Harbeth Cone Excursion - a theory?

[Diminish]

Hello again to all at the HUG. I would like to share some observations that I have noted and ask if they agree with what others observe, and whether or not this is integral to the design of Harbeth loudspeakers. What I'm talking about here is cone excursion, sometimes referred to as "X-max". It is the distance that the cone travels in it's efforts to produce sound. It is this motion that excites the adjacent air molecules and produces sound. Although I'm uncertain of the exact mathematical relationship relating cone excursion to sound pressure level (SPL) in a Harbeth, I have seen a formula for infinite baffle (sealed cabinet) enclosures that would suggest that excursion is proportional to SPL. Lower frequencies with longer wavelengths require more excursion.

You can go to You Tube and watch videos of "bass tests" where some pretty incredible (and dangerous) things happen as a result of extreme SPL's at low frequency. There is one where CD's are violently shattered by placing them near a subwoofer's port, and another where a woman's glass eye pops out. One particular subwoofer listed a whopping 5" of cone excursion! Now, while it might be impressive to see a speaker cone jumping around like the third monkey on the gang plank of Noah's Ark, this is diametrically opposed to the goal of natural sound reproduction. Going a little further in my search on You Tube, I encountered some videos of B&W's Kevlar drivers which also displayed impressive X-max. I have previously owned speakers from Vienna Acoustics, B&W, Rogers, and Paradigm, and noticed quite a bit of travel in their woofers when playing low bass at high volume.

My Harbeth M-30, on the other hand, barely seem to move at all. Now let me be clear about this from the beginning: I am not recommending that anyone perform a "bass test" on their Harbeths. I am not encouraging users to play low bass at high volumes, nor am I saying that a loudspeaker's excursion is any measure of it's sound quality. This is merely an observation. My Harbeths are every bit as musically satisfying as any of my previously owned models, in fact, I feel that the bass response is better than any of the aforementioned. I'm just wondering: #1) If this is something that anyone else has noticed, and #2) If low excursion was a goal of the designer.

I know that Harbeths are engineered to sound best at moderate volumes. Is this, in any way, related to shorter cone excursion? Do the RADIAL material and the thin wall cabinet design allow for more efficient operation in terms of converting the cone's motion into sound?

[garmtz]

Maybe your earlier speakers had much smaller cones? Please bear in mind that the bigger the cone, the less excursion is needed for a given sound pressure level. Also, these extreme excursions are normally only 'needed' when electronic music with very low bass content is played (with the occasional large symphony orchestra or large acoustic drums as exceptions to the rule).

The problem with the youtube content, is that you don't get an idea of the playback levels that are being used. I know the movie where the lady loses a glass eye and we are talking levels way over 120 dB (sometimes in the lethal range, 140-160 dB), something that no Harbeth speaker (or untreated car) will survive. Extreme excursions call for extreme levels and if you don't play at excessive levels, the cone will not move excessively.

[Diminish]

Undoubtedly, longer excursions do translate to higher levels of sound pressure. I assume that this would be true even for planar speakers. I still think that my M-30's are able to produce deep bass at relatively high sound pressure levels (SPL) without much in the way of excursion. The other speakers that I've owned have all been 2 way, bass reflex designs with similar configurations. The Vienna Acoustics had a 6 1/4" full range driver and all of the others had 8". My Paradigms, in particular, displayed quite a lot of excursion. Even with their 93 dB sensitivity, I don't feel that they did a better job of reproducing bass than the Harbeths.

I realize that there are several factors at work here. Generally speaking, longer excursion (X-max) will produce higher SPL's. Longer wavelengths (lower frequencies) require longer excursions as well. It is the current supplied by the power amplifier to the voice coil that causes the cone to move. * Well, I know that a loudspeaker driver responds to current, but I would have to think that the product of current and voltage, Power, is more aptly responsible for this motion.

In the product literature for the P3ESR, they talk about the new 5" RADIAL driver being a "long throw" design. If I'm not mistaken, this refers to the voice coil travel. A smaller driver would be required to move much farther to displace a similar amount of air compared to a large woofer. If you look at the surround on the smaller RADIAL driver against the 200mm unit, you can see that the it seems to be designed to accommodate more travel. Again, looking at the 12" woofer on the Monitor 40.1, it is obvious that only a relatively small excursion is required to move large quantities of air. Another advantage is that it liberates the 200mm driver from reproducing low frequency content.

Another factor that can influence SPL with regard to cone excursion is loading. A horn loaded driver's cone need only move a small fraction of the distance a dynamic cone would have to move to produce an equal SPL. Despite their shortcomings which result in tonal colorations and "beaming", some horn loaded drivers can boast efficiencies of 115dBW/m.

lthough I don't have any measurements to support the claim that the 200mm RADIAL driver in my M-30's doesn't seem to move nearly as much as other HiFi speakers that I've observed, I am sure that this is the case. Since Harbeth recommends listening with the grilles on, many of you on the HUG may have never seen your drivers moving. I was wondering if anyone could confirm this observation, and if it is intentionally designed into the speaker. Horn loading is one technique that uses a much smaller cone excursion to produce high SPL's. I noticed that the full range driver in the M-30, M-20, P3ESR, and M-40.1 are all mounted behind the baffle. Does this provide a loading effect? Does the thin wall cabinet design allow for more efficient sound transfer? Is the RADIAL material itself capable of this on it's own?

[A.S.]

Undoubtedly, longer excursions do translate to higher levels of sound pressure...

I suspect the answer to your conundrum is contained in your own words ....

The other speakers that I've owned have all been 2 way, bass reflex designs with similar configurations... [one] in particular, displayed quite a lot of excursion. Even with their 93 dB sensitivity, I don't feel that they did a better job of reproducing bass than the Harbeths.

The common thread here is that all three speakers in question are vented, that is, they have an opening of some shape and size. In a vented cabinet the woofer in motion at low frequencies will 'bounce' the air inside the box as an acoustic spring and sound will (in a certain relatively narrow frequency range) be phase inverted and emanate from the port.

As with any mechanical system, and very much like the design of a car suspension, the designer can balance many variable. He can design an F1 racing car with a taught suspension system in which the driver feels in intimate contact with the contours of the road. Such a design is only comfortable providing that the road surface is glass-like smooth and barely usable on conventional roads. Another compromise would be the Rolls Royce suspension: a big, wallowing, superbly soft suspension that soaks up every lump and bump and is slow to respond making high-speed cornering rather tricky. In the first case if the road-surface's undulations (at a particular travelling speed) exactly coincide with the suspension's natural resonant frequency then there will be violent, poorly controlled up and down motion of the car. In the RR example, it may not even be possible to identify the natural resonant frequency because it is so well damped.

So, what I'm suggesting is that the violent excursion you've witnessed is probably only evident over a relatively narrow low frequency range and power level. It reflects a particular design balance, possibly even a cultural preference for a different listening 'driving experience'. If you use a turntable as a source with its inevitable LF rumble and groove tracing noises this cone excursion effect will be magnified by the presence of lots of VLF to tickle the speaker's vent. That's really all there is to this! No new breakthrough! No reinvention of physics!

[A.S.]

Undoubtedly, longer excursions do translate to higher levels of sound pressure...

I suspect the answer to your conundrum is contained in your own words ....

The other speakers that I've owned have all been 2 way, bass reflex designs with similar configurations... [one] in particular, displayed quite a lot of excursion. Even with their 93 dB sensitivity, I don't feel that they did a better job of reproducing bass than the Harbeths.

The common thread here is that all three speakers in question are vented, that is, they have an opening of some shape and size. In a vented cabinet the woofer in motion at low frequencies will 'bounce' the air inside the box as an acoustic spring and sound will (in a certain relatively narrow frequency range) be phase inverted and emanate from the port.

As with any mechanical system, and very much like the design of a car suspension, the designer can balance many variable. He can make a racing car taught system in which the driver feels in intimate contact with the contours of the road. Such a design is only comfortable providing that the road surface is glass-like smooth. Another compromise would be the Rolls Royce suspension: a big, wallowing, superbly soft suspension that soaks up every lump and bump and is slow to respond making high-speed cornering rather tricky. In the first case if the road-surface's undulations and travelling speed exactly coincide with the suspension's natural resonant frequency then there will be violent, poorly controlled up and down motion of the car. In the RR example, it may not even be possible to identify the natural resonant frequency.

So, what I'm suggesting is that the violent excursion you've witnessed is probably only evident over a relatively narrow low frequency range and power level. If you use a turntable as a source with it's inevitable LF rumble and groove tracing noises this effect will be magnified by the presence of lots of VLF to tickle the speaker's vent. That's really all there is to this!

I'd suggest that