Forum Discussion

BFL13's avatar
BFL13
Explorer II
Nov 30, 2018

Adding VAs Gen Running Things Question

I had the gen out again doing a few tests. One thing that I don't understand is this:

RV fridge on 120v (a heater) by itself VA (and watts same) 355
MW by itself 1426 (watts 1340--was some PF with the MW)

So to run both at once, you would expect 1426 + 355 = 1781
But on the Kill-A-Watt, it was 1650 VA. How come?

Is it perhaps an example of how running a resistive load with the MW affects things? I only have this one example. Or is it some other reason?

40 Replies

  • Wow, look at this.

    While I'm with family of electrical and all sort of engineers, when I run my genny and it trips, I just say "oops" and restart it again.

    Can't help it guys, sorry. LOL!
  • true but not part of the OP's question

    inverter for RV, not involved in this test

    generator is inverter type for power output
  • Gdetrailer wrote:


    The cleanliness and accuracy of the generator WAVEFORM also plays a key roll in just how accurate your "Killowatt" meter will end up.. Poor quality sine waves OR Modified Sine waves do not "play well" with the Killowatt meter..
    My observation has been different, finding that the kill-a-watt gives reasonably accurate RMS values with MSW and mangled sine waves. The only shortcoming I found with MSW is that the frequency reading doesn't work.
  • Gdetrailer wrote:
    MrWizard is spot on.

    AC "electronics" and DC "electronics" ARE two different creatures when it comes to mathematical computations..

    To fully understand and make the math work can often lead to the use of Calculus.. You cannot simply add and subtract to make the numbers work in AC voltages due to Inductance, capacitance and resistance that your AC loads WILL present.

    The cleanliness and accuracy of the generator WAVEFORM also plays a key roll in just how accurate your "Killowatt" meter will end up.. Poor quality sine waves OR Modified Sine waves do not "play well" with the Killowatt meter..

    Additionally, you ARE using an instrument to measure AC that I suspect is NOT very highly "calibrated" and trying to use it as the source of "truth".

    The "Kilowatt" meter is effectively like trying to use a sledge hammer and a railroad spike to to fix a Rolex watch, just not enough FINE DETAIL and most likely a lot of ERROR to work with..

    Rather than spinning your wheels, sometimes you just have to "accept" that the results may be less than what you want due to variables like how accurate your meter is, how accurate the gen waveforms are and how reactive your loads are..

    Nearly 30+ yrs of electronic tech experience talking..


    exactly. in the OP's case to "add them together" you need to realize that each value has a magnitude and angle associated with it. It's called "vector" math as compared to "scalar" math, single values you add, substract, multiply, divide.

    and you have to accurately determine what the magnitudes and angle is for each variable.

    and if you have equipment to "measure" it you need to know the equipment limitations. The kilo watt is a nice qualitative tool, gives you a good idea of the power draw. Works well to give a reasonably good idea of current draw, power factor, VA and watts to help answer questions or solve problems. But it is NOT a tool to give accurate verifiable results. It can tell you that "yup you are likely pushing the limits of or have margin left in a generator, breaker etc."


    It is NOT a good quantitative tool, one that gives a precise accurate answer under all conditions.

    Consider that a good power quality meter that will give accurate answers for the same conditions run's in the thousands if not 10's of thousands of $$$.
  • Reason to know is if this would help reduce the VA being pulled when I am running my converter/charger. It was suggested that I try a small resistive load at the same time.

    I don't know how small would have the effect in proportion to what the converter pulls in VA, and don't want to add more VA than I would "save". I will play around with that and see what happens.

    I am running the RV battery bank down so I can do some full tests tomorrow.
  • MrWizard is spot on.

    AC "electronics" and DC "electronics" ARE two different creatures when it comes to mathematical computations..

    To fully understand and make the math work can often lead to the use of Calculus.. You cannot simply add and subtract to make the numbers work in AC voltages due to Inductance, capacitance and resistance that your AC loads WILL present.

    The cleanliness and accuracy of the generator WAVEFORM also plays a key roll in just how accurate your "Killowatt" meter will end up.. Poor quality sine waves OR Modified Sine waves do not "play well" with the Killowatt meter..

    Additionally, you ARE using an instrument to measure AC that I suspect is NOT very highly "calibrated" and trying to use it as the source of "truth".

    The "Kilowatt" meter is effectively like trying to use a sledge hammer and a railroad spike to to fix a Rolex watch, just not enough FINE DETAIL and most likely a lot of ERROR to work with..

    Rather than spinning your wheels, sometimes you just have to "accept" that the results may be less than what you want due to variables like how accurate your meter is, how accurate the gen waveforms are and how reactive your loads are..

    Nearly 30+ yrs of electronic tech experience talking..
  • Two possible factors:
    (1)Could it be that the voltage is lower with both devices running together?
    (2)With a small generator, running the microwave will result in the waveform changing. More so if it's not an inverter generator. Adding a resistive load results in lessening the waveform corruption. Can see this only with a scope. Measuring each device individually while both are running might yield a small clue or two.
  • fridge heater is resistive load

    MW is an inductive & capacitive load

    combined they are a 'reactive' load

    you are trying to do simple math on a complex equation

    need some electrical engineering classes to do those equations

    and you need to know the values of all the 'variables'