Measuring with cheaper socket and ballast

[EPdeHaas]

I've been taking some measurements for some lights using a smart plug from Nedis. The downside is that this plug doesn't measure below 3W. To overcome this I set up each bulb alongside an incandescent bulb I still had laying around (know it would come in handy at some point), measured just de incandescent bulb, then measured both bulbs together and substracted the draw of the incandescent bulb to get the draw of the smart bulb.

For me personally this seems to be accurate enough, and I thought I'd share it here to help others setting up lineair/calibrate configs.

I was also thinking of creating some lut files and was wondering if this would be seen as bad practice / to inaccurate.

[bramstroker]

@EPdeHaas This should work (also see #371 where the OP is using some resistors, which has the same effect).
The measure tool provides an option to measure the dummy load first and than subtract this from the actual measured power when writing the CSV files.

However as you correctly indicate I am wondering about the resolution of this particular smartplug being insufficiant. As lights typically have a power draw of < 10W and the plug does not measure below 3W out of the box.
Most other plugs we tested (for example shelly plug S) do not measure correctly below 0.4W, which is sometimes also problematic but can measure almost complete spectrum quite accurately

[EPdeHaas]

After playing around with it a bit more, it seems that you're right on the accuracy. And in my use case it's not a big issue. I just want to get a better idea of what I'm using. I don't have solar or batteries to worry about. But I have to agree that te csv I generated isn't sufficient enough to be added to the list. And as I found out that it wouldn't be accurate enough, I didn't do a full spectrum measurement, but just some point and then used a js script to fill in the rest. And the same but cubed for the hs list.