Bass Crossover Point - Extreme Slope

[raypenn]

What is the best bass crossover point for Khorn, LS and Belle, especially with an extreme slope crossover?

Do you want to be as high as the woofer will permit (assuming the midhorn can take it)?

Or do you want to be as low as the midhorn can go (assuming the woofer can take it)?

Or do you want to be somewhere in-between?

The reason for my confusion is that many seem to crossover close to the same point but higher than the midhorn "name" (i.e. Trachorn 400, Kilpsch 400, Klipsch 500, Altec 511).

For example, Al K suggests that you use his ES400 only for Khorn - Al, why would the ES400 not work for your Trachorn 400 in the Belle or LS? What factors would cause somebody to use the higher crossover points such as the ES500 or ES600 use?

Thanks.

[Hardhead]

This question is worth sending back toward the top for an answer.

[D-MAN]

The Khorn uses a nominal 400Hz point. Recently, Klipsch has moved it up to 450Hz, perhaps a little optimistically. Staying with the traditional 400Hz point is entirely justified.

There are 3 reasons for this:

1) the Khorn folds themselves are a limiting factor in the upper-bass reponse ; distortion (coloration) occurs above that point

2) the K33E driver upper response corner fall-off at about 400Hz.

3) the lower limit of the midrange driver/horn (fc = size) combination; traditionally the crossover 400Hz uses a 1rst order (6 db) slope. This means that both the bass driver/horn combination is producing the same frequencies as the midrange/midhorn combination at the same time, albeit at an reduction and corresponding increase in output, respectively, to approximate a reasonably flat response at the specified crossover point.

The Extreme slope crossover (in the case of the Khorn at 400Hz) allows for a very rapid cutoff at 400Hz, so the woofer and midrange drivers do not produce that same frequency at the same time; the frequency overlapping is greatly reduced if not eliminated for our purposes.

Benefits would include a decrease in comb-filtering distortion (intersecting waveform distortions of the same frequency from differing sound paths).

I would also add that the peception of "speed" or "timeliness" of the bass horn is made more seemingly "quick" (perceptually speaking) probably due to the lack of noticable smear between the two drivers, bass horn (longer sound path) vs. the midrange (shorter soundpath). Since the same frequencies are not shared in the case of the ES network, the difference in timing delay is not as perceptually apparent as with a lower order crossover. This is a rather subjective call on my part but I believe the effect to be noticable. Certainly mid-and-upper bass is "cleaner and clearer", with less muddiness overall.

The Belle and LS both use the same crossover point (400Hz) although many owners have gone as high as 800Hz with a 6db slope. In the cases of the shorter Heritage horns (Belle/LS) as long as the 800Hz limit is not exceeded (the throat slot size then becomes the absolute limitation at about 950Hz), the main limitation is the midrange/horn combination's capabilities.

DM

[boom3]

Dman, this is all good info. What are the slopes of the current Klipschorn network?

[Al Klappenberger]

Ray,

What D-MAN says is all true.

The choice of crossover frequency is not as critical for the Klipsch speakers becasue the Bass is horn loaded. If they were direct radiators you would want to get up to the midrage ASAP -- OR -- Another school of thought is that the crossover should be at the frequency where the disperssion between the woofer and squawker become the same. In an all-horn speaker, that's not an issue.

My thinking is that the lower you cross over the larger the part values in the network become. This means larger inductors with more DCR in the woofer path. This suggest a higher crossover is a good idea. The Khorn starts to go down hill above about 350 Hz. That is why it's crossed at 400 Hz. The Belle woofer goes nicely up to 700 Hz and the LaScala will also. To answer this, I build the ES500 network with AWG #12 wire and the ES600 and ES700 using #14 wire.

The other factor is how far the squawker will go down. The extreme-slope netwroks will allow any squawker driver to be used safely below its normal range safely becasue the extreme low frequencies normaly let through by the normal crossover filter are just not there! The other factor is the low range of the squawker horn. If you are using the Altec 811b "VOT" horn in a Belle or LaScala you need the ES700 network. The 511b horn will go below 500 Hz allowing the ES500 to be used. If you have my Trachorn you can cross at 400 Hz or higher. The Trachorn will operate safely well below 400Hz. This is just to allow for totally safe operation down to 400.

The ES400 network also has a slightly more gradual slope than the ES500, 600 or 700 networks. This is to keep the group delay below the limits published in the B&L paper. I personally do not believe that group delay is of any concern at all, but I wanted to be below this limit just to be sure. Group delay is nothing more than a measure of phase linearity. It is NOT the same as time or propogation dealy. For this reason, I would avoid the ES400 for any application EXCEPT the Khorn. It is a special compromise specifically for the Khorn.

AL K.

[D-MAN]

If I remember correctly the KlipschType-A is a 6, 12 and 18 db slope network, respectively.

Speaking of how low the woofer goes, it just occured to me that the K33E is "balanced" for the extreme lows and that's why it tends to drop off at the fairly low 400Hz or so.

As I mentioned elsewhere in an old post about other bass horn drivers, I found that drivers tend to be "weighted" in their response bandpass either one way (lows) or the other (mid-and/or-upper bass). I suppose that some of the more expensive drivers could have a wider and more well balanced bandpass, but I don't know for sure.

PWK clearly went for a bass horn driver (K33E) that leans towards the lows (understanding the limitations of the Khorn folding methodology and its limits) and compensated for its upper frequency corner roll off by adjusting the throat cavity opening size, network, and midrange driver/horn combination.

DM

[boom3]

Ummm, no, I meant the current (2005) Klipschorn network....

[D-MAN]

It's the same slopes as above; circuits changes notwithstanding.

DM

[boom3]

Thanks!