 2000 Plus
o GpsMap 130 NMEA v 2.0 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 4000 (*)
o GpsMap 276C software 2.7 with AH/ST 7000
o Gps V software 2.5 with AH/ST 4000
o Gps III Plus with AH/ST 4000

. Tests witch NOT showed the 23 degrees problem:
o GpsMap 130 NMEA v 1.5 with AH/ST 4000(*)
o GpsMap 176C software 4.0 with AH/ST 4000
o GpsMap 176C software 4.0 with AH/ST 6000
o GpsMap 76 software 3.8 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 5000 Plus (software #14)
o Gps III Plus with AH/ST 5000 Plus (software #14)
o Gps Magellan with AH/ST 2000+
o Gps NavMan with AH/ST 6001

(*) My owned equipment

I have obtained the NMEA OUTPUT messages sent by some of the gps
involved in the tests, using the HyperTerminal software, and the
result was:

. Gps witch SEND the GPBWC message: GpsMap 76, GpsMap 176C,
GpsMap130 with NMEA v 1.5
. Gps which NOT send GPBWC message: GpsMap276C, GpsMap 60C, GpsMap130
with NMEA v 2.0, Gps V, Gps III Plus.


As a sample, see below the messages sent by GpsMap 76 and GpsMap 276C:

GpsMap 76 NMEA output:

$GPRMC,002040,V,1248.678,S,03827.627,W,10.0,344.7, 210205,23.0,W,S*0E
$GPRMB,V,0.00,L,AIC,BA01,1248.413,S,03827.698,W,0. 275,345.4,10.0,V,S*2F
$GPGGA,002040,1248.678,S,03827.627,W,8,10,2.0,18.6 ,M,-10.5,M,,*5C
$GPGLL,1248.678,S,03827.627,W,002040,V,S*40
$GPBOD,345.6,T,8.6,M,BA01,AIC*04
$GPBWC,002040,1248.413,S,03827.698,W,345.4,T,8.4,M ,0.275,N,BA01,S*7D
$GPVTG,344.7,T,7.7,M,10.0,N,18.5,K*47
$GPXTE,V,V,0.00,L,N,S*11

GpsMap 276C NMEA output:

$GPRMC,124752,A,1248.704,S,03827.581,W,0.0,103.9,2 40205,23.0,W,A*34
$GPRMB,A,0.22,L,,G1,1247.472,S,03827.840,W,1.260,3 48.4,,V,A*01
$GPGGA,124752,1248.704,S,03827.581,W,1,11,0.9,2.7, M,-10.5,M,,*61
$GPGLL,1248.704,S,03827.581,W,124752,A,A*41
$GPBOD,358.3,T,21.4,M,G1,*0B

Notes:

1. As I have only the PC/USB cable that not allows the use of NMEA, I
had made an adaptation in the original power/data cable witch came
with the unit to connect it to a serial DB9 port in my laptop in order
to obtain via HyperTerminal, the NMEA data sent by the gps.
2. The manual of the GpsMap 76 do not says that it sends the BWC
message, but my tests showed that it sends it, at last in the 3.8
version, witch works OK with the autopilot.
3. I became very surprised when I have taken a close look at the
GpsMap 276C manual, as it says, in page 90 that the unit ALWAYS
sends the following: RMC, GGA, GLL, BWC, VTG, XTE, and RMB, but In
page 107, the manual says it sends RMC, GGA, GSA, GSV, GLL, GPBOD,
RTE, and WPL. (nothing about RMB and XTE and BWC)


Considerations


1) Looking at the tests results, we can see that the GpsMap 276C (as
the GpsMap 60C, GpsV, Gps III Plus, and the GpsMap 130 NMEA 2.0) does
not send the GPBWC, witch I think, is the message that works well with
the old autopilots.
2) The manual of the AH/ST 4000 Plus states that the autopilot gets
the Waypoint name and the Bearing to Waypoint from messages APA,
APB, BWR, BWC and RMB, in this order, and I think that it uses the
RMB only if the BWC is not received.
3) We had made some tests with the AH/ST2000+ with at last two
Magellan gps receivers, witch sends APA, APB etc, not only RMB, and
the interface worked OK.
4) I believe that there is an incompatibility or software error in
these old AH autopilots software, which interprets the wrong way the
RMB message, mistaking the True Bearing as if it was Magnetic Bearing,
and so, applying the magnetic deviation correction (W23 = - 23) witch
is causing the problem.
5) It is impossible to correct the software of these old units,
because their design does not allow this and because all these units
are out of production.
6) The only unit in the test that does not showed the 23 degrees
problem was the AH/ST 5000 Plus which is in reality, a brand new unit,
with last year updated software to version 14, made by RayMarine , not
an old Autohelm unit, so probably it has the software corrected for
the problem with RMB message. I had tested it with GpsMap 276C and
with a Gps III Plus.

Final Conclusions

1) As Garmin GpsMap 130 NMEA V 1.5, GpsMap 76 V3.8, and GpsMap 176C
4.0 all works OK with the old AH autopilots listed, and these units
also sends the BWC message, I think that the problem with the RMB
message could be easily bypassed if the gps sends the BWC.

2) So, I am asking that Garmin analyses this and make ASAP, in a next
software update, a change in the GpsMap 276C software so that the BWC
NMEA message is included in the batch of messages always sent by
GpsMap 276C, as stated well clear in page 90 of the manual.

3) As a better idea, why the "Advanced NMEA Setup" of the GpsMap 276C
could not be similar of the GpsMap 3006/3010, witch allows the
selection of each sentence that the gps unit will send? This is the
obvious solution for the problem of compatibility in the communication
with other NMEA devices, as stated in this Garmin Manual. See pages
73-74 of the 3010 manual where it says: "check any other equipment's
owner manual to see what sentences are required to be enabled for
functionality".


============ The END =============

Thank You very much Wout, I will folow your instructions and return
with my results. I have some hope that Gamin will fix this, but the
Mux is an atractive solution witch will add much more flexibility to
my "network". I must think a litle more on this.

Regards

Pascal





"Wout B" wrote in message ...
Have a look he http://brookhouseonline.com/nmeamux.htm or he
http://brookhouseonline.com/pdf%20fi...%20Editing.pdf for
details about the on the fly editing.
Wout


"Wout B" wrote in message
...
Nice analysis Pascal!
A few suggestions:
First of all, to verify if your conclusions are correct, download a (free)
trial-copy of a program called "Nematalker". You'l find the site with a
Google search. Under "Custom" you can define your own GPS sentences. By
generating a set of sentences to control your autopilot you can experiment
and see how your autopilot reacts if you leave the RMB sentence out and
use
WBC instead or if you change/delete parameters in the RMB.
An important question is wether the Magnetic Variation used by your
autopilot is the value as received from the GPS in the RMC sentence, or if
it is the internally stored value for variation, defined with setup.
If the autopilot uses the variation from the GPS, one solution would be to
set var in the RMC to 0.
If you purchase a NMEA multiplexer from Brookhouse (standard model:
US$149.00) it will include a number of features that allow you to adapt
or
modify the NMEA data from the GPS. The "editing on the fly" feature can be
used to set magnetic variation in the RMC to 0 or, if that does not work,
you could "convert" the RMB to a BWC sentence. Some of the parameters for
BWC are not available in RMB, but the autopilot probably only looks at
lat/long, distance and bearing.
You can try this all out in advance by simulation using Nematalker. Once
you
know what changes are necessary in your Garmin output to make your
autopilot
work correctly, we can see what mux "edit on the fly" directives are
necessary to achieve this.
If you can convince Garmin to add the BWC sentence, that would be great,
but
I doubt it.
Regards,
Wout

"Pascal Goncalves" wrote in message
om...
I do not know how many people here could have the problem I will talk
about, but I have a 36 ft sailboat, several Garmin Gps and an Autohelm
Autopilot, and I have used the NMEA interface since 1997, not without
problems.

The fact that worries me more is that the problem is recurring every
time Garmin launches a new gps model, appearing that, every time, they
forget the need of maintain the compatibility with older units. Now,
the problem reappeared again, and this is the third time I will have
to fight with their technical support people because this problem.

Please, read the following problem report, and perhaps, if you have
similar hardware and application, do the tests I have done and report
the results here, and if possible, join me in the fight with Garmin,
so that they correct the problem in a more definitive way, like I am
proposing, and to do so, open yourself a new problem with Garmin
technical support, based in your own reasons.

Thanks

Pascal Goncalves

S13 00/W 038 27

GpsMap 276C NMEA Interface Problem

Background

I am a Garmin customer since 1992, and I had owned more than 10 Garmin
gps models since that time. My use of gps is mainly in marine mode,
since I have an 36 feet sailboat witch have an Autohelm ST4000
autopilot working interfaced with my 1997 Garmin GpsMap 130. At
that time, I have found a problem with this interface, as I observed
that the bearing to next Wpt (BRG) showed by autopilot , was 23
degrees less than the waypoint BRG showed on the gps, what caused the
initial error of equal magnitude (23 degrees) to the left, on the
course of boat.

As the local magnetic declination here in Salvador - Bahia - Brazil
(S13 /W 038) is 23 degrees West, I concluded that the autopilot was
incorrectly applying the Magnetic deviation correction to the True
Bearing, thinking it was Magnetic Bearing. One day, I have changed
the NMEA version used by GpsMap 130 from 2.0 to 1.5 and the problem
disappeared.

Four years ago, in 2001, I have purchased a new GpsMap 76, and one
year later, when Blue Chart finally supported the South America East
Coast, I decide to use it interfaced with the same old autopilot, and
them, the problem of 23 degrees was found again, and we had reported
this problem along others to Garmin.

Because a problem with the limitation on the length of waypoint names
supported by the old autopilots (max 4 characters) witch not allowed
the use of trackback of the GpsMap 76 with old autopilots, Garmin
make an improvement on the autopilot compatibility, introducing in
version 2.90, the so called "Advanced NMEA Setup" and after upgrading
the software of my Map76 the problem of the 23 degrees disappeared
again. I ended thinking that Garmin finally changed something that
was in error with the GPRMB NMEA message sent by the gps, the same
time it introduced the ANS.

Last year (2004), I had purchased a new Garmin GpsMap 276C and a few
weeks ago (28/jan/2005), I have finally interfaced the new
Map276C with the AH autopilot, and for my surprise, the old 23
degrees problem reappeared again.




Test Results

I had spent the last 2 weeks analyzing the problem, making several
tests, using several autopilots and gps combinations, and displaying
the NMEA messages with Windows HyperTerminal on my laptop, in order to
determine what is causing the problem.

. Tests witch resulted in the 23 degrees problem:
o GpsMap 60C with AH/ST 2000+
o GpsMap 276C software 2.7 with AH/ST 2000 Plus
o GpsMap 130 NMEA v 2.0 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 4000 (*)
o GpsMap 276C software 2.7 with AH/ST 7000
o Gps V software 2.5 with AH/ST 4000
o Gps III Plus with AH/ST 4000

. Tests witch NOT showed the 23 degrees problem:
o GpsMap 130 NMEA v 1.5 with AH/ST 4000(*)
o GpsMap 176C software 4.0 with AH/ST 4000
o GpsMap 176C software 4.0 with AH/ST 6000
o GpsMap 76 software 3.8 with AH/ST 4000 (*)
o GpsMap 276C software 3.1 with AH/ST 5000 Plus (software #14)
o Gps III Plus with AH/ST 5000 Plus (software #14)
o Gps Magellan with AH/ST 2000+
o Gps NavMan with AH/ST 6001

(*) My owned equipment

I have obtained the NMEA OUTPUT messages sent by some of the gps
involved in the tests, using the HyperTerminal software, and the
result was:

. Gps witch SEND the GPBWC message: GpsMap 76, GpsMap 176C,
GpsMap130 with NMEA v 1.5
. Gps which NOT send GPBWC message: GpsMap276C, GpsMap 60C, GpsMap130
with NMEA v 2.0, Gps V, Gps III Plus.


As a sample, see below the messages sent by GpsMap 76 and GpsMap 276C:

GpsMap 76 NMEA output:

$GPRMC,002040,V,1248.678,S,03827.627,W,10.0,344.7, 210205,23.0,W,S*0E
$GPRMB,V,0.00,L,AIC,BA01,1248.413,S,03827.698,W,0. 275,345.4,10.0,V,S*2F
$GPGGA,002040,1248.678,S,03827.627,W,8,10,2.0,18.6 ,M,-10.5,M,,*5C
$GPGLL,1248.678,S,03827.627,W,002040,V,S*40
$GPBOD,345.6,T,8.6,M,BA01,AIC*04
$GPBWC,002040,1248.413,S,03827.698,W,345.4,T,8.4,M ,0.275,N,BA01,S*7D
$GPVTG,344.7,T,7.7,M,10.0,N,18.5,K*47
$GPXTE,V,V,0.00,L,N,S*11

GpsMap 276C NMEA output:

$GPRMC,124752,A,1248.704,S,03827.581,W,0.0,103.9,2 40205,23.0,W,A*34
$GPRMB,A,0.22,L,,G1,1247.472,S,03827.840,W,1.260,3 48.4,,V,A*01
$GPGGA,124752,1248.704,S,03827.581,W,1,11,0.9,2.7, M,-10.5,M,,*61
$GPGLL,1248.704,S,03827.581,W,124752,A,A*41
$GPBOD,358.3,T,21.4,M,G1,*0B

Notes:

1. As I have only the PC/USB cable that not allows the use of NMEA, I
had made an adaptation in the original power/data cable witch came
with the unit to connect it to a serial DB9 port in my laptop in order
to obtain via HyperTerminal, the NMEA data sent by the gps.
2. The manual of the GpsMap 76 do not says that it sends the BWC
message, but my tests showed that it sends it, at last in the 3.8
version, witch works OK with the autopilot.
3. I became very surprised when I have taken a close look at the
GpsMap 276C manual, as it says, in page 90 that the unit ALWAYS
sends the following: RMC, GGA, GLL, BWC, VTG, XTE, and RMB, but In
page 107, the manual says it sends RMC, GGA, GSA, GSV, GLL, GPBOD,
RTE, and WPL. (nothing about RMB and XTE and BWC)


Considerations


1) Looking at the tests results, we can see that the GpsMap 276C (as
the GpsMap 60C, GpsV, Gps III Plus, and the GpsMap 130 NMEA 2.0) does
not send the GPBWC, witch I think, is the message that works well with
the old autopilots.
2) The manual of the AH/ST 4000 Plus states that the autopilot gets
the Waypoint name and the Bearing to Waypoint from messages APA,
APB, BWR, BWC and RMB, in this order, and I think that it uses the
RMB only if the BWC is not received.
3) We had made some tests with the AH/ST2000+ with at last two
Magellan gps receivers, witch sends APA, APB etc, not only RMB, and
the interface worked OK.
4) I believe that there is an incompatibility or software error in
these old AH autopilots software, which interprets the wrong way the
RMB message, mistaking the True Bearing as if it was Magnetic Bearing,
and so, applying the magnetic deviation correction (W23 = - 23) witch
is causing the problem.
5) It is impossible to correct the software of these old units,
because their design does not allow this and because all these units
are out of production.
6) The only unit in the test that does not showed the 23 degrees
problem was the AH/ST 5000 Plus which is in reality, a brand new unit,
with last year updated software to version 14, made by RayMarine , not
an old Autohelm unit, so probably it has the software corrected for
the problem with RMB message. I had tested it with GpsMap 276C and
with a Gps III Plus.

Final Conclusions

1) As Garmin GpsMap 130 NMEA V 1.5, GpsMap 76 V3.8, and GpsMap 176C
4.0 all works OK with the old AH autopilots listed, and these units
also sends the BWC message, I think that the problem with the RMB
message could be easily bypassed if the gps sends the BWC.

2) So, I am asking that Garmin analyses this and make ASAP, in a next
software update, a change in the GpsMap 276C software so that the BWC
NMEA message is included in the batch of messages always sent by
GpsMap 276C, as stated well clear in page 90 of the manual.

3) As a better idea, why the "Advanced NMEA Setup" of the GpsMap 276C
could not be similar of the GpsMap 3006/3010, witch allows the
selection of each sentence that the gps unit will send? This is the
obvious solution for the problem of compatibility in the communication
with other NMEA devices, as stated in this Garmin Manual. See pages
73-74 of the 3010 manual where it says: "check any other equipment's
owner manual to see what sentences are required to be enabled for
functionality".


============ The END =============