Does the RF-7 defy physics??

[Klipschguy]

I'm curious. How does the RF-7 achieve 102dB 1m/1w with an F3 of 32Hz using a bass reflex cabinet loaded with dual 10's? Neither Cornwalls nor VOT's can do it by utilizing a 15" driver. Isn't there a trade off between size, efficiency and low frequency extension?

I'm actually considering taking the plunge on a pair of these.

Andy

[meuge]

Surface area of a single 15" is ~176 sq. in

Surface area of 2 10s is ~157 sq. in.

So it can't be the surface area of the drivers.

However, we don't know the mass of the drivers, and the differences in the voice coil construction, which could be very different in the ole' Cornwalls vs. the RF-7s.

But I have a sneaking suspicion that the lion's share of the responsibility for such a high sensitivity lies with the horn.

[D-MAN]

The mouth of a Khorn (for instance) is 4 sq. feet (approx) or 576 sq. in. versus the overall "acreage" of 2x10" drivers or 157 sq. in.

Now which one moves the most air? we are talking area.

How much extension is needed to have 2x10" drivers achieve a usable acoustic watt of output? Klipshorn needs .16 inches of extension to accomplish that (118db SPL)- Dope from Hope quote:

The sensitivity rating is very questionable - need frequencies measured, power applied and distance at which it was measured.

See any of those? Probably not.

Also, remember that the more extension, the higher the distortion (IM and TIM).

If you've made it this far without getting PO'd thinking that I'm bad-mouthing your choice of speakers, I'm not. Everyone has the right to make their own choices based on there own needs and requirements, and SHOULD feel good about it. But that MAY fly in the face of physics!

These are physical laws that are immutable. Consider that, too.

DM

[DrWho]

Well the klipsch specs state:

102dB @ 2.83 volts/1 meter

- http://www.klipsch.com/product/product.aspx?cid=362&s=specs

The frequency response is also rated at +-3dB so at worst the sensitivity of the system would be at 99dB at the lower end of the spectrum (like 32Hz). Throughout the midrange, lower mids, upper lows I have no doubt that it's pushing right around 100dB sensitivity no problem...

I like to think about two aspects of speakers, the electrical efficiency and the mechanical efficiency. EVERY 10" driver is going to move the same distance to provide the same SPL. Now how many watts it takes is a completely different story. The RF-7 is using very light drivers which means it is essentially more efficient, thus allowing for less watts to produce the same SPL. But let me stress that the drivers are still moving the same distance as any other 10" driver so the frequency modulation distortion is going to be the same. Having two 10" drivers will reduce the cone excursion in half. With the two drivers, each one is running with about 96dB sensitivity which is by no means an unreasonable number for a 10" driver (I can think of many speakers that meet this).

At the same time, the RF-7 uses a really powerful motor which allows for a good deal of linear cone excursion...giving you that deep extension into the 30's (and doing it loud). I think many people notice a huge increase in sound quality when a subwoofer is introduced after setting the mains to small...which relieves the mains of the LF information, thus reducing the majority of the cone movement and thus lowering distortion.

so no, the RF-7 isn't defying physics (klipsch is really good about posting real specs).

[D-MAN]

Direct radiators produce a much higher level of distortion than horns do for the same SPL, they also exhibit dynamic compression at all frequencies due to the inefficiency of moving a cone diaphragm against the atmosphere (i.e., inefficient coupling).

DM

[Dylanl]

I asked Klipsch Tech. support that same question long ago and never got a satisfactory answer. That reading is hard to believed.

[Klipschguy]

I agree, a 10" driver can do 96dB1m/1w - just look at musical instrument drivers. They may be 96 dB efficient, but they have a free air resonance of about 65Hz, not 35Hz.

The physics is tough to explain on these. I am, however, eager to hear a pair.

[Hardhead]

----------------

On 7/26/2005 7:32:30 PM Dylanl wrote:

I asked Klipsch Tech. support that same question long ago and never got a satisfactory answer. That reading is hard to believed.

----------------

Actually, it's easy to believe if you want to badly enough. When you get tired of believing it, though, you can click your ruby slippers together 3 times and you'll be back in Kansas where the laws of physics apply.

[3dzapper]

You are neglecting the output from the excited air column in the port from the equation. At the tuned frequency there is very little cone movement.

Rick

[DrWho]

Well here are a few drivers I found after a very quick search. The sensitivities for these drivers are a bit on the low side (closer to 93dB average), but when compared to the RF-7 woofers I don't think it'd be that hard to imagine another 3dB of output considering the robust motors and superlight cones...

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=290-362 (SPL: 96dB, Fs: 40Hz)

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=295-550 (92dB, 30Hz)

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=297-580 (92dB, 28Hz)

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=290-426 (95dB, 46Hz)

One thing I wanted to note about the eminence driver is that they are after far more power handling capabilities instead of low frequency extension. A little looser on the suspension and that 40Hz mark would easily reach down into the 30's with the sacrifice of max SPL (but who needs 130dB at home anyway?)

I just wanted to comment that horns are not always lower distortion when compared against a similar performing direct radiator system, but now we get into the realm of which compromises are acceptable. Holding volume as the most restrictive aspect of the speaker, horns do better in the higher frequencies whereas they do poorer in the lower frequencies. For example, build yourself a 20Hz bass horn and then using the same volume build yourself a direct radiator system. Of course the horn will be more efficient, but amp power is rather cheap and the direct radiators will be capable of far more output when keeping distortion levels the same. (or you can use less direct radiators and use up a lot less space).

[MrMcGoo]

Folks forget that the spec is for 2.83 volts which would be 2 watts on a four ohm speaker or 1.5 watts on a 6 ohm speaker. S&V rated the RF-7s as a 6 ohm speaker. IMO the woofers are closer to 4 ohms.

Independent tests tend to show that the woofers are not keeping up with the horn as others have speculated. I cut the response of my horns by about 4 db at 8 kHz to get a flat response.

In my room, the RF-7s tend to fall off more than 3 db below 40 Hz. However, the RF-7s have better bass than the vast majority of conventional cone speakers that I have heard.

I would not be concerned about the sensitivity of the RF-7s. However, their impedance curve is cause for concern due to a minimum impedance of 2.8 ohms. The bass response of the RF-7s will not be good without a good amp IMO.

Bill

[Mark Hardy]

Let's see, where to start.

As stated 2.83V is 1W at 8 ohm, but 2W at 4 ohm. It's good that they spec's sensitivity to the applied voltage, BTW.

If 1 10" woofer is 96 dB/2.83V/1m, then two should give 99 dB.

The woofers on the Cornwall or VOT have accordion surrounds which besided increasing Fs considerably compared to a foam or rubber surround, will markedly limit xmax. I dunno, but I am guessing that the speaker in question may be capable of considerably more excursion?

[DrWho]

Two 8 ohm woofers wired in parallel give you a 6dB increase in output due to the fact that twice the current is flowing (impedance drops in half) and you double the surface area (doubles efficiency). So two 8 ohm 96dB drivers wired in parallel give you 102dB for the same voltage (or 99dB with half the voltage if you want to insist on using one watt). But since amps are voltage sources, you're going to get 102dB for the same spot on the dial...and you need to use the 102dB level when trying to match levels to the HF horn.

Running the numbers and ignoring any electrical properties, the RF-7 woofers would need about 50mm excursion P-P (or 25mm one way) in order to achieve 120dB of output...that's about 1" one-way for a single driver. But using two drivers cuts excursion in half for the same output so we're talking only 0.5" in order to reach 120dB levels. This number will of course vary a bit depending on the exact tuning/volume of the cabinet (the 25mm comes from the 40Hz region where excursion is likely to be at its greatest because the port vastly reduces excursion). Again, these numbers are far from being unreal so they're not defying physics.

[djk]

Posted by djk (M) on July 25, 2005 at 00:55:59

A 2 cu ft 2nd order sealed box with an F3 of 30hz can only be 0.18% efficient (84.55dB).

A 2 cu ft 4th order vented box with an F3 of 30hz can only be 0.36% efficient (87.56dB).

A 2 cu ft 6th order vented box with an F3 of 30hz can only be 0.90% efficient (91.54dB).

Classical music at mid hall can reach levels of 0.2W at 30hz.

This requires 112W, 56W, and 22.4W for the above systems.

To keep FMD below an annoying level (3%) would require a 300hz crossover and a nominal frame size of 15", 10", and 8", with a Bl of 29.6, 13, and 6.7

All data calculated per DB Keele and PW Klipsch.

------------------------------------------------------------------------------

The RF7 is at best 3.5 cu ft net. If it was 4 cu ft it could be 90.56dB/W or 93.56dB/2.83V/1M to be -3dB at 30hz (it's a 4th order vented design). This is for half space, you could add 3~4dB for a small sealed room with solid walls (not 1/2" sheetrock).

The 0.2 acoustic watts maximum output is about 105dB (in half space) and requires 6.3mm x-max (probably the linear limit of an RF7 woofer). In reality an RF7 will have much more than the stated 3% FMD because of the 2.2Khz crossover point (FMD is proportional to the square of the bandwidth in octaves, about 7dB higher from what I calculated in the above (based on a 30hz~300hz bandwidth).

I would be more inclined to listen to a pair of RB75 with a real electronic crossover and a pair of good woofers (a real beefy 18 or a pair of 15s per side). I did this many years ago with an 8" two way (Kg2) and JBL pro woofers, and in many aspects the sound was better than Klipschorns. The Kg series tweeter couldn't take it, but the tweeter in the RB75 can.

Note:

DB Keele worked for Klipsch (after EV and before JBL).

http://www.dbkeele.com/papers.htm

At the USPTO:

4,580,655 Defined coverage loudspeaker horn

4,308,932 Loudspeaker horn

4,071,112 Horn loudspeaker

[MrMcGoo]

I have always believed and advocated that the RF-7s be run as small to minimize distortion. The only question is where to set the crossover. Sixty Hz is best as near as I can tell.

It is better to let the subwoofer handle the low bass.

Bill

[TBrennan]

As usual ole DJK knows his business.

Small box, low bass, high efficiency---pick any two. This reminds me of a conversation recently at AA about another "magic" speaker, the Zoo Drooid.

[dragonfyr]

It's rather surprising and a bit sad that so few are aware of who DBKeele, Don Davis, Dick Heyser, Dr. Gene Patronis, Peter DiAntonio, Russ Berger, Sam Berkow, etc., etc., etc., are. Let alone, Olson, Boner, and and ever growing list of notables!

Especially when references to them and their work often is met with derision or dismissal. And even more especially when dismissed by such knowledgeable 'audiophiles' who seem to be aware sufficiently enough only to complain of Bose and Monster!

Oh well...

[DrWho]

djk...

why are those sensitivities so low? Surely you're going to get closer to 100dB of output with 2.83V. I just can't see klipsch overrating by 10 dB, so what gives? I have a feeling that you're talking about "mechanical efficiency" or at least that's the term I use for it...dunno what the proper term is if there is one that describes output versus cone excursion. The way I see it, the RF-7 has overkill motors which give more cone excursion per watt, so though not improving the mechanical nature of the system (thus not reducing distortion or anything like that) the system is still 102dB efficient because it will put out that much SPL for 2.83V

For the record, I've seen the RF-3II woofers playing and they can do a good .5" of excursion one-way and still sound relatively clean (they could probably do more, but my ears were giving out)...I can only imagine the RF-7's are capable of more while even sounding cleaner.

[TBrennan]

Dr Who---It's not a matter of midband efficiency or output but a matter of how low the speaker will go, the f3. Box size and TS specs are involved there, not excursion.

Do a little playing around with a bassbox calculator (they're available free online) and some drivers, you'll see. There's no free lunch. Like I said; small box, high efficiency, deep bass. Pick any two. Also known as Hoffman's Iron Law.

[DrWho]

Ah yes, I've been modelling drivers right and left the past few months...and there are plenty of 10" drivers out there that can go into the 30's with similar cabinet volumes but of course with less sensitivity. Btw, excursion is very much related to the low frequency output of the driver (as it takes more excursion to produce lower frequencies)...though I'm looking at overall output when I mention that because it seems to me that the RF-7 gets its fancy numbers due to the electronics (like a low impedance which greatly increases current flow and other things like that)