Hello,
I am running an LC29H(DA) (Waveshare LC29H(XX) GPS/RTK HAT) as an RTK rover and cannot obtain a FIXED solution under conditions where I would expect one. I would like to request the latest LC29H(DA) firmware, and to ask whether a fix-convergence improvement is expected from it.
Module
$PQTMVERNO,LC29HDANR11A03S_RSA,2024/03/19,13:55:31
$PQTMCFGRCVRMODE,OK,1 (rover)
$PQTMCFGSVIN,OK,0,0,0.0,0.0000,0.0000,0.0000 (survey-in off)
$PAIR081,0 (navigation mode: normal)
Fix rate 1 Hz. Corrections are written directly to the module UART at 115200.
Correction stream (commercial VRS network, Trimble caster, RTCM 3.4 MSM4)
Decoded from the actual byte stream reaching the module:
1074 GPS sats 9 signal IDs 2, 10, 16, 24, 32
1084 GLONASS sats 3 signal IDs 2, 3, 8, 9
1094 Galileo sats 8 signal IDs 5, 10, 16, 20, 24
1124 BeiDou sats 6 signal IDs 2, 8, 24, 32
plus 1005, 1230, 1032, 1033, 1007, 1021/1023/1025
So GPS L5 (24), Galileo E5a (24) and BeiDou B2a (24) are all present.
Conditions
- Age of differential corrections (GGA field 13): steady at 1.0 s over many minutes.
- Baseline to the reference station decoded from 1005: 7–9 m (virtual reference station at the rover position).
- RTCM3 frames: several thousand received, 0 CRC failures (validated before the UART).
- 34–37 satellites, HDOP 0.52–0.60, multi-band helix antenna on a ground plane, stationary.
- C/N0: Galileo E5a up to 48 dB-Hz, GPS L5-Q up to 48 dB-Hz.
Behaviour
GGA quality is 5 (RTK float) continuously and never reaches 4 (RTK fixed), over sessions of many minutes. The float solution does converge — measured scatter while stationary is about 0.10 m north / 0.07 m east (1σ) — but it never resolves.
Questions
- Is there firmware newer than LC29HDANR11A03S_RSA for the (DA), and does it change RTK fix convergence? Could you send the package and any release notes?
- The network encodes GPS and Galileo L5/E5a as the combined 5X (I+Q) MSM signal, while the module reports tracking L5-Q. Does the RTK engine use 5X corrections when tracking the Q component, or are 5I/5Q required for the second frequency to be usable?
- $PQTMEPE reports values far larger than the observed scatter — for example 7.736 / 7.736 / 14.738 while measured 1σ scatter was under 0.10 m. Is the EPE expected to be this conservative in RTK float?
- Is there any non-Windows flashing path (Linux tool, documented protocol, or source for QGPSFlashTool)? All my systems are Linux.
Thank you.