Out of phase tests on Khorns

[Kudret]

I have ben complaining about the sounstage I am getting from my Khorns especially when they are in the corners. Following some suggestions from this thread (http://forums.klipsch.com/forums/thread/764355.aspx) I used an AVIA DVD to see if my Khorns were out of phase and it "appears" that they are not. However, if only the tweeters were out of phase, can this be determined by a test tone or is there another way to individually check the phase of individual drivers? The reason I am asking is, when I reversed the connections on one of the tweeters, the speakers still passed the test.

Then I reversed the connections on the tweeter and mid driver on one speaker and repeated the test. I was expecting the speakers to be out of phase, but I was surprised to see that they still passed the test. I also played a few songs and I cannot hear any difference between before and after reversing the drivers on one speaker. What gives?

Kudret

[Al Klappenberger]

Kudret,

I don't think you can hear the phasing on a tweeter. Remember that hook-up phasing represents +-180 degree increments. The wavelength at tweeter frequency is typically only a couple inches ore less. All you need to do is turn your head and the distance change will shift the phase by more than 180 degrees! From what I have seen, the phasing left to right of the woofer and squawker will be audable. The phasing between the drivers within a single speaker also seems to effect the frequency response glitches near the crossover. I'm note sure this is udable either. I think what's imortant is that the same phasing between the same drivers in opposite speakers (both squawkers and both woofers) is what is important.

AL K.

[psg]

It's really strange that your KHorns don't image. Remind me how far they are separated? It might also be room acoustics. I just moved and was disappointed with the imaging in the new place until I noticed I was seated too far away, behind the crosshairs of the two speakers. The speakers are separated only a few inces less than before, a tad under 18 feet. It's almost back 100% when seated at the correct location, and I assume the rest is due to the semi-finished basement. It should be back 100% after I finish the basement next fall. (Should I use suspended ceiling or not?)

[Kudret]

Thanks, Al.

I think what's imortant is that

the same phasing between the same drivers in opposite speakers (both

squawkers and both woofers) is what is important.

I tried the reverse phasing between the same drivers in opposite speakers (both

squawkers and both tweeters). They sound the same as before. Anyway, at least I know that poor soundstage is not caused by some drivers being connected in reverse.

Kudret

[LarryC]

There was a thread heavily involving Meagain that discussed testing for out-of-phase drivers, as opposed to entire systems being out of phase. Someone, mdeneen I think, said that a tone generator (audio oscillator, I believe) could generate frequencies limited to the middle of each driver's range, and that would quickly show which if any pair is out of phase. BTW, I don't agree with AlK that out-of-phase tweeters can't be heard.

I'd suggest that, if you think it's OK with your amps, that you listen only to each woofer pair at the same time, then the MR pair, then the tweeter pair, to see if you get the hole-in-the middle sound that (1) feels like it's literally trying to pull your ears apart and (2) suddenly jumps from one speaker to the other as you walk back and forth. Others likely have good tests, too.

[DrWho]

Try sitting on the floor right at the crossing point of their coverage pattern (basically be able to look straight down both squakers).

[Erland]

May I suggest the test and sampler CD from Stereophile.

Track 2 on the CD has helped me a lot when working on phase, imaging and listeningposition. The track contains examples of a Fender base guitar played first in separate channels, and then out of phase and last in phase. Since the test "singal" contains not only a tone, but a real instrument, it is so much easier to determine when things fall into place.

I am sorry to say I do not have it with me, and will not have it in my possesion for another 8 days, but if you can wait that long, I can send it to you by email.

Regards
Erland

[colterphoto1]

Well, you could use the D cell 'thunk' test on the woofers, but you have to see them for that to work.

Is it possible that something is awry in one of your crossovers? Try running your source test cd directly to the woofer, bypassing the crossover altogether and see how it reads.

Michael

[tigerwoodKhorns]

May I suggest the test and sampler CD from Stereophile.

Track 2 on the CD has helped me a lot when working on phase, imaging and listeningposition. The track contains examples of a Fender base guitar played first in separate channels, and then out of phase and last in phase. Since the test "singal" contains not only a tone, but a real instrument, it is so much easier to determine when things fall into place.

I am sorry to say I do not have it with me, and will not have it in my possesion for another 8 days, but if you can wait that long, I can send it to you by email.

Regards

Erland

Could I get a copy too? I am having a bass issue in a room and want to check phasing of the individual drivers.

Thanks,

Chris

[psg]

It's really strange that your KHorns don't image. Remind me how far they are separated? It might also be room acoustics.

Found it: My room is 17x34 and I have my Khorns along the short wall.

Have you tried putting them on the long wall at various spacings, using a board of some sort as a false second wall? It seems to me that 17' should be enough, but at least you could rule out spacing between speakers if you tested a bunch of them.

[Erland]

I guess I have to be careful not to get the editor of Stereophile on my neck here, but yes you can have a copy of the soundfile.

I am awhare that not many is willing to put their email adress on the net, so how do we do this? I am new to this forum....

[Kudret]

Thank you all for your suggestions. I intend to get to the bottom of this or sell my Khorns soon. I made a lot of investment into my audio setup and I want to start enjoying listening to some music.

Peter, Initially they were in 17' corners. I prefer to sit at least 9' away from the front wall. Then I moved them out of the corners 12' apart and toed them in to the sitting position which is about 11' away from the front wall. This reduced the bass substantially but at least removed the "sitting on the stage" effect. I have my HT set up at the other end in the same room and placing them along the long wall does not look feasible.

Larry, I wish the search function was operational. Actually that was the first thing I tried before posting. I believe woofers are ok, as I can hear as well as feel he bass when they are in the corners. The problem with testing the tweeter and squawkers is that they pass

the test even when they are completely out of phase on one speaker.

DrWho, I

will wait for the CD from Erland. Maybe the tests on that CD will be

more conclusive.

Erland, Thank you so much. I will take you up on your offer.

Michael, Woofers are ok. I also used some test tones to check the level of both speakers and I am reading off the same dB for the same test tone from both speakers. I believe crossovers are ok. I was wondering if there was a D cell type test for squawker but I guess the answer is no.

On the other hand, what could be the other causes for poor sounstage? Should I start looking into room treatment? I do not know where to start though.

Kudret

[PrestonTom]

Thanks, Al.

I think what's imortant is that the same phasing between the same drivers in opposite speakers (both squawkers and both woofers) is what is important.

I tried the reverse phasing between the same drivers in opposite speakers (both squawkers and both tweeters). They sound the same as before. Anyway, at least I know that poor soundstage is not caused by some drivers being connected in reverse.

Kudret

WAIT, WAIT, WAIT....

If you reversed the phase between the squawkers between the left & right cabinets, it should sound very different (compared to before). Something is very wrong here!

When the squawkers are in phase (L vs R), & if you are right in the middle then the vocals (and percussion usually) should sound smack in the middle (when in phase). dead center between the 2 cabinets. If they are out of phase, the image will be diffuse and if you lean a little bit to the right, the image will be mostly on the right, and if you lean a little to the left, the image will be mostly on the left. Don't try this without sitting directly between the cabinets and back about 10 ft

When they are out of phase it will be a very "fragile" sweet spot. The percept and the difference should be quite noticeable and not require a special CD.

Let us know,

-Tom

[D-MAN]

Long wall/short wall has little to do with what you are describing. My speakers are 15ft apart on the short wall and they work just fine.

I had some polarity/phase issues when I swapped in a set of new-out-of-the-box drivers - tell me that isn't the case here. Your speakers are not brand new?

Here's the mono-check:

First, you are going to have to work with one cabinet in MONO and get it right for sure.

That means IGNORE the sweet spot and do your istening exactly on axis of ONE SPEAKER.

Then you get to do the same to the other speaker ONE AT A TIME to ensure that you have correct wire phasing. This means speaker cables, crossovers and internal wiring to each driver.

When you have both correctly wired (make absolute sure on this!), then go to MONO on BOTH SPEAKERS AT THE SAME TIME, staying on axis.

If you do not get an very consistant top-to-bottom "fixed" single image in front of you, then you have a ROOM problem. That is, if any portion of the MONO soundstage is not always directly between the two speakers , or any portion of the image (or any instrument, etc)moves in any direction, then you have a room problem.

I'm very suspicious of the fact that you can change polarity on the midrange drivers without a very noticable change occuring - this usually is an indicator that it is the room, not the speaker.

Let me ask this, if it is the room, and not the speakers, then what other speakers are you going to work better in that room?

DM

[Kudret]

Tom and D_MAN,

Yes, I just reversed the phase between the squawkers between the left &

right cabinets and did the test again and I am hearing exactly the same thing in both tests - 1.5 seconds of sound in the center and 0.5 seconds of sound which is difficult to localize.

When the squawkers are in phase (L vs R), & if you are right in

the middle then the vocals (and percussion usually) should sound smack

in the middle (when in phase). dead center between the 2 cabinets. If

they are out of phase, the image will be diffuse and if you lean a

little bit to the right, the image will be mostly on the right, and if

you lean a little to the left, the image will be mostly on the left.

Don't try this without sitting directly between the cabinets and back

about 10 ft

I played several songs that I am familiar with both, squawker in and out of phase. In both cases it is difficult to find the sweet spot. The vocal seems to be coming from a wide area. In fact this has been one of my major problems as I've been moving around looking for the sweet spot which seems to be shifting from one song to the other.

D-MAN, Speakers and the drivers are all stock. I have some mono LPs and I will try what you suggested and report back.

Thank you both.

Kudret

[WMcD]

Gee, as this thread progressed almost all my bright ideas were stolen by others.

I do suspect overall that it is a room problem that they don't image better. The problem may be keeping sound off the side walls by toe in, or putting up some absorptive material on the walls.

Usually reversed polarity shows up most in the bass where there is obvious cancellation and reduction in level.

I found my polarity problem in the midrange image of the Quartets when they were well away from the wall. With correct polarity the sweet spot was directly between them and the sound was centered. Moving the head side to side didn't alter things much.

With incorrect polarity there was not as good an image. The real kicker was that moving the head a bit seemed to make things different, but not better.

I wonder about the nature of the test signal. If it is not restricted to just the midrange, the bass and tweeter are sounding off too. That could be an issue.

I don't know how the machine test is done. Is this using a microphone and internal listening computer? I would think this can only listen for cancellation, and in a live room, this will only occur in the bass region.

It is a puzzler.

Gil

[PrestonTom]

Kurdet,

D-Man and I are pretty much saying the same thing. He is advocating a mono recording and I am suggesting listening to the vocal (which is essentially "mono" since it is typically equal in both the L & R channels). The mono would be better if one is available or if your pre-amp can perform this. The vocals contain enough energy to present a well-imaged location from the squawkers. The diffused image (if L & R are out of phase) would be quite obvious. If the woofers are out of phase then the perceived level would vary as you moved around the room (as the summed waveforms go in & out of interference). Although you can get a similar "varing in level" from room issues, so the diagnostic value is sometimes suspect. I am starting to agree that a test CD may be better at sorting these problems out.

D-Man may be correct about the room issue (if the experiment ultimately points that way); however, I would be quite surprised. Usually an out of phase mid-range (out of phase re: L vs R) would have a much stronger influence than the room, but ...... who knows.

The sweet spot should be more stable, if the things are not out of phase (L vs R). If it is not a phase issue, then the room is problematic.

Keep us up to date,

-Tom

[tigerwoodKhorns]

Erland:

nevadamail@hotmail.com

Thanks,

Chris

[Kudret]

I listened to the speakers separately with a mono LP. I have to say that this was a very useful test. Both speakers sounded fine in mono but the right speaker had sounded slightly different than the left one. I compared individual drivers separately in the same manner and found out that tweeters sounded slightly different. Then I played some test tones and measured the spl and found out that right tweeter was about 3 dB lower. I have a pair of spare K77 and used one of them to preplace the right tweeter. I also swapped the crossovers as well as the K55 drivers just to make sure that they sound the same. Now both speakers sound similar.

Then I hooked up both of them and played the same mono LP. Vocals and instruments are all in the center but everything sounded too big. No matter what I did I could not find the sweet spot that would reduce the size of the vocals and the instruments. Despite loosing deep bass I gave up on corner placement idea for the moment.

Then I moved the speaker out of the corners and 12' away from each other while sitting 9' from the front wall. Speakers are toed-in towards the sitting position. I payed the same mono LP and now everything sounded cleaner (compared to corner placement). Soundstage also improved. I increased the attenuation of mid drivers from -3.9dB to -6.0dB and this helped further reducing the size of the instruments and vocals.

Now, with stereo music I am getting a more detailed soundstage. It is still not perfect though as the soundstage is not very detailed. I also had a test LP which has a drum test for checking the location of the drums. All I can hear is some drums on the left or right, and I am not able to further localize them. I think the test CD should help a lot identifying the remaining problems, but at least I can listen to some music now while I wait for the test CD (Erland: kkibrisli@hotmail.com).

Thanks for everyones help. I will keep you posted as I am not done yet.

Kudret

[PrestonTom]

Kudret,

I am glad the mono source was helpful.

The frequency response of the mid-range will change if the squawkers are not facing you directly (at a tanget). When you toe them inward, you will get that effect along with different room acoustics (speaker-room interaction).

Still I am confused by some of your comments, and this may simply reflect that words are difficult to apply to percepts.

The sweet spot refers to the spot in the room where you listen that you get a nice stereophonic effect. That is instruments will lie on an imaginary line extending from the left speaker to the right speaker. Typically, the vocals are recorded so that they appear in the middle and many times the percussion will be at that spot also with most jazz & pop music. However, on some recordings, esp in the 70s & 80s, the percussion was mic'ed & mixed so that it had a "huge" sound and was not always restricted & precisely localized in the middle. Forgive me for explaining what you probably already know, but I am concerned that we are using words differently or that you are using some unfortunate recordings. This aspect of the image can also be degraded if the room is not symmetrical, although for this to destroy the stereophonc effect and the sweet spot, it would need to be very asymmetric.

The term, soundstage, is tricky it can refer to "width" of the image but more typically its appearent "depth". This aspect is also prone to poor room acoustics but can be enhanced if there is "some" reverberation in the room or in the original recording, Sound stage is tricky, the percept is not as vivid or "convincing" and can frequently be degraded by a poor recording (in fact, it may be absent) or a poor listening room.

I am not sure what you mean when you say the the sound is too "big". Does this mean the instruments are spread too wide, or that an individual instrument seems "large"? Again it is difficult to use words for these things. If the instruments are localized left to right but their position seems to jump around, then that is typically room acoustics.

It is worth the effort....

-Tom