Tweeter placement?

[NcHalfrican]

I often see tweeters placed off-center. Why is this?

 

[matevana]

A possible means of lessening the diffraction effects, particularly off-axis.

 

[godfrey]

Two reasons I can think of:
1) Diffraction/reflection from cabinet edges: Having the various edges different distances away from the tweeter spreads the bumps and dips in the frequency response around.
2) Apparently you get better stereo imaging if the left and right tweeters are closer together than the woofers. Hence a lot of panel speakers e.g. Apogee and Maggies have the treble ribbon on the side rather than down the center of the speaker, with the suggestion that the speakers should be placed with the ribbons on the "inside".

 

[DBMandrake]

A possible means of lessening the diffraction effects, particularly off-axis.

Offsetting a tweeter to one side a few inches might reduce the amplitude response variations caused by diffraction (especially directly on-axis, where its usually the worst) however it doesn't lessen the total amount of diffraction, it just redistributes it more in time. (something often overlooked or misunderstood) Re-radiated signal from the baffle edges is still just as bad as before.

Offsetting the tweeter horizontally on a baffle also introduces asymmetry in the horizontal polar pattern at treble frequencies - something to be avoided IMHO.

Better solutions for controlling diffraction are one or more of the following:

* Radius the edges of the baffle with a sufficiently large radius for the frequencies concerned (and avoid any bumps on the surface of the baffle as much as possible so diffraction doesn't happen before reaching the curved baffle edge)

* Add some sort of soft absorption on the baffle surface so the wave traveling along the baffle surface is mostly absorbed before reaching the edge (only practical above several Khz)

* Use a tweeter with some sort of wide band directivity control, such as a wave-guide. This will reduce the amount of signal traveling along the baffle to the edge by as much as 10-20dB and drastically reduce diffraction problems, provided the wave-guide is smoothly terminated to the baffle itself.

I know a lot of speakers over the years have used offset tweeters, and some of them may be very good, but I think it's now generally accepted that a speaker that is axially symmetric down the middle (drivers in a vertical line, left and right side of the speaker mirror images of each other) is a fundamentally better approach, and that diffraction should be treated at the source rather than offsetting the tweeter in an attempt to make the on-axis response look flatter.

 

[sreten]

Hi,

You can do all the diffraction reducing ideas, and offsetting can still
be a good idea. Assymetric left and right responses are not bad per
se, its all about the drivers responses, and what suits the c/o used.

rgds, sreten.

 

[Ared]

Hi!

My new speaker project ended up having two "right side" cabinets. (The guy that made the cabinets was in a hurry and only noticed his mistake later on ..) The tweeters are located to the left pretty much in accordance with the golden ratio. Make no mistake hi's going to fix this eventually, but I'm forced to listen to the speaker like this for a while. Now I'm wondering how bad is this generally? I hear human voices sound weird at times, I need to sit left from the misaligned speaker to make them sound okay for me and generally whilst music really sounds amazing with it I feel problems with stereo imaging. Is this effect of this mostly placebo, am I imagining this, or left sided tweeters really suppose to sound weird listening from the right side?

 

[tvrgeek]

Done for diffraction by DIY, done in OEM to say "me too". It does cause tilt in the polar response. My recent work says this is not the best way to deal with diffraction. I have gone back to in-line and am using increasingly large radius edges and going back to dense felt pads. Reducing the effects seems better than spreading them around. My thinking changes with every few pairs I build.

 

[Ared]

Done for diffraction by DIY, done in OEM to say "me too". It does cause tilt in the polar response. My recent work says this is not the best way to deal with diffraction. I have gone back to in-line and am using increasingly large radius edges and going back to dense felt pads. Reducing the effects seems better than spreading them around. My thinking changes with every few pairs I build.

The question is not whether this is a good way to deal with the problem or not, but if in my case the cnc machine made two identical front panels for the cabinets and as such I'm curious how much better it will be when it gets fixed The speakers themselves are built from plans found online.

 

[nas3000]

The question is not whether this is a good way to deal with the problem or not, but if in my case the cnc machine made two identical front panels for the cabinets and as such I'm curious how much better it will be when it gets fixed The speakers themselves are built from plans found online.

I am just a rookie here, but will put my 2 cents. If you look at the former soviet block made speaker cabinets with off-center tweeter, almost 90% of their models were identical for L and R. Have listened to many of them and big part of them are rubbish, not for this reason of course... but sticking to your question: you will hear the difference only if the angle of the L R tweeters vs L R mids (woofers/subwoofers does not matter at all) is very different - in simple words your listening distance is close to the speakers. If it is a bookshelf speaker, you can experiment even by putting one speaker upside down or on the side: it may be, that at your listening location you will not hear a difference at all. So, lets say that if we are talking about 20x30cm bookshelf speakers, I would guess it is audible if your listening distance is less than 2-2.5meters, and the distance between speakers is similar. Also I would guess, that even if it is audible it may be not automatically bad.