GPS baud rate setting

Hi all,

I have a GPS puck that acquires/computes a GPS fix every second (1Hz). This GPS then transmits this data to a Vario (V7) and then PDA.

The default baud rate of the GPS is 4800 baud but I have the ability to increase it to 9600 or 19200 bauds. Is there a benefit of a higher transmission baud rate for the GPS? Does a 4800 baud rate allow the full transmission of a GPS NMEA 0183 sentence every second? I am not sure how much data is contained in a NMEA 0183 sentence, and if it can "all go through" every seconds at 4800 bauds. (Note: The vario and PNA can receive a 19200 bauds). What are the drawbacks of a baud rate set too high? Transmission errors?

Given the GPS limitation (1Hz fix) I understand that, best case, my vario (V7) and PDA cannot get a fix more frequent than one per second. I want to set the baud rate high enough where the GPS 1Hz fix is the limiting factor, not the speed of communication between the GPS and the Vario/PDA.
Thanks all for your inputs.
B.

On Tuesday, May 6, 2014 9:16:17 AM UTC+12, Bastoune wrote:
Hi all,



I have a GPS puck that acquires/computes a GPS fix every second (1Hz). This GPS then transmits this data to a Vario (V7) and then PDA.



The default baud rate of the GPS is 4800 baud but I have the ability to increase it to 9600 or 19200 bauds. Is there a benefit of a higher transmission baud rate for the GPS? Does a 4800 baud rate allow the full transmission of a GPS NMEA 0183 sentence every second? I am not sure how much data is contained in a NMEA 0183 sentence, and if it can "all go through" every seconds at 4800 bauds. (Note: The vario and PNA can receive a 19200 bauds). What are the drawbacks of a baud rate set too high? Transmission errors?



Given the GPS limitation (1Hz fix) I understand that, best case, my vario (V7) and PDA cannot get a fix more frequent than one per second. I want to set the baud rate high enough where the GPS 1Hz fix is the limiting factor, not the speed of communication between the GPS and the Vario/PDA.

Thanks all for your inputs.

B.

You've got it right.

GPSs always started with 4800 BPS because there's too much data for 2400. If you run it at 19200 then you'll be sure to get the up to date data with less than 1/4 of a second delay.

The chances of error at 19200 are higher, but still minuscule with a short, well shielded, cable in an electrically quiet environment. You could probably run at 1 Mbps like ancient 3rd party AppleTalk stuff did on RS422 (if your gear supported it) and seldom see errors. And anyway, the NMEA sentences have a checksum which will pick up most errors.

Just run it at the higher speed :-)

On Mon, 05 May 2014 15:51:08 -0700, Bruce Hoult wrote:

On Tuesday, May 6, 2014 9:16:17 AM UTC+12, Bastoune wrote:
Hi all,



I have a GPS puck that acquires/computes a GPS fix every second (1Hz).
This GPS then transmits this data to a Vario (V7) and then PDA.



The default baud rate of the GPS is 4800 baud but I have the ability to
increase it to 9600 or 19200 bauds. Is there a benefit of a higher
transmission baud rate for the GPS? Does a 4800 baud rate allow the
full transmission of a GPS NMEA 0183 sentence every second? I am not
sure how much data is contained in a NMEA 0183 sentence, and if it can
"all go through" every seconds at 4800 bauds. (Note: The vario and PNA
can receive a 19200 bauds). What are the drawbacks of a baud rate set
too high? Transmission errors?



Given the GPS limitation (1Hz fix) I understand that, best case, my
vario (V7) and PDA cannot get a fix more frequent than one per second.
I want to set the baud rate high enough where the GPS 1Hz fix is the
limiting factor, not the speed of communication between the GPS and the
Vario/PDA.

Thanks all for your inputs.

B.

You've got it right.

GPSs always started with 4800 BPS because there's too much data for
2400. If you run it at 19200 then you'll be sure to get the up to date
data with less than 1/4 of a second delay.

The chances of error at 19200 are higher, but still minuscule with a
short, well shielded, cable in an electrically quiet environment. You
could probably run at 1 Mbps like ancient 3rd party AppleTalk stuff did
on RS422 (if your gear supported it) and seldom see errors. And anyway,
the NMEA sentences have a checksum which will pick up most errors.

I think a basic GPS which doesn't have a route set in it just sends a
GPRMC message every second. This says where the receiver is and contains
65 characters. So, a GPS running at the default 4800 baud with 8 bits + 1
stop bit per character, which can send just over 533 chars/sec can easily
send one every second. This is exactly what a simple, 'blind' GPS such as
the Garmin GPS35, sends and all it can send.

A GPS that will allow you to set a route into it, such as most of the
hiking GPS units, will send a few more message types so the instrument
receiving the messages can know where its going as well as where it is.
The route is one or more waypoints, so in our terms that can be anything
from the next turnpoint to a whole predeclared task. This adds three more
message types:

GPRMB - 66 characters, says where you're coming from and going to,
distance, bearing and speed toward
destination, cross-track error
GPROO - 24 characters + 9 per turnpoint in the route, lists the
turnpoints that define the route
GPWPL - 33 characters, waypoint details

Each second the GPS sends the GPRMC and GPRMB messages together with
*one* of the other messages, which cycle so that a complete route
containing 'n' waypoints repeats every n+1 seconds.

In other words, the most characters that would be sent in a second would
be 171 for a route with four waypoints and this would take about 320 mS
to transmit, so even this still uses only 1/3 of the available time at
4800 baud.

Just run it at the higher speed :-)

Indeed, though in practise 4800 baud will still handle almost any task
we'd want to fly.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

On Monday, May 5, 2014 2:16:17 PM UTC-7, Bastoune wrote:
Hi all,



I have a GPS puck that acquires/computes a GPS fix every second (1Hz). This GPS then transmits this data to a Vario (V7) and then PDA.



The default baud rate of the GPS is 4800 baud but I have the ability to increase it to 9600 or 19200 bauds. Is there a benefit of a higher transmission baud rate for the GPS? Does a 4800 baud rate allow the full transmission of a GPS NMEA 0183 sentence every second? I am not sure how much data is contained in a NMEA 0183 sentence, and if it can "all go through" every seconds at 4800 bauds. (Note: The vario and PNA can receive a 19200 bauds). What are the drawbacks of a baud rate set too high? Transmission errors?



Given the GPS limitation (1Hz fix) I understand that, best case, my vario (V7) and PDA cannot get a fix more frequent than one per second. I want to set the baud rate high enough where the GPS 1Hz fix is the limiting factor, not the speed of communication between the GPS and the Vario/PDA.

Thanks all for your inputs.

B.

Martin,

What is your take on a 4800 baud rate system that also transmits PowerFlarm and Vario sentences two ways.

Richard
www.craggyaero.com

On Thu, 08 May 2014 08:23:01 -0700, Richard wrote:

What is your take on a 4800 baud rate system that also transmits
PowerFlarm and Vario sentences two ways.

I don't have enough information to make any estimate. To do that you need
to know:

- baud rate and character format

- format of the various messages exchanged with the FLARM and vario

- the frequency and order in which the messages are exchanged.

and I don't have any of this stuff.

I am able estimate the spare capacity in a GPS receiver's output stream
because I found the NMEA sentence formats and character format (8N1) on
the 'net [1] and spent some time with a Garmin GPS2+ and a GPS35 hooked
up to the serial port on a PC, watching the data stream and working out
how it changes as the GPS settings are changed, i.e. no active route,
active routes with different numbers of waypoints.

[1] http://aprs.gids.nl/nmea/ gives some NMEA sentence formats: enough
for our purposes and the NMEA 0183 specification gives baud rate and
character encoding. However, this stuff is quite hard to find because it
seems that the standards that define NMEA sentence formats and ordering
are not exactly an open source document. The full spec will set you back
around $2800.

BTW, it seems that 38400 baud is also a standard NMEA transmission rate.
It is used for AIS (Automatic Identification System), an automatic
tracking system used on ships and by vessel traffic services (VTS) for
identifying and locating vessels by electronically exchanging data with
other nearby ships, AIS base stations, and satellites.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

On Thursday, May 8, 2014 2:42:52 PM UTC-4, Martin Gregorie wrote:
On Thu, 08 May 2014 08:23:01 -0700, Richard wrote:



What is your take on a 4800 baud rate system that also transmits

PowerFlarm and Vario sentences two ways.



I don't have enough information to make any estimate. To do that you need

to know:



- baud rate and character format



- format of the various messages exchanged with the FLARM and vario



- the frequency and order in which the messages are exchanged.



and I don't have any of this stuff.



I am able estimate the spare capacity in a GPS receiver's output stream

because I found the NMEA sentence formats and character format (8N1) on

the 'net [1] and spent some time with a Garmin GPS2+ and a GPS35 hooked

up to the serial port on a PC, watching the data stream and working out

how it changes as the GPS settings are changed, i.e. no active route,

active routes with different numbers of waypoints.



[1] http://aprs.gids.nl/nmea/ gives some NMEA sentence formats: enough

for our purposes and the NMEA 0183 specification gives baud rate and

character encoding. However, this stuff is quite hard to find because it

seems that the standards that define NMEA sentence formats and ordering

are not exactly an open source document. The full spec will set you back

around $2800.



BTW, it seems that 38400 baud is also a standard NMEA transmission rate.

It is used for AIS (Automatic Identification System), an automatic

tracking system used on ships and by vessel traffic services (VTS) for

identifying and locating vessels by electronically exchanging data with

other nearby ships, AIS base stations, and satellites.





--

martin@ | Martin Gregorie

gregorie. | Essex, UK

org |

Typically varios send one sentence a second containing the lift data and some
other stuff. Sentences indicating things that don't change that often (MC,
ballast, bugs, whatever) are sent less frequently, or only when those values
change. FLARM data volume depends on the connection speed. At 19.2K it
sends a pretty large amount of data describing all the targets in view. At
slower speeds you just get the warnings.

Matt

On Thu, 08 May 2014 13:04:39 -0700, mattmckrell wrote:

Typically varios send one sentence a second containing the lift data and
some other stuff.

Some (mainly older ones?) like the SDI C4 don't send anything - they just
receive GPRMB,GPRMC,GPROO,GPWPL - the set needed to calculate wind and
final glide. Presumably 'other stuff' includes IAS.

Sentences indicating things that don't change that
often (MC, ballast, bugs, whatever) are sent less frequently, or only
when those values change.

Makes sense. How big are the sentences? Are their formats generally
available?

FLARM data volume depends on the connection
speed. At 19.2K it sends a pretty large amount of data describing all
the targets in view. At slower speeds you just get the warnings.

I'd heard that, but have seen any documentation of either those messages
or the FLARM additions to their IGC log files. IIRC both message sets are
regarded as proprietary/undocumented.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

On Thursday, May 8, 2014 6:31:58 PM UTC-4, Martin Gregorie wrote:
On Thu, 08 May 2014 13:04:39 -0700, mattmckrell wrote:
FLARM data volume depends on the connection
speed. At 19.2K it sends a pretty large amount of data describing all
the targets in view. At slower speeds you just get the warnings.

I'd heard that, but have seen any documentation of either those messages
or the FLARM additions to their IGC log files. IIRC both message sets are
regarded as proprietary/undocumented.

The FLARM message sets for traffic and collision warnings are very
well documented and used by at least a dozen other manufacturers
and software packages to display FLARM info. See "Data Port Spec":
http://www.flarm.com/support/manual/index.html

See the "Sentence PFLAA" section for explanation that 19200 is
required to receive traffic information.

Hearsay is dangerous (and rampant on RAS)...
Best Regards, Dave "YO electric"

NMEA 0183 is 4800 by definition, except for binary (AIS is 38400 for example).

Of course, there may not be enough band-width for large amounts or frequent data, so people use faster speeds in various applications.

Your GPS puck is just fine at 4800. If you have a 10Hz GPS, flarm data etc, you will want something faster

On Thu, 08 May 2014 15:42:42 -0700, Dave Nadler wrote:

The FLARM message sets for traffic and collision warnings are very well
documented and used by at least a dozen other manufacturers and software
packages to display FLARM info. See "Data Port Spec":
http://www.flarm.com/support/manual/index.html

Thanks for the link, Dave. Bookmarked.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |