[time-nuts] The home time-lab
attila at kinali.ch
Mon Jul 25 20:07:59 EDT 2016
On Mon, 25 Jul 2016 12:28:52 -0500
David <davidwhess at gmail.com> wrote:
> I have never seen one which did not use a class-D output with L-C
> filtering. Total efficiency is in the 84 to 92 percent range.
Actually, the better ones have 3 level or 5 level inverters (or even more).
Ie their output stages can not only produce +340V and -340V, but
also 0V (for the 3 level) and +/-170V (for the 5 level). For obvious
reasons, this gives a nicer output waveform that needs less filtering.
While the 3 level is easy to build, the 5 levels and more require
a flying capacitor (or multiple for more than 5 levels). The handling of
this capacitor is kind of tricky as you need to ensure that the flying
capacitor has the right voltage under all conditions. For those interested,
quite a few papers on that topic are publicly available. Please excuse me
not linking to any of these, but I know very little of inverters, so I do
not know which one of the papers would be the right ones to read.
Any simple idea will be worded in the most complicated way.
More information about the time-nuts