RF-7: nice image, natural sound

[leok]

Since last Thanksgiving, when I purchased a pair of RF7s, I've been tinkering with both the speakers and amplification to achieve a particular goal for which I believe the RF7s are well suited. That goal is the reproduction of classical instruments, especially small, chamber ensembles, in a way that the instruments sound natural and the soundstage is well defined and stable, and at amplitudes that are similar, in my living room, to amplitudes I experience in live performances.

Having finaly achieved that goal I want to thank members of this BB for their generous help, and encouragement. This is a difficult and subjective task that requires exchange of some fact and mostly opinion. Since opinion is easily criticized, I consider it particularly valuable.

Finally, as a result of this BB, I am becoming aware of stereo sound capabilities that are well beyond my initial goals.

Following is a list of what I found most critical, generally in order of impact, to the success of the project. Certainly, for others, the list might be different.

1) Speakers: The RF7 is a remarkable speaker. For $2200/pair, they are beyond remarkable. I don't know of a better deal today, in any product. They are capable out of the box, and respond very well to nominal modifications.

2) Amplifier: I listen to these speakers at a fraction of a Watt. I doubt I have ever put more than 3.5 Watt peaks into them. Low distortion at very low power is very important for both the natural and the imaging parts of the objective. Two types of amp: tube amplifiers and a custom switching amp based on the Tripath TA1101B have been succesful. Both of these amps avoid the nonlinearity and crossover distortion that plagues transistor designs at low levels. Elimination of global negative feedback provides a tremendous improvement in the upper harmonics. Many amps are available with this "feature" (or lack of it).

3) Speaker Modifications: I applied Dynamat "Extreme" to the horns and upgraded the crossover components. Replacement crossover components are Hovland Musicaps in series with the horn and Solen air core inductor in series with the cones. Also, I substitued Solen metalized polyprop. caps in parallel with the cones and in the RLC circuit that parallels the horn. The Solen caps, in the parallel circuits, which I changed just recently, made these already smooth speakers much smoother. I suspect that, at my low listening levels, the reduction in dielectric absorbtion, provided by the Solens, addresses something similar to crossover distortion produced by less expensive stock caps. If you upgrade the crossovers, don't skip the parallel circuits.

4) Speaker Connection: There was a link, somewhere on this BB (I can't find it now) to an article that recommended, for improved imaging, NOT bi-wiring, for tube amps (that also applies to my Tripath amp). The logic, from an engineering viewpoint was reasonable, so I tried it. My speakers have always been bi-wired. So I re-installed the metal jumpers and tried a single Monster MCX-IS set (Litz bundles wrapped around a "flux tube") to each speaker. The imaging improved. Instruments now clearly stay in position, but there is a feeling of air, or space around each one .. they have become so distinct. Also, the overall sound of the RF7s became more crisp .. just enough of that "dryness" I find so appealing in my Forte IIs.

Beyond my initial goals: Well, there is the elimination of global negative feedback that I mentioned in the Amplifier paragraph. Next, I am in search of an opportunity to try a high quality Single Ended 2A3 or 300B amplifier. I need to hear Cornwalls (with Alnico magnets) driven by a 2A3 SET because I've heard so much about that setup. Finally, I'd love to hear a set original Klipsch corner horns with an Al Klappinger crossover, driven by a very high quality tube amp.

Meanwhile, the music I hear in my living room has the magic I remember from when I first heard "high fidelity" tube sound in 1960. But now it is much more natural sounding, and it's stereo.

[leok]

mdeneen .. your note reminded me of an additional factor.

5) Room Treatment: To reduce reflections from my rather low ceilings, I have 1ft x 1ft panels of acoustic absorbing material protruding like square baseball caps from the top of the RF7 cabinets. I have also absorbed reflections from a brick fireplace adjacent to the right horn. Except for casual listening, I always remove the grills.

Amp distortion .. I don't know. I considered taking a few distortion analyzer measurements, but considering the very low level of the amplifier output signal I anticipated a measurement quagmire that could quickly become larger than the set of audio questions the measurements are to shed light upon. Being told that I'm much more tolerant of even than odd harmonic distortion has added to my uncertainty.

Perhaps you could suggest a measurement setup that you consider effective, and I will attempt to perform the measurement.

For the benefit of others reading this post: what mdeneen and I are refering to here is the problem of measuring audio excellence. Actually, I suspect both of us know we're not really talking about that, but casual readers might not.

mdeneen, I'll try to contact you by email so we can work out some of the protocol.

[firejim]

leok...It is nice to see some positive attributes on the RF7's,especially achiveing the sound you are after.

I have had a pair of Cornwall's for aproximately 18 years that were in a flood and am considering replaceing them with the RF7's,due to water damage.The Cornwall's were never connected to tube amps,but were connected to very high end solid state electronics...they sounded great!

I have had a couple of listening sessions with the RF7's and am extremely impressed with the imageing,tonal balance,wide sound stage,and inner detail that the 7's offer.They were not connected to tube amplification.The RF7's seem to offer better imageing and finer inner detail that my Cornwall's did not produce.

I wish I could have a side by side listening test that would help my decision on whether to restore the Cornwall's or to go with the RF7's.I am leaning towards the RF7's especially after reading post's like yours.

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

--

[gamer]

Havent had a chance to hear the 7s yet but not due to lack of trying. Only place round hear that sells Klipsch is Good Guys and all they have set up are the 5s. I bet they are awesome!! Whats with the dynamat on the horn? Please explain?

[leok]

Several items here .. I'll go in reverse order:

gamer, The dynamat is to damp horn resonance. I believe it is effective. Instrument timbres are cleaner, and more uniform from note to note. It seems particularly effective with woodwinds. I cut pieces to fit all of the horn surfaces.

firejim, I'm glad the post helped. I think the RF7s, are overlooked by many classical listeners, because they haven't heard them driven properly. This is generally true of all horn speakers, but I think the RF7 is quite an achievement.

mdeneen,I'll pursue the measurements. What you describe is where I'm starting to bog down in my SET investigation, and investigation into improvements in general. I'm beginning to suspect I'm running out of low hanging fruit. I'm not going to spend $2K and 2 months of building without listening, and 2A3 SETs aren't that easy to just borrow for a while.

I do have access to distortion specs for the Tripath TA1101B. They look very tube-like. Below 5 Watts, THD+N is under .03%. Above 5 Watts the distortion goes through the roof. See Tripath.com. Go to the site map and look under data sheets. TA1101B is the first one, I think. (If you go to the TA1101B product page, the data sheet there is for an eval. board, which is also interesting). My circuit is much like the data sheet version except I have the input DC coupled. Also, I use nicer components that they do on their demo.

[gamer]

Thanks for the info Ill have to give that a try!

[TroyTN]

For best stereo imaging how do you have your RF-7's? Pointing directly at the listening position? I get better separation pointing them just to the left and right of the listening position. If I point directly at me they do not seem as dynamic. Then again if I had a larger room it might help. My room is only 14' feet wide. RF-7's are 9' apart and I sit 12 feet from them.

The manula states to face them forward and slightly tune them inward..

Troy

[bwm]

Just as a suggestion, if you haven't already done so, buy a CD test/setup/configuraton disk from Reference Recordings or Chesky or Stereophile.

My XLO Test/Burn-In Cd from Reference Recordings has tracks to test your imaging, phase, and absolute phase. It is truly revealing. It let's you pinpoint the absolute best speaker positions and toe-in for accurate imaging. I was never able to get good imaging until I used this disk.

Changing position of a speaker only a fraction of an inch can make a big difference. You simply have to keep adjusting until it's right, according to the imaging track...there's no textbook answer. And if you move the speaker for housecleaning, you can easily re-image by listening to the test track.

quote:

---

Originally posted by TroyTN:

For best stereo imaging how do you have your RF-7's? Pointing directly at the listening position? I get better separation pointing them just to the left and right of the listening position. If I point directly at me they do not seem as dynamic. Then again if I had a larger room it might help. My room is only 14' feet wide. RF-7's are 9' apart and I sit 12 feet from them.

The manula states to face them forward and slightly tune them inward..

Troy

---

[leok]

TroyTN,

I have the speakers pointed directly at the listening position. I think it minimizes contributions from the room and lets me hear the on-axis characteristic of the speaker. This can also be the most difficult, because it is the most revealing of imperfections. But I think when it works, the image is what the recording engineer intended.

BWM, The XLO Test/Burn-In Cd is a great idea. I think I'll give it a try. I get real tired of wondering, when something doesn't seem right, if it's me or it. Thanks.