Ryan Malloy
c93bbef26d
Extract Gabe Emerson's Trav'ler rotor scripts into a proper Python library with firmware protocol abstraction (HAL 2.05 + HAL 0.0.00), Hamlib rotctld TCP server, Click CLI, and isolated leap-frog algorithm with the elevation copy-paste bug fixed.
221 lines
12 KiB
Markdown
221 lines
12 KiB
Markdown
# Travler_Rotor
|
|
Multi-axis antenna rotor using Winegard "Trav'ler" satellite dish
|
|
|
|
Make sure to check your Winegard ODU firmware / HAL Version!
|
|
This variant of the software (v2.0B) works with the newer Winegard Trav'ler firmware; HAL 2.05.003.
|
|
It may work with other versions as well. Key differences between my other Trav'ler Tracker branch (2.0)
|
|
for HAL 0.0.00: The newer HAL uses a different initialization sequence and sends different status messages,
|
|
so the init script looks for those before cancelling the search routine. The newer HAL also uses different
|
|
submenu commands, namely "motor" (vs "mot" in HAL 0.00.00).
|
|
|
|
Gabe Emerson / Saveitforparts 2025. Email: gabe@saveitforparts.com
|
|
|
|
Video demo of version 1: https://youtu.be/X1hnReHepFI
|
|
|
|
Video demo of version 2: (in progress)
|
|
|
|
**Introduction:**
|
|
|
|
This code acts as an interface between Gpredict / Hamlib Rotctld and a Winegard brand "Trav'ler" satellite dish.
|
|
NOTE: Specifically one using HAL firmware version 2.05.003
|
|
These dishes are designed for vehicle use, but can be mounted to any stable surface. They can sometimes be found
|
|
2nd-hand from used RV dealerships, Craigslist / Facebook marketplace / etc.
|
|
|
|
The Winegard Trav'ler consists of an Outdoor Unit (ODU) or motorized "turret" with a standard Ku-band TV dish attached,
|
|
and an Indoor Unit (IDU), or small black box that has the power input and display screen. You will also need a DC power
|
|
supply, if the original is missing then something around 48-52VDC should work.
|
|
|
|
Note there are several versions of the Trav'ler dish and this code has only been tested with the LG-2112. Versions with
|
|
a GPS search may not respond to the init code that looks for a "NoGPS" flag from the firmware.
|
|
|
|
This code is experimental and will probably void the warranty on any antenna you try it with. Use at your own risk!
|
|
|
|
|
|
|
|
|
|
**Applications:**
|
|
|
|
This antenna works as a general-purpose az/el rotor, and could have skew added to the code if desired. I have been using
|
|
my Trav'ler dish as an S-band LEO weather satellite tracker. It can also be used to point at GEO satellites, which is closer
|
|
to the original intended use.
|
|
|
|
The code could be adapted for other purposes like sky surveys, Wifi or other RF surveys, etc. See my other Github projects
|
|
for some example sky surveys / radio telescope implementations using similar antennas.
|
|
|
|
This could possibly also be used for DIY radar applications, drone tracking, point-to-point Wifi, etc. Note that the dish
|
|
has some speed and motion limits which might not make it suitable for every application.
|
|
|
|
**Hardware Requirements:**
|
|
|
|
The stock LNB should be replaced with a feed appropriate to whatever frequency you intend to use. For example, I use 3-D
|
|
printed helical feeds designed by DerekSGC for L and S-band. These can be found at https://www.thingiverse.com/thing:4980180
|
|
|
|
|
|
|
|
|
|
The internal coax cable should not be used unless you disable / bypass the onboard power injector. Otherwise 14-18VDC will
|
|
be supplied to the feed/LNA, which will kill 5VDC equipment. Technically the internal wiring is the wrong impedance for SDR use,
|
|
but I have not noticed any issues mixing 50 and 75-ohm wiring at S-band frequencies.
|
|
|
|
I use a Microcircuits ZX60-242LN-S+ Low Noise Amplifier connected to my helical feed, powered from a 5v USB source. Below that
|
|
is an RTL-SDR Wideband LNA, powered by my SDR's bias-tee. I use a HackRF One for S-band. I keep the HackRF's onboard amp turned
|
|
off to minimize noise. This seems to work well for strong satellites like NOAA and DMSP, but not for weaker ones like HINODE. The stock
|
|
33"x23" dish is probably too small for weaker signals. It might be possible to add a slightly larger reflector, but YMMV.
|
|
|
|
Interfacing with the Trav'ler serial port requires an RS-485 cable and 6-pin phone connector (RJ-25). I have included a
|
|
photo of the USB adapter chain that I use. It includes a USB-to-Serial cable, a DTECH RS232-to-RS485 converter, and an RJ-25 jack
|
|
wired as follows (looking at the bottom of the phone connector with the tip up):
|
|
- Pin 1: GND
|
|
- Pin 2: T/R-
|
|
- Pin 3: T/R+
|
|
- Pin 4: RXD-
|
|
- Pin 5: RXD+
|
|
- Pin 6: Not used
|
|
|
|
|
|
|
|
|
|
|
|
|
|
**Software Requirements:**
|
|
|
|
This code was developed for Python 3. The code uses serial, socket, and regex. If not already installed, use "pip install -r
|
|
requirements.txt"
|
|
|
|
If your computer uses a different serial port for the RS-485 adapter (such as COM3 for Windows or /dev/ttyACM0) you will need
|
|
to edit line 17 of travler_rotor.py and line 10 of travler_init.py
|
|
|
|
**Trav'ler Dish Basic Firmware Info**
|
|
|
|
You can interact with the Trav'ler firmware directly by connecting the RS-485 cable chain to the "Factory Only" port on the
|
|
IDU. Then run "screen /dev/ttyUSB0 57600" (or appropriate port, you may need to check lsusb or equivalent command on your
|
|
local computer).
|
|
|
|
Once connected to the firmware, some basic commands are:
|
|
- "?": List available commands
|
|
- "motor": enter the motor submenu
|
|
- "a" (from within the motor submenu): Show current dish position, or set desired position by specifying motor # and degrees.
|
|
- "g" (from within motor submenu): go to a specified azimuth/elevation/skew (Some firmware has a typo listing az/sk/el).
|
|
- "q": exit the current submenu
|
|
- "nvs" - enter the non volatile storage (for changing long-term settings).
|
|
- "os": Enter the OS submenu
|
|
- "tasks": list running tasks (from within os submenu)
|
|
- "kill [name of task]": Kill a task (such as "kill Search" to disable the TV satellite search movement routine).
|
|
|
|
The firmware has a lot of options, not all of which I understand. There are several ways to do various things, including
|
|
multiple motor-related submenus that all behave slightly differently.
|
|
|
|
|
|
**Notes on Trav'ler limitations and quirks**
|
|
|
|
Physical and Firmware limitations:
|
|
|
|
The Tra'ler can physically move past 0 degrees and 90 degrees in elevation, but the firmware doesn't like to go below 15
|
|
degrees. This is likely a built-in safety feature to keep it from impacting other objects on an RV roof. There are two
|
|
options to address this:
|
|
|
|
1: I have put in a soft-coded limit of 15 degrees elevation in my code, which you can enable by un-commenting lines 116,
|
|
117, 124, and 125 of travler_rotor.py. I recommend telling Gpredict that the rotor's minimum elevation is 15 degrees.
|
|
|
|
2: You can edit the nvs settings in the firmware. Make sure your dish won't physically collide with anything at the new
|
|
minimum elevation! You will need to enter the firmware as described above, then the "nvs" submenu. Type the following:
|
|
"e 125 0"
|
|
"e 127 0"
|
|
This sets both the search and safe minimums to 0 elevation. Type "s" to save the new setting, "q" to return to the main
|
|
menu, and then "reboot" to use the new minimum elevation.
|
|
|
|
When using the "g" method to operate the motors, the dish will halt/abort movement if a new command (or any keystroke /
|
|
character) comes in.
|
|
|
|
When using the "a" method to operate motors, the dish will wait until the current motor stops moving before accepting a new
|
|
motor command (in the case of the AZ and EL motors. I believe the Skew / SK motor can run simultaneously with one
|
|
of the others).
|
|
|
|
Initial Calibration:
|
|
|
|
When first powered on, the Trav'ler goes through a series of calibration movements to establish position and wrap limits.
|
|
Afterwards, the default behavior is to search for a TV satellite, which for our purposes is a waste of time. The travler_init.py
|
|
script connects to the dish and waits for the calibration to complete, then kills the search in the firmware's task manager.
|
|
|
|
Stowing:
|
|
|
|
The Trav'ler dish has a built-in "stow" command which is intended to fold it flat against the roof of an RV or trailer.
|
|
I have not used this command in my code and I tend to ignore it. The modified L-band feed that replaced my LNB would
|
|
likely not survive a stow procedure.
|
|
|
|
Cable Wrap:
|
|
|
|
The dish also has a cable wrap system that prevents it from tangling its own internal wiring. This seems to be a somewhat
|
|
variable position, but is usually either at 360 or 455 degrees. Thus the dish can spin from 0 degrees, past 360, but will halt
|
|
and reverse to the other side of the wrap position upon hitting its limit. The wrap position can be found with the "a" command
|
|
in the firmware's "mot" submenu. I usually address this by using Gpredict to manually run the dish through the range of motion
|
|
required for the upcoming pass. This helps ensure it is on the correct side of the wrap position. If the wrap position is near
|
|
the start or end of a track, I will simply disengage Gpredict during that part.
|
|
|
|
Meridian Crossing:
|
|
|
|
The dish will cross the 0/360 position during tracking *most of the time. I have sometimes encountered issues with this where
|
|
the dynamic wrap position gets set to 0. In these cases, the dish will approach 0, then stop, reverse direction and come at
|
|
the new position from the other side. This can spoil a live track of a satellite. It's possible there's a way to avoid or
|
|
address this. I just never got around to implementing it and don't encounter the issue often enough.
|
|
|
|
Possible GPS issue:
|
|
|
|
Some versions of the Trav'ler have a GPS subsystem that assists with acquiring TV satellites. Other users have reported that
|
|
attempting to disable the GPS puts the IDU and ODU in a state where they no longer communicate, effectively bricking the dish.
|
|
I have not encountered this myself, as my dish does not have GPS, but it's something to watch out for.
|
|
|
|
IDU / ODU cable
|
|
|
|
If the cable between the IDU and ODU is cut (common for used units pulled from RVs), the wires to the IDU should be:
|
|
|
|
Top row: Green, Yellow, Orange.
|
|
|
|
Bottom row: Red, Brown, Black.
|
|
|
|
|
|
|
|
|
|
**Positioning the antenna**
|
|
|
|
The baseplate of the Winegard Trav'ler is marked with arrows and the word "BACK" at the 0/360-degree position (North).
|
|
Typically I place my dish so that these arrows are aligned with true North. If using the dish as a portable unit, not bolted
|
|
to a roof or vehicle, make sure to secure the base so that the dish cannot wiggle or fall over.
|
|
If the stow command / feature is used, North / "Back" is the position at which the dish will "faceplant" onto the ground/roof.
|
|
|
|
**Setting up rotor in Gpredict**
|
|
|
|
In Gpredict's rotor settings, you will want to create a new rotor at 127.0.0.1:4533, with 0->180->360 mode, minimum elevation
|
|
of 15 and maximum elevation of 90.
|
|
|
|
**Example setup and use procedure**
|
|
|
|
Your procedure may differ depending on the software and setup you use. My basic procedure for tracking a LEO satellite with
|
|
the Trav'ler is as follows:
|
|
|
|
- Connect serial cable to Trav'ler's IDU, run "./init.sh" on computer
|
|
- Power on IDU. Trav'ler will initialize and home, then init script will jump to rotor script.
|
|
- Power on the first LNA with external bias-tee
|
|
- Open SDR++, turn on SDR and activate second LNA with SDR's bias-tee function
|
|
- Set gains as desired (may need to max these out for faint signals)
|
|
- Keep onboard amp (for HackRF One) turned off.
|
|
- Set Gpredict Antenna Control for desired satellite, select the correct rotor.
|
|
- If rotor script is not already running and waiting for Gpredict, run "./rotor.sh"
|
|
- Prep SDR to record with desired bandwidth, baseband, int16
|
|
- Activate tracking and engage rotor in Gpredict.
|
|
- Record the pass (I like to keep an eye on the dish in case it does anything weird).
|
|
- Stop recording and disengage the rotor when the signal gets too low to be usable.
|
|
- Run the baseband recording through Satdump.
|
|
|
|
Some of these steps could be combined with Satdump, which can also handle the rotor control, recording, and live decoding.
|
|
|
|
Personally I like to track the current satellite in N2YO.com alongside the other windows, just to see where it is. I
|
|
also have a security camera aimed at my dish so I can watch it moving from my computer.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|