3 channel question

[Tom Mobley]

Holy cow. skip this thread for a day and look what happens.

Steve, I'll PM you about the box. Sorry about the delay, I've been a little snowed under here with stuff.

Do you want to run a Y cable from the mini-box to the two SET, or would

you rather have two RCA out? Somehow the Y cable deal is not very

appealing, seems better to make two outputs.

[Tom Mobley]

Incidentally, I was using an updated and modded HF-81 to run a three

speaker array. I ran the Khorns off the 8 Ohm taps as normal,

hooked up my center Belle to the + terminals of the 4 Ohm

outputs. Seemed a little strange to me, but worked OK and got me

by till I built a minibox. Reallly good to have the extra

flexibility of the center channel volume pot though. I found that

often different CDs wanted different level on the center.

Probably something about the way the individual engineer nixed down the

tracks.

My Khorns are 23 ft apart, the chair is 12 feet across right in the

middle. I usually noticed the center hole effect until I finished

and installed the center Belle. right now i have a very

minimalist deal set up: Sony 333ES player feeds L & R input

of the minibox, three outs go to the three inputs of a TEAC AL700

Tripath amp. A little crude maybe but very effective.

System just kicks butt on any decently recorded material.

Don't want to be putting any poorly recoded stuff in there

though. It's faults will be displayed prominently.

[steve]

Tom,

the way I understand it is I only need one output from the minibox, and do not need the "Y" at all. I can rewire the SET to be a monoblock, with a change of 2 wires per the manufacturer..so if the minibox takes the L+R inputs (2 inputs) and mixes them down to one output (with a volume control in there somewhere) that's all I need. I only need one RCA from the mini to the SET..the inputs on the SET are wired in parallel, and either will work when i rewire the SET for mono.

Please email/PM me with price. Thanx again!

Steve

EDIT..the SET is re wired with the speaker outputs..when I'm done I use the + from one side and the - from the other..

[Istari]

You may want double RCAs for the output. You use one for the center channel and one for a subwoofer. Thats how i have my PWK 3 channel set up.

[Al Klappenberger]

Shawn,

Take some time to really think about all this. You will eventually see what's going on with the bridged speaker and the inverted channel. It IS L+R. To explain it any better would require a lot iof illustrations and time. I'm sure you will see it in your own good time without all that.

Al K.

[sfogg]

Al,

"Take some time to really think about all this. You will eventually

see what's going on with the bridged speaker and the inverted channel.

It IS L+R."

Klipsch's setup over the 4 ohm terminals is *different* then the one

you suggested. In no way shape or form is his L+R. It is L-R...

period...end of story. Re-read his paper if you doubt it.

The config. you posted requires a pre-amp with balanced outputs that

inverts phase on the second signal connection on the output. Not

everyone will have that. And further not all pre-amps with XLR

outputs actually put an inverted copy of the signal on the second

signal leg.

Shawn

[Al Klappenberger]

Shawn,

I see that I will have to write up the detailed explanation of why the bridged speaker IS L+R. I don't have time this moring, but I'll do it later this evening.

I find it very hard to believe the a class outfit like McIntosh would provide ballanced connections through XLR connectors that were not truely balanced. I don't have the schematics on the MC275 or C2200 though to prove it though. "Ballanced", of course, means equal amplitued signals of opposite phase one each wire with respect to the third common (ground) wire. The normal implementation is to use active inverters stages to generate an inverted signals of equal amplitudes to provide the balanced outputs. Older implementations use a floating transformer winding. There is simply no good reason to use XLR connectors otherwise. This would be no different then the standard RCA connections (unbalanced).

Al K>

[3dzapper]

From Dope From Hope Vol14 #4:

Speaking of improved localization: "This is because the center speaker is fed a mono left plus right channel mix. It closely recreates what would have been picked up by a soloist mike at center stage".

Hmmm,,,,, I just noticed that Volume 11 #3, the original treatise on derived center channels, is missing from the DFH download we got from Justin.-(

Rick

[sfogg]

"I see that I will have to write up the detailed explanation of why the bridged speaker IS L+R."

No you don't. Your setup is L+R... the one I have been talking about in just about every post on this topic is not your setup. That is why I have continually said Klipsch or PWKs method... not Klappenberger's method.

I have been talking about the setup from Klipsch in the paper you referenced. You can try to explain how that is L+R but you are going to fail to be able to. Klipsch's setup over the 4 ohm connection of an amp is L-R. There is no phase inversion on one input to the amp and no reversal of phase to the speaker. What Klipsch recommended (and in fact said is better then the mini-box in that paper) is from running a center speaker off the difference signal. That doesn't make sense. I don't know how I can make that any more clear.

"I find it very hard to believe the a class outfit like McIntosh would provide ballanced connections through XLR connectors that were not truely balanced."

I have no idea if McIntosh does this or not. I simply said not everyone provides a phase inverted duplicate on the second leg of the XLR. That is correct... not everyone does. Read reviews where this is tested and you will find that statement to be correct.

"The normal implementation is to use active inverters stages to generate an inverted signals of equal amplitudes to provide the balanced outputs. Older implementations use a floating transformer winding. There is simply no good reason to use XLR connectors otherwise. This would be no different then the standard RCA connections (unbalanced)."

Actually that isn't correct... though everyone thinks it is.

The source does not need to pass a phase inverted signal over the other leg to get the benefits of common mode noise rejection (CMNR). All that needs to happen is that both legs of the signal be balanced.... IOW have the same impedance.

If you have that and the receiving end has a normal balanced input (subtracts the second leg from the first leg) you still gain the benefits of CMNR over that interconnect even if the signal is only passed on one leg.

Before you argue it... think about what happens in that situation. The signal is running on a single lead... it hits the balanced input on the amp (or next device down the chain). The input in effect substracts the second leg from the first leg. Since there was no signal on the second lead basically nothing changes on the signal and it has made the connection into the amp.

Now consider what happens to noise that gets into the interconnect. If both signal leads are balanced (IOW impedance matched) the noise will couple into both leads equally and be in phase between both leads. That noise hits the balanced input of the amp... what happens..... the second leg is subtracted from the first leg... the noise cancels while the signal still got through per above.

Shawn

[sfogg]

Rick,

"Speaking of improved localization: "This is because the center speaker is fed a mono left plus right channel mix. It closely recreates what would have been picked up by a soloist mike at center stage"."

Which method is that refering to in the paper you quoted from. The mini-box is L+R as I have said.

"I just noticed that Volume 11 #3, the original treatise on derived center channels, is missing from the DFH download we got from Justin.-("

The AES paper Al referenced is in there.

In that paper the setup from Klipsch bridging a center speaker across the four ohm taps of an amplifier is L-R.

Anything in phase/equal amplitude (IOW central mono material) between both channels appears equally at both four ohm taps.. as such no signal flows through the center speaker. Anything out of phase between both taps allows signal flow through the center speaker because between the four ohm taps there is a voltage differential. (One is positive the other is negative WRT the same signal)

Shawn

[Al Klappenberger]

Shawn,

OH! I thought you were talking about the scheme I proposed! Yes, the scheme shown in figure 7 of the paper is L-R as shown. It does NOT show any phasing beween the two channels leaving you to assume there is no phase inversion anywhere. That phasing wil determine if it is L+R or L-R. It is NOT a better scheme than the minibox method becasue it has no provision to adjust the center level. It also does not provide a good load to the amp. My scheme doesn't either though. I only used that scheme until I got my hands on a mono amp. I quickly reverted to the "minibox" method. But like I said earlier, L-R actually DOES provide some activity through the center on stereo programs that have wide ambience. It just fails totally on a true "mono" program. L+R acts like you would want it to on mono programs. I have even seen L-R used for a rear channel.

Al K.

[sfogg]

Al,

"It is NOT a better scheme than the minibox method becasue it has no provision to adjust the center level. It also does not provide a good load to the amp."

I agree, but in that paper Klipsch claimed it was the better method. It is a wacky configuration IMO.

"But like I said earlier, L-R actually DOES provide some activity through the center on stereo programs that have wide ambience."

Of course it will... but like I said earlier it doesn't make any sense to put that in a center speaker. If you listen to how L-R material images in two channel (again using the phase tests I have talked about) it is anywhere but front and center.

So why would you then put the information to a speaker located front and center and have conflicting imaging queues between L/R and Center?

"I have even seen L-R used for a rear channel."

Again... that basic setup is the 'Hafler Matrix.'

The L-R is diffuse and 'surrounding' (in 2 channel listening) is a fundamental basic concept of just about every matrix encoding/decoding method out there. That is where/how it is used very very very very very very very often. Read some of the papers I have referenced. Using L-R for rear makes sense... as it keeps a mix 2 channel compatible. The imaging of L-R is diffuse and 'surrounding' so using that for surround speaker(s) is logical.

Shawn

[sfogg]

" is a fundamental basic concept of just about every matrix encoding/decoding method out there."

For a 'lighter' read then the earlier links I posted for an explaination of the basics of matrix encode/decoding read the section on 'Dolby Stereo' here:

http://www.hometheaterhifi.com/volume_8_1/dolby-prologic2-3-2001.html

Shawn

[3dzapper]

"Which method is that refering to in the paper you quoted from. The mini-box is L+R as I have said."

The paper recommends and shows the schematics for both the low level and high level mini box..

Below is my low level box meant for an integrated center channel amp. A simple resistive summing circuit.

Rick

[sfogg]

Rick,

"The paper recommends and shows the schematics for both the low level and high level mini box.."

If the speaker level schematic is the one where he uses resistors to drop the level back down to line level to connect to a third amp channel that too is L+R.

"A simple resistive summing circuit."

Exactly, it is just a passive mixer.

Shawn

[Al Klappenberger]

Shawn,

I haven't bothered to look into the verious surround encoding schemes. I think I remember seeing L-R someplace as being part of the old Pro_logic method. I dittled with L-R for a little while as a center. It definitly didn't do what I wanted it to do and I dumped it real quick. I just took delivery of a nice new Cornwall and wanted to do something with it until I get the right stuff to run it properly!

The resistor box Rick posted is the same thing again as with just 2 resistors and a pot. It's an attempt to set the levels right on each channel without gain controls. I really don't like that idea. Every setup is different and requires level controls on each channel to do it right. The amps I am using have gain controls on each channel, so I just use those. It simpliies the minibox to just 2 or 3 resistors. I use the 3rd just to lower the level and the impedance for the center amp input.

Al K.

[DoubleJ]

I've been waiting for a long time for this question to be discussed but now I'm not so sure if it was worth the wait. I thought I had a handle on how to pull off this but now I have no clue what or how to do this.

Here is what I have..

1. Pr of K-Horns

2. 3 Cornwalls

3. Pr of Belles

4. Scott 299c

5. Dynaco ST-70

6. Teac AL-700

7. A DRBill home made Blackbox

8. JM-Peach

Room size will roughly be 25.5 X 19.5.

Can a 3 channel set-up be configured from the items listed above?

Please Advise

The how question will come later down the road.

Thanks

JJ

[PrestonTom]

Double J ,

It looks like you have all the pieces.

A pair of amps (a stereo amp or a pair of monos) and third amp (either a mono or one channel of a stereo amp)

A pair of K-Horns and a Belle for a center. Or use the 3 Cornwalls

A passive summing box (cheap and easily constructed).

A nice big room

I hope the above discussion has not dissuaded you. It really is quite simple to set up. After this, a few evenings for tweaking levels etc.

Give it a try......

Good Luck,

-Tom

[Al Klappenberger]

JJ,

Sure! Why not use the odd Cornwall as a center between a set of Khorns. This is the a scheme PWK used for a while. The three Cornwalls by themselves would do it too.

Al K.

[Parrot]

Discrete three-channel recordings played through three speakers are vastly superior to playing two-channel recordings with a fake center speaker. Making a comparison would be laughable. Just thinking of it makes me LOL!

There are really only two minor problems: 1) Three channel recordings are limited to Mercury Living Presence and RCA Living Stereo SACDs, about 50 titles at present. 2) These 50 titles are all classical (a problem if you prefer other categories).

I've toyed with a fake center a few years ago, with a fake signal for a Heresy or a Belle Klipsch, and found the results satisfactory when not sitting near the sweet spot, but of no value when sitting in the sweet spot.

For the last couple of years a fake center signal hasn't been necessary anyway because there is no hole in the middle in my listening room. But boy do I love a center channel in multichannel recordings.