3 way vs. 2 way

[RFK]

Paul Klipsch originally designed the K-Horn as a 2-way speaker. If you take a look at the K-horn Paul Klipsch is standing by on the Klipsch Website it is a 2-way model. Steve Phillips and others at Klipsch have stated that Paul Klipsch originally designed the Klipschorn as a 2-way speaker and fealt that the truest sound was achieved from a 2-way speaker (better imaging etc.). Shortly before Paul Klipsch died Klipsch was going to release a Special Edition 2-way K-horn and apparently they got rerouted on other product when Paul died. On this forum you constantly hear that a 2-way speaker is asking too much of one horn to reproduce too many frequencies-NOT TRUE. I own many Heritage speakers, Legend Speakers and Reference Series speakers and I can tell you that although different 3 way is not superior to 2 way or visa-versa. Both designs have their own sound and both designs are great! A K-horn sounds great and so does an RF-7-they sound different yes but still great! Next time somebody on the forum says 2-way is not "True Klipsch" you can correct them because Paul Klipsch had high regard for 2-way AND 3-way design!

[SCOOTERDOG]

I always fantasized about a 3 way.

[damonrpayne]

I'm not a speaker designer, but I would think 3 way is almost always better than 2-way.

The laws of physics are the laws of physics. In the realm of "accurately reproducing sound", a certain sized horn or woofer is going to have a great degree of accuracy only within certain frequency ranges. I have read that only increasing complexity and power consumption of crossovers make loudspeakers with 4 or more different sized drivers undesirable. I wouldn't know anything about how n-way speakers behave differently in terms of imaging. Sounds like a good topic for debate.

[prodj101]

I think it is safe to say that 2 way speakers are not as good as 3 ways. Like damon said, physics lean in favor of a 3 way design, but that does not mean that a 3 way will always be superior. Infact it is often the opposite because good 2 way crossovers are much easier and less expensive to design well, and because of this they are often the better performing of the two. Crossovers are an extremely important part of a loudspeaker, and are in fact even more important, according to many designers, than the drivers themselves. This is because beyond just splitting the signal in half (or however many ways you need it) a properly designed crossover should also include baffle step compensation if needed, and also an L-Pad if necessary to even out the frequency response between drivers.

Beyond crossover design, driver placement on the front baffle is also very important and has a large impact on imaging and depth. When soundwaves (specifically midrange and high frequencies) travel from the driver they not only go forward, but also off to the sides along the baffle. After the sound waves reach the edge of the baffle, the space they have to radiate turns from 2pi to 4pi, which can cause up to a 6db inrease or decrease in sound level. By off setting a driver to the right or left of the center line this effect can be decreased to an extent, but baffle step compensation is often needed. This is why on many high end speakers the tweeter is offset from the rest of the drivers, or in it's own enclosure entirely (note B&W nautilus line). Calculating all of these factors is very difficult and much trial and error is necessary, and when you listen to cheap 3 way speakers at best buy it is obvious that this attention to detail is missing. However, if you listen to a very high end 3 way speaker i.e. Meadowlark Audio, B&W Nautilus, or Verity Audio I can guarantee you that you'll see the light

[RFK]

A good example of what I've observed in listening to my RF-7's and my LaScalla's is with my RF-7s the sound seems to blend and project from the entire face of the speaker and with the LaScalla's I notice more of a separation-in other words I am more aware of a separate sound coming from the tweeter and midrange etc. A good example would be the 2-way Epic Series that Klipsch produced with the patented "Controlled Focus Technology"-the imaging was better than my LaScallas but the sound was less lifelike in the higher frequencies. In the 2-way RF-7 design Klipsch was able to brighten the sound dramatically with a titanium tweeter vs. the Epic's aluminum tweeter and make the bass tighter with the Ceremetalic woofers vs. the Epic's carbon graphite woofers. I think that is why Klipsch has been producing 2-way instead of 3 recently with the exclusion of the Heritage Series and the new Reference Premiere Series that is supposed to be a 3-way design. The Reference Premiere Series will have a .75 inch "supertweeter" and a 4 inch fiberglass midrange horn driver. I believe this new design will prove to be a little less bright sounding than the RF-7 because of the Refernce Premiere using fiberglass woofers rather than Ceremetallic and a .75 inch tweeter instead of a 1.75 inch titanium tweeter-less "metally sounding" than the RF-7-I happen to like the metal sound however.

[dougdrake]

FWIW, the last project Paul devoted his time to was a 2-way speaker - the Jubilee. Hours upon hours of research and design working to get it right...

[TBrennan]

RFK---I think you might be missing the very important point that the old KHorn 2-way was crossed at 500hz and the midrange was entirely horn-loaded, either by the basshorn or mid-treble horn.

The classic Altec and JBL 2-way horn systems also crossed-over low, generally at either 500 or 800hz. Thus even Altecs and JBLs that used direct-radiating woofers had much of the midrange horn-loaded.

But the RFs cross at 2000hz and even higher. The midrange is from the direct-radiating cone woofer and not from a horn-loaded compression driver. The horn is merely a tweeter.

But yeah, 2-way horn systems can sound very good, IMO the finest horn systems are 2-ways. But not 2-ways done like RFs. I'm talking 2-ways done like Altec VOTs; A7s, A7-500s and A5s.

Note that aluminum diaphragm compression drivers generally have more extended high frequency response than titanium ones for a given size diaphragm. Titanium was first used by JBL for compression drivers and for it's reliability advantage over aluminum, not because it sounded better. Titanium compression driver diaphragms have less tendancy towards fatique cracks in the surround than aluminum ones.

[RFK]

I know what you are saying Tom but according to Steve Phillips the ceremetallic woofers allow for a higher crossover due to their transparency and excellent ability to produce lower midrange frequencies. You have to admit ceremetalic woofers are able to do things because of their lightness and rigidity that paper coned or even carbon graphite woofers cannot-I'll even add fiberglass woofers as being less transparent than Ceremetallic.

[RFK]

I think a good way to illustrate this point is to take a look at what Klipsch has released regarding the new Reference Premiere speaker. The Reference Premiere is said to be a 3-way design consisting of a .75 inch super tweeter, a 4 inch fiberglass midrange driver for the horn and fiberglass woofers. Why a 3-way for this design? Fiberglass is not suseptable to denting like the Ceremetallic cones are and with the new-style molded front baffle plate you would have to pull off the whole front of the speaker to replace a woofer. Thus Klipsch has made a stronger, less transparent woofer that is supposedly 10inches wide that because of its make up would require mid-range support because the 10 inch fiberglass woofer would not sound as detailed under 2200hz for lower midrange frequencies as the Ceremetalic does-enter the 4-inch fiberglass midrange driver. The RF-7 has a huge magnet for a 1.75 inch tweeter that I bet will equal the new Premiere's tweet and mid magnets combined. I think Klipsch went with the fiberglass woofers because of warrenty issues that would result from having the Ceremetalic less accessible for replacement due to the molded front baffle and that some people think the Ceremetallic sound is too bright/harsh. You may disagree but I think this is a compelling argument or observation depending on how you look at it.

[michael hurd]

Relative density of a material used to make a cone, varies greatly. You have aluminum, titanium, polyproplene, paper, and all sorts of other composites. They all have many ups and downs, but it all comes to intended usage, and price.

Regardless of warranty issues, a metal cone ( aluminum or titanium ) can help to draw heat out of the voice coil and gap. Compressed paper has to be one of the stiffest material to weight ratios, unfortunately unless treated, does not play well with water. The molded front baffle came about for better performance.

The proposed midrange driver is a 4" cone, loaded via a tractrix horn ( a la Genelec ), not a compression driver. Regardless, I am sure that it works fine.

[michael hurd]

Regarding the 2-way vs 3 way: At high frequencies, mass works against you. It takes a strong magnetic field, and a lightweight diaphragm to produce high frequencies. This is the reason that there aren't many tweeters sporting phenolic diaphragms these days. ( except for k-77 )

There are some that use titanium, aluminum, and also beryllium. If a phenolic diaphragm hits the phase plug in a comression driver, it will likely make some noises, but survive. A titanium diaphragm will not be as rugged. PWK even tried a beryllium diaphragm driver that exploded into dust when the diaphragm hit the phase plug.

A three way design allows for more power handling, and allows a tweeter that is relatively fragile, to live a long life. Every time you lower the response an octave, excursion goes up 4x, at the same spl, with the same diaphragm. This is why the tweeter in the reference series is crossed over around 2000 hz, to protect it from damage. In the same 2-way, you need a woofer that is capable of extending to 2kz, without nasty anomalies.

[RFK]

I am not saying that some of your points are not valid but as far as power handling the 2-way Epic CF-4 handles 300 watts continuous with 1200 watt peaks-compare that to the 3-way KLF-30 that handles 200 watts continuous with 800 watt peaks or even the 2-way RF-7 for that matter that handles 250watt continuous and 1000 watt peaks. Many of Klipsch's 2-way designs handle more power than my 3-way LaScalla's or the Klipschorn.

[doudou]

"The laws of physics are the laws of physics."

i agree.

theoricaly the 3 way should sound better.

but don t forget that in a 3 way speaker the crossover are more difficult to make and there are more electrical components.

an electrical component is never perfect, so the more components you intruduce in the crossover, the more difficulties to make a good crossover there are.

and crossover is not the only problem, with more ways, the frame of the speaker is more difficult to make, it has to be more robust...

in a nutshell, it s easier to make a good 2 way speaker.

it means that if you have two speakers of the same price, one 2 way and one 3 way, the 2 way will always sound better.

[RFK]

Between 3 way or 2 way it really comes down to personal preference. If we are talking the laws of physics then the truly perfect speaker would be a one-way speaker with a driver that could produce the lowest lows all the way up to the highest highs. In physics the sound of an explosion is a bunch of frequencies blended seamlessly into one sound-thus the best and most realistic way to reproduce the sound would be for a driver to produce all frequencies. An explosion or any other sound for that matter has no mid range driver, a tweeter or a bass producer-the sound emits from one source. How could you point to the wind and say from what inch of space the whistle emits from?

[michael hurd]

John, power handling is a funny issue. There are thermal limited power handling specifications and mechanical limitations. The former has to do with how efficient ( how much waste heat% ), and also how fast the driver can dissipate heat. A close magnetic gap allows faster heat transfer, one piece metal coil former and cone construction, pole piece venting and also the construction of the frame itself.

Nowdays, we have some drivers that will approach their mechanical limitations ( X-max ) long before they would fail from excessive current through the voice coil. The k-33 drivers used in the heritage products have short coils and tight suspensions. They will stay within their designed travel when horn loaded or loaded in a reflex enclosure. However, excessive continuous power will usually melt the coil lead connection under the dustcap.

Whether a speaker designer says 100 w continuous/ 400 max, doesn't really mean that much. I have driven some speakers with a lot more power than recommended, and have never had a failure ( burned coil ). OTOH, you could damage the speaker by feeding it a continous high power low frequency sine wave, even if it is within the recommended power handling.

What it really boils down to is to know your limitations. If you feed a reflex enclosure a high power signal below it's tuning frequency, cone motion becomes uncontrolled. This can damage the suspension, tearing the spider from the frame, the coil can come out of the gap and strike the pole piece and begin to rub.

[doudou]

i agree with you RFK.

the only problem is that there are very few good wide band speakers that can reproduce sound fron 20 to 20000hz.

with an only way, no crossover, and the sound comes from an unique source, like a real sound.

[RFK]

I know what you mean as far as rated power. I agree that specs can be misleading as an example in a Twin Cities Ultimate Electronics store a friend of mine and me tested an old LaScalla on a rather inexpensive 200 watt per channel amp-(I think it was an Adcom) we redlined the amp and it was clipping madly into the LaScala and the LaScala could not be destroyed. I am sure there are many speakers with higher rated power handling than the LaScala that would have almost instantly blown.

[RFK]

That is true. They definately have yet to invent a single driver that could properly produce a 20-20khz bandwidth.

[michael hurd]

A one way driver as you suggest would not work. To produce low frequencies, with low distortion, you need a large piston diameter. The trouble is that as the frequencies increase, the wavelengths become comparable to the dimentions of the driver, then the sound becomes concentrated, and begins to beam. In other words, the polar response changes. Add to that fact that mass begins to work agains you, then you start to see why we have multiple drivers in a single enclosure.

[doudou]

that why some people says the head phone is the better way to listen to stereo tracks:

one unique way that reproduce sounds from 20 to 20000Hz, the room is perfectly known (it s your ears), so no room cancelation or resonance, a very low distortion because it don t need a lot of power...

last but not least:no neighborgs(neighbourgs/neighbors i don t really know how this word should be writen, someone can help me?) or wife screaming when we listen to the music too loud

only advantages