I am trying to understand what is the main difference between inversions and voicings, The way I see it is that the inversion will establish the lowest note on a chord, while the voicing is the specific arrangement of harmonics used when playing that chord.
But this becomes blurry in my mind when using extended chords. Can someone please explain how inversions work with extended chords?
One just gets more inversions. With 7th chords, there are 4 inversions; any note can be in the bass. Inversion of chords does not specify the disposition of upper voices (in general.) There is an exception in that 9th chords almost always have the 9th above the root. For chords with more than 5 tones, it's not always the case that the object is a chord; it could be a simpler chord with non-chord tones added (not uncommon).
Exactly the same way as with simple chords, just more possible inversions.
Having said that, anything beyond the 7th in a 'pile-of-thirds' type extended chord is almost always going to be in the upper structure, not as the bass note. C, C/E, C/G, C/B♭ (call it C7/B♭ if you like) are commonplace. C9/D, C7(#11)/F# and C13/A are less likely.
If you're thinking in guitar terms, and there's a bass player, there may be little point it thinking of 'inversions' in what YOU play, just 'voicings'. It doesn't matter so much whether you're playing C E G, E G C, G C E or an open voicing if the bass is hitting (say) an E. Whatever YOU do, musically it's a first inversion.
The way I see it is that the inversion will establish the lowest note on a chord, while the voicing is the specific arangement of harmonics used when playing that chord.
That right.
But this becomes blurry in my mind when using extended chords.
Let's first point out that seventh chords can put any chord tone into the bass including the seventh. Ex. C: V4/2 would be F G B D with the seventh in the bass, typically resolving down to the third of the tonic chord in I6. Without question that is totally conventional harmony.
Now let's consider extensions to be the intervals high than the seventh: ninths, elevenths, and thirteenths. The seventh should be present with those extensions so they aren't mixed up with simple intervals add2, sus4, add6, etc.
The conventional treatment would be to place the extensions above the chord root. (Walter Piston's Harmony is my source for that point. His text is concern with roughly Bach to Brahms with later Romantics like Grieg and Dvorak, so fairly conservative.) I think the rationale is that putting the extensions in the bass can give the impression of other chords with some added non-chord tones or their own extensions!
Consider a basic G7 and the extensions above. Put the ninth in the bass and it sounds like an Am9 on the bottom with two appoggiaturas, put the eleventh in the bass and it sounds like a Cmaj9, put the thirteenth in the bass and it sounds like an Em7 with a b9 or a non-chord tone of some kind.
So for all practical purposes, conventional handling sets the inversion "limit" at third inversion. And for clarity's sake put the extensions above the root, but you probably have a lot of freedom with the specifics of the voices above the root.
In theory, inverting an extended chord works the same as inverting a simple triad, except that more inversions are possible. In practice, however, notes are usually omitted from extended chords to avoid jarring dissonances, so not all of the inversions are seen in the wild.
13th chords are usually in root position because their inversions are academic. Because all 7 notes of the scale are present in a 13th chord (at least in theory), inverting it has the effect of turning it into a different 13th chord.
To touch on the distinction between inversion and voicing: A chord's inversion will only tell you which note is in the bass. It doesn't say anything else about it. A chord's voicing is a more complete picture that would include the inversion as well as other characteristics. The term voicing is often used in the context of a particular instrument, e.g. piano voicings or guitar voicings, and therefore include a consideration of what is idiomatic for that instrument.
Other characteristics which factor into the voicing are (not necessarily exhaustive):
There is some crossover. Inversions (and root position) will tell what note is at the bottom - in C major triad, C as bass will be root, E as bass will be 1st inversion, and G as bass will be 2nd inversion. It matters not which order the other two are, above the lowest note.
The more notes added to that triad (7ths, 9ths, 11ths, 13ths, and their ♭/♯ alterations), the more inversions there can be. Obviously, with a 5 note chord, there are 5 different choices for which note is at the bottom. Again, order of other notes (or even lack of them) won't affect the naming of that inversion. Some may not sound good - but that's a different kettle of fish!
Voicings - open or closed? Closed voicings occur when each and every note is as close to the next (up and down) that it can be. Open voicings use whatever notes are appropriate, but will not use every consecutive note from that chord. All voicings are subject to root or inversions. They can have any of the approriate notes, in any order. They can double or treble certain notes. On guitar, it's common on chords such as open E to play three root notes.
So, whilst being closely related, they're not the same, although they do both contribute to information about a chord, they are not synonymous. Every voicing will also have a label for its inversion, but an inversion tells us little about the voicing.
Check out slash chords, drop voicing. And be aware that in certain circumstances, 3rd, 4th, 5th inversions won't necessarily sound good. In others, they work really well...
