Short range NWS radar loops. It currently only supports composite reflectivity. I’ll add support for more radar types. Radar stations can be found by clicking on this map.
Short range NWS and Mosaic (regional and entire continental US) radar loops with support for multiple radar image types.
Place files into path/to/haconfig/custom_components/nwsradar/ or use HACS to install (preferred).
I had to install libjpeg-dev to my OS before I could get this to work.
What card do you use for the camera? I would prefer something where you could zoom & pan to focus on a particular area. This is the first time I have used a camera in HA.
The base topo image is in jpeg format, I bet this causes the requirement. I don’t know if there is a workaround.
I just followed the builtin bom camera platform, so I don’t profess to be an expert on this. As far as I can tell, it uses the same logic as bom camera platform.
I tested it with picture-entity card. I cannot zoom with it on my phone anyway. I’m not sure how that would work, the platform just supplies a gif as a binary python object.
- type: picture-entity
entity: camera.my_stn
After more testing on my end, it seems that the radar only updates if the page is refreshed. Sometimes there is an error in the log during the update process too. I’m not sure how to get around this.
If anyone knows how to periodically update the image in the camera platform without a refresh, I would be interested to learn. I should test the bom platform to see if the same behavior occurs.
I have updated this to poll images and to precompute a merged overlay image rather than apply all the static overlay images every time. This significantly speeds up loading time, particularly the polling.
I’d like feedback on what a good polling time interval is. Right now it updates the image every 5 minutes. This seems like a good balance of having up-to-date images but not downloading too frequently. Each image is approximately 10 kB, and 6 images are downloaded by default, so this comes to ~17 MB a day or ~506 MB a month.
The custom integration has also been reconfigured to work with HACS, but you can still manually download it to install. The files to copy are now in custom_components/nwsradar/ folder. There is PR to include this as a default integration repo in HACS, but it isn’t merged in yet. You can add as a custom repo in HACS in the meantime.
Use the map link in the first post, and click on your area. It should take you to the local radar station. Look in the top left of the page for the station ID inside the arrow navigators.
For example, near Mobile, AL, the code is MOB.
camera:
- platform: nws
station: MOB
If you meant to click New Orleans area you can use the arrow keys to find a better station.
I got really fed up with topo background and found the undocumented Standard style radar images and loops. This is now in version 0.3.0 available. This is now the default. The topo is still available as Enhanced style.
Have you tried the latest version with the standard radar? It might not need this dependency, but I haven’t tried. The standard radar comes as premade gif format, so there are no jpeg images. I have no idea whether handling gif images only also requires it.
I may get some ideas from this in the future but I have moved on to openHABian on my Pi 3 for Z-Wave functionality and stability. I have been running their testing version and it is much more stable than the released versions here. In fact I upgraded last weekend with zero issues!
I’m not quite sure what’s going on, but the image for LWX in lovelace hasn’t been updated in 28+ hours. Going to the NWS site for LWX shows current images. A screen refresh to clear the cash (CTL+F5) did not work, and there’s nothing in the system log that’s relevant to the integration.
I’m currently running Home Assistant 0.111.4 with Supervisor 227 (both latest as of this posting).
I’m glad it is working now. Did you check the Enhanced or standard style radar on the website and was it the same as what you used in ha? The server might have images available for one but not the other.