The advantage of Kapton on most polyesters is its temperature resistance! But ... there are polyesters that are able to support more than 220 °C for a thickness of ...12 μ!
Thank you Solhaga for the movies ! Have you established a relationship between the low cutoff frequency and the movements of the diaphragm ? Do you think this cutoff frequency might be a priori determined based on the diaphragm/magnetic field specs ?
Well, Bernt, first of all thank you for these precisions. The knowledge of the Bl may be of some interest. For me, the value should be low... For example, one strip and 10 pleats of 10 cm, like in your own designs, we have 1 m more or less of conductor, the magnetic field at the center is not very high by construction, then the resulting value will be low. Despite this low value, the sensitivity of most AMTs remains high, somewhere in between 90 and 95 dB and sometimes more. I obviously know Heil's theory, about the moving air out of the pleats, but it remains some discrepancy between the movements of the diaphragm and the acoustics levels obtained by these drivers.
The present thread is getting more and mode interesting. I guess some pros are reading these posts with some interest !...
Thank you Solhaga for the movies ! Have you established a relationship between the low cutoff frequency and the movements of the diaphragm ? Do you think this cutoff frequency might be a priori determined based on the diaphragm/magnetic field specs ?
Well, Bernt, first of all thank you for these precisions. The knowledge of the Bl may be of some interest. For me, the value should be low... For example, one strip and 10 pleats of 10 cm, like in your own designs, we have 1 m more or less of conductor, the magnetic field at the center is not very high by construction, then the resulting value will be low. Despite this low value, the sensitivity of most AMTs remains high, somewhere in between 90 and 95 dB and sometimes more. I obviously know Heil's theory, about the moving air out of the pleats, but it remains some discrepancy between the movements of the diaphragm and the acoustics levels obtained by these drivers.
The present thread is getting more and mode interesting. I guess some pros are reading these posts with some interest !...
Last edited:
I don't think the membrane gets that hot. I haven't been able to detect with a pyrometer any temperature rise at all.
I will conduct some tests in order to get the Cms up. Perhaps they will shed some light on the issue.
I'm more concerned by the irregular and low Cms.
Yes. Me, Bernt and Joppe are all subscribers
.Well Solhaga, Cms is normally related to the "suspension" of a driver... For the AMTs there isn't such a part isn't ? The diaphragm has a "natural structural stiffness" and no damping is needed. As stated by Heil himself, this structural stiffness is given by the thickness and the pleating of the diaphragm...
Of course there is somewhat difficult to get a structural stiffness but still lightweight diaphragm !
Thus I think that the question is : what is the best stiffness for a given thickness ? Or, what must be the thickness for the highest stiffness !…
Of course there is somewhat difficult to get a structural stiffness but still lightweight diaphragm !
Thus I think that the question is : what is the best stiffness for a given thickness ? Or, what must be the thickness for the highest stiffness !…
Last edited:
Thanks for mwmkravchenko your insights. They are of great value for the AMT DIY:ers.
I also filmed an AMT at 150 Hz, but perhaps you saw that one as well.
My conclusion on the AMT-movements and the distortion made by their lack of linearity and also the low frequency cut-off, is that one has the work with the Cms. It is by far too low and irregular.
I have some ideas, but I'll take them later in my thread.
Thanks again, Bernt for the passionate work you are doing (for us).
It's a gold mine of research. We all love you for that.
BTW:
The Aurus Cantus AMT you'r referring to for sure cannot be this one?
As far as I have tested, yes all the Arum Cantus membranes are the same construction materials, just different in size.
Nice video.
I recently was able to aquire some software that should allow me to expand on what you have done. I am interested in pushing the current design cycle of AMT's.
There are repeated mistakes in the commercial models. They make for simple manufacture. But very serious resonance peaks.
This is never shown in the measurements, as AMT's and most tweeters are measured in the near field or closer.
Measured in the normal listening positions on the Mundorf type of structure. This is used by Hygeia or Tianle which manufacture the vast majority of the AMT's sold today.
I am happy to see that the AMT's that are on this thread and the other thread do not follow that design flaw.
As are most things in audio, people repeat the designs of others. And rarely question the reasons for the design choices, or simply ignore the design flaws.
I have never been that type of a designer. I like to make things as good as is possible. Even looking to other disciplines for insight as to what may work.
A little note on the temperature requirements of the diaphragm materials.
Looking at real world spectral density, or music content in the range of use in an AMT shows you that you are asking the driver to consume tenths of a watt to just a couple on serious peaks due to the inherent efficiency of the driver.
Chasing a high temperature substrate is not as important as it may seem.
This of course depends on the application. A prosound application is very different indeed. They require all the bells and whistles to be added to both survive the temperature required, and dissipate the heat as quickly as can be accomplished.
That black color is an etch resist.
Underneath the black is the surprise.
And I will keep it as a surprise.
The membrane that supports the metal trace is the real key to why they sound so good.
And I just happen to know exactly how to reproduce that after working on it for a few years.
I have quite a few diaphragm samples in stock. Both folded and flat.
WHat I can tell you is quite in agreement with what you are saying.
And I can add this.
That the ends of the pleated diaphragm are set in a compound to keep the diaphragm acting as a monopole.
They are the source of the CMS figure in that they dominate any rigidity in the diaphragm pleats. More simply put the ends of the diaphragms become much more stiff than the diaphragm itself.
If you pay attention to the type of end termination you use you can tailor the CMS to a limited degree.
The BL is almost funny to calculate. Even the diaphragms that are a series parallel set of tracks like the ones done on the commercial offerings are still a very low BL transducer.
Magnetic my friend.
Think cutlery. As in Knives and forks, spoons.
I have given you the grade and type of steel that fits the bill.
It is magnetic stainless steel.
If you look at this post you'll see what I mean.
Sorry for the cross-posting, båndsei. But I think it is good for us all to benefit from the momentum in this thread right now.
I understand your aim Solhaga, but you're after the consequence not the cause, from my personal point of view.
As soon as "stiffness" is of concern, two physical parameters are of paramount importance :
- first, the Young's Modulus
- secondly, the density d.
There is a very important ratio derived from this two parameters : E/d^2
Further investigation on this ratio shows that one of the best materials seems to be "expanded polystyrene". But, the films used for AMTs are doubled partly with aluminum. One way to improve the above ratio is to "correctly" cover the active walls of the diaphragm by the aluminum strips. This will solve much of the issues and indirectly address the Cms you're trying to "improve"...
We have yet to define what "correctly" means in the above sentence, thus measurements of FR, IR and distortion are to be done with different width of the strips. For a given film, one might find an optimum value for a given weight of the whole diaphragm.
May I put some dirt into the water to make it less clear.
Muddy the waters is the expression.
The best sounding AMT diaphragm I have used is the Arum Cantus. I have friends that have analysed the ELAC AMT and found it to be a great performer as well.
Having personal experience with the Arum Cantus I can tell you that the end termination is almost a stiff as if hot melt glue was applied to the ends to afix them in place on the rectangular carrier.
But. The diaphragm material is exceptionally pliable. Very, very soft.
Similar to the construction methods you gentlemen have used.
I have worked with and supplied commercial diaphragms and they are not anywhere near as soft as what you are using.
You guys are onto something useful. Run with it.
Muddy the waters is the expression.
The best sounding AMT diaphragm I have used is the Arum Cantus. I have friends that have analysed the ELAC AMT and found it to be a great performer as well.
Having personal experience with the Arum Cantus I can tell you that the end termination is almost a stiff as if hot melt glue was applied to the ends to afix them in place on the rectangular carrier.
But. The diaphragm material is exceptionally pliable. Very, very soft.
Similar to the construction methods you gentlemen have used.
I have worked with and supplied commercial diaphragms and they are not anywhere near as soft as what you are using.
You guys are onto something useful. Run with it.
I personally think that whereas in dynamic drivers the voice coil is attached to the cone, which afterwards drives the air, in the AMT the voice coil, in certain sense, directly drives the air !...
Since the elements of the diaphragm are of very low weight, the transient response remains very good up to very HF values.
Then, there is no break up of the membrane, like in cones, and the harmonic distortion might be very low.
But, Mark the "structural stiffness" is a quite different parameter than the stiffness you are speaking about. The fact that the diaphragm remains exceptionally pliable is the reason why it's completely dampened from any internal resonance... This explains why the AMT doesn't sound like paper, metal or plastic !... i. e. the sound isn't colored by the diaphragm material. To be perhaps more precise, graphene has a Young's modulus of 1003 GPa whereas Mylar-PET is in between 2 to 2.7 Gpa. Even with a very high value of E for Graphene, if you use a very thin layer of this material, it will remain much more pliable than a thick Mylar-PET laminate...
Finally, the multiple pleats emitting sound in an independent way from each other, there is a sort of averaging of every peaks and dips, resulting in a smooth response for all their FR.
Again, these are my personal conclusions on the AMTs after carefully reading Heil's patents. May be am I wrong, but the present thread is probably the best place to exchange our views, in order to derive AMTs, with reproducible and optimized specs.
Since the elements of the diaphragm are of very low weight, the transient response remains very good up to very HF values.
Then, there is no break up of the membrane, like in cones, and the harmonic distortion might be very low.
But, Mark the "structural stiffness" is a quite different parameter than the stiffness you are speaking about. The fact that the diaphragm remains exceptionally pliable is the reason why it's completely dampened from any internal resonance... This explains why the AMT doesn't sound like paper, metal or plastic !... i. e. the sound isn't colored by the diaphragm material. To be perhaps more precise, graphene has a Young's modulus of 1003 GPa whereas Mylar-PET is in between 2 to 2.7 Gpa. Even with a very high value of E for Graphene, if you use a very thin layer of this material, it will remain much more pliable than a thick Mylar-PET laminate...
Finally, the multiple pleats emitting sound in an independent way from each other, there is a sort of averaging of every peaks and dips, resulting in a smooth response for all their FR.
Again, these are my personal conclusions on the AMTs after carefully reading Heil's patents. May be am I wrong, but the present thread is probably the best place to exchange our views, in order to derive AMTs, with reproducible and optimized specs.
Last edited:
Well yes and no.
The AMT wires are attached to a diaphragm so they are in fact moving the air by means of pushing the diaphragm.
So the cone and coil analogy is still applicable.
Breakup of the membrane is what causes the distortions and more correctly nonlinearities.
The Mundorf were very easy to measure in this regard.
They use a very stiff diaphragm material and it displays a defined higher frequency breakup characteristic.
There are relationships between force from the conductor in the magnetic circuit and the resulting distortions as reported by solhaga. Now is it a combination of two effects/
He found a wider conductive layer caused greater problems. Then after empirical tests were conducted has settled on a
conductor diaphragm ratio that he has measured to be the best balance.
So when there is a difference in the conductor size and the frequency contamination or distortion the question is what is the source.
I cannot say definitively.
But I can guess to some degree.
The foil is not infinitely stiff. Nor is the diaphragm or the method used to bond the two together.
The larger foil conductor size definitely has a higher mass.
And a wider trace definitely has a higher stiffness as the cross section of a beam calculation would give you as a reference.
When it comes to AMT's there are no real studies as to the physics involved as far as I know. I would really like to read some if anyone has some links.
This is a niche type of audio transducer.
When working with knowledgeable colleagues we are always amazed by how these little drivers can cause us more headaches!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.