Cornwall 180deg out of phase with Khorn

[KhornKerry]

A few months back I was testing the low frequency phase relationship of my home entertainment system to make sure that I didn't have any of my 6 speaker wires accidentally reversed. I was using 2 identical microphones and a dual-trace oscilloscope. I applied the same sine waves to the speakers in the frequency range of 50 to 200 Hz. I first placed both microphones next to each other and at the side of one of the Khorns bass bin, and used lissajous pattern to ensure that both microphones and scope channels gave the correct response ( a 45 deg line increasing from left to right / ). My system has 4 Khorns and 2 Cornwalls used as the center channel. All 4 of the Khorns showed to be in phase with each other but the Cornwalls were 180 deg. out compared to the Khorns. The Cornwalls were in phase with each other. The Khorns are 1984 and 1978 models the Cornwalls are 1981.

I very carefully checked my speaker polarity wiring and everything was correct. I ended up reversing the speaker wires to the Cornwalls so all speakers would be in phase.

I was wondering if this was done intentionally by Paul for the 3 speaker wiring as seen in the 1975 sales brochure showing 2 Khorns and a Cornwall center, or not important back in the days of just stereo.

I would also like to know if the La Scala, Belle and Heresy speakers are our of phase with respect to the Klipschorn.

[Klewless]

I expect what you are seeing is NOT that the Cornwalls are out of phase BUT that the five foot length of the Khorn creates the effect of phase difference. The upper end however does not exhibit this behavior quite as drastically since those horns are MUCH closer to each other in length. It is possible that your have now put the midrange more out of phase while fixing the bass.

What is one to to do in order to get to hi-fi heaven!!!!

This might even be the case for having a center that does not go down quite as low as the mains.

[mark1101]

For what it's worth I have experienced the same issue with my Las and corns (see models/years below). When I got my Scott 299 recently I connected a Las and Corn in parallel on each channel. I started fooling with the phase reversal switch and got weird sounding results that really bothered me. It sounded like an improvement and a degradation simultaneously (if that's possible).

I disconnected the speakers and ran 1 pair at a time. The Las were perfect and sounded worse with phase reversal. The corns sounded better with the phase reversed, not as good in phase.

Well, that led me to tear down the corns and check every connection possible. There was nothing wrong.

Here's how I fixed it. The problem went away after repositioning the corns in the room. I was very surprised about this, but I only needed to move one of them about 18 inches. I have a large room and have a las and corn on each side. I moved the corns a little further apart. That's all.

I have since obtained an additonal Scott amp and now run the corns off it (1 external preamp into 2 scott amps).

What puzzles me about your situation is that I assume you are actually seeing the sine waves out of phase as monitored by the mikes at each driver. Is that right? Are the mikes the same distance from each driver?

[WMcD]

I'd like to add to the above.

All the wiring diagrams I've seen show the horn woofers to have the normal polarity, as does the Cornwall.

However, with the bass horn, you do have a propagation delay of probably a little bit more than 5 feet.

Therefore, whether a single test tone arrives at the mouth at any phase depends very much on the frequency used. It might be interesting to just use the dual trace scope in two channel mode, if possible. Have one channel on the woofer electrical input and the second on the microphone. As you change the frequency, you should see the phase at the microphone change due to the time delay.

I'm going to guess at 15 inches up the front, 24 inches to the back and 24 inches toward the front for a total of 63 inches path length. A 214 Hz tone has a wavelength of 63 inches, so there would be an in phase condition.

On the other hand, the two first distances total to 42 inches. This means that the woofer is 42 inches "behind" the midrange driver. A full 42 inch wavelength is 350 Hz which is just a tad below where the woofer stops working and the midrange turns on. I believe PWK built in this aspect. As you can see, the bass horn delay at the crossover frequency means the bass and mids are in phase just at that point.

Best

Gil

[KhornKerry]

First let me make it clear that I did not change any wiring in or on the Cornwall or Khorn speakers, all I did was reverse the speaker wires at the Amp. The wall with the Cornwall center which is flanked by Khorns is 16 feet corner to corner.

I guess my question is this, should I reverse the leads going to the Cornwall? With it reversed it seems to have a little more low end bass. The wave length at 40Hz is about 22.5 feet so 5 feet of travel in the Khorn is about 90 degrees. I would like to see the phase relationship of a La scale verses a Cornwall both microphones at the front plane of each speaker.

I also listen to SACD multichannel, the inserts in some of the Sony disk suggest that the best configuration is for all 5 speakers be of the same type. So if the speakers were all the same they would have to be in phase with each other.

[PAR69]

Another option might be to change the phase of the Cornwall woofer by swapping the leads off of the crossover (positive to negative/negative to positve).

Any one see a problem with this?

Paul

[WMcD]

My point is that the midrange driver is 2 feet behind the front face and the bass driver is 5 feet behind the front face. Changing polarity is not going to change that fact and the issues (if any) which arise from it.

So I'd think you're best off leaving everything in the original polarity.

Smile.

Gil

[KhornKerry]

Gill

Does "normal polarity" mean that if you put the + of a 1.5V AAA battery on the + speaker terminal and the - to the - that the woofer cone would move away from the magnet?

Kerry

[WMcD]

Yes, that is the "convention." You should find that the electrical terminals are marked, somehow, to indicate at least a positive terminal. I wouldn't hesitate to use a 6 volt battery on the woofer. I have used even a 12 volt to tickle it, but not long term. Just enough to see which way things are moving.

I'd be happy to write at greater length, maybe with a scanned sketch, about the effect of the 5 foot (that is about a 5 mS time delay). In any event, if any source (driver) is 5 feet farther away than another, there are going to be frequenies where they are effectively 180 degrees (1/2 wavelength) different in phase, and some where they are 360 degrees different (1 wavelength) different in phase.

In the first case this will occur when 5 feet is 0.5, 1.5, 2.5, etc wavelengths, the second case is when 5 feet is 1.0, 2.0, 3.0, etc, wavelengths. And we're only here talking about the nodes and anti-nodes. At the in between frequencies the phase differences are going to shift accordingly.

Now, I'm not saying that reversing the polarity of one driver is not going to have an effect. Rather, all it will do is swap the "first case" above with the "second case" above.

Now just maybe, there is some overall acoustic problem with the bass of the room and set up which makes the overall response of the bass better with the second case and it seems like a possible choice. Fine.

Now the next question is whether to switch just the polarity of the bass driver in the center (by swapping the leads to the bass driver only). That may be okay, but it will mess up the phase relation of the mid and bass at the crossover frequency and the crossover was designed to have the woofer and mid driven phase (same polarity). Now response at the crossover point is not the original . . . and probably poorer.

So, you say, I'll switch the polarity of the entire box at the input to the crossover, to avoid that crossover issue. However, now the center channel overall is out of polarity at all frequencies (all three drivers) with the flanking units.

Is this necessarily bad? Early on, PWK maintained that center channel polarity is not audible; or at least that L - R was not that much different than L + R in the center. With the change in overall polarity you've got -L + -R. Part of PWK's reasoning was that the Lissageau (sp?) pattern displayed by L on the vertical channel and R on the horizonatal channel, showed a random phase relation, pretty much a circular hash. That was most likely from live stereo recordings.

Later, his suggestions for a center channel mixer honored the phase relation. It may be that with the use of mixing boards and the like, the center channel information on a recording (the singer is mixed "mono") really did not work well and would disappear with L - R. Of course that is not to say that -L + -R is not going to work. However, it may well be a lot different than what the recording engineer has set up in his little mixing studio.

Overall I'd say you certainly should play with the settings and find what you like the best. One suspicion is that keeping all the connections "normal" will be the best overall compromise.

Best,

Gil

[g3dahl]

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

On 9/4/2003 4:08:49 PM PAR69 wrote:

Another option might be to change the phase of the Cornwall woofer by swapping the leads off of the crossover (positive to negative/negative to positve).

Any one see a problem with this?

Paul

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

Yes, there would be a problem. Reversing the leads between the crossover and the woofer would cause a response null at the crossover frequency. The relative polarity of the drivers is part of the crossover design, so you can't change it without redesigning the crossover.

Kerry, if it were my system I would reverse the polarity of the Cornwall, as you did. The idea is to connect things so that the listener receives the same part of the waveform from all three woofers. Because of the longer path length in the LS's, this is achieved with the Cornwall (or both LS's) connected in reverse polarity.

Gil is probably right, though, about it not being terribly critical in this application, as PWK wrote. You should feel comfortable making your final decision by ear. The answer might not be particularly clear, because the phase relationships between the mids and tweeters will be all over the place anyway. I would base my decision on the woofers, which are covering the frequency range in which the ear is most sensitive to phase relationships.

Gary Dahl

[LynnOlson]

Don't forget the path-length difference between the speakers in the corners and the center speaker, in addition to the 5 foot bass delay in the corner horns.

In theory, the most accurate correction would be a HT controller that provided an adjustable delay for the center channel between 5 and 8 mSec. (You need a tape measure to set the delay correctly. At sea level, 1 millisecond is 13.8 inches long.) The bass would then be path-length-synchronized for all three channels, and everything could be wired in-phase.

It's true the center mids would then be out of sync by five feet, but then, that's the case for the horns anyway, so the audibility would probably be minor.