The accuracy specification in the WM-1 manual states that the accuracy is "5% of full scale and 8% worst case". (Makes you wonder why they just don't just say 8 % ???)
So on the 20 w scale that is +/- 1 watt (using the 5% accuracy figure).
So, if you read 3 watts on the WM-1, it could still be within specification for an actual power anywhere between 2 to 4 watts.
I read their manual and it has some really interesting stuff in it and there is a big discussion of accuracy. If I use a power meter, I just use it for relative readings.
Oscilloscopes are getting pretty cheap now. If I want more accuracy, I like to look at the waveform across a dummy load and do the math to calculate the power.
Finally, if the power indicator on the Argo is exactly between '3' and '5', that may not be '4' due to the non-linearity of the meter. Plus, the Argo wattmeter also has an accuracy tolerance, I'm sure.
You can't win.
"I did a bit of "domestic archaeology" in the attic this afternoon and