weewx-nuvoler

A powerful WeeWX extension that automatically uploads comprehensive weather station data to Nuvoler.com.

License Python WeeWX

Overview

weewx-nuvoler is a WeeWX extension that automatically monitors archive records from your weather station and uploads comprehensive meteorological data to Nuvoler.com, a weather data collection service supporting 10+ weather parameters.

The extension features advanced wind minimum estimation using a 3-parameter Weibull distribution with intelligent fallback methods, ensuring accurate wind speed statistics regardless of your station’s data availability.

Features

Requirements

Installation

Step 1: Register Your Station

Visit Nuvoler Sign Up to register your weather station and obtain your station ID and password.

Step 2: Download the Extension

wget -O weewx-nuvoler.zip https://github.com/rc-chuah/weewx-nuvoler/releases/latest/download/weewx-nuvoler.zip

Step 3: Install

For WeeWX v4 and earlier:

wee_extension --install weewx-nuvoler.zip

For WeeWX v5:

weectl extension install weewx-nuvoler.zip

Step 4: Configure

Edit /etc/weewx/weewx.conf and add the required configuration:

[StdRESTful]
    [[Nuvoler]]
        station_id = YOUR_STATION_ID_HERE
        station_pass = YOUR_STATION_PASSWORD_HERE

Optional parameters:

[StdRESTful]
    [[Nuvoler]]
        # Station credentials (required)
        station_id = YOUR_STATION_ID_HERE
        station_pass = YOUR_STATION_PASSWORD_HERE
        
        # Enable or disable extension (default: true)
        enabled = true
        
        # Nuvoler server URL (default: https://www.nuvoler.com/data/recibir.php)
        server_url = https://www.nuvoler.com/data/recibir.php
        
        # Upload interval in seconds (default: 60)
        post_interval = 60
        
        # Maximum backlog of records to upload (default: unlimited)
        max_backlog = 10
        
        # Data upload timeout in seconds (default: 60)
        timeout = 60
        
        # Maximum upload retry attempts (default: 3)
        max_tries = 3

Step 5: Restart WeeWX

sudo systemctl restart weewx

Check that the extension loaded successfully:

tail -f /var/log/syslog | grep -i nuvoler

Expected log output:

nuvoler: version is 0.1
nuvoler: Data will be uploaded to https://www.nuvoler.com/data/recibir.php

How It Works

The extension operates through the following workflow:

  1. Monitor - Listens for new archive records from your WeeWX weather station
  2. Extract - Retrieves all weather parameters (temperature, humidity, pressure, wind, precipitation, UV, dewpoint)
  3. Convert - Transforms all values to metric SI units (Celsius, mbar/hPa, mm (L/m²), knots for wind)
  4. Estimate - Calculates wind minimum using 3-parameter Weibull or fallback methods
  5. Upload - Sends the data to Nuvoler.com via an HTTP GET request

Supported Weather Parameters

Parameter Unit WeeWX Field Description
temperature °C outTemp Outdoor temperature
rh % outHumidity Relative humidity
mslp mbar/hPa barometer Mean sea level pressure
wind_dir ° windDir Wind direction (0-360°)
wind_avg knots windSpeed Average wind speed
wind_min knots (estimated) Minimum wind speed (3-param Weibull)
wind_max knots windGust Maximum wind speed/gust
precip mm (L/m²) hourRain Hourly precipitation
uv Index UV UV index (unitless)
dewpoint °C dewpoint Dew point temperature

Wind Minimum Estimation

Method Hierarchy

The extension uses an intelligent 4-level hierarchy to estimate wind minimum, automatically selecting the best method based on available data:

1. 3-Parameter Weibull Distribution (PRIMARY - Best Accuracy)

Used when both average wind speed and wind gust (maximum) are available.

Mathematical Formula:

θ (location) ≈ wind_avg × 0.30    (minimum threshold)
k (shape)    = 2.0                (Rayleigh-like, typical for surface winds)
λ (scale)    = (wind_max - θ) / (-ln(0.01))^(1/k)  (spread parameter)

wind_min = θ + λ × (-ln(0.05))^(1/k)  (5th percentile)

Advantages:

Example:

Input: wind_avg = 4.86 m/s, wind_max = 7.78 m/s
Estimated parameters:
  θ = 1.458 m/s (minimum threshold)
  k = 2.0 (Rayleigh-like)
  λ = 2.946 m/s (spread)
Result: wind_min ≈ 2.125 m/s (4.13 knots)

2. 2-Parameter Weibull with Variance Modeling (FALLBACK 1)

Used if 3-parameter Weibull encounters numerical issues, or when explicitly selected.

Formula:

variance_scale = 1.253 × 1.65 = 2.067
estimated_std = (wind_max - wind_avg) / variance_scale
wind_min = wind_avg - (0.7 × estimated_std)

Advantages:

3. Climatological Mean-to-Min Ratio (FALLBACK 2)

Used when only average wind speed is available.

Formula:

wind_min = wind_avg / 1.65

Basis:

Advantages:

4. Extreme Value Theory Gust-to-Min Ratio (FALLBACK 3)

Used when only maximum wind speed is available.

Formula:

wind_min = wind_max / 2.5

Basis:

Advantages:

Auto-Selection Logic

If (wind_avg available) AND (wind_max available):
    Use 3-Parameter Weibull
Else if (wind_avg available):
    Use Climatological Ratio
Else if (wind_max available):
    Use Extreme Value Theory
Else:
    Raise error (at least one wind parameter required)

Unit Conversion

The extension automatically converts wind speeds from m/s to knots using WeeWX’s built-in conversion:

knots = m/s × 1.94384449

All wind values are rounded to 1 decimal place.

Upload Protocol

The extension uploads weather data to Nuvoler using HTTP GET with the following format:

GET /data/recibir.php?station_id=YOUR_STATION_ID_HERE&station_pass=YOUR_STATION_PASSWORD_HERE&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=9.4&wind_max=15.1&wind_min=4.1&precip=2.4&uv=5&dewpoint=14.2 HTTP/1.1
Host: www.nuvoler.com

Parameters:

Unit Conversion Details

The extension converts your station’s native units to the following for Nuvoler:

Category Field Conversion Result
Temperature outTemp Fahrenheit → Celsius or pass-through °C
Humidity outHumidity Pass-through %
Pressure barometer inHg/Mb → mbar/hPa mbar/hPa
Wind Speeds windSpeed, windGust mph/km/h/m/s → m/s → knots knots
Wind Minimum (estimated) 3-param Weibull → m/s → knots knots
Precipitation hourRain inches/cm/mm → mm (L/m²) mm (L/m²)
UV Index UV Pass-through Index
Dew Point dewpoint Fahrenheit → Celsius or pass-through °C

All conversions are automatic and transparent to the user. The extension handles US, Metric, and MetricWX unit systems seamlessly.

Troubleshooting

View Extension Logs

Check WeeWX logs for nuvoler messages:

tail -f /var/log/syslog | grep nuvoler

Or on systemd systems:

journalctl -u weewx -f | grep nuvoler

Enable Debug Logging

To see detailed debug information, enable debug mode in /etc/weewx/weewx.conf:

debug = 2

Then restart WeeWX:

sudo systemctl restart weewx

Debug logs will show:

Test the Extension Manually

Run the built-in test suite to verify correct conversions and wind estimation:

cd /usr/share/weewx
PYTHONPATH=bin python bin/user/nuvoler.py

This executes three unit system tests and detailed Weibull analysis. Expected output:

================================================================================
Test 1 - Purely US Units (weewx.US)
Input: US units (°F, mph, inHg, inches)
================================================================================
https://www.nuvoler.com/data/recibir.php?station_id=50&station_pass=12345&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=9.5&wind_max=15.3&wind_min=4.2&precip=2.4&uv=5&dewpoint=14.2

================================================================================
Test 2 - Purely Metric Units (weewx.METRIC)
Input: Metric units (°C, km/h, mbar/hPa, cm)
================================================================================
https://www.nuvoler.com/data/recibir.php?station_id=50&station_pass=12345&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=9.4&wind_max=15.1&wind_min=4.1&precip=2.4&uv=5&dewpoint=14.2

================================================================================
Test 3 - Purely MetricWX Units (weewx.METRICWX)
Input: MetricWX units (°C, m/s, mbar/hPa, mm (L/m²))
================================================================================
https://www.nuvoler.com/data/recibir.php?station_id=50&station_pass=12345&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=9.4&wind_max=15.1&wind_min=4.1&precip=2.4&uv=5&dewpoint=14.2

================================================================================
3-PARAMETER WEIBULL ESTIMATION ANALYSIS
================================================================================

Test Case: wind_avg=4.86 m/s (~9.45 knots), wind_max=7.78 m/s (~15.1 knots)

================================================================================
METHOD 1: 3-Parameter Weibull Distribution
================================================================================
Estimated parameters:
  θ (location) = 1.458 m/s (minimum threshold)
  k (shape)    = 2.000 (distribution curvature)
  λ (scale)    = 2.946 m/s (spread parameter)

  5% percentile (wind_min): 2.125 m/s (4.13 knots)
  10% percentile (wind_min): 2.414 m/s (4.69 knots)
  25% percentile (wind_min): 3.038 m/s (5.91 knots)
  50% percentile (wind_min): 3.911 m/s (7.60 knots)

Theoretical statistics:
  Mean = 4.069 m/s (observed avg: 4.86 m/s)
  Std Dev = 1.365 m/s

  Estimated minimum (5th percentile): 2.125 m/s (4.13 knots)

================================================================================
COMPARISON: All Available Methods
================================================================================
3-Parameter Weibull (PRIMARY):  2.125 m/s (4.13 knots)
2-Parameter Weibull (FALLBACK): 3.871 m/s (7.53 knots)
Climatological Ratio (FALLBACK):2.945 m/s (5.73 knots)
Extreme Value Theory (FALLBACK):3.112 m/s (6.05 knots)
Auto-selected method:           2.125 m/s (4.13 knots)

================================================================================
EXPECTED OUTPUTS (for all three unit tests - should be identical):
================================================================================
temperature=22.5 (°C)
rh=65 (%)
mslp=1013.2 (mbar/hPa)
wind_dir=180 (°)
wind_avg=9.4 or 9.5 (knots)
wind_max=15.1 or 15.3 (knots)
wind_min≈4.1 to 7.5 (knots from advanced estimation)
precip=2.4 (mm (L/m²))
uv=5 (index)
dewpoint=14.2 (°C)
================================================================================

All three unit tests should produce nearly identical results, confirming correct unit conversion across all WeeWX unit systems.

Common Issues and Solutions

Issue Cause Solution
No uploads occurring Credentials missing or incorrect Verify station_id and station_pass in /etc/weewx/weewx.conf match your Nuvoler account
Missing parameters in upload Weather station lacks sensors Extension only uploads available parameters; ensure your station has required sensors (temperature, humidity, pressure, wind, etc.)
Connection errors Network or firewall issue Test connectivity: curl -v "https://www.nuvoler.com/data/recibir.php?station_id=YOUR_STATION_ID_HERE&station_pass=YOUR_STATION_PASSWORD_HERE&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=12.5&wind_min=8.0&wind_max=16.0&precip=2.4&uv=5&dewpoint=14.2"
Incorrect unit conversion Wrong unit system configured Check /etc/weewx/weewx.conf for correct unit_system setting (US, Metric, or MetricWX)
Wind minimum unrealistic Insufficient wind data Wind minimum estimation improves with both avg & max; if only avg available, climatological ratio used; if only max, extreme value theory used
Extension not loading Python path or import error Verify WeeWX installation; check logs with debug = 2 enabled

Test Manually with curl

curl -v "https://www.nuvoler.com/data/recibir.php?station_id=YOUR_STATION_ID_HERE&station_pass=YOUR_STATION_PASSWORD_HERE&temperature=22.5&rh=65&mslp=1013.2&wind_dir=180&wind_avg=12.5&wind_min=8.0&wind_max=16.0&precip=2.4&uv=5&dewpoint=14.2"

Dependencies

This extension uses only Python’s standard library. No external packages are required.

Python modules used:

WeeWX modules used:

Architecture

The extension uses a thread-safe queue-based architecture inherited from WeeWX’s RESTful framework:

  1. Main Thread - Captures new archive records and queues them
  2. Background Thread - Processes queued records, performs calculations, and uploads to Nuvoler
  3. Queue - Thread-safe FIFO buffer for decoupling data capture from uploads

This design ensures that weather data uploads don’t block WeeWX’s main weather station loops, maintaining responsive station operation.

Key Components

License

Copyright © 2026 RC Chuah

Distributed under the terms of the GNU General Public License (GPLv3)

Credits

Mathematical References

This site is open source. Improve this page.