A U D I O... P A P E R
PRINCIPLE OF THE DECWARE PHASE GUIDE
NOV 2003
by Steve Deckert
Our new High Definition Tower speakers are
using a modified Fostex full range driver, model FE206E. A
respectable sounding driver in it's stock form, a giant killer with the
proper modifications. It comes stock with a whizzer cone and
vented dust cap. On more expensive speakers that reproduce mid and high
frequencies it's common to find what is typically called a "phase plug"
located in place of the dust cap. The bullet shape of the
plug is designed to reduce the inherent phase cancellations that occur
between the dust cap and speaker cone. By reducing the phasing
issues at these higher frequencies it is possible to improve and
flatten frequency response. From a more subjective perspective it
helps add depth and clarity to the sound stage with less beaming.
Many Lowther and Fostex type full range
drivers also use phase plugs of some sort to accomplish these goals.
Some have whizzer cones to extend high frequency response above
8K, some do not. Either type will be effected the same by
the addition of a phase plug. The the only modification would be
to the shape of the plug itself to better compliment the steeper angle
of the (whizzer) cone.
You will notice many variations on the shapes
and application of phase plugs, especially if you look at Lowther
speakers. What makes this different is a small hollow cavity that
resonates. The resonance only happens at mid-range frequencies
(where it's needed) and by resonating it creates a pressure node at the
tip that acts as a wave guide that varies in shape with the music.
The
hollow reservoir resonates to create high and low pressure nodes at the
tip.
In the illustration above, you can see
a cross sectional view of the main components involved with the
application of this device. They are as follows:
A) Stock steel pole piece that it
attaches to.
B) The voice coil bobbin that it sets
inside.
C) Shown in red - the phase guide
itself.
G) The speaker cone.
The phase guide displaces the dead air
space above the pole piece. This area above the pole and between
the inside walls of the bobbin create the same effect as an
empty silo in a barn yard. If you walk inside the empty silo
and play music, the reflections make it hard to understand what you're
listening to. There are two ways to correct the reflections
inside an empty silo. Extensive modification to it's shape and
absorbing wall treatments or fill the silo. By filling it you
reduce the air and the distance between surfaces to a point where there
is little or no sound.
The high frequency energy (resonance)
created by the voice coil that is not absorbed by the cone travels
along the surface tension of air laying on the phase guide. The
high pressure of air molecules that form on the surface of the phase
guide create a skin effect or surface tension that acts as bending wave
transducer. Put a different way, the energy from the voice coil
bobbin uses the surface tension found on the phase guide as a
bending wave transducer. This is illustrated in D) of the above
picture. This wave guide offers organization and stability to the
air molecules that are bending across it's surface.
When the energy reaches the end of the
phase guide it launches off the rim into the air and out into the room.
Since the phase guide and it's rim are fixed and solid, they do
not move. In part E) of the picture above you can see this makes
for a solid platform for the energy to launch from. (Aids in the
linear projection of sound) The alternative is removing the
phase guide and allowing the energy to launch of the chaotic end of the
voice coil bobbin, a point that moves. This is the weak spot with most
moving coil speakers that do not have the voice coil bobbin trimmed
perfectly flush with the cone. Anything that sticks past this
point resonates with ugliness. The dust cap is often used to
block this nasty sound from being heard.
The most interesting part of the
Decware phase guide is how the hollow cavity modifies the actual shape
of the plug as it relates to the dispersion or organization of
sound waves.
Specifically the shape or tip of a
phase plug determines how well it corrects the phase errors between the
cone and the plug. With this phase guide, the hollow cavity
resonates just as any Hemholtz resonator would, with the frequency and
Q determined by the length, diameter(s) and port. (See part F) At
frequencies that are below resonance, the air pressure at the tip is
zero making the shape of the tip true to it's physical dimension -
flat. At higher frequencies, the piston of air in the cavity will
resonate creating an alternating high and low pressure at the tip.
When the pressure is high, the tip takes on the shape of a bullet
with respect to low pressure around it. This in turn sets up a wave
guide for the sound that surrounds it. This means that the shape
of the tip is constantly being modified in real time by the sound
itself. That translates into different phase angle correction at
different frequencies!
Hearing is believing. When you
compare a fixed plastic non-hollow phase plug to this it is easy to
realize that the fixed bullet shape will only modify phase angles one
way. The angles will be ideal for some frequencies and less than
ideal for others. To demonstrate this, you can take some
gum and cover the opening of the Decware phase guide and listen.
It won't matter if you make gum hemispherical or bullet shaped or
even flat. If you plug the hole you will hear a significant
reduction in high frequency extension and clarity when the opening is
blocked. You can even do this with you finger. Unblocked, the
sound is wonderfully more open with noticeably better detail in the top
end. You can amaze yourself for hours by going back and forth.
I worked on tweaking the Fostex driver
for about 3 days. The majority of attention was on finding the
right phase plug to replace the dust cap. The time was spent
trying every shape and style phase plug I could make or find, including
the very light bulb I used to see by. This is when I came up with the
Decware phase guide.
This is the primary modification that
we make to the Fostex FE206E driver. I have also done this with
other drivers both with and without whizzer cones and get consistent
results with all of them. In listening to test drivers with
conventional phase plugs, this phase guide was very audibly superior.
In practice it is also superior because you don't have to try to
computer model the ideal shape for any given cone geometry. The
only variable with this design is length which must be determined for
each application.
As for it's altering the magnetic
properties of the pole which is obviously going to change the shape of
the magnetic flux in the gap and alter driver specs, so far no negative
effects on the sound could be detected. If the alteration of
these specs actually makes the driver sound better then it stands to
reason there will be a driver somewhere that will react in a negative
way. My observations so far indicate the change is minimal.
Some may notice that the opening to the
hollow cavity in the phase guide is actually square. The interior
of the cavity is round. This eliminates the standing waves that would
otherwise rest between the two parallel rims and cause problems
with everything else.
Steve Deckert
This technology is used in our High
Definition Tower speakers
that are ideal for lower power SET amplification.
Listening tests were focused around our
signature loudspeaker, the HDT starting with the stock FE206E
drivers. While the HDT cabinet design goes a long way to maximize
the Fostex drivers performance it still wasn't perfect when
compared to our reference full range tube driven electrostats.
There was still a touch of dryness and beaming in the
mid-range. The top end sparkle made famous by the
electrostatics or even a good pair of ribbon tweeters simply did not
exist. In fact I found myself frequently thinking that adding a
ribbon tweeter at 10K would really make the FE206E's sound great.
After modifying the drivers with the phase guide, the dryness
went away, the beaming and slight shout in the mids vanished, and the
high frequencies extended higher than I can hear. I listened for
many weeks and the desire to add a ribbon tweeter went completely away.
I honestly believe this modded driver sounds more musically
correct and has far flatter and more accurate frequency balance than
most if not all Lowthers. I also estimate that the new PM6A if
modified with this same phase guide would also impress
me. The PM6A is the only Lowther driver I like well enough
to live with, but they cost a bunch. I think our modified
FE206E is better in enough ways that many people may actually prefer
it's sound. At 1/3 the price it makes more sense for everyone.
NOTE: The FE206EM has now been improved
and is called the DFR-8
UPDATE on Phase Guide - Aug 2006
In a
relentless pursuit for ways to machine these phase guides in low
quantities at an affordable cost I finally reached my limit and decided
to see if I could get similar results from a non-magnetic version of
it. A cast resin copy was made and tested. The results are
too close to be measured, but subjectively the top end seems even
smoother. During the end of the test sweeps I compared it in a
DFR-8 against a stock fostex with no modifications. The stock
fostex had a serious ring at just past 1 kHZ and several other peaks.
The DFR-8 is simply peak-less. Glass smooth throughout the
entire sweep. By comparison the stock driver would sound broken
and unlistenable.
Many people have been starting to want
these phase guides to become available for sale so they can modify
their own drivers. This will now be possible as a result of this
latest improvement.
HOME
Decware is a trademark of High Fidelity
Engineering Co.
Copyright � 1996 1997 1998 1999 2000 2001
2002 2003 2004 2005 2006 2007 2008 2009 2010
by Steve Deckert
|