Experiment for lobes & comb filtering and all that nasty stuff?

[Coytee]

Just for kicks and for me to try to learn something at home...

Is there anything I can program into my Dx-38 (and I might add, without moving my speakers around) that I can do or undo, that might help exaggerate any of these negative effects so I might better hear and understand them?

What if anything can I do to maximize some lobes? or maximize some comb filtering or any of these other nasties??

I can only presume that I've heard them but out of ignorance, didn't know the name of that which I was hearing. I'd like to try to force my system to create as exaggerated example as I can so I can start trying to put names with effects.

Thoughts?

(other than I've got too much time on my hands)

[mas]

There is a very simple experiment you can do to demonstrate this dramatically.:

In the enclosed doc, under "comb filtering amplitude" they define an experiment that was firt demostrated at SynAudCon workshops 20 years ago. It can be done very well with two 2-3" speaker cubes such as what Bose commonly sells. Also, a pair of speakers like the Minimus7s that Radio Shack used to seel work Very well!

In their experiment you need only feed both speakers with identical pink noise.from a split signal - or the LR pair froma receiver playing the track equally in both channels. Sit in a chair in front or off to the side of the speakers and have another person slowly slide only the top speaker back..

You will experience what appears to be a classic rotating air raid siren effect. - similar to a Leslie organ speaker produces. What you are experiencing is the lobe 'moving' as the speaker is moved back.

If you have an RTA, you can display the comb filtered effect in the frequency response.

The-Digital-Delay-Advantage.pdf

[mas]

Or, if you get really bored, go over and check out the "Presence Channel" thread in home theater and find out how many folks are going out of their way to add the feature thinking it is a positive feature!

Can we say: I hope Bose doesn't show up and want royalties! As this is the exact same thing that was done by Bose in their direct reflecting systems where they intentionally introduced slightly delayed signals by means of longer path lengths due to reflections.!

And depending upon the thread you read, it is either 'inaudible', marvelously effective in adding - well something, or....Pardon me while I consult with teh scarecrow in the Wizard of Oz for diretions once again!

I give up....

[DrWho]

Didn't you mention a certain track that brings out the differences between the steeper slopes? Try playing that track, but set the crossovers on your EV to 6dB/octave (and make sure you don't crank it too loud). Listen to that track and walk around the room and note how different it sounds everywhere....that would be polar lobing in the crossover region.

Another simple experiment is to put both of your speakers right next to each other and sitting in the middle of the room (this works better outside). Now walk around and note how the highs change faster than the lows. Now spread the speakers out and pay attention to the lows...they're gonna change faster than when together. This again is the effect of polar lobing.

Comb-filtering is really the same thing as polar lobing, but it just means there is a lot more of it (so more densely packed). Sometimes engineers intentionally add polar lobing so that the main lobe is narrower (and the side lobes smaller). I think it might be better called "line array theory".

And if you don't want to do this with a particular track of music, pink noise works just as well too.

[WMcD]

I would say no. The lobes and comb filtering occur only because of what goes on after the voice coil, or actually, after the movement of the diaphragm. So nothing ahead of it can change the effects.

Lobing occurs because of what I'll call "antenna" effects. It would happen in an anechoic chamber. We are typically looking at a single frequency, at least at one time. The diaphragm or horn throat acts as an antenna and the surrounding box can add to it. Basically you can think of the diaphragm as being made of smaller elements all moving in phase (we'd like that anyway).

Suppose you are sitting on-axis and a good distance away. There are different path lengths to you. But they are pretty much the same. You can use your pointing finger and middle finger to point to the edges of the diaphragm and imagine that the paths are equal. The path to the center is not too different from these.

Suppose you move 10, or 20 or more degrees off axis and point again. You realize the pathlengths are different. When the difference is 1/2 wavelenght (180 degress out of phase from each other), you get a null. They can cancel down to 0 dB.

But you can see that depending on angle and frequency, sometimes the differences are 1 wavelenght different and there is 360 degrees of phase difference. This is equivalent to two signals in phase at that frequency. They add to 3 dB gain.

So this is where radiation patterns with "fingers" or lobes come from (no relation to you pointing to the speaker, natch). You can't change them with feeding a louder or softer signal because the phase thing is always the same.

Again this is even so in an anechoic chamber and in fact this is typically where radiation patterns are measured. It avoids room effects.

- - - -

Room effects are ususally what causes comb filtering.

Let's start with an ominidirectional radiator. And take it outdoors and put it about 3 feet above a hard driveway. You move away a good distance. Now there are two paths to your ear. One is direct. The other is the bounce off the driveway. The bounce path is always longer in distance and time and therefore there is a phase difference. Again, at your ear, depending on distance and frequency, you get the 180 degree null, or the 360 degree peak.

Of course, the 180 degree issue and 360 degree issue arise at interger multiple of those wavelengths .

There is not much you can do except you a directional speaker (horns are good) so that no energy is aimed at the driveway.

= = = = =

The next step is that we move the speaker indoors. Now instead of just the floor (driveway) we have multiple walls. Now there are multiple reflections. Some of them can be altered with room treatment. But again, all these events happen after the sound is away from the speaker. Driectional speakers can help avoid some of the issues of course.

- - - - -

It is worth pointing out that you might have a speaker with a favorable directional pattern but oddball frequency response. Now there, you can touch up the frequency response with equalization.

- - - - -

If you want to get an idea of room effects: Find a test disk or computer program which will give you sine waves at a given frequency. Play it though one of the pair of speakers.

Moving around the room you will hear the nulls and peaks. One impressive effect is in the bass region. Try 100 or 200 Hz to start, At some locations, you will find a null which makes you think someone turned off the amp. You could map those with a piece of masking tape on the carpet. If you shift the frequency (wavelenght) by even 20%, the nulls have moved to other locations.

Note, the null is caused by an out of phase addition (actually - subtraction by adding negatives, in a way - aka "interference"). You could double the power to the speaker. That wouldn't change a thing because the signal along all the paths have increased and maintain the null.

Basically this is why PWK said, speaking of interference at high freqs,, "you can equalize for your left ear, but not your right ear." Pretty much the same with lower freqs (longer wavelenghts).

Sorry if I'm beating a dead horse here.

Gil

- - - - - -

[mikebse2a3]

Is there anything I can program into my Dx-38 (and I might add, without moving my speakers around) that I can do or undo, that might help exaggerate any of these negative effects so I might better hear and understand them?

Richard you can change/adjust the delay of the K402/K69-A HF Horn relative to the Jub LF Horn with the EV DX38 and the comb filtering and lobing patterns will change/shift in the frequency ranges that both Horns reproduce/share. This effect would be the easiest to hear using pink noise and if you can get your wife or someone to adjust the Delay while you stay in one location you will hear the Tonal shifts and colorations created as the combfiltering/lobing gets worse.

mike tn[]

edit: I would also suggest you use only one channel (either the left or right jubilee) while trying this Richard.

[mas]

Comb filtering and polar lobing can occur between any two (or more) acoustic sources, be they drivers , speakers, real and virtual sources (speakers & surfaces, susrfaces and surfaces, etc) - and they will manifest themselves as Doc said as different aspects of the same thing.

Comb filtering is simply looking at the phenomena in the frequency domain. While polar lobing addresses the spatial distribution of the effects in the acoustical space.

The underlying phenomena that results in both comb filtering and polar lobing is called superposition, whereby 2 or more signals are 'superimposed' (I'm not sure where the letters "im" got lost in the descriptive term!). And this can certainly occur in source material prior to the reproduction of the signal. But one must understand the diferences. It is much the same as with the differences between induced distortion and the recording and playback of distortion.

In the playback of a destructively superposed signal, you will hear the frequency variations due to comb filtering reproduced (assuming no additional superposed effects are generated), but you will not have the accompanying polar lobing.

The additional comb filtering and polar lobing will be of a different 'order of magnitude' resulting from the interaction of both the acoustical playback devices themselves and also with the interaction of the room, where reflections behave as additional delayed virtual sources.

Does this distinction make sense?

And Richard, to do the experiment, just follow the guidelines explained in the PDF - only adjust the delay electronically instead of physically moving the speakers! I hope that it is apparent that this accomplishes the same thing. ;-)

Richard, if you want a really great demonstration...stop by Stokely Athletic Center, the Thomson-Boling Athletic Arena, or the downtown Coliseum prior to a major concert. If you ask to speak with the Road Manager and subsequently the FOH engineer typically around 1-3pm (not during the actual sound check as things become a bit crazier with the'arteests'), I think any engineer who is worth their salt (and doesn't think you are there to get a pass) will be glad to demonstrate comb filtering and polar lobing for you - as you walk in a giant arc from left to right in the audienc space. You will not areas where certain frequencies ae either exaggerated (you won't motice them nearly as much as where) and where certain frequecies seem to simply disappear - which they have! - through phase cancellation!

The larger spacing of the lobes and nulls due to the greater angular dispersion by virtue of the greater distance from the speakers on the floor make the effect very pronounced and dramatic - and you will be forever leary of ever again saying a concert sounded great in your life-and instead you will be preoccupied with identifying the particular lobes and nulls in which your seat was located unless you get a seat immediately in front of the FOH mix position! And some people wonder why I can't stay seated during a show!

**Or you can simply buy a Yamaha receiver and implement their "presence channel"!!! ROFLMAO! You see, what some see as a problem is marketed as an 'enhancement' to others!

And they say you can't fool audiophiles! Or was that audiophiles are fools? Where else can one spend so much for such utter scientific failure marketed as the next best thing?

[Coytee]

Comb filtering and polar lobing can occur between any two (or more) acoustic sources, be they drivers , speakers, real and virtual sources (speakers & surfaces, susrfaces and surfaces, etc) - and they will manifest themselves as Doc said as different aspects of the same thing.

I will admit to not having read all this (overslept & gotta get moving towards work)

Anyway, this caught my eye... let me throw this idea out there...

Since I'm crossing over at 500 hz, could I take the DX and in order to 'hone in" on a narrow frequency, use the DX and chop everything below say 400 hz off that is going to the woofer. At the same time, release its crossover point so it is playing up to its limit (near 1000?)

Secondly, take the 402 and leave it starting at 500 hz and also chop it off at say, 1000. (now both halves are playing a narrow range)

Point being... try to narrow down the sound to just the overlap area & get rid of any extraneous sounds (distractions?) or would I NEED the extraneous sounds to help clarify things?

Or is that entire idea wacko? (but it is an illustration of my comment of can I do something to help exaggerate (or narrow) the effect...even if its a BAD illustration [])

[Blvdre]

I worked briefly at a video duplication facility in the master editing room. The master tapes had the obligatory color bars and , I believe, 1khz tone for calibration purposes. I could crank that tone up, move my head to a particular position, and hear virtual dead silence. Perfect phase cancellation! Not a desirable effect in music reproduction.

[mas]

Richard, call me.