Phase Plug?

I cant think of a reason to have one unless you are playing kilohertz stuff. The reason for a phase plug is to redirect high frequencies produced near the center of the woofer, strait out into the room instead of bunching up before they leave the apex of the cone.


Could I rephrase that, and tell me if its correct?

Without a phase plug, the high frequencies from the center of the cone will tend to ripple outwards on the surface of the cone as going forward. But, with the phase plug, it avoids this, and the signal tends to go straight out without this distorting effect?

Correct?
 
There is an other effect that is easy to measure : on some badly designed drivers the dust cap creates a resonating cavity that will add a nice peak before the roll off. Many times this seems to be due to break up modes of the cone (and it is...), but many times also, when removing the dust cap, no more peak...
So this happens not because of the phase plug but because of removing the dust cap.

Any confirmation/contestation about this ?
 
SEAS FAQ

What’s the difference between a dust cap and a phase plug?

Why use a phase plug?

• Reduced moving mass
• No gluing or other interaction with the cone
• Better cooling of the voice coil
• No cavity resonances
• More consistent high frequency response

Why use a dust cap?

• Avoid air leakage and “chuffing” noise
• Larger cone area
• Conceal a small voice coil with a big dust cap
• Cosmetical reasons
 
Mission: Phase Plug

What are the advantages of having a phase plug for a mid/woofer?

Phase Plugs are found in compression drivers normally loaded by horns. You can adapt an open frame driver for horn loading by surrounding it with an enclosure that provides a back cavity and a front cavity equipped with a phase plug. Some obstructions placed in front of these O.F. drivers, while called phase plugs, are not!
The purpose of a phase plug is 4-fold:
1) Reduce front cavity volume [Vc]
2) Reduce effective throat area [St] 'seen' by the driver through use of radial or circular slits placed in close proximity to the driver's diaphragm.
3) Mitigate standing wave modes in the front cavity through the judicial placement of a select number of these slits.
4) By these means, a high compression ratio [St]/[Sd] may be achieved to improve conversion efficiency, flatten frequency response, and extend it as well.

Regards,
WHG

P.S. Here are the abstracts for the three most important articles I have found to date. There are additional articles by Dodd, Henricksen and others, available at the AES website.

File: AESP1384.pdf
Date: Nov-78
Title: An Application of Bob Smith's Phasing Plug
Author: F. M. Murray
Affiliation: Jamse B. Lansing Sound, Inc., Northridge, CA
Publication: AES-P, No. 1384, Cnv. 61 (Nov-1978)
URL: AES Publications: Preprint Search
e-Library:
Abstract: The war of the phasing plugs still rages after more than 25 years. Compression driver phasing plugs have vacillated between annular rings, salt shakers, teardrops, and now radial slots again. When Bob Smith provided simple design criteria for optimization of the annular ring type, little did he realize how studiously he would be ignored.
Abstract: His design is now incorporated into a large compression driver capable of operating to the high frequencies where this design is important


File: ASAJ025-0305.pdf
Date: Mar-53
Title: An Investigation of the Air Chamber of Horn Type Loudspeakers
Author: Bob H. Smith
Publication: ASA-J, Vol. 25, No. 2, Pg. 305, Mar-1953
Affiliation: Division of Electrical Engineering, University of California
URL: Cookies Required
Abstract (1): The front air chamber design is treated as a boundary value problem which yields a solution of the wave equation for the general case in which the horn throat enters the air chamber in a circumferentially symmetrical manner.
Abstract (2): The following specific cases are analyzed: (1) the case in which the horn throat enters the air chamber by means of a single orifice, (2) the horn throat enters the air chamber by means of a single annulus of radius [r] and width [w], and (3) the horn throat enters the air chamber in [m] annuli of radii [r1],[r2],...[rm] and widths [w1],[w2],...[wm].
Abstract (3): The analysis reveals that the radial perturbations caused by the horn throat excites higher order modes. At the resonant frequencies of these modes the horn throat pressure becomes zero and the loudspeaker does not radiate. By suitable choice of annulus radii and widths the first [m] modes may be suppressed and the corresponding nulls in the output pressure eliminated.

File: AESP-7258.pdf
Date: Oct-07
Title: A New Methodology for the Acoustic Design of Compression Driver Phase-Plugs with Concentric Annular Channels
Author: Mark Dodd
Affiliation: GP Acoustics, Ltd., Maidstone, UK
Author: Jack Oclee-Brown
Affiliation: GP Acoustics, Ltd., Maidstone, UK
Publication: AES-P No. 7258, Cnv. 123 (Oct-2007)
URL: AES E-Library A New Methodology for the Acoustic Design of Compression Driver Phase-Plugs with Concentric Annular Channels

Abstract(1): In compression drivers a large membrane is coupled to a small horn throat resulting in high efficiency. For this efficiency to be maintained to high frequencies the volume of the resulting cavity, between horn and membrane, must be kept small. Early workers devised a phase-plug to fill most of the cavity volume and connect membrane to horn throat with concentric annular channels of equal length to avoid destructive interference [1].
Abstract(2): Later work, representing the cavity as a flat disc, describes a method of calculating the positions and areas of these annular channels where they exit the cavity, giving least modal excitation, thus avoiding undesirable response irregularities [2]. In this paper the result of applying both the equal path-length and modal approaches to a phase-plug with concentric annular channels coupled to a cavity shaped as a flat disc is further explored.
Abstract (3): The assumption that the cavity may be represented as a flat disc is investigated by comparing its behavior with that of an axially vibrating rigid spherical cap radiating into a curved cavity. It is demonstrated that channel arrangements derived for a flat disc are not optimum for use in a typical compression driver with a curved cavity. A new methodology for calculating the channel positions and areas giving least modal excitation is described.
Abstract (4): The impact of the new approach will be illustrated with a practical design.
 
Last edited:
I agree with most of the above comments. On a woofer a "phase plug" isn't really a phase plug in the performance sense. The original ussage and the name comes from compression drivers where wavefront phase uniformity is achieved by careful manipulation of the path lengths from all parts of the diaphragm. For woofers there is a bit of high end fashion to it because some classic drivers (the Lowther) had one, so "it must be good".

The addition of a dust cap gives a variable for adjustment of the woofer's high-end response. You can usually tame the top peak with the right dustcap.

David S
 
Nearly all of my woofers are made with a phase plug. There are several benefits in the way that it is done. First, it is part of the cooling system. The VC windings sink heat into the aluminum former. The former sinks heat into the full copper sleeve on the pole quite quickly. There it is dissipated into the large steel pole. The phase plug then pulls heat from the steel pole where it can be sent to the external air instead of into the enclosure like in a traditional vented pole. This whole process increases the power handling of the driver and lowers power compression.

The biggest benefit though is to extend the clean upper end response. This is more about eliminating the dustcap than adding a phase plug though. The dustcap is always responsible for some kind of breakup mode. The frequency it is at depends on the size, thickness, and stiffness of the dustcap. I've measured drivers, the 18sound 6nd410 for example which had a breakup peak of nearly 15dB at 4500hz due to the dustcap! Removing the dustcap got rid of that resonance. Even if crossing a full octave below with a high order crossover, the resonance is still problematic. Especially in a driver with high 2nd harmonic distortion. The harmonics of the fundamental tones below the crossover will excite the resonance to levels higher than the fundamental in some cases. EQ can do nothing to prevent those resonances from being excited. Damping the dustcap will shift or lower the magnitude of the resonance but it will still be there. Eliminating the source of the resonance, dustcap in this case, is the only clear solution.

When a dustcap is put on, it also creates a chamber of air at the top of the pole. This chamber of air has a resonance to it which often times can be in in the area the driver is intended to be used. With driver excursion this chamber of air changes. It gets larger on the outward stroke and smaller on the inward stroke. With a solid non-vented pole piece this chamber of air expanding and contracting creates a continually varying loading on the cone which leads to very high distortion. The solution is to typically vent the pole. The problem is now you have a chamber of air that is "tuned" to a certain frequency by the "port" or pole vent. This creates an even more significant resonance. Then as the cone moves in and out the chamber of air is changing in size while the port stays the same size. The resonance created by this tuned port is continually varying in frequency based on the excursion of the driver. Inward excursion the tuning goes up. Outward excursion the tuning goes down. It is a continually variable resonance that nothing can be done with EQ to correct. The most logical solution is to get rid of the chamber of air and any resonance is now eliminated.

There is also some benefit in a wideband woofer to help smooth the upper end response. The TD15M for example shows a smoother response from 3KHz to 4.5KHz with the phase plug than with simply removing it completely. There is also slightly better off axis response in the 1KHz to 2KHz range with the phase plug on.

John
 
A lot of generalities there about dustcaps having and causing strong resonances. I have more frequently found the opposite, that an appropriate chosen dustcap will reduce driver top end resonances. Of course many dustcaps are simply chosen for good looks.

I designed a 10" sub for Snell with a heatsink on the front of the core pole. This, of course, meant that no dustcap could be used. Cooling was improved but the air rush noises from the leakage path were unfortunate.

The variable air chamber resonance problem sounds like a "straw man" arguement. Really?

David S,
 
interesting

Ive often considered removing the dustcap on a pair of visaton AL130, with the aim of improving things a little.

What has stopped me is the possibility of making things worse. Doesnt removing the cap, result in air leaks between the vc former and pole pieces? surely this creates a whistling resonance? I always thought there was another purpose for the cap besides dust protection. wouldnt this rule out a sealed box design?
 
Hi guys, just to throw this in:

A phase plug of sufficient size would, I expect, give better off-axis performance as it stops interference from different parts of the cone, by acting as a reflector. Certainly seems to help my Fostexes.

From Planet10hifi.com
The phase plugs work primarily 2 fronts. 1st they fill the hole left when the dust cap removed. The air trapped in the cylinder inside the voice coil and above the pole piece causes what is called an "oil-can resonance". This invariably causes some midrange distress. Removal of the dustcap and filling the hole pretty much eliminates this source of coloration. The 2nd thing the phase plug does is to effectively halve the diameter of the cone. Any side-to-side standing waves, reflections, etc are pushed up about an octave. Related to this is an improvement in high frequency dispersion, widening the sweet-spot, and reducing the laser-like hot spot some full-range drivers suffer from. The smaller effective cone diameter means that the point at which beaming starts is pushed up and due to the shaped nature & the hardness of the plugs some HF energy is redirected.

Chris
 
Interesting read,

So for woofers, it is a cooling device (makes sense) For mids it pushes beaming up an octave. I have a set of Infinity midranges from the late 90's, the dust cap is very tall with the peak of it at the level of the edge of the cone. I'm assuming it attempts to perform the same function although it does XO at 3 KHz.

The only speaker I have with an actual phase plug is those Tang Band 3" full ranges, it is aluminum so I'm assuming it both cools the voice coil, limits beaming and for strictly marketing reasons--looks cool.
 
I have more frequently found the opposite, that an appropriate chosen dustcap will reduce driver top end resonances. Of course many dustcaps are simply chosen for good looks.

If the cone has a resonance, the cause of the resonance should be looked at first and corrected. Adding a dustcap may help to some extent, but it's not a solution. A broad damping of the entire cone or selection of proper material for the bandwidth is going to be much more effective than gluing on a dustcap. In a cabinet you can add a brace in a long wall to shorten the distance on the panel. This creates separate resonances that are pushed uphigher in frequency, but on a woofer it doesn't work so well. You are now creating 2 separate resonances in the cone, one the corresponds to the distance from the coil to the dustcap and another that corresponds to the distance from the dustcap to the edge of the cone. In addition you're adding a third resonance from the dustcap itself. Depending on the material properties you may have multiple resonances that are all higher in magnitude than the original you started with.

I designed a 10" sub for Snell with a heatsink on the front of the core pole. This, of course, meant that no dustcap could be used. Cooling was improved but the air rush noises from the leakage path were unfortunate.

I've never had a single issue with air noises from the leakage path. Keep in mind that air does now flow in one direction but is pushed back and forth multiple times per second. If the path is long and narrow enough the box pressure will never be high enough to create an issue. As with any fluid flow situation the resistance is based upon the ratio of surface area to the amount of free flowing area. With enough resistance there is essentially now air flow to be concerned with.

The TD15's for example have a VC former that is about 3.5" long. The gap inside is .017" wide. There is a huge amount of surface area on the ID of the former and OD of the pole and very little area for air to flow between. In order to have leakage not only does the air need to go through that tiny long resistive chamber, but it first has to go through the venting under the spider, down through the .020" thick x .75" tall gap plate, into the cavity inside the magnets, and then take a 180degree turn up through the VC former. The tighter the gaps both inside and outside the coil, the more resistive to air flow and the more "sealed" the enclosure is.

I have seen issues with other phase plug drivers and it is typically poor design. If a spider is not porous at all and there is no venting under the spider in the frame, you can have much higher pressure differences with excursion. This can force much more air through the gaps than just the internal box pressure alone. I've also seen very wide gaps in drivers with phase plugs and very short VC's with thin top plates at the same time. There is little resistance to any air flow and as a result there can be substantial noise with higher excursions.

The variable air chamber resonance problem sounds like a "straw man" arguement. Really?

Yes really. Any time you have a chamber of air in a woofer you will have a resonance that you can see on an impedance curve. Unvented chamber under a spider is the most common example if the spider is not porous enough to allow air to pass through easily. The dustcap is the most significant. Take a typical 3" coil woofer. As an example figure in in a 200mL chamber of air and add in a 1" x 3" long vent. You can calculate what the resonance of that chamber will be and you'll see a blip on the impedance curve the directly corresponds to it. As the cone moves in and out, the pole is essentially becoming a piston in a compression chamber. Vary that volume from 100mL to 300mL depending on excursion of the driver while keeping that 1" x 3" long vent constant. See what happens to the tuning of that chamber.

John
 
diyAudio Moderator
Joined 2008
Paid Member
Last month I cut out a dustcap. Unfortunately I broke the rules and altered the crossover (slightly) at the same time. The difference was subtle but very nice, like not changing the tonal balance but cleaning things up.

Here is an on axis plot (sans crossover) of : Green = before, Orange = no dustcap, and Red = a simple cylindrical plug. CSD is before and after.
 

Attachments

  • plug.gif
    plug.gif
    7.1 KB · Views: 1,684
  • plug2.gif
    plug2.gif
    18 KB · Views: 1,789
If the cone has a resonance, the cause of the resonance should be looked at first and corrected.

If a cone has a resonance? All cones have many resonances. Even with todays design tools, the acoustical output of the many mechanical modes of any wooofer are still randomly addressed. Dustcaps are one good variable for controlling top end frequency response.

I've never had a single issue with air noises from the leakage path. Keep in mind that air does not flow in one direction but is pushed back and forth multiple times per second. If the path is long and narrow enough the box pressure will never be high enough to create an issue.

It is the narrow passage that leads to turbulence and hence air noise.

Yes really. Any time you have a chamber of air in a woofer you will have a resonance that you can see on an impedance curve. Unvented chamber under a spider is the most common example if the spider is not porous enough to allow air to pass through easily. The dustcap is the most significant. Take a typical 3" coil woofer. As an example figure in in a 200mL chamber of air and add in a 1" x 3" long vent. You can calculate what the resonance of that chamber will be and you'll see a blip on the impedance curve the directly corresponds to it. As the cone moves in and out, the pole is essentially becoming a piston in a compression chamber. Vary that volume from 100mL to 300mL depending on excursion of the driver while keeping that 1" x 3" long vent constant. See what happens to the tuning of that chamber.

John

A 3 to 1 variation in an under dustcap area? Thats a pretty good excursion.

The blips I've seen in impedance curves are more typically from the surround resonance, one of the lowest frequency resonances. I can't recall anything related to a back chamber Helmholtz resonance. If such a resonance exists I wouldn't worry about it varying with frequency but would just treat it. Many drivers have either venting through the pole or damping under the dustcap to deal with cavity resonances.

Lots of myths flying here about "phase plugs". Some facts: these plugs do not act as phase plugs in the true sense: they do not improve response by equalizing path lengths or propagation times. They are generally not large enough to act as "reflectors" or even occluders of sound.

They may have a minor impact on frequency response. Simply leaving the dustcap off has the greater effect and removes a major variable that driver designers use to improve top end frequency response.

Removal of the dustcap always creates an air leakage path through the woofer. This may or may not create serious air noise at high excursion, depending on factors of coil and gap design, spider material, etc. You may get lucky or you may not.

David S.
 
Last month I cut out a dustcap. Unfortunately I broke the rules and altered the crossover (slightly) at the same time. The difference was subtle but very nice, like not changing the tonal balance but cleaning things up.

Here is an on axis plot (sans crossover) of : Green = before, Orange = no dustcap, and Red = a simple cylindrical plug. CSD is before and after.
I did a similar experiment with one of my older spare Coral Flat 8's which had a damaged dust cap, (which are aluminum domes crimped on the voice coil former with a gauze covered vent hole in the middle) and to be honest I don't think removing the dust cap is all that people crack it up to be...

My measurements mirror those of yours pretty much - simply removing the dust cap actually makes the cavity resonances between 1-2Khz measure much worse because you still have a cavity and now that it's exposed it radiates directly instead of being largely obscured by the dust cap with only a small amount of the resonance "leaking out" through the hole and by re-radiation from the dome.

With the dust cap simply removed the result was unsatisfactory, you now have to fill the hole with something to eliminate that cavity resonance - eg a "phase plug". I tried a variety of different lengths, shapes, like cylindrical, bullet etc, and all of them have the same effect of eliminating the cavity resonance between 1-2Khz provided they are a close fit.

Where they all differ is in the 4-8Khz region, where even small changes in the shape of the "phase plug" have a large effect on the response. Although I'm sure it's possible with just the right shaped taper to get a relatively flat response, with trial and error I could not get a response anywhere near as flat from 4-8Khz as the original dust cap being in place.

My conclusion was that for that driver, a phase plug is a mistake. Whilst filling the cavity did improve the resonance between 1-2Khz, it was relatively minor to begin with when the dust cap was in place, and without going to a lot of trouble to develop just the right shaped plug the response from 4-8Khz was all over the place. It's not just a matter of sticking something in that looks like a bullet and hoping for the best ;)

Treble response from 10Khz up whilst slightly flatter was a lot lower in level, and with far worse off-axis response than the original dust cap. On these drivers the dust cap is acting as a horn loaded dome radiator. (With the curvilinear whizzer cone acting as a horn for the dome) Removing the dome means all the treble radiation is coming from the whizzer cone only, so response drops a lot >10Khz, and treble becomes a lot more directional, even with the phase plug in place.

What could have been done instead, is adding soft damping (cotton wool ?) behind the dust cap to eliminate the 1.5Khz cavity resonance and just leave the dust cap in place. It really is an important and active part of the driver, responsible for a lot of the treble response and dispersion, not just a shiny embellishment...

I've found the ~10Khz "oilcan" resonance of the aluminum dust cap can also be largely eliminated without removing it simply by putting a small dent in it near the edge with a small flat blade screwdriver - a mod that I've done to my other units as well. You can't see it unless you look closely, the treble response is not significantly reduced (unlike removing the dust cap) and the response is significantly flatter, with a much better CSD ~10Khz.

Although I can see certain specific cases where a "phase plug" might be useful, I definitely don't think they're a "must do" modification that many people think they are, and unless done just right and only on the right driver, can actually make the performance much worse...
 
Last edited: