Jul 25 2011

Pique the Geek 20110724: Loudspeakers Part II

I was running out of time last week as the publication hour was neigh, so I was not able too include everything that I wanted to include.  In the meantime, I have gained some additional information that refutes one of the points that I made.  Remember, The Geek likes to be corrected when he is wrong.  We shall get to that after the fold.

This installment has to do with some experimental or otherwise not widely used loudspeaker designs, but also something much more personal to me.  It has to do with a good friend from almost 40 years, and his interaction with Paul Klipsch and others involved in the company.

First we shall make the correction, then talk about other loudspeaker designs, and then the more personal stuff.  Ready to go?

First, for the correction.  I perpetuated notion that Klipsch Heresy loudspeakers were called that because of being the first cone loaded loudspeaker built by them.  My dear friend for decades and Kossack justasabeverage pointed out to me in a personal conversation that the term Heresy did NOT have to do with that model being the first cone loaded loudspeaker that Klipsch made, although by coincidence it was that as well.

According to the account that justasabeverage sent me, the real heresy was building them with a straight back, so that they did not fit into a corner.  Up until that time all Klipsch loudspeakers were designed to fit in corners to allow the walls of the house to help reproduce lower frequencies.  A conventional flat back loudspeaker was the heresy.

He also had a couple of stories about Paul Klipsch.  Once when my friend was working at the airport for a general aviation company, Klipsch flew in to Fort Smith from Hope where the plant was.  The weather turned bad, so IFR certification was in effect.  My friend and Klipsch spent most of the day playing cards and talking before another aeroplane and two IFR certified pilots could make it to Fort Smith.  My friend asked Klipsch about his IFR certification, and Klipsch replied that he had one, but it needed amperes.

My friend was a bit mystified, and Klipsch explained to him that amperes were a measure of currenct, and that his IFR certification was out of date.  Thus, he was not current.  Klipsch was an electrical engineer to the core.  I love geeks!

My friend also told me that Klipsch was wearing one his trademark terms on his tie.  It was very stylized and difficult to read unless the tie was held in just such a manner, and then you could make out the word “bullshit”.  Klipsch often wore a button with the same word or the tie, and when competitors started making unsubstantiated claims, out came the button or tie.  By the way, Klipsch lived to be 98 years old, publishing his last paper at 96.

Now let us get to a couple of topic that I did not have time to cover last week.  A very interesting type of loudspeaker is the magnetorestrictive design, originally designed for the Navy for sonar applications.  Magnetorestrice loudspeakers are very small, and you can barely hear them.  But if you put a pair of them on a material capable of resonating (think of the wooden soundboard of an acoustic guitar), they become quite effective.  They are used in some home stereo applications, but lots of applications are commercial.  For example, they can be attached to windows in stores and the window appears to produce sound.  Actually, it does.

The principle behind the way that they operate is really pretty simple.  Ferromagnetic materials all have some magnetorestrictive properties, and material have been developed to maximize these properties.  In simple terms, when a magnetorestrive material is subjected to a magnetic field, it changes shape slightly.  This can be used to reproduce sound if the magnetic field is modulated with the signal intended to be amplified.  This is very much like the piezoelectric loudspeakers that we discussed last time, but instead of a changing electrical field, it is changing magnetic field that activates them.

Another very interesting design in the plasma arc loudspeaker, that uses a plasma as the medium to be modulated.  Since plasmas are just ionized gases, they are very low mass and thus easy to modulate in a wide range of frequencies.  They are particularly good at very high frequencies for reasons discussed last time.  However, they have a plethora of problems.

One problem is the production of ozone as oxygen in the air is converted to a plasma.  Ozone is like oxygen on steroids.  Although oxygen is an extremely powerful oxidizing agent, for quantum mechanical reasons it is slow to react at room temperature because it contains two unpaired electrons, causing reactions with matter will only paired electrons (by far the most common type of matter) to proceed slowly because they are spin forbidden.  Ozone as all of its electrons paired, so reacts extremely fast.  This has been addressed by using a flow of helium as the plasma, but then you have to get new tanks of helium from time to time.  They are mostly museum pieces now.

Ribbon loudspeakers are sort of straightforward.  They are basically a thin metal foil inside of a static magnetic field.  The signal is sent through the foil, and it moves as it interacts with the magnetic field.  They tend to have good high frequency response because the foil is so low in mass, but suffer other technical problems, including incompatibility with most amplifiers.  The foil is also delicate and easily torn.  Akin to them are loudspeakers that have a conventional voice coil embedded in a nonconductive ribbon.  Once again, they are in a static magnetic field but the signal is produced by conventional voice coil techniques.  They are more properly caller planar magnetic loudspeakers.  There is not really any great advantage of these two types over more conventional horn loaded ones.

Ohm Acoustics makes a loudspeaker based on the bending wave principle and it is not really very clear how they work or if they are better than other designs.  This design is pretty much like a conventional cone driver, except it is claimed the the cone is intentionally less stiff than in a normal cone, and that the stiffness increases towards the edge of the driver.  They are expensive, and I have no real feel for whether or not they are technically better.

Before we leave this topic, it is important to discuss loudspeaker enclosures.  You can have the finest drivers on the planet, but without them being properly enclosed, fidelity suffers.  For example, cheap portable radios sound “tinny” not just because the loudspeaker is cheap, it it probably not enclosed properly, if at all, in a cheap unit.  To understand the importance of the enclosures, let us look at some physics.  In a driver, let us use a cone as an example because it is easier to visualize, equal amounts of sound are produced by both the front and the back of the cone.  However, the wave produced by the front of the cone is almost exactly 180 degrees out of phase with the one produced from the back of the cone.  Here is why.

Let us imagine that the front of the cone is producing sound by moving towards us, thus compressing the air in front of it.  This is called a compressive wave, because the pressure is higher than that of the ambient air.  However the back of the cone is decompressing (rarefying) the air to the same extent.  This is a rarefication wave, and when it is combined with the pressure wave, they cancel each other, so the exact same two sounds are 180 degrees out of phase.  Overall sound quality is diminished, and overall volume is also reduced.  Because of reflection from different surfaces, the waves are never exactly 180 degrees out of phase, but close enough to introduce significant distortion.  Hence the need for an enclosure.  This is most important for electrodynamic drivers.

The enclosure also serves to support the loudspeakers rigidly to that they do not flop about the room.  A naked woofer at high volume can leap about a bit.  The simplest enclosures are just wooden or other sealed cabinets, with the working ends of the drivers mounted to face the listener.  Heavy construction is the hallmark of a good enclosure, because any loose connextions vibrate.  There are some drawbacks to a sealed enclosure, but overall they give excellent performance.  It is beyond the scope of this piece to get into the fine details of the acoustical engineering involved.

Now, the sound from the rear of the driver can be reflected back through the cone, also adding distortion.  Some enclosures are packed with sound dampening material, like glass wool or some such, but it is also important to consider the heating of the voice coils as they oscillate, so it is possible to add too much damping material, particularly at high power levels.  There are many variations in enclosure design, but the most important variation is the bass reflex.

In this design the midrange and tweeters are separated from the woofer in the same cabinet.  Those components are now in their own sealed enclosure, and the woofer in a different enclosure in the same cabinet.  A port is drilled into the woofer enclosure and a tube of a length dictated by the specifics of the enclosure is inserted.  When properly designed, the sound produced from the rear of the cone is phase shifted to reinforce the sound from the front of the cone, and thus the bass response sounds louder.

That about does it for loudspeakers.  Next week we shall take on a new topic.

Well, you have done it again!  You have wasted many more einsteins of perfectly good photons reading this loud topic.  And even though John Boehner admits that he really does want to push the Nation into default just to make a point when he reads me say it, I always learn much more than I could possibly hope to teach by writing this series, so keep those comments, questions, corrections, and other feedback coming.  Tips and recs are also welcome.  I shall remain here tonight as comments warrant, and shall return tomorrow after Keith’s show (I am still waiting for his producers to contact me about that science adviser gig) for Review Time.

Warmest regards,

Doc, aka Dr. David W. Smith

1 ping

Comments have been disabled.