The 40.2, 30.1 and P3ESR have their bass drivers mounted from the inside of the front baffle, while the SHL5 and C7ES3 have their bass drivers mounted from the outside of the front baffle. I doubt this was a random choice. Was there a specific reason for this?
The issue is not of extra protection at all. It concerns the most workable phase relationship between the bass unit and tweeter given all the other considerations of the design. Let's suppose we had to describe in precise words (not pictures) to someone on the phone the arrangement of the drive units on the speaker's baffle. Visualise the complete speaker cabinet suspended in the air on a hanging chain without the grille so we can clearly see the drive units; now we can talk about the position of those the two (or three) drive units in 3D space according to the X, Y, Z principal.
Imagine as you look at the speakers from the face (the baffle) the position of the voice coil of the tweeter and voice coil of the bass/mid unit. The X plane describes the position of these voice coils on a line drawn from left to right: in the the case of all Harbeth speakers the two voice coils are at the same position in the X plane, intentionally. The Y plane describes a line drawn through the middle of the cabinet from top to bottom: as the tweeter is above the woofer it has a higher number in the Y scale.
Now, there is another plane to consider: the Z plane. As you face the speaker this describes the position of the voice coils of the drivers as distance away from you. Clearly, the tweeter is nearer to you than the bass/midrange unit and although this distance is small (say, 5cms) it is quite an appreciable distance acoustically because sound waves travel so very slowly.
So, at the design stage I have to adjust the crossover and/or physical mounting of the woofer (I can't change the position of the tweeter, it must always be nearest to you) to compensate for the positional differences between the drivers. Shuffling the position of the bass/mid driver (relative to a fixed tweeter position) by even 1cm or so (i.e. 20%) can give an extra degree of design flexibility.
Rear mounting does slow down production though which increases cost and selling price.
Wow. That's some good digging Don. Thanks for finding that!
I haven't even considered the Z axis driver location as an issue with respect to the tweeter's Z axis position. But now that I think about it, I am aware of some speaker designers that call their speakers "time aligned" with respect to the proper listener position. Time aligned I think means that the signals from the tweet and woof arrive at your ear at the same time when speakers are properly positioned and listener is properly seated in front of the speakers. Typically to do that, as I recall reading about it, is they must use first order crossovers, and they must locate the woofer cones in front of the tweeter cones by a few centimeters in the Z plane (this having to do with the inertia or mass of the cone itself). Sometimes you can discern such a design because the baffle is itself angled--like a triangle or pyramid. I note also that some reviews I've read of Harbeths (Stereophile specifically) call Harbeths neither time aligned nor time coincident, as evidenced by the impulse response graph given as standard procedure on that magazine's tests.
I was originally concerned about this aspect of Harbeth speakers until I heard the C7ES3s and the SHL5+'s for myself at my dealer for the first time. And then all of these concerns vanished. These two models I've listened to, in my opinion, offer every bit as much depth and width in their sound reproduction as do those that are calling themselves time aligned.
I'm not sure what to make of this, except to note that some designers go to A LOT of trouble to make their speakers time aligned, but where apparently it is not that much of an issue in real life room listening. I'm not sure why this is so, but I think it has a little to do with the human filtering of extraneous "signals" (reflections, etc) so that your brain is able to piece together what the original recording venue/acoustics was mostly supposed to be. Just postulating though, as I am not a scientist in this area.