Diy AMT

[båndsei]

http://www.unitedhomeproducts.com/id101.htm
The Kitharo amt from Oscar Heil is able to go down to 700 hz.
Is the diaphragm the same as the one used for the great ESS.?
It is a different magnet system.
Does anyone knov how it is made?
Can Mylar keep its shape if heated in a jig?
Many questions,but it could help me to make a AMT go deeper.

Bernt

 

[ondesx]

Yes it's the Kithara that I tested... Well, you'll probably find some additional info here :

www.precide.ch

I hope this helps ...

 

[båndsei]

I have found a dechription in German about modification of AMT diaphrame:

Bernt

 

[automaticmojo]

The ESS Great Heil can also go down to 700hz.
Part of the problem you will run into is that at this range the rear radiation is essentially in phase with the front, (along with the inevitable room reflections) causing cancellations, and account for some of the nasal qualities experienced at this low crossover level.
This will still happen with any dipole Heil, regardless of the type of diaphragm used.
The total sensitivity, and some minor peak/valley smoothing, along with perhaps a little different bit of voicing (do to differences in diaphragm material and pleat count) seems to be the main differences sonic wise with the non ESS Heils
Adding a transition driver to cover the fairly narrow range from 250 to 2500hz is the way to go.
The AMT3 crossover is the ideal model IMHO (and one can easily tweak the values/etc) , it sports some fairly generous driver overlaps and 12db slopes, but with the right drivers throws a wide and convincing sound stage that images for days.
No honkiness, no mid bass issues, it all falls together quite nicely.
If the speaker system is well thought out, and integrated properly, IMHO there is little advantage going 2.5 way vs 2 way.
The additional advantage of the 2.5 way is you can good to say a 10" woofer and really have some nice tight bass response.
In my modified AMT3's I used the Dayton ST255-8 with excellent results, along with the Alpha 6A/Aurasound.

 

[båndsei]

I have a midrange speaker to cover the range:100-1100.,and a subwoofer .
I am using active crossover and EQ:Mini DSP.

This is my messurement of the great Heil,messured at 40 cm distance and crossed at 500 hz.
It performs to 700hz. But distorsion rises much below 1000-1100 hz.
I dont belive I can make a AMT better than a great Manufactor,but it could be fun.

Bernt

 

[båndsei]

I have just made a very nice looking diaphragm with plastic made for laminating and alu/paper for coil :

I have made this tool from heatzink from CPU:

It don´t sound ,nor messure nice:

Now i am looking for some new material.The plastic i used was 60 my.
I have read about some plastic/mylar 12my and alu coated 12my.Where can i get some??

Bernt.

 

[miklos]

"Adding a transition driver to cover the fairly narrow range from 250 to 2500hz is the way to go."
How about from 110 to 2500Hz? Then one could have a powered sub plus a two way.
I have a couple of candidates for it, an Altec 414, or a Goodmans Axiom 10.

 

[automaticmojo]

Where the transition driver crosses over to the Heil is more critical IMHO.
With a 12db/octave crossover there is some driver overlap, so matching the drivers is fairly critical. Smooth response curves, and impedance peaks out of the passband, along with high output are beneficial.
The stock AMT3 crossover is a bit strange when you break it down-a hybrid between a series and parallel. The woofer along with the midrange high pass is in parallel arrangement, but the midrange low pass, and tweeter high pass are series, with the midrange being out of phase.
There is some debate, with the stock AMT3 setup on how the speaker sounds with the mid in phase, or reversed. I prefer it reversed, along with blocking the sides of the Heil with foam.
The first schematic is a breakdown that makes it a bit easier to make up a hard wire version.
The second view is the stock O.E. schematic-which can be a bit confusing if you don't look very carefully, especially regarding the inductor values, as the print is poor.
The values can be tweaked depending on your particular drivers, but IMHO this network gives you a very easy and successful way of fully implementing the Great Heil in a DIY design, without having to deal with active crossover/amps.
Using low inductance/ESR air coils and caps, and large gauge wiring you further reduce the resistance and distortion in the network, and add a db or 2 in efficiency.
All good things IMHO.

 

[automaticmojo]

Also the woofer should drop off cleanly on the low end, without any doubling or artifacts of "pretend" bass.
Unless your looking for HT excitement, a clean, non doubling musical response, with room gain to 34-36 should be sufficient IMHO.
It not only cleans things up, but prevents any wasted power on non musical artifacts.
Using/matching subwoofers to a Great Heil speaker, is not unlike trying to get a good subwoofer match to a Magnepan.
Woofers with poor transient response aren't really up to the task and IMHO should be avoided, as I can guarantee you won't like the results.

 

[miklos]

Also the woofer should drop off cleanly on the low end, without any doubling or artifacts of "pretend" bass.
Unless your looking for HT excitement, a clean, non doubling musical response, with room gain to 34-36 should be sufficient IMHO.
It not only cleans things up, but prevents any wasted power on non musical artifacts.
Using/matching subwoofers to a Great Heil speaker, is not unlike trying to get a good subwoofer match to a Magnepan.
Woofers with poor transient response aren't really up to the task and IMHO should be avoided, as I can guarantee you won't like the results.

Thank you for your advise. In this case i have to try the 414 for the full lower band. What do you think about active xover? THX.

 

[automaticmojo]

An high quality active crossover may be the way to go, but you have a lot of added expense and equipment and fiddling.
A good passive network done right will sound just fine. The problem with many of the passives is they aren't done right-high ESR, excessive inductance impedance mismatches etc.
Despite what some may say, in my experience an active crossover does not always guarantee the best sound, if you really did your home work with the passive.
I have heard good, and not so good examples of both. Some electronics crossover seem to have a narrow sound stage, and add a hard electronic haze to the sound, and some frankly are just plain unmusical IMHO. Part of this is the unavoidable added electronic processing, an added noise and potential electrical interference.
Personally I would do what ever your comfortable with, and go from there.

 

[miklos]

Some electronics crossover seem to have a narrow sound stage, and add a hard electronic haze to the sound, and some frankly are just plain unmusical IMHO.
Why is that? After all a filter is a filter ,before or after the amp, and amp is the same in both cases.

 

[automaticmojo]

Hard to say- poor quality parts, electrical interference. With the purely digital crossovers, poor quality ic's, power supply issues, etc.
I have seen more them a few hot-rodded active crosovers-so others must have felt the same.
A filter is a filter spec wise, but they can perform differently sonic wise. Capacitors, resistors, inductors all can impart a signature-good, bad, or indifferent.
As mentioned, these are my opinions and experiences-try one out and see what you think, as that's all that really matters. Don;t take my word for it-go out and find out for yourself-DIY.
Some people really like them, in my experience I'm luke warm to them so far-stock wise from what I have heard.
I also prefer the traditional passive setup, with less fiddling, amps wires etc.
Other this doesn't bother.
IMHO I would take the plunge-and see what happens-you can always sell it, or maybe it will be the best thing your ears have heard.

 

[båndsei]

I just foun this on the internet:PSP Marine Tapes : Sail Repair Tapes
Mylar selfadhedsive tape 25my or 50my. 50mm or 150mm wide.
Have ordered 50my 150mm wide.For diaphragm,might have ordered 25my 50mmwide.

Bernt

 

[SEdwards]

I'm following this thread very keenly, since I'm considering a horn-loaded AMT for my wedding speakers next year.

A question for the AMT gurus. What determines the lower rolloff frequency more, given identical pleat depth -
1/ total area
2/ height
3/ width

Do you need a half wavelength~ lambda/2 to fit in the height of the pleat for good wave launch from the diaphragm? The TPL is about 15 cm high which is about half the wavelength at 1 kHz.

 

[automaticmojo]

http://www.diyaudio.com/forums/planars-exotics/171441-hornloading-heil-amt1.html

Ess hd12 - AudioKarma.org Home Audio Stereo Discussion Forums


Horn vs. Heiled Horn?

As mentioned earlier part of the difficulty one runs into with crossing over the Great Heil low (1k) has to do with the in phase lower frequency form the rear causing cancellation. Some call it a bit of a "honk" sound, some consider it more of a midrange dip. This still happens to a smaller degree with the AMT3, which in practice crosses over fully @ 2500hz.
Some of this is in phase output from the rear wrapping around the Heil housing, and interfering with the overlap frequencies of the transition driver, and the front Heil output-the foam blocks on the side eliminates this. With the AMT3 it's a subtle, but noticeable improvement, as the cancellation is very mild to begin with.
IF you are completely blocking off the rear output with the horn setup, you would avoid much of this. You still may have a bit of honk/tonality issue-but I think the dip would be minimized, and perhaps precise horn tuning would gain some improvements, aside from output gain.
If do block of the rear, you will lose the airy quality of the Heil. Even without the contribution of the rear output, the output would obviously rise do to the (larger) horn loading.
IMHO the airy quality of the Heil is a added bonus, along with the correct tonality and awesome imaging the driver already has.

 

[SEdwards]

Thanks for the info and links automaticmojo. I have more reading to do!

One of the reasons that getting some replacement diaphragms or making some is so interesting to me, is that they can serve a dual purpose as tweeters and also HQ headphone drivers. Learning how to make them would be the way to go in this case.

 

[Grumpy_Git]

There was a thread on DIYA somewhere that discussed the technicalities of Air motion transformers, and pleat depth was mentioned as being the major factor in high end response, cant remember where but a search should throw it up.

I would think that total surface area would be the ultimate decider of the amount of low end you get (subject to having enough folded width to get the air out - too tightly spaced and you would get nothing!

Bernt I think you will need a thinner material for your homemade diaphragms. look at the 2-10 micron mylar sheets that aeroplane modellers use to cover their wings
eg: HERE

Nick

 

[tomtt]

i'd suspect, with more surface area,

the lower the frequency to be had.

Oskar was a brilliant man- I think he probably would have figured it out.

seems he did

the Transar woofers.


ESS - YouTube


seen an unmounted pair at a audio store in the late 80's.

 

[SEdwards]

That would also be my first guess (up to the dipole/bipole wrap around effects mentioned above). The pleats do break the symmetry though. It would be interesting to have 2 diaphragms of the same area but different dimensions.

For those with AMTs and measurement capabilities, may I request a couple of experiments?
1/ What happens when the front of the driver is enclosed like an ear chamber for headphones? With the planars in my Fostex headphones there is a natural rolloff in free air at about 1 kHz. With sealed earpads output is actual higher below 1 kHz than below. I wonder if the resonance of the AMT gets in the way of full range headphone use.

2/ What happens when you expose less of the diaphragm at the front? Besides a reduction in output, perhaps there is better linearity. In an ideal universe you could have a giant diaphragm and expose only part of it. If it is big enough the part in the center doesn't really 'see' the boundary.

I would think that total surface area would be the ultimate decider of the amount of low end you get (subject to having enough folded width to get the air out - too tightly spaced and you would get nothing!
[/URL]

Nick