Klipsch 3-Channel 'Back in the Day' - I don't Understand?

[meagain]

This has been bothering me since I began lurking here and unfortunately since the search engine is down - can't look up the answer (if available)......

I understand that PWK designed speakers decades ago to be used as a center for Klipschorns, but looking at the time period this was done, puzzles me. I thought center channel speakers were a relatively new deal. For HT/surround. Forgive my ignorance as I don't know my audio timelines, but say, in the 50's, how was a center channel accomplished?

Assuming recordings were made for 2 speaker stereo, how would a center have been physically run, and what music would come out of it? Were all 3 speakers pushing the same music? Or were certain things somehow filtered separately to the center?

I totally don't understand how a center could be used for music TODAY, and certainly don't understand how it could have been pulled off Soooo many years ago with the recordings & gear available. I asked my husband about this and he's as puzzled as I. Can someone explain? I'd really appreciate it.

[Frzninvt]

The center channel signal was derived from taking the + from R channel and the + from the L Channel and connecting a center speaker to it, basically.

PWK had a couple of designs for this 3 channel box, one was speaker level, and the other was line level.

The Heresy, Cornwall, La Scala and Belle were all designed to augment the Klipschorn's and were never really intended to be stand alone speakers. Using a derived center channel makes a drastic difference in filling the "hole" in-between two stereo speakers.

I used the same method in my IASCA competition car audio vehicle using a Vifa 3" soft dome midrange for a center and controlled it with an L-Pad.

Nowadays the center channel is a discrete signal and not a derived one when running in the digital domain.

Confused?

[JJkizak]

The center channel might have been inspired by Bell Labs when running tests in a 25 ft X 25 ft room (or was it 30 ft x 30 ft) and did test a center speaker with the K-horns but I can't remember if it was 1935 or 1955. It was stated that with the speakers that far apart that a fill-in was needed. Somebody has to have that article somewhere as I remember reading it in a magazine. Maybe Popular Science or Popular Mechanics. Then again it could have been Stereo Magazine.

JJK

[Ray Garrison]

Hi there.

Think about this for a minute. "Back in the day", when a performance was miked live, the right channel mike would capture sound that was mostly on the right side of the stage, and the left mike would capture sound that was mostly on the left side of the stage. Some of the sound captured by both mikes would be the same - that is, a sound source sitting in the middle of the stage would be captured equally loudly and in phase by both mikes. (The phase relationship between sounds captured on the left and right extremes has a big factor in the reproduction of the soundstage and such, but that's another topic for another day...)

What would happen if you came up with a simple little circuit that was able to extract the signal that was common between both channels (the center of the soundstage) and you sent this signal to an amp that was driving a speaker sitting in the middle between the left and right speaker? It turns out is helps solidify the left to right imaging, and in general provides a more realistic portrayal of the original performance.

Note that this won't work this way for modern multi-miked recordings where there is no center channel info captured in the first place.

An interesting twist on this is taking the *difference* signal between the left and right channels - that is, everything on the left channel that does not appear on the right, and everything on the right channel that does not appear on the left, and doing something with this. It turns out if you delay this difference signal a smidge, and play it back behind you, you get an amazingly realistic portrayal of a 3D soundspace. This was called the "Hafler" circuit, after the guy who designed it. I used to use it all the time back in the day...

[CaptnBob]

Many preamps "back in the day" had a "L+R" mono output (the Marantz 7T and McIntosh MX 110 for instance) you could hook up to a mono amp.

I know at one point Dynaco had "Dynaquad" for rear speakers which was basically taken off the left and right channels' positive terminals. PWK had a system which ran the negative leads from the left and right speakers to the positive terminal of the center speaker and then go from the negative terminal of the center speaker back to the amp. It worked ok, sort of, but some amplifiers took a real dim view of the proceedure.

Some speakers - like the big Universities - had a dual voice coil, so you could hook up two amps to the same driver. (or, if you were so inclined, connect the voice coils in series for 16 ohms, or parallel for 4 ohms.)

[Johnny dB]

Most of the vintage stereo Scott integrated amps, ie. 222, 272, 296, 299... have a center ch. pre-output. Very handy when adding a subwoofer.

[Hardhead]

It's not my intention to get into a shouting match, but the derived 3-channel setup has worked extremely well for every vinyl and CD recording--modern multi-miked and otherwise--that I've played on my setup for the past 33 years.

[Tom Mobley]

I have khorns on a 24 ft wall, the room is 13' wide. I perceived a hole

in the center effect so I built myself a center Belle in matching wood

for the setup. I was (and am) amazed at how much difference it

made. Rock solid imaging with depth and presence. A small

splitter box can be built that sums the left and right. It goes

between the pre-amp and whatever power amps you have. You need at

least 3 channels of power amp one way or another.

I also got usable results hooking the center to the 8-ohm terminals

while the Khorns are hooked to the 16 ohm terminals. (My Khorns

are 1960 and labeled as 16 ohm speakers.) This was on tube amps

that had multiple speaker connections for different impedance speakers.

[Seadog]

It's not my intention to get into a shouting match, but the derived 3-channel setup has worked extremely well for every vinyl and CD recording--modern multi-miked and otherwise--that I've played on my setup for the past 33 years.

I also get excellent results, using the derived center out signal from my Scott 299D to a 99D powering my center Cornwall.

[Hardhead]

As Tom says:

"I was (and am) amazed at how much difference it made. Rock solid imaging with depth and presence. A small splitter box can be built that sums the left and right. It goes between the pre-amp and whatever power amps you have. You need at least 3 channels of power amp one way or another."

[meagain]

OK - Maybe I'm slow, but I'm still trying to 'get it'. You are basically taking mixes designated for 2 channel/2 speakers, stealing sound from the 2 mains and shooting that to the center to run 2 ch. stereo thru 3 speakers. Yes?

I just wonder what that sounds like because it wasn't designated for that. Still thinking this through. And if it sounds OK, is this only something one does for very widely spaced mains? Or does anyone also do it for a small room setting? I'm just curious about this.

[Seadog]

You are not actually "stealing" sound from the R and L speakers. Those still sound exactly the same. The line level signal from the R and L is used to create a line level derived center signal which is amplified and sent to the center speaker via a separate amp.

The effect improves imaging and enlarges the listening sweet spot in the room (at least that is what it does for my system).

My Scott 299D actually has amplified center channel out terminals as well as the line level. I tried running the center channel Cornwall using the speaker terminals, but it took too much power away from the mains (also Cornwalls). The low end seemed to suffer; not as much "punch" in the bass. The separate center amp cured that problem and it sounds great. I run the center about 5 - 6 dB less than the mains.

[WMcD]

Its too bad the search engine is so messed up. Over the years I've posted all of PWKs articles and those reporting the tests by others he relies upon. Essentially they show:

Bell Labs in the 1930s experimented with "stereo". That term means "solid". Bell Labs found that two channel recording with L and R placement of microphones and speakers can create a decent image. But 3 channel recordings with three microphones and three speakers were very much better. Call this 3-3-3.

There were tests with microphone mixing and recording and playback. E.G. take three mikes (the third being center) and mix it down to two tracks with the Center mike being equal in the left and right. Then at the play back, let the L and R speakers get their channel playback, but then have a a mixed L plus R in the center. Call this 3-2-3. This worked well enough and you didn't have to have a three channel recording.

In those days getting even two channels of recording, transmission, or storage, was a big problem, alone.

Its not entirely clear to me what they were doing but the 1940 production of Fantasia had five or six channels of sound with surround sound. The tracks were recorded on a separate film camera and played back over multiple speakers. The play back "projector" with the sound tracks had to be synchonized with the picture. A very big deal. It was only available on the coasts, but there was a travelling show too. Equipment would be shipped around in trucks.

Eventually, stereo sound came to the fore in consumer audio in the '60s with two track tape recorders, multiplex FM, and microgroove stereo vinyl. Here at least we had two channels of storage and transmission. Call that 2-2-2.

PWK was a pioneer in the consumer market, and long time fan of Bell Labs. Note that the K-Horn was already established before consumer stereo.

His tests just confirmed what Bell Labs had found before. Adding a center channel speaker did wonders even if it was just fed with L plus R (which was there in the recording of discrete L and R). It was very much recognized by the manufacturers of high end equipment. This is why the classic McIntosh and Scotts adopted it.

In the number game, we don't know what is going on at the recording end, it could be described as ?-2-3.

Of course there was a problem for the average consumer, I infer. Two channel stereo was expensive enough and some recordings were just hoaky demonstrations of ping pong and trains going by.

It was difficult to convince most consumers that best results could be had with yet a third speaker and third amp and the special preamp, or PWKs home made mixer box. What? another expensive gadget?

For a sign of the times, please note that the early Beatles recordings were mono. Two channel was still too experimental.

In my view, the good effects of the center channel got swamped out by other audio developements over the years.

There were 45 rpm singles (mostly mono), record changers, 8 tracks, cassettes, and Quad (4 channel sound). These had their own merits (plus and minus) but were a distraction from the basic issue.

One return to the idea of center channel came with the home theater revolution and Ray Dolby's work of surround sound in theater. Ray had done good work in many areas.including cutting his teeth in video recording, and later a compression and expansion noise reduction systems which in many forms were known as Dolby A, B, and C.

There was much overlap but essentially, theaters were using surround and a center channel based only on a two track recording. Star Wars had it. VHS tape units started using a form of FM modulation on the audio tracks. Very good sound for the day. Food for tweeking.

So early home theater was sold most as having surround sound. This used Left minus Right fed to side speakers with a 10 millisecond delay. But the rigs (at least some) also had the holy grail of our theme. A center channel which was just L plus R. Now things were returning to the earliest credos of Bell Labs.

People didn't know diddly about center channel; they wanted surround sound. and the 0.1 of 5.1 which was a subwoofer. But the 5.0 was L, CENTER, R and two surround speakers (which were only getting mono L minus R). These all came from just two channel recordings on the VHS track, and the two channel on the theater film!

Things broadened out. now there are multi channel recordings everywhere.

Still, mixing L and R to put into a center channel speaker works wonders on many old or new recordings. There is good "information" in old recordings which pops out with the mixing and placement of the center channel play back.

In my view, this is not just a wall of sound "filler". Rather, it allows extraction of long lost information in the recording.

It is not "new science". Rather, it is old science re-discovered. PWK said that when he demo'd to Bell Labs. He said he was "bringing coals to Newcastle."

Best,

Gil

[psypathic]

So would it be possible to just hook up a third RF-7 to the L+&R+ and get all the that was recocorded and send it to a third chanel in conjuction with the left and right(or is that where the box comes in)?

[DRBILL]

About every year and a half as best I can figure it. There is absolutely nothing that can be added to this subject!

Or maybe it is like a bedtime story of which the children never tire.

I'm guilty, too. I recently asked a serial number question!

DRBILL

[Hardhead]

meagain--

I agree the derived center channel concept is counterintuative. Adding a mono channel makes stereo sound better? Nevertheless, it works. Somewhere in the Dope From Hope papers the reason it works is explained correctly. It may help you to understand it if you consider that the derived center channel needs to be 6-8 dB's lower than the left and right speaker levels; if the derived center channel is louder than -6 dB's compared to the L and R, it tends to ruin the stereo image.

[JJkizak]

I might add listening to Fox the other day on some football game in 5.1 that the center channel seemed to have a hole in it until the announcers would speak, then their voice was exclusively in the center. Also all of the commercials came from the center channel only.

JJK

[meagain]

Thanks everyone for your help. I think I understand it now. Whew!

I also printed out the Dope from Hope and just found some entries on this so will be reading it tonight.

[Mark Hardy]

The "Hafler" circuit and "Dynaquad" are one in the same. That'd be the late David Hafler, founder of Dynaco (and later Hafler). I don't think there was any delay in the Dynaquad box, tho'... strictly passive IIRC.