If Alan Shaw doesn’t see a problem, then what's the problem? It will be fine.
40 responses Add your response
with most crossover slopes you get better phase agreeance between the tweeter and woofer when listening below the tweeter axis.
Get about 5-7 feet away from your speakers and sit on the floor. listen to the phase, listen to the mids and upper mids get a lot more snappy and clean. In most cases, that is.
I think teo audio mentions the important point. The pattern of vertical dispersion is generally something like a wide fan, with a 90 degree width and a 10 or 15 degree height. The fan is aimed slightly downward from the tweeter. Imagine a cross section of this pattern that looks like a super narrow triangle, the tip of the triangle is the tweeter, and one of the long sides is parallel to the floor and the other long side is angled down by 10 degrees or 15 degrees from level. So when the speaker is vertically oriented with the tweeter on top, the pattern is just a bit below the tweeter; flip the speaker and the pattern will reverse, it will be ever so slightly upward. So now you need to position yourself slightly above the tweeter. The worst is when you rotate the speaker and set it on its side. Now its a a narrow fan offering 10-15 degrees of width and a very wide vertically, splashing HF on the ceiling and floor. Try it at home, its dramatic how a speaker on its side means that you move just a little bit left or right off axis darn near makes the tweeter almost go away. This is a good demonstration in your house of HF pattern. This works with almost every kind of conventional two way or more cone loudspeaker.
Thanks for the replies! @millercarbon – he did not say that. I misinterpreted one of his responses on the Harbeth forum. He was referring to a speaker in a confined space. Though he did say that it shouldn’t matter (in terms of bass) whether the woofer is on top or bottom.
@teo_audio – I follow you, though it is interesting that Shaw, who certainly is science oriented, insists that Harbeth perform best when the tweeters are at ear height. Presumably he has designed that way.
Brad – thanks for the interesting input.
Id say close to zero [conventional 2way/3 way] speakers designed to be vertical can work properly sideways (rotated 90 degrees or horizontal). The reverse is also true, very few designed to work horizontal can work vertical. You must be aware of the speaker dispersion pattern and the designed orientation. Think about headlights in your car that are designed to be wide and not "tall"-same idea. Waveguides and driver interaction at crossover (when they are both playing the same thing) determines this dispersion pattern and tweeters usually need to be physically directly above the driver it crosses over to. There are some exceptions, but few. Upside down is at least vertical, so you can at least figure out someway to make that work- most of the time. But beware of energy hitting the ceiling just as you would energy hitting the floor in the "right side up" orientation (why rugs make so much difference). Reflected energy is usually the source of most problems with speaker/room interaction and severely affect speaker imaging.
I paired some old, small Tannoys (603) with an NAD D3020 in a traveling system. The speakers had been loaned to a friend and I had not listened to them for a couple of decades. After setting up the system 1500 mi from home, I was not happy- harsh highs with a weirdly recessed midrange. Flipped them on a whim and the overall sound improved noticeably - more open, natural midrange and smoother highs. My wife noticed the improvement and advocated leaving them “upside down” for the three month stay, (even though these speakers have a definite top and an unfinished plastic bottom).Your ear is the final arbiter.
Some years ago, I purchased a pair of sight-unseen 24" Target stands for use with my Salk Veracity HT1s from an acquaintance who had them stored at his parent's house, instate. They turned out to be 28" stands, so I faced the same dilemma. Always helpful, Jim Salk saw no issues running them tweeter down and I was very pleased with the results. Later, I've been able to right the system with a pair of Paradigm Ultracube 10v2 subwoofers beneath the Salks with IsoAcoustics ISA-L8R200 stands, carefully adjusting for height and tilt (arrival times.)
The results in my little studio cube are quite spectacular...nearly 20 years of room tuning and sophisticated DSP...first, do no harm /// or, if minor harm be unavoidable, be certain that it is heavily outweighed by success. The golden paradigm of "straight wire with gain" I've found to be a misnomer in the digital world. After giving up a hard worked - full blown SOTA/Fidelity Research fx64/vacuum/electronic flywheel rig two decades ago, overall, this is decidedly better. Good luck, wipsaw! Happy tunes and More Peace. Pinthrift
Not sure what crossover order the Harbeths use but these two links give a visual on the different lobing patterns. Merely for visual, no claim by me to the content accuracy.
There should be no problems with mounting a speaker upside down, unless the woofer magnet assembly is being supported by an internal brace that is relying on gravity. That would be an unusal setup (most drivers are not supported internally. Harbeths are not supported this way).
You can certainly try inverting the speaker. However, there might be some complications as far as the sound is concerned. This might increase the problem with floor bounce cancellation of lower midrange frequencies. The higher the woofer is off the floor, the greater the chance of floor bounce being an issue. To some extent, designers try to account for floor bounce, but, they never account for someone inverting the speaker and increasing the height of the woofer.