The destructive behaviour you talk about does cause an issue. But it is limited to a dip whose magnitude is related to the width/depth ratio of the box. As you go off-axis this dip decreases into there is no dip when the drivers are on the sides.
Ignoring that, one hears the full frequency range, but you only hear the lowest stuff from the back driver.
It has become popular to do a castle “bipole” with the 2nd driver on the top.
Ignoring that, one hears the full frequency range, but you only hear the lowest stuff from the back driver.
It has become popular to do a castle “bipole” with the 2nd driver on the top.
It's useful to think of air as a stretchy fluid. If the box design is fairly symmetrical, pressure waves from the back (or front) don't really 'go' much past 90deg left or right because the matching pressure waves from the other side will act like a reflective wall out to infinity. However, that's not exactly the same as "flush wall mounting" because we still have to count half the depth of the speaker box.
The efficiency of the "horn loading" is doubled, and this could be further improved if we add more drivers.
The efficiency of the "horn loading" is doubled, and this could be further improved if we add more drivers.
My brain isn't spinning in the least. That graphic says exactly what I said verbally this whole thread.
The two drivers act independently. Their outputs sum. This is the principle of linear superposition. It's taught in freshman physics classes, and reiterated in graduate acoustics classes, of which I took many in grad school.
The two drivers act independently. Their outputs sum. This is the principle of linear superposition. It's taught in freshman physics classes, and reiterated in graduate acoustics classes, of which I took many in grad school.
I'm seriously not trying to be contentious, but I think if you go back and read my very first post, it's not exactly ambiguous. From the jump I was referring to wavelengths that are long relative to box dimensions and it feels like that didn't really sink in, but I didn't grasp what was happening.
I honestly just wanted to know if there was some phenomenon going on that very basic acoustic theory couldn't explain.
Thank you for the conversation . . . I did learn quite a bit about baffle step effects in the transition zone, although that's not a particular concern of mine right now. I will file it away for future use!
I honestly just wanted to know if there was some phenomenon going on that very basic acoustic theory couldn't explain.
Thank you for the conversation . . . I did learn quite a bit about baffle step effects in the transition zone, although that's not a particular concern of mine right now. I will file it away for future use!
If you're looking for other benefits, perhaps cancellation of mechanical vibrations. Most speakers would have some dipole action anyway, with destructive panel interference at say -40dB. A bipole seems a more elegant fix than the usual brute-force method of adding as much mass as can be tolerated.
This is a huge boon, the more so the lower you go. I like doing this with woofers anytime i can. The push-push we did was able to be dome in 15mm BB since so much box load is removed by the active reaction energy cancelation. Tysen V2 is another good example (here before veneer):
http://www.planet10-hifi.com/boxpix/tysenV2-passive.jpg
dave
I'm looking at doing a bipole right now. With bipoles both drivers transition to forward radiation as mirror images so they do provide baffle step. Really it should perfectly cancel out the baffle step better than a filter can. However you will still have the extra high frequencies from the rear driver unless its filtered off. So its somewhat different.
You can do a quick experiment. Take two bookshelf speakers and place them back to back and wire them in series. Normally series would be 0dB net gain if aimed forward (double area but half current). But now, the bass should be more apparent below the baffle step. For circa 7in to 8in wide cab the baffle step is about 900Hz or so. If you wire in parallel, sensitivity should be +6dB plus more apparent bass. If you have a mic and REW, measure some polars in both configs. All of this is very accurately modeled in programs like Akabak or VituixCAD.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.