I am programming a python application for car rallys. You drive a 60 mile course in speed classes on closed roads and whoever is closest to the perfect time for that class wins. Example - in the 120mph class the winning time would be 30 minutes 0.0000 seconds. It is highly competitive, the winning time is often less than 0.10 seconds from perfect.
I think the SparkPNT GNSS Flex Module - ZED-X20P meets my requirements
PPS for accurate lag measurement to feed real time correction calculations
L band for positional accuracy with reasonable latency using Ublox L band correction service
Fallback to SBAS when L band is not available
RTK available as a backup method (cell service unavailable and don’t want to set up a base station unless absolutely necessary.
USB connection for gnss data connection to Windows 11 laptop
HERE ARE MY QUESTIONS
can I turn off nmea sentences I don’t need using the u-center 2 software? I don’t see how to do that in the software documentation
do you sell cases for boards? The board will be going in a race car, it needs protection and needs to be mounted securely.
What is the L Band update rate in Hz? Please confrim the ZED-X20P will work with Ublox L band service today
I am not an expert in gps technology, let me know if there is anything else I should be considering besides PPS and L band technology for race car position and time tracking.
The sales team directed me to post this on the forum, sorry if this pre-purchase question does not belong here. I can communicate by email if that is more appropriate.
Thank you for your assistance, I appreciate your help.
Yes, you can selectively enable messages and / or protocols on individual ports. You can disable all NMEA with a single setting if you need to.
Sadly, no. We have some cased RTK products, but we don’t sell cases for individual boards. There are some excellent cases / enclosures out there that are easy to modify. E.g. Hammond.
The X20P L-Band service is not yet available. Details are sparse, and it will almost certainly require a firmware upgrade and subscription to activate.
I’m not sure what you mean. Can you clarify what you mean by lag in L-Band?
Sorry, I have no idea. I’ve been burned several times when buying dual and triple band antennas which promised the Earth and then failed to deliver - spectacularly. You can’t go far wrong with the SparkFun L1/L2/L5 Surveying Antenna. I know it’s not a great match for vehicle applications, but it works really well.
I’m intrigued. How do you time each run? Are you logging data in the car, and post-process it to find when the car crossed the start and finish lines? Or are you using radio / telemetry to transmit the car’s position in real time? Either way:
GNSS gives you superb timing accuracy. Each message contains a very tight position and time solution. If you have the messages (solutions) from before and after crossing the line, you can interpolate the crossing time with great accuracy. Do you need RTK position accuracy? Maybe you do if you’re chasing (say) sub-1/100th of a second absolute accuracy. But don’t forget that a basic, single band GNSS receiver will get you time accuracies below 100ns.
Thank you for your prompt response Paul, it is very helpful.
I will do more research on RTK positioning and reply to this thread with questions. I’m not a programmer or gnss expert so it might be a while and I might wait until L band becomes available.
This might be a good application for the free Galileo HAS E6, instead of SBAS. After initial convergence, HAS position and timing accuracy might be the ticket? HAS improves both the Galileo and GPS(USA) systems.
PointPerfect Global is “expected” in the 2’nd Quarter of 2026. The Existing L-Band service will end at the end of 2025.
My top choice would be a Network RTK (RTN), but you mentioned no cellular service for NTRIP corrections. Is a Starlink Mini an option for the vehicle? If so, you’re off to the races.
RTK would be the next choice in accuracy, but a 60 mile course could be problematic for a single Base.
I’m not sure that’s how it works. The NEO-D9S was a 2400 baud modem for the Inmarsat signals, and I’d suppose the ZED-X20P will just commit a channel in the receiver. It just comes as a continuous stream of data, at relatively low bandwidth.
But, it’s a MODEL of how the signals are behaving rather than real-time observations, so it’s life span runs many minutes.
Anyway promises are for 2026Q1, and frankly the drop-dead for the old SPARTN L-Band is the end of the year, so probably strong motivation there.
A rally course could be a challenging environment for L-Band. The deprecated u-blox L-Band service uses geostationary satellites, which are low in the sky at the higher latitudes. I’m guessing the new X20P service will also be geostationary. It may not be a good match for a rally course, especially if it is in forest or a hilly region.
I wonder if standard dual-band or triple-band GNSS would be good enough?
Let’s say you survey your start and finish lines. You could get the ends of the lines to within a couple of centimetres.
Then, let’s say your car is doing 100mph when it crosses the finish line. Roughly 45 m/s.
If your GNSS is running at 10Hz, you’d be getting a fix every 4.5m.
Let’s say that - with some hand waving - you’d get better than +/-1m positioning accuracy from dual/triple-band GNSS.
For the two messages (solutions) logged either side of the finish line, you can interpolate the time the car crossed the line, based on position, distance and speed. (If the speed is changing, you can interpolate that too.)
With a position accuracy of 1m, your crossing time interpolation would be accurate to say 20-25ms. Is that good enough? Probably not if winning times are less than 0.1s from perfect…
Galileo HAS - as suggested by @rftop - would get you to about 20cm and towards [more hand waving] 5ms. Perhaps good enough?
Correct, we expect PointPerfect Global will be delivered over L-Band from an Inmarsat (or similar competitor) bird.
That’s a great point by @PaulZC regarding the delivery method (L-band verses E6 band) !
Zooming around at 100-120 mph, I’d lean towards using HAS on the E-6 band (directly from GNSS sats) verses a L-band delivery (Geostationary Inmarsat) that can be fairly low in the sky depending on your location.
PP Global will be a PPP-AR product with a marketing focus towards the X20P initially, but the free Galileo HAS service is also PPP-AR and is currently operating (at Service Level 1) for various compatible receivers (not brand/model specific).
[just a personal opinion]: I don’t think HAS is ready to compete in the RTK game, but this might be a great application.
With either product, once initial PPP convergence has occurred - your device could maintain an accurate variable of Distance/Time (average speed) that’s updated “continuously” to prevent the driver from busting checkpoints. This sounds like bracket racing on a long road course …very interesting .
.