Announcement

Collapse
No announcement yet.

120V to 120/240V. Transformer or other ideas?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #61
    190914-1132 EDT

    Some additional thoughts.

    When a generator starts up, if it sees no existing terminal voltage, then it says I am the master and sets an internal memory saying so. Thereafter it is in a master mode that defines frequency, and no load voltage. Both the frequency and no load voltage can be invariant values, but their values are defined within this master generator.

    Then the inverter of the master can synthesize an internal impedance that is relatively non-dissipative. This is done by changing the generated voltage as a function of load current. This would be some reasonable value. Possibly 5% from no load to full load. If an external power source parallels this master, then the voltage would rise because the external source is supplying some power, and the master would supply less power and current.

    When a slave generator is added, then a self balancing occurs based on the synthesized internal impedances. This I have to think about a little more.

    .

    Comment


      #62
      190914-1212 EDT

      fastline:

      Since you have a Honda inverter generator you could run some of the experiments I described. This would possibly help us better understand if any of our ideas are correct.

      It even might be possible to simulate slave operation by using power company supply as the master. One would do this by creating an internal impedance from the power company supply to a load, for example 1 or 2 ohms in series with a 1500 W heater. Then parallel the heater with the Honda. The choice of the resistor would be based on what the internal impedance of the Honda looks like, and what the experiment is to show.

      1000 ft of #12 copper wire is about 1.5 ohms if memory is correct.

      .

      Comment


        #63
        190914-1409 EDT

        I went on-line looking for Honda internal impedance. Haven't found anything.

        But ran across this https://www.electro-tech-online.com/...output.152935/

        Further searching and I ran across this very good discussion on many things. I have only scanned thru it.
        http://www.screenlightandgrip.com/ht...​​​​

        .
        Last edited by gar; 09-14-19, 02:34 PM.

        Comment


          #64
          190915-1527 EDT

          Nobody has responded with any quantitive information on Honda internal impedance, or a description on how the inverter generators balance load. The patent reference did not adequately address this question.

          Tried a new search string and found this result that is very useful and informative, but does not provide an answer. Google is a rather poor search engine these days. If I include internal impedance in my string, then don't bring up results that don't include that string as the first results.

          The useful site found is http://www.screenlightandgrip.com/im...r_%20Part2.pdf . A continuation of the site referenced in my previous post. This site does not answer my question, but has much useful general information, and has scope pictures, and FFT plots.

          .

          Comment


            #65
            The search string "parallel inverter load sharing" brings up quite a few papers on a Google search. The few papers I viewed were not directed at inverter generators specifically. But I think they are applicable to PWM inverters that typically have an H-bridge output stage and drive a passive load (by that I mean not a nearly voltage source type load such as the POCO in solar applications). Such inverters are used in UPS applications, inverter generators, and other places. So there may be useful information in these papers that's relevant to the present discussion.

            A few papers are:

            https://www.researchgate.net/profile...​​​​

            https://www.sciencedirect.com/scienc...74667016440358

            Comment


              #66
              190915-1715 EDT

              synchro:

              Thanks. Good reference. Probably what Honda is doing.

              I would like to see some experimental data on operation of Honda generators.

              Assuming the droop method is used by Honda, then I believe my suggestion of an autotransformer driven by two generators, one on each half of the autotransformer, would be a bad idea. Transformer impedance would probably degrade balancing between the two generators. Therefore, both generators need to be directly paralleled, and drive 1/2 of the autotransformer. Thus, the larger autotransformer is required.

              .

              Comment


                #67
                190927-1029 EDT

                fastline:

                What experiments or new information can you provide us on your Honda?

                .

                Comment

                Working...
                X