This is an interesting and useful posting. But sometimes the readings from a noisy supply will be inacurate because of the noise on top of the DC output, and possibly on top of the input as well. An easy solution is to add filtering and read the voltage outside the filters. Usually the output voltage will be different with the noise filtered out, either higher or lower. Likewise the current, although it will usually be lower. This does reflect that noisy supplies are not as good as quieter ones.
This article seems to ignore the fact that most (those without active PFC) AC-DC power supplies are non-linear loads, and therefore the traditional concept of power factor doesn't apply. Current drawn by a typical SMPS (like a computer power supply) is in phase with the voltage since the bulk filter caps charge at the peak of the voltage waveform. The resultant waveform is severly distorted however. This is not the case if the SMPS uses active power factor correction.
For non-linear loads, make sure to use a wattmeter designed to measure distorted waveforms. You can check for distortion by measuring the Crest Factor of the waveform using a DMM that has a peak capture mode and can measure true RMS. A pure sinewave will have a crest factor (Peak/RMS ratio) of 1.414. Any variation greater or lower indicates distortion. Typical SMPS current draw waveforms will have a very high crest factor, generally around 5. This indicates the typical "peaky" current waveform when caps charge at the peak of the voltage waveform. [voltage waveform distortion - especially in SMPS - tend to have crest factors less than 1.414 due to voltage flattopping]
An SMPS with active PFC will have sinusoidal (well, very nearly sinusoidal) voltage and current waveforms, and they will be in phase so input power is simply VrmsxIrms.
If you need a wattmeter, an excellent and economical choice is the Weston 310 series. These are the "Lab Standard" instruments in the mahogany cases rated at 1/4% accruacy. Since they are electrodynamometers they measure true watts and they have no permanent magnets that can change strength over time. You can check their calibration with DC using an accurate voltmeter and ammeter on the potential and current terminals, respectively, but it is best to flip polarity to check both ways.
These instruments were very expensive when new (>$1000), but are now available quite cheaply on the used market. And they are real works of art.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.