Thread: HAFLER CIRCUIT: passive surround sound

Posts: 29
Page: prev 1 2 3

Post by Windsurfer January 30, 2007 (21 of 29)
Beagle said:


WHY BOTHER? Well the Hafler idea is a lot of fun, it is even 'natural' by some definitions.

I believe I read (way back in the 1960s) that if you used the Hafler set-up to play a record miked with a crossed figure of eight pair, as was done with some early HMV recordings, that you would actually be playing back exactly what the mikes were putting on the tape with respect to spatial (hall characteristics) effects. Yep I call that eminently natural!

Post by notherdaysgoneby July 19, 2009 (22 of 29)
Beagle said:

This little note will doubtless make MCH people ask why I don't jump into multichannel itself, and it will make 2-channel people ask why I bothered, but...

I finally gave in to my curiosity (and a bit of encouragement from 'fafnir') and hauled the unused Castle Pembrook speakers downstairs and set them behind the Listening Couch/Divan/Settee.

Then (1) I ran a single wire from the right-positive post on the power-amp to the positive post on the right rear speaker. Then (2) a single wire from its negative post to the negative post on the left rear speaker. Then (3) I completed the circuit with a single wire from the left rear's postive post back to the left-positive post on the power-amp. (Oh yes, an on-off switch stuck anywhere in the circuit; alternatively a volume control.)

In shorthand: R+ to R+, R- to L-, L+ to L+.

That's the famous Hafler Circuit, given to the world by the famous designer David Hafler (and picked up and made into millions of dollars by Mr Dolby). The effect is subtle; it "warms up" the acoustic feel of the room. It isn't time-delay reverb, just an out-of-phase 'difference' between left and right, which itself is neither left nor right.

The cellist wife approves, so the Pembrooks are downstairs to stay.

This is the 4 speaker Hafler, do you know how to do the 5 speaker? I knew someone who had a 5 Speaker with the 5th speaker being the center speaker. It sounded just like 5.1 Dolby almost to me. I have a 5.1 Dolby surround R.C.A setup and it works cool shit but I have another stereo in my bedroom and I was trying to set something up. But I really would like the center speaker.

Post by Fitzcaraldo215 July 19, 2009 (23 of 29)
notherdaysgoneby said:

This is the 4 speaker Hafler, do you know how to do the 5 speaker? I knew someone who had a 5 Speaker with the 5th speaker being the center speaker. It sounded just like 5.1 Dolby almost to me. I have a 5.1 Dolby surround R.C.A setup and it works cool shit but I have another stereo in my bedroom and I was trying to set something up. But I really would like the center speaker.

I too never forgot the experiments I did with the Hafler/Dynaquad circuit in the early '70's. I thought it had potential, but I never followed through with a full time hookup. I think you might need to put a potentiometer in the circuit so that you can get proper front/rear balance. That's the way the original circuit was published ages ago.

I never heard of a 5 channel version, but the only way it makes sense is with a simple summed left/right in the center. So, you wind up with mono summed signal into the center and a mono difference signal into the surround or surrounds. Trying to do it all with just 2 amp channels might also lower the impedance and damping factor your amp "sees". So, it might not be all sonic upside.

This is, of course, a crude, poor man's, simulated multichannel, which, I suppose, might be beter than nothing. Matrixed multichannel simulated from 2 channel sources with today's gear includes all these Hafler circuit tricks plus a whole lot more. And, by not doing it with speaker level signals, it will not create load problems for the amps, which are one per speaker channel these days. Most importantly, with even a simple Mch AVR, you also get to handle discrete hi rez Mch from SACD and other sources that are simply the most true to life recordings anywhere.

The useful thing about this experiment is that you get a glimpse of one of the reasons Mch is superior. Out of phase information in the hall is simply cancelled electrically in stereo, so you never hear it. Discrete Mch preserves this information and plays it back from multiple angles, so some of it can be heard in reproduction, as it is in real life.

Post by Beagle July 21, 2009 (24 of 29)
notherdaysgoneby said: do you know how to do the 5 speaker?
As Fitzcaraldo says, a simple merging of the front speaker signals will equal a mono centre signal: (R+ & L+) to (+), (R- & L-) to (-). You can find numerous variants of the Halfler if you google it, including diagrams which will make more sense than the text here.

Elsewhere in this thread I mention that I inserted an old integrated amp into my Halfler, thus giving me separate control over sound levels in my 'rears'. I admit that I could have added a centre speaker to this set-up, but unlike you, I am quite happy with the sound-image I have with the two 'fronts'. The perceived effect of the 'inaudible' rear speakers is to wrap the stereo image around to something like 180+ degrees, i.e. generally enhancing the stereo effect.

And as Fitzcaraldo suggests, 'true Mch' no doubt does things which my 'Half-mulitichannel' does not do. Again it is my happiness with my current equipment which continues to delay any move to 3-, 4-, 5-, 6- or 7.1-channel. Whether one is rich or poor, the Halfler Circuit is an amazing way to employ surplus components!

Post by Food,Water&Zeppelin August 23, 2009 (25 of 29)
notherdaysgoneby said:

This is the 4 speaker Hafler, do you know how to do the 5 speaker? I knew someone who had a 5 Speaker with the 5th speaker being the center speaker. It sounded just like 5.1 Dolby almost to me. I have a 5.1 Dolby surround R.C.A setup and it works cool shit but I have another stereo in my bedroom and I was trying to set something up. But I really would like the center speaker.

I learned about the halfer set up about a year ago and have been using it ever since but i recently added the final peace of the puzzle a center channel i did this by runing two wires from the right and left positives on the amp to a single speakers negative and positive.

p.s.This set up may be a poor mans set up but its the same set up the high priced amps are using.

Post by emaidel August 24, 2009 (26 of 29)
This is quite an interesting thread. I can recall quite vividly my first experimentation with the Hafler/Dyna circuit, and the mixed opinions of those who listened to it. I was in the service at the time, living in an off-base apartment in Italy, and connected four KLH-6 loudspeakers via the system. I found that some records had some truly amazing effects played back with the four speakers, especially Led Zeppelin's "Whole Lotta Love," while others had little that was worth the effort, and expense, of connecting two additional speakers. At the time it was referred to as early "four-channel stereo," and not "multi-channel." Many of my fellow soldiers who listened to the setup were unimpressed, though I was.

Over the next several years, upon my return to the U.S., "quad" started to develop with the release of the three competing (and incompatible) systems: QS, SQ and CD-4. I went gun-ho quad myself, and especially liked the pseudo-quad effects I was able to obtain from Lafayette Radio's "Composer A" circuitry, but ultimately tired of it, and chose to spend my hard-earned bucks on superior fidelity, rather than surround effects. Getting rid of a Lafayette-branded quad system, and replacing it with the very same Dahlquist DQ-10's that I'm using to this day, and a Luxman amplifier was quite an ear-opener.

Today I would love to have a decent multi-channel system, but have neither the space nor the financial wherewithal for one, being retired and on a fixed income. Then, of course, there's the question: Where in the world would I put FIVE DQ-10's?

Still, I'm sure it's wonderful to have a decent multi-channel system, and if I had one, I'm sure a flood of memories would come back to me.

Post by krisjan August 24, 2009 (27 of 29)
The Hafler circuit always works best with minimally mic'd classical recordings. Those early Mercury's, RCA's and some of the London/Decca recordings really come off as spendid. It really doesn't do much for studio R&R recordings other than provide an occaisionally intersting effect. But with minimally mic'd classical, the soundstage can become quite expansive. Once I'd implemented the Hafler circuit in the early 80's, I never looked back to conventional two-channel. I continued using it in various implementations until a few years ago where I simply inserted a modern AVR to generate the surround effect (primarily using algorithms based on Hafler's idea) from the two-channel signal. I find it very satisfying.

Post by Kunst_des_Fugue August 19, 2012 (28 of 29)
I hope that nobody minds seeing an old thread resurrected, but there is more detailed information available about extracting ambient information in this article from a Swedish audio accessories company. It seems quite sensible to me. The link is http://www.sonicdesign.se/matrix.htm
I hope that someone will benefit from reading it. I have not tried it yet.

Post by Dennis Quad December 11, 2012 (29 of 29)
I actually saw this version of a Hafler circuit ahead of the other ones described already. Imagine that you have four preferently identical loudspeaker systems and a stereo amp that has a common ground between the left and right channels. Place your speakers in a diamond around you: left, front, right and back. Connect the back speaker between the left and right hot terminals of the amp, connect the + lead of the left speaker to the left hot terminal of the amp, connect the + lead of the right speaker to the right hot terminal of the amp. Now, connect both of the - leads from the left and right speaker to the + lead of the front speaker and the - lead of the front speaker to the common ground of the amp. You now have something like the regular matrix except that the sound field is rotated 45 degrees (No more left front, right back etc.)
The configuration where the left back and right back are connected between the hot wires is really only three channel because the backs are carrying the same signal only inverted. What makes them sound so good is that sounds from the left front arrive at the right ear at about the same level as sounds from the right back but at roughly opposite polarity (you have to factor in some time delay for midrange sounds that take a little more time to take the trip around your head which lends the soulfulness that makes this arrangement often sound as good as any).
The diamond arrangement described in my first paragraph was actually seized upon by the original SQ patent except that they would have the front speaker favor signals that the left channel lead the right by 90 degrees and the back visa versa. They later realized that the public had its heart set on placing speakers in the corners of the room which had probably an equal number merits and drawbacks.
If you were to electronically shift the sound field 45 degrees before amplification and then played them through four speakers connected like in my first paragraph but pushed to appropriately to the corners, you could with just a 2-channel amp arrive at something like Peter Scheiber's original Regular Matrix.

Page: prev 1 2 3

Closed