Garmin GPS Moving Positional Accuracy

For five + years I have used two Garmin GPS-19X HVS receivers for precision navigation in high-speed car racing using custom software with great success. However, there is always an ambiguity of my exact location that translates into milliseconds of time error at the finish line. This small error is very significant in these events.

NMEA PGRME and GPGGA report the horizontal accuracy and HDOP with the GPS-19X at about 10-13 feet under clear sky conditions. This causes a +/-075 mS error which can easily lose a race.

I have evaluated the WAAS and differential options and none of them seem appropriate for the racing environment. So, I now have two Garmin GPS-24xD HVS receivers. These receivers are L1 and L5 capable. Unfortunately, Garmin never published a significant manual on the technical aspects of these receivers. I am extrapolating from the GPS-19 manual. I record GPRMC and PGRME at a 10 Hz rate for data logging.

The GPS-24xD units consistently report a horizontal error estimation of 3.3 -3.9 feet under clear sky. This is a significant improvement from the GPS-19. A one-hour parked data logging on Google Earth shows that GPS-24 static position stability is much better than the GPS-19. The GPS-24 position shows as a dot while the GPS-19 position wanders around the parking area.

However, I need an improvement at 150+ mph. So static improvement doesn’t help. The GPS units are mounted on the outside rear-view mirrors of the car, 62 inches apart. I have created a driving database over the last few weeks with 200K + points using two GPS-19’s and two GPS-24’s.

These driving tests were all at highway speeds, 70-80 mph. It is difficult to find a 150-mph venue.

With the two GPS-19’s, I expected to see the distance between points from the two receivers with the same time stamp from GPRMC to vary about +/- 10 feet. The second GPS unit will show a driveline position ahead or behind the first GPS unit as you drive along. The measured results matched this scenario with a standard deviation of about 3 feet.

Next I repeated the driving tests with the two GPS-24xD receivers. I expected to see a substantial improvement in the distance variation due to the L1/L5 capability and the reported HDOP and horizontal error estimation improvement. There is no improvement. The raw data and the statistical analysis of the GPS-24xD data for moving positional accuracy is almost identical to the L1 only GPS-19X receiver.

So, I have two GPS-24xD receivers that seem to offer no moving accuracy improvement. I have searched for any other methods or studies that would demonstrate the moving positional accuracy of a GPS receiver with no success. I realize this is a narrow niche use for these units, but if you have experience in this area, I would like to hear from you.

 

**************************************************************************************
The 10Hz updates are essential for controlling my speed on approach to the finish line. I watch an analog ribbon display similar to an ILS airport approach for the last few seconds. But changing to 1 Hz may be worth trying as an experiment.

Switching the dynamics mode is a good idea. I will reprogram the units and make some highway speed drives.

I am really surprised that Garmin never published a serious technical document on this high-end device.

Thank you for your reply. All experiences and suggestions are appreciated.

 

**************************Update*****************************************************

Most of the setup options available for the GPS-19 are grayed out and not available for the GPS-24xD.

*******************************************************************************

 

I connected with the GPS-24 using the GPS-19 option on the Garmin configurator. I tried a lot of combinations that are allowed for a GPS-19 but grayed out for the GPS-24. I uploaded the GPS-19 configurations, then removed the power for 20+ seconds. The GPS-24 ignored all the GPS-19 allowed changes. I am able to change from NMEA 4.11 to NMEA 2.3 and change the Talker ID. That's not much. I will repeat my road tests with NMEA 2.3 enabled, I don't expect that to improve matters.