Understanding Air Motion Transformers - diyAudio
Go Back   Home > Forums > Loudspeakers > Planars & Exotics

Planars & Exotics ESL's, planars, and alternative technologies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 12th February 2010, 06:19 PM   #1
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
Default Understanding Air Motion Transformers

Moved from Multi-Way
________________________________________________

I have recently become interested in Air Motion Transformers. However, VERY little data exists if one is interested in understanding their behavior aside from Dr. Heil's original patent and the 2 articles available through the AES E-Library.

What effect does the fold geometry have on the response (ex. fold depth, number of folds, etc)? How does this affect the impedance? Intuitively, I would assume it is related to the transformation ratio.

What properties of the film do we seek to maximize/minimize?

What effect does the magnetic structure have on the response? After analyzing my Beyma TPL150s, it appears that half of the surface area of the folds lie immediately behind bars. I would think the bars dimensions are too small to produce diffraction. However, I can't help but assume a reflection is taking place at this surface.

Thanks,
Thadman
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."

Last edited by Pano; 15th February 2010 at 05:35 PM. Reason: moved
  Reply With Quote
Old 15th February 2010, 10:14 AM   #2
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
C'mon, I know some of you are interested in the underlying mechanisms
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."
  Reply With Quote
Old 15th February 2010, 11:23 AM   #3
diyAudio Member
 
Inductor's Avatar
 
Join Date: Feb 2006
Location: Cascais
Quote:
Originally Posted by thadman View Post
C'mon, I know some of you are interested in the underlying mechanisms
There was one guy here that copied a Quad electrostatic a few (many) years ago (better one?), so then he was sent (invited) to the factory in England to do some presentations.
__________________
NEXT-proaudio Portugal

Last edited by Inductor; 15th February 2010 at 11:46 AM.
  Reply With Quote
Old 15th February 2010, 03:38 PM   #4
diyAudio Member
 
Join Date: Nov 2007
Location: Adelaide
Thadman, you ask some good questions; although AMT's really belong in the "Planars and exotics" forum. I think many of us would benefit from being able to read a treatise on the pleated diaphragm loudspeaker but we are still waiting for someone to write one!

Perhaps we need to define a few terms. The word pleat is more often encountered than fold. The pleat depth to pleat width ratio is central to the transformation process. Depth dimension divided by width dimension gives Heil's mass reduction factor (f)

In theory the pleat depth dimension puts a limit on the highest frequency that can be radiated. Output will fall away above a frequency where the depth excedes Lambda/4. If we want to radiate 20kHz where the wavelength (Lambda) is 17mm the depth should not excede around 4mm. In practical electrodynamic designs (yes, electrostatic designs have been proposed long before the Heil patents) the depth has to be kept small as it is related to the magnetic gap.

The dimensions of the diaphragm assembly in both the horizontal and vertical planes will determine the directivity of the AMT just as the acoustic size does with cones,domes,ribbons etc. Once having decided the dimensions of the magnetic structure, the number of pleats that can be "crammed" into it will define the f factor and the efficiency (watts in for output SPL).

More pleats and a high f number do not guarantee more efficiency as the ammount of force that can be generated is a killer aspect. In the patent Heil mentions an f of 11 but in real world production AMT's figures of 4 to 5 are common. Force is the product of IBl. Current (I)xflux density(B)xlength of conductor (l). Compared with a moving coil motor we are immediately challenged. The gap is many times wider and the conductor length many times shorter.

It is the transformation of the force/motion that loads the motor, the more so the higher the f number. Transformation can be equated with leverage, but from the motor's perspective the leverage is around the wrong way. Recall that lever theory mentions a "moment" (force times distance from the fulcrum). If we want to lift a heavy weight we place the fulcrum near the weight and a large motion and modest force gets transformed to a large force and small motion. In the case of the AMT we get hold of the lever where the weight was and may be lucky to move the lever let alone do any work (move air).

The AMT pleats are like a "box of air" with one side missing. Applying a force to the two larger surfaces causes the air to be pressurised and at the same time the force has to create a vacuum in adjacent pleats. Of course this is no different to a cone,dome,ribbon etc but they do not have any transformation of the motion. The electrical impedance of the AMT tends to be almost entirely due to the resistance of the conductors with very little inductance and the most benign resonance behaviour of any transducer. If you read the AES Klaus Heinz paper you may recall that he used an expanded scale on the magnitude axis of the impedance plot lest you might miss the reonance peak! At www.diyaudio.com/forums/planars-exotics/153220-electrostatic-amt.html you can see a discussion I started on the ESAMT. Sadly ES means cannot generate the force required to power an AMT, answering a question previously remaining unanswered. Maybe some of the patent holders knew it didn't work but neglected to tell the rest of the world.

Keith
  Reply With Quote
Old 15th February 2010, 06:56 PM   #5
el`Ol is offline el`Ol  Germany
diyAudio Member
 
Join Date: May 2005
Location: Bavarian Forest
If an AMT is a foil with a meandering conductor that is folded, why does it produce positive pressure when current is flowing in one direction and negative when it flows in the other?
Attached Images
File Type: jpg amt.JPG (9.9 KB, 747 views)
  Reply With Quote
Old 15th February 2010, 07:03 PM   #6
el`Ol is offline el`Ol  Germany
diyAudio Member
 
Join Date: May 2005
Location: Bavarian Forest
Are the folds asymmetric?
Attached Images
File Type: jpg amt2.JPG (11.3 KB, 739 views)
  Reply With Quote
Old 15th February 2010, 11:30 PM   #7
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
el ' Ol,

I believe the pleats are symmetric. Here is a diagram of traditional pleat geometry.

Click the image to open in full size.

"Target Modes in Moving Assembles of Pleated Loudspeakers"
AES E-Library: Target Modes in Moving Assemblies of Pleated Loudspeaker
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."
  Reply With Quote
Old 16th February 2010, 12:30 AM   #8
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
Quote:
Originally Posted by Keith Taylor View Post
Thadman, you ask some good questions; although AMT's really belong in the "Planars and exotics" forum. I think many of us would benefit from being able to read a treatise on the pleated diaphragm loudspeaker but we are still waiting for someone to write one!

Perhaps we need to define a few terms. The word pleat is more often encountered than fold. The pleat depth to pleat width ratio is central to the transformation process. Depth dimension divided by width dimension gives Heil's mass reduction factor (f)

In theory the pleat depth dimension puts a limit on the highest frequency that can be radiated. Output will fall away above a frequency where the depth excedes Lambda/4. If we want to radiate 20kHz where the wavelength (Lambda) is 17mm the depth should not excede around 4mm. In practical electrodynamic designs (yes, electrostatic designs have been proposed long before the Heil patents) the depth has to be kept small as it is related to the magnetic gap.

The dimensions of the diaphragm assembly in both the horizontal and vertical planes will determine the directivity of the AMT just as the acoustic size does with cones,domes,ribbons etc. Once having decided the dimensions of the magnetic structure, the number of pleats that can be "crammed" into it will define the f factor and the efficiency (watts in for output SPL).

More pleats and a high f number do not guarantee more efficiency as the ammount of force that can be generated is a killer aspect. In the patent Heil mentions an f of 11 but in real world production AMT's figures of 4 to 5 are common. Force is the product of IBl. Current (I)xflux density(B)xlength of conductor (l). Compared with a moving coil motor we are immediately challenged. The gap is many times wider and the conductor length many times shorter.

It is the transformation of the force/motion that loads the motor, the more so the higher the f number. Transformation can be equated with leverage, but from the motor's perspective the leverage is around the wrong way. Recall that lever theory mentions a "moment" (force times distance from the fulcrum). If we want to lift a heavy weight we place the fulcrum near the weight and a large motion and modest force gets transformed to a large force and small motion. In the case of the AMT we get hold of the lever where the weight was and may be lucky to move the lever let alone do any work (move air).

The AMT pleats are like a "box of air" with one side missing. Applying a force to the two larger surfaces causes the air to be pressurised and at the same time the force has to create a vacuum in adjacent pleats. Of course this is no different to a cone,dome,ribbon etc but they do not have any transformation of the motion. The electrical impedance of the AMT tends to be almost entirely due to the resistance of the conductors with very little inductance and the most benign resonance behaviour of any transducer. If you read the AES Klaus Heinz paper you may recall that he used an expanded scale on the magnitude axis of the impedance plot lest you might miss the reonance peak! At www.diyaudio.com/forums/planars-exotics/153220-electrostatic-amt.html you can see a discussion I started on the ESAMT. Sadly ES means cannot generate the force required to power an AMT, answering a question previously remaining unanswered. Maybe some of the patent holders knew it didn't work but neglected to tell the rest of the world.

Keith
Thanks for the detailed response

Air Motion Transformers and Waveguides / Horns are fundamentally similar as they both function as acoustic transformers. Perhaps some analogs can be drawn between the two technologies.

I believe waveguides / horns exhibit an upper limit related to the mass of the system. As a result, high Bl transducers are required to extend the HF response.

Intuitively, I would assume maximizing the number of pleats / depth of pleats leads to a similar phenomenon.

How can we relate this to HF rolloff of an AMT?

Also, the optimum ESL surface displacement would be defined by mode shape J=0. However, I believe an AMT with a symmetric pleat geometry whose displacement is defined by mode shape J=0 would result in a non-optimum response. The pleats in the center (anti-nodes) of the membrane will experience a higher displacement than the pleats at the edges (nodes).

This may not be significant for direct radiator applications. However, I believe wavefront shape is VERY significant for waveguides.
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."
  Reply With Quote
Old 16th February 2010, 03:04 AM   #9
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
Is it possible that turbulence could play a role at high velocities (high SPL)? Assuming it does, we might further restrict the solution space by incorporating a flow component (upper limit) in the optimization process.

I'm not sure its possible to analytically calculate the state of the flow considering the complex geometry (pleat geometry, magnetic structure geometry). CFD might have to be used.
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."
  Reply With Quote
Old 16th February 2010, 04:11 AM   #10
thadman is offline thadman  United States
diyAudio Member
 
Join Date: Dec 2005
Location: west lafayette
Quote:
Originally Posted by Keith Taylor View Post

In theory the pleat depth dimension puts a limit on the highest frequency that can be radiated. Output will fall away above a frequency where the depth excedes Lambda/4. If we want to radiate 20kHz where the wavelength (Lambda) is 17mm the depth should not excede around 4mm. In practical electrodynamic designs (yes, electrostatic designs have been proposed long before the Heil patents) the depth has to be kept small as it is related to the magnetic gap.
I would assume this effect is a result of destructive interference from the secondary wavefronts generated at the aperture exit (ie helmholtz resonances). If we restrict the solution space (front cavity =/< 4mm depth, rear cavity =/< 4mm depth), an optimum ratio should then exist for pleat depth / magnetic structure depth.

The magnetic structure geometry will define the aperture impedance (resistance/reactance). Will geometry be significant with regards to the impedance (resistance/reactance) and/or flow state (laminar/turbulent) if the dimensions are insignificant with respect to wavelength (ie =/<4mm)? Should the volume of the bars be maximized (rectangular) or should roundovers be incorporated?

What effect does the bar<-->membrane proximity have on the final response? Can we quantify this reflection? Assuming the membrane mass is insignificant with respect to the air load, can we assume it is acoustically transparent? If the bars were staggered, would the reflection lead to the pressure wave passing through the membrane and leaving the opposite side of the transducer or would it bend around the bars?

I believe the optimum design would minimize reactance and allow resistance to approach a constant value wrt frequency.
__________________
"It is a profound and necessary truth that the deep things in science are not found because they are useful; they are found because it was possible to find them."
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Air Motion Transformers (Heil's) and New Diaphragms EchoWars Planars & Exotics 159 11th October 2013 11:55 PM
ESS AMT-1 Air motion transformers ?! promitheus Planars & Exotics 167 6th August 2010 09:21 PM
Mundorf air motion transformers on ebay audioartspeaker Swap Meet 0 20th November 2007 09:07 PM


New To Site? Need Help?

All times are GMT. The time now is 12:00 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2