Hi Guys,
Excellent discussion of an interesting topic!
Here is a great link explaining the wall bounce cancellations in easy language that everyone here can understand:
Wall behind the Loudspeaker Cancellation
I am working on a speaker project and have been looking at on wall versus in room/free space issues. Taking measurements in both situations and trying to adapt the crossover accordingly. Here is an example of the wall bounce nulls that I am getting and a couple photos of a measurement run.
One idea that I have been wondering about is… could you fill in the nulls by adding a second woofer at a different distance from the wall? So you would end up with a stepped baffle. Have a TMM or an MTM speaker with a TM looking regular but the second M (or W) farther out. Not exactly like this, but maybe you get the idea:
http://www.troelsgravesen.dk/speakertalk_files/ellam-flex-1.jpg
Look at the first woofer’s distance from the wall and where the nulls will be. Then calculate the distance that the second woofer needs to be from the wall in order to fill in the nulls. The second woofer will also have nulls, but they will be where the first woofer has good output to the listener. Maybe even have the two woofers with their nulls and received output well offset from each other to make, ideally (but probably not reality) a flat response. Could an idea like this work? Has it been done?
speaker dave- The VS300 looks quite interesting. How did you get all the needed EQ with only a passive network?
Some other ideas that have been mentioned here and might be worth my mulling over are:
-A very shallow and very wide cabinet. That maintains volume for the woofer and, effectively starts to create a flush mount speaker. There will still be nulls but perhaps they are easier to deal with? (Maybe more in the crossover region?)
-A baffle that is angled a little, left to right, so that the nulls from each side of the cabinet occur at different frequencies, to smooth things out a bit. Or maybe it spreads the problem around and makes it more difficult to deal with? The sound may then also be directed too far away from the listener.
-Have a second woofer on a delay from the first, to fill in the null. That was mentioned here. Could work in theory. I don’t know how difficult that would be in practice.
Thanks!
Excellent discussion of an interesting topic!
Here is a great link explaining the wall bounce cancellations in easy language that everyone here can understand:
Wall behind the Loudspeaker Cancellation
I am working on a speaker project and have been looking at on wall versus in room/free space issues. Taking measurements in both situations and trying to adapt the crossover accordingly. Here is an example of the wall bounce nulls that I am getting and a couple photos of a measurement run.
One idea that I have been wondering about is… could you fill in the nulls by adding a second woofer at a different distance from the wall? So you would end up with a stepped baffle. Have a TMM or an MTM speaker with a TM looking regular but the second M (or W) farther out. Not exactly like this, but maybe you get the idea:
http://www.troelsgravesen.dk/speakertalk_files/ellam-flex-1.jpg
Look at the first woofer’s distance from the wall and where the nulls will be. Then calculate the distance that the second woofer needs to be from the wall in order to fill in the nulls. The second woofer will also have nulls, but they will be where the first woofer has good output to the listener. Maybe even have the two woofers with their nulls and received output well offset from each other to make, ideally (but probably not reality) a flat response. Could an idea like this work? Has it been done?
speaker dave- The VS300 looks quite interesting. How did you get all the needed EQ with only a passive network?
Some other ideas that have been mentioned here and might be worth my mulling over are:
-A very shallow and very wide cabinet. That maintains volume for the woofer and, effectively starts to create a flush mount speaker. There will still be nulls but perhaps they are easier to deal with? (Maybe more in the crossover region?)
-A baffle that is angled a little, left to right, so that the nulls from each side of the cabinet occur at different frequencies, to smooth things out a bit. Or maybe it spreads the problem around and makes it more difficult to deal with? The sound may then also be directed too far away from the listener.
-Have a second woofer on a delay from the first, to fill in the null. That was mentioned here. Could work in theory. I don’t know how difficult that would be in practice.
Thanks!
No equalisation will cure this situation because it originates from interference; adding amplitude at the dip frequency will also boost the reflection and thus their sum remains the same.
Though I do not agree completely with this statement I do make the same conclusions as Genelec in the link above as to what can be done about it. The null can't be sufficiently compensated for by adding eq. The only thing the eq will do is add delayed energy at that frequency into the room coloring the sound. Eq works in the signal domain. The wall reflection operates in the room domain and will behave fundamentally different because of this. Only the bass reinforcement can be compensated for in the signal domain with near wall placement.
Adding a second woofer closer to the wall would also be a solution as long as the power response is held the same.
Though I do not agree completely with this statement I do make the same conclusions as Genelec in the link above as to what can be done about it. The null can't be sufficiently compensated for by adding eq. The only thing the eq will do is add delayed energy at that frequency into the room coloring the sound. Eq works in the signal domain. The wall reflection operates in the room domain and will behave fundamentally different because of this. Only the bass reinforcement can be compensated for in the signal domain with near wall placement.
Adding a second woofer closer to the wall would also be a solution as long as the power response is held the same.
Thanks for the measurements! Did you verify that the huge dip you're seeing is caused only by the back wall? You could move the speaker a bit forward and remeasure. The dip should change in frequency.
Did you also try eliminating the dip with absorption behind the speaker?
Controlled directivity is the enabler to place loudspeakers next to room boundaries. Direct radiating drivers are then the issue--not room boundaries. Geddes and Danley can help you there.
There are controlled directivity loudspeakers that virtually eliminate boundary diffraction issues, loudspeaker cabinet edge diffraction issues, and high modulation distortion of direct radiators.
There are controlled directivity loudspeakers that virtually eliminate boundary diffraction issues, loudspeaker cabinet edge diffraction issues, and high modulation distortion of direct radiators.
Flaesch-
Those Stage Accompany speakers look great. Well, they are kind of ugly. But yes, they have the shallow and wide cabinet that I mentioned. Plus the gradual slope from the baffle to the wall. I wonder how much less the wall bounce nulls look with those. Thanks for the link.
Another variation on that shallow, wide, and slanted idea. Thanks for the link. OK, good to see that the idea is out there. My German is very little, but it looks like they measured the idea and the magnitude of the nulls has decreased to only -10 dB but they are still there. Post 55:
Projekt "Schwachmatt" - Seite 4 - Visaton Diskussionsforum
So that cabinet design can contribute to an overall solution.
Aha! Controlled directivity. I did not think about that and it’s something that I know little about. It now gives me something to think about. Here is an excellent link that I found for the topic:
Constant directivity loudspeaker designs
Thanks. I did not do a gradual stepping away from the wall. I just measured at the wall and then in the middle of the room. The plot below is the crossover model using the in room measurements. It’s a shorter gate time here, so it’s not an apples-to-apples comparison. I might have the in room, long gate time measurements that I could dig out. (Although my wife has since commandeered that computer.) But as I recall, yeah, those nulls are only there with the wall. And the big null at ~290 Hz corresponds to the cabinet depth. (I think. I’m not finding a good frequency to distance calculator at the moment.)
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.