In the past, very special people have known about HARTLEY speakers. Audiophiles, engineers, musicians, recording technicians, scientists and other professionals who use sound or consider sound to be a very important part of their lives or careers. No matter what the price, HARTLEY speakers are considered among the best available anywhere in the world. Meticulous hand-crafting and unique problem solving of the physics of loudspeakers are the reasons for our long standing reputation. We don't sell a lot of speakers that way, but we don't have to compromise our standards either. In fact some of our drivers require over fifty hand operations to build.

Only in recent times do we find companies using cone materials other than paper: polypropylene, kevlar, aluminum, synthetic polymers and other plastic formulations to name a few. Their more recent revelations was our understanding over FIFTY years ago! Starting with polmerization to paper cones, in 1956 Dr. Harold Luth, chief engineer and Master Chemist developed the World's First Synthetic Loudspeaker Cone. Until that time the only non-paper cones were that of bakelite paging speakers used by the Navy. To accomplish this truly revolutionary feat Dr. Luth had to invent a new synthetic to which the world had never seen. A series of monomers cross-linked became the solution which led to other problems and solutions. Platos' loudspeaker cone is one that has: no mass, infinite stiffness, perfect geometry and ultra-fast sound transmission time. Dr. Luth choose a molding process with a special fabric carrier and internally molded ribs.

With his physicist hat on (and pipe) he designed a superior geometry never before seen. The cones were shallow, incredibly light and stiff and could withstand high molecular pressure without edge noding and bending.

Most speaker driver manufacturers use foam, treated cloth or butyl rubber as their compliant surround. Foam, untreated

will deteriorate inside-out from humidity and moisture in the air. Treated cloth is much less compliant, and exhibits more

resonances. Butyl rubber a much better choice will break-down due to ozone in the air. The solution Dr. Luth determined

due to his extensive background in chemistry was silicone rubber. Totally inert, extremely low resonance, proper

durometer, impervious to moisture, as well as UV rays and a lifetime of greater than 99 yrs.

Every other manufacturers surrounds are glued to the cone. This can lead to spurious resonances and even separation at

the joint. Dr. Luth's solution was to change the chemical mix during molding from polymer to silicone rubber thus eliminating a joint completely. A seamless molding solution!

220 MSG---207 MSG

The dual voice coil in the HARTLEY speakers (220 MSG  &  207 MSG) is based on a principle first developed by the British physicist, A.C. Barker, in 1938. Barker's "duode" coil consisted of two coaxial windings, isolated by a plastic film.

By transformer action, high frequency signals imparted to the copper windings were induced on the aluminum shorted turn.

The aluminum tube could move independently of the copper windings to vibrate the speaker cone to which it was connected.

 Barker's voice coil, although sound in principle, was never fully successful because he was never able to fully isolate the two sets of windings with the materials available at that time. However, developments in synthetic chemistry have enabled us to

produce a highly compliant silicone compound, only 4 mils thin, which isolates the windings effectively and permits the

inner aluminum tube to move independently at high frequencies. The HARTLEY aluminum tube is a complete circuit not slotted. It should be noted that in the HARTLEY speaker, the aluminum and copper windings are each connected to respective sections of a dual cone. In the Barker voice coil, the aluminum tube was the only part of the assembly fastened to the cone.

 Electrically, the dual voice coil may be considered an air-cored transformer, modified somewhat by the presence of a small amount of iron on the pole piece of the speaker magnet structure. The copper windings and aluminum tube comprise a voltage step-down transformer, in which the copper turns are the primary and the aluminum tube the secondary. Signals fed to the copper windings of the voice coil are induced by transformer action on the aluminum shorted turn. Because it is a step-down transformer, the induced voltages are lower and the current higher than the primary voltages and currents.

Air-cored transformers are quite inefficient at low frequencies. Consequently, the current induced in the aluminum tube begins to drop sharply at 2000cps and virtually disappears at 1000cps. An elegant way of eliminating a crossover with its losses and distortions.

 220 HS---218 HS---224 HS

 All HARTLEY HS woofers employ a unique "heat sink" which is actually the aluminum tube the voice coil is wound on.

The tube is cut so that a significant portion protrudes through the apex of the cone allowing the greatest concentration of heat (around the coil windings) to be dissipated through the tube outside the cone and into the air. During the coil winding a special high temperature epoxy, manufactured by HARTLEY, is applied to all layers of the windings and baked in with a temperature of over 450 degrees F. This extreme temperature far exceeds the rating of the copper wire itself!tem description

In the first HARTLEY speakers in the 1920's our spiders were made out of bakelite and had four legs that were bolted to the magnet structure. The shape defined the name "spider". Today we still employ the same design but with materials to meet 21st Century standards. Spiders for the polymer series are made from tri-laminate fiberglass which is impervious to adverse weather conditions and strong enough to withstand parallel and perpendicular forces.

Most driver manufacturers use stamped steel frames. These stampings are inexpensive but they exhibit unwanted

resonances. Ringing is often used when referring to this steel frame. All HARTLEY polymers drivers use a sand-cast aluminum frame, light and strong, inert and then polished. Yes, an expensive way to make a speaker frame.......also the best!

RESTORING  FORCE

Since the invention of the moving coil loudspeaker one of the major problems has been to get the moving coil with cone, surround and spider attached back to its starting position rapidly after an electrical signal has driven it out.

More precisely, return the moving mass with no overshoot! In the past, attempts have been made using the air compression in a sealed enclosure behind the speaker diaphragm. Whether it is isothermal or adiabatic compression of air the results are too slow. Driver manufacturers have also designed spiders and surrounds to act as a spring to aid in the movement of the cone assembly.

A moving coil loudspeaker reproduces sound as a result of an electrical signal flowing in a coil in a magnetic field, which causes the coil to move in the field carrying the diaphragm with it, the direction being determined by the polarity of the signal. The signals are generally derived from two basic forms: a single d.c. transient which pushes in one direction only; and the sine wave which pushes in two directions alternately. In musical sounds you get both and infinite combinations of them.

Fig. 1 shows a representation of a simple sine wave. The arrows symbolize the electrical signal pushing on the coil and speaker cone, first in one way and then the other. The length of the arrows represents the push on the coil and cone. From point a to point b the amount of push is increasing and from b  to c  it lessens to zero, then it turns around and does the same thing over in the opposite direction. Now from b to c  and from d to e  as the arrows get shorter and the push lessens, something has to move the coil and cone backward, which is what is called the "restoring force". What does the restoring force have to do? Looking at Fig. 1 again, the distance from point a to point e   is time. At 50cps it is 1/50th of a second. Therefore, the restoring force must be able to move the coil and cone from point b  to point c  in ¼ of that time or 1/200th of a second. The motional velocity of the cone will depend on the frequency and the distance the cone travels on its amplitude. Since the loudspeaker is a velocity device, to maintain constant acoustical power output, the excursion of the cone must vary as the inverse square of the frequency. At 25cps, although the frequency has been cut in half, the amplitude or excursion has to quadruple. So although the cone has to move from b  to c  in only 1/100th of a second, it has to go four times as far....its motional speed must be doubled.

The above data would seem to indicate that the restoring force would not be so important at the higher frequencies because amplitude varying by squares would be in reverse direction, so motional speed would be going down.

Unfortunately, when the frequency goes up, the matter of inertia becomes a question. At 3000cps the electrical signal has to start up a mass, the coil and cone, from no velocity and get it out to its stopping point in a split second. The restoring force has to bring it to a dead stop and start it in the opposite direction and stop it at the starting point. This situation worsens with increasing frequency. It has been calculated that at 60cps with a cone excursion of 1/4 inch, the forces involved are around 93 times the acceleration of gravity, and at 10,000cps it is up around 2060 g's. With a large magnet, this "g" element can be overcome but getting it back is another matter.

Why is this restoration speed essential? Since the outward thrust is extremely rapid (due to the magnetic action of the coil and magnet), by the time the coil has begun to return to its starting position, the next signal is ready to perform work; on the other hand, a voice coil which is never back in time to take full advantage of it.  But we are speaking of only a simple signal.  What happens when the voice coil has to deal with complex signals or transients which are so important for achieving natural sound?  The problem becomes infinitely greater.

We ran some experiments with a conventional coil to see how fast the restoring force would make it move.   Pulling the coil to full amplitude with a strong light thread, then connecting the coil to an 8 ohm load to match it and placing an oscilloscope across same for observation, we burned through the thread and noted that the coil moved about 1/4 inch in 1/350th of a second.  That is fairly fast and this speed can be kept up to about 90 cps, but once we get much above that, the coil starts lagging behind the signal considerably.  Next we used a d.c. transient at 10 cps and it was possible by utilizing the oscilloscope and expanding the sweep frequency to 100 cps to determine the time required for the voltage to rise to its maximum point. It was 1/1300th of a second!  To reproduce this signal the coil must move out to maximum excursion in 1/1300th of a second, be stopped and moved back in 1/1300th of a second.  With this coil moving at best in 1/350th per second, this was obviously an impossibility.

From this simple experiment, a logical question arises:  If the magnetic thrust is faster than the mechanical restoration

force, why not employ magnetic restoration?  This is exactly what Dr. Harold Luth, Chief Engineer of the Hartley Products Corp. evolved and patented under the name of Magnetic Suspension.  It was mentioned earlier that a fixed magnet helps thrust the voice coil outward.   In the Hartley-Luth Magnetic Suspension speakers, this same magnet is used to return the coil back to its neutral position.  This is accomplished quite uniquely as follows:  A thin film of iron, backed by a thin strip of plastic, is placed around the coil form beneath the coil winding.  The geometry of the film is very carefully adjusted for maximum axial force, in both positive and negative directions.  Because there is no mechanical suspension (in the form of a spider or accordion pleated concentric disc), the voice coil with its iron band rests at a neutral point within the magnetic field of the fixed magnet.  From this position, it resists motion in either direction.  Because the fixed magnet is quite powerful, to move the coil and cone from this "rest" position to the maximum outward thrust would require a force of nearly two pounds.  But this would be true only if we used a mechanical force.  Here we have a magnetic force generated by the coil and aided by the force of the fixed magnet.

Another simple but elegant solution.

Previously discussed was the speed of a conventional coil suspension and it was found to move from its maximum excursion to its rest position in 1/350th of a second.  By a similar method, Dr. Luth determined the speed of Magnetic Suspension to be 1/1600th of a second.  This is truly fast, but there were other interesting observations.  In Fig. 2, the traces noted on the oscilloscope of the conventional coil and the Magnetic Suspension coil are super-imposed.  On the solid line curve for the MS coil, the voltage collapse curve pretty much follows that normally obtained in a inductor.

But in the dotted curve for the mechanical suspension, the voltage collapse curve contains a number of intermediate peaks. In an amplifier this would probably be called ringing; in loudspeakers it's usually called hangover, but in any event it is the result of the mechanical spring bouncing.

This magnetic spring is an engineer's delight.  It is non-mechanical, it never changes, it never fatigues, and is unaffected by temperature changes.  What's more, it doesn't bounce!  When the coil moves to its maximum excursion and is released, it flies back to its neutral position and stops exactly at dead center.  There is no overshoot and the action is linear.

It has been emphasized that one of the great barriers to faithful reproduction of sound is a sluggish voice coil.  We also know that a pure unadorned sine wave signal is rarely encountered. With the voice coil now in a position to perform with lightning speed, the results are translated into superlative transient response.  It is this quality of magnetic suspension which makes it possible to convey every nuance within the kaleidoscopic spectrum of music.

Figures 1 & 2

From the beginning,  it has been evident that speaker magnets are structurally asymmetric......and it follows that the magnetic field in the structure is asymmetric. Conventionally, we speak of  "the lines in the gap". The magnetic lines of force above and below the top plate of the magnet were always unknown for the very good reason that no practical way was available to measure them. 

A common method of measuring the magnetic field in a speaker magnet utilizes a search coil connected to a ballistic galvanometer.  The coil is placed in the gap and quickly pulled out of the gap, thus the turns of wire in the coil cut the magnetic lines of force generated, and the electrical current is measured with the galvanometer.  Unfortunately, this measures the average field in the gap.  You cannot pin-point the lines. 

Magnetic lines are defined in gravimetric units, which is to say, force equals mass times acceleration (of gravity).  With this basic physical law as a starting point, Dr. Luth designed an instrument with a coil whose length was measured in thousandths of an inch, which we could place in a precise point in the gap.  A current passing through the coil generated a force which was balanced with a fixed mass.  We measured the field in terms of its definition - an absolute value. 

The next step was to design a dynamic search coil, in which a single turn of wire five thousandths of an inch in diameter could be set at any place in the gap and at any point above or below the gap.  This coil is then vibrated at a very high velocity with an amplitude of motion less than one ten-thousandths of an inch.  The voltage generated in the single turn is amplified and measured on a cathode ray tube, and calibrated against the gravimetric balance.

 Using our new measuring instrument, we found that the magnetic field in all magnets was indeed asymmetrical.

Contrary to what one might expect in looking at the structure of the magnet, there were many more lines above the top plate than below. With such a structure, there is no way to get an absolute uniform forward and backward motion of the speaker cone.  Amplitude distortion is inevitable.  

Our new EQUALIZED FLUX MODULE is a totally balanced magnet, precise tooling and design, providing mirror imaging of the magnetic field above and below the top plate of the magnetic assembly.  The results are a more efficient gap, and the elimination of amplitude distortion.