The Littlest Heil

[bimasta]

When it comes to Tweeter phase, count me out. Assuming I can still hear to 12KHz at my age, I couldn't tell if it's in phase or out. "Hertz" used to be called CPS (Cycles per Second, which had a common-sense meaning, not just a German name) and its meaning was clear. A driver moved through +/– cycles— ie, the cone/dome/diaphragm moved in/out so many times per second, in accord with the AC music signal. At 14KHz, each "+" and "–" is 1/28,000 of a second. Can I hear a difference? No.

Can I hear if it's in or out of phase with a Mid which is at + or – for 1/5,000 of a second, or a Woofer at 1/100 of a second? No.

I can probably hear if the "image" is blurred in the Bass and Mids, and maybe by changing the phase (switching the +/– speaker wires) then Voila, I fixed it! ("Absolute Phase" is another matter entirely.)

But High Frequencies will sound the same to me. I've tried it. Makes no diff. (I'd rather have everything be playing "correctly" of course, so if the Xover reverses phase, I want to reinstall the Tweeter right.)

I once saw a 15" Woofer being run at 1Hz with a Frequency Generator. It took the big cone One Second to move in/out one time. It was funny to see, but impressive — inaudible of course, but totally accurate.

(There's a street in Heidelberg called "Hertz Strasse" — ironically it's One-Way.)

 

[toddalin]

Yes I am. Never heard of a Heil diaphragm wired in reverse.


Measurements are what they are-they only tell you what something might tend to sound like, but does not take into account varying room acoustics/listening distance etc.


Never heard of the suck out when hooked up correctly, in but IMHO just as easy to swap and listen if it's a concern.

As mentioned-some like it one way-some the others. Maybe they have their drivers wired up wrong to begin with-explain the difference.

What is "reverse"? I.e., what color wire should be on the positive to produce an outward/forward movement?

Is it possible for the diaphragm to be installed with either side forward, and if so, how would one know what side each color wire should be on and if it is correct?

Measurements are measurements and can easily determine the difference between proper and inproper polarity by how the drivers reinforce/cancel each other through the crossover range. And my measurements do account for room acoustics because I set the microphone on the couch where I sit. The ear is the liar and is easily deceived by the media being reproduced.

You don't hear a sickout just like you don't see a black hole. You only hear its effects when it actually comes into play.

 

[puppet]

I remember the 1st gen diaphragms were single sided and the foil was forward. I still put the double sided in that way but always check the phase with measurement. You don't want them to be out of phase with the adjoining driver(s) unless it's a third order butterworth, in which case they'll be 90* out of phase either way you phase them. In other orders/types, the frequency cancellation will kill the joy. (killjoy ... get it bimasta?) You may not be able to hear 12khz but you can certainly hear 1-1.5khz (or whatever xo frequency) ... that's were the problem would be.

 

[automojo]

What is "reverse"? I.e., what color wire should be on the positive to produce an outward/forward movement?

Is it possible for the diaphragm to be installed with either side forward, and if so, how would one know what side each color wire should be on and if it is correct?

Measurements are measurements and can easily determine the difference between proper and inproper polarity by how the drivers reinforce/cancel each other through the crossover range. And my measurements do account for room acoustics because I set the microphone on the couch where I sit. The ear is the liar and is easily deceived by the media being reproduced.

You don't hear a sickout just like you don't see a black hole. You only hear its effects when it actually comes into play.

No-they are installed correctly, foil forward.
The Heil diaphragm squeezes together in and out like a accordion-it doesn't move in and out like a conventional ribbon-that's why it's unique in how it's projects the sound and it's efficiency in doing so..
So the polarity of it, vs the other drivers isn't as important like conventional drivers
Some claim cancellation issues from the rear output of the Great Heil-which is in phase with the the front because of the squeezing action of a single pleated diaphragm-but apparently changes as it wraps around the Heil housing.
I have always wondered about this claim
RonC noticed this years ago and recommended the foam blocks.
Myself have switched a few times -but currently have them unblocked. After some time-I think your probably simply hearing less air, and side radiation, and mainly direct radiation-then cancelation effects.
The out of phase wiring of the transition driver as I recall had more to do with the crossover region between it and the woofers-which I was surprised by at first-till you remember the Heil is a unconventional driver-and the woofers/mids are conventional cone transducers.
I have some literature from 1974-1975 were ESS explains this-try to locate it.
Its been quite some time.
I understand measurements and their value, as well as how a suck out can manifest in undesirable ways-I was generalizing for convenience, fairly obvious if you choose to read it that way.
But I haven't found there being any sonic drawback, with my 2 pairs of AMT3's with the driver polarity wired as from the factory-sorry.
I messed around with them years ago-have to look at my notes-but decided on leaving them as wired from ESS. RonC probably has some good additional input on this.
As mentioned-my biggest issue with the AMT3 was the bass/woofers matching up well IMHO to the rest of the drivers-(IMHO the mids/highs were some of the best I have heard-they way the projected into the room in a highly musical/natural way) and most of my mods concentrated on that early on-but no real changes to the basic crossover design-or polarity of the individual drivers. To get the bass to have the same quality, and extension (and better level matched) as the Great Heil.
The cap changes were a experiment, as I had a extra pair of crossover to play with. I was blown away by the difference.
Listen to alot of speakers over the years-and these modded 3's are some of the best I have heard period.
Perfect? Nope. No speaker is. But they convey a musically organic enjoyable listening experience (now) top to bottom at any level I choose, that I can listen to for hours-and this includes friends and family. Deep clean bass is not a issue now-no need for subs.
Maybe my ears are lying? Who knows?
A lot of people I have been able to help in a similar way with their AMT3's would agree.
And that's the other cool thing-these are from from the run of the mill 3's-or any other speaker on the market. Pretty awesome IMHO.
Very few true full range speakers out there with clean sub 30hz response.
I think there are some that take issue with the Great Heil-the dips in response, too bright-whatever. That's great-I get it. But also sorts of speaker, transducers are a matter of taste-regardless of others enthusiasm for them. Is what it is-as all things audio.
The only thing we can all really prove is we exist as thinking things.....so the rest is a complete waste of time anyway.....

 

[puppet]

The heils output is dipole not bipole. One sides radiation is out of phase with the opposite side. As the diaphragm pushes air out of the pleats on one side, the pleats on the other side are drawing in air to be expelled. The phase of the acoustic wave(s) are therefore 180* opposite of the other. The destructive cancellation occurs, in space, half way between the two. This condition is local to the heil.

A similar condition has an impact on the heils interaction with other drivers in the speaker, in the xo region, if the heil wavefront is out of phase with those of the other driver(s). The destructive cancellation occurs at some point in front of the speakers, on and off axis, in the transition frequency region. eg. if the transition is set at 2khz, both the heil and the adjoining driver produce a 2khz frequency 180* out of phase of each other. The result is a vastly reduced 2khz signal intensity. Depending on the xo slopes, the cancellation can be extended beyond the transition frequency. The shallower the slope, even more cancellation occurs above and below the transition frequency. The measured result is what's termed as "suckout" and will appear as a null in the measure response.

Proper driver summing will produce a nice, combined, forward "lobe" between the two drivers. Conversely, the forward "lobe" will be destroyed, leaving little output @ the transition. One sounds better than the other.

 

[toddalin]

No-they are installed correctly, foil forward.
The Heil diaphragm squeezes together in and out like a accordion-it doesn't move in and out like a conventional ribbon-that's why it's unique in how it's projects the sound and it's efficiency in doing so..
So the polarity of it, vs the other drivers isn't as important like conventional drivers

While this is how many/most people think the Heil works, it is wrong, and there is a forward/rearward motion that does create a distinct polarity.

Yes the pleats move in and out like an accordion, but if you watch an accordion player, you'll see that he doesn't squeeze from the outter edges straight through a central line. He arcs the bellows one way then the other as he pumps them.

Consider an air cleaner on a car. It is pleated, but the outside curves out while the inside curves in. So the spaces between the pleats are larger on the outside than the inside. Now make it linear rather than circular and curve one side out (and the other side in) as the pleats are compressed then reverse so the curves are reversed this as the pleats are expanded.

 

[toddalin]

The heils output is dipole not bipole. One sides radiation is out of phase with the opposite side. As the diaphragm pushes air out of the pleats on one side, the pleats on the other side are drawing in air to be expelled. The phase of the acoustic wave(s) are therefore 180* opposite of the other. The destructive cancellation occurs, in space, half way between the two. This condition is local to the heil.

A similar condition has an impact on the heils interaction with other drivers in the speaker, in the xo region, if the heil wavefront is out of phase with those of the other driver(s). The destructive cancellation occurs at some point in front of the speakers, on and off axis, in the transition frequency region. eg. if the transition is set at 2khz, both the heil and the adjoining driver produce a 2khz frequency 180* out of phase of each other. The result is a vastly reduced 2khz signal intensity. Depending on the xo slopes, the cancellation can be extended beyond the transition frequency. The shallower the slope, even more cancellation occurs above and below the transition frequency. The measured result is what's termed as "suckout" and will appear as a null in the measure response.

Proper driver summing will produce a nice, combined, forward "lobe" between the two drivers. Conversely, the forward "lobe" will be destroyed, leaving little output @ the transition. One sounds better than the other.

You got/get it.

 

[automojo]

Ok-

You got/get it.

Got cha
I stand corrected @ puppet. Much appreciated for the clear to the point description.
Probably explains the cancelation from the sound waves wrapping around the Heil housing @ RonC mentioned. Thanks
And goody goody for you!! Jumpin' some else's train...another merit badge for the man cave!!
As mentioned IMHO the 3's sound as wired from the factory-so you maybe missing something in the crossover. And most of the others I have worked closely with and there mods feel they same.
It's not like others would have not realized it over the years till you post.
Let alone HF review back in the day-which was very favorable.
And it's very possible others that preferred the opposite of the factory wiring had there drivers reversed.
Who knows.
But I will use your finding the the AMT3 database- thanks!

 

[automojo]

The heils output is dipole not bipole. One sides radiation is out of phase with the opposite side. As the diaphragm pushes air out of the pleats on one side, the pleats on the other side are drawing in air to be expelled. The phase of the acoustic wave(s) are therefore 180* opposite of the other. The destructive cancellation occurs, in space, half way between the two. This condition is local to the heil.

A similar condition has an impact on the heils interaction with other drivers in the speaker, in the xo region, if the heil wavefront is out of phase with those of the other driver(s). The destructive cancellation occurs at some point in front of the speakers, on and off axis, in the transition frequency region. eg. if the transition is set at 2khz, both the heil and the adjoining driver produce a 2khz frequency 180* out of phase of each other. The result is a vastly reduced 2khz signal intensity. Depending on the xo slopes, the cancellation can be extended beyond the transition frequency. The shallower the slope, even more cancellation occurs above and below the transition frequency. The measured result is what's termed as "suckout" and will appear as a null in the measure response.

Proper driver summing will produce a nice, combined, forward "lobe" between the two drivers. Conversely, the forward "lobe" will be destroyed, leaving little output @ the transition. One sounds better than the other.

Thanks for the explanation.

 

[automojo]

Ok!
Just to confirm my years ago finding with the AMT3's factory wiring/phase scheme.
Reviewed my many photo's of the crossover and wiring-it's correct to factory spec.
Checked polarity of the mid range and woofers-with my damping tester-woofer cone moved in/mid cone out.
I swapped the wires on the Heil-blue to purple-purple to blue.
Spun a track of Cure classic Jumping' Someone Else's Train..
It did not sound good. No air-sound of the cymbals very diminished. Dull sounding/etc.
Confirmed this with my audio buddy visiting and wife.
Swapped back over-purple to purple, blue to blue-Wala!-the good sound returned. Air, correct sound of the cymbals, energetic and musical sounding-the norm.
Wife and audio buddy came to the same conclusion.
So, as mentioned earlier, assuming your rebuilt crossovers etc are wired correctly-and mid/woofer driver polarity observed correctly, I would have to recommend running the transition driver out of phase the Heil and woofers, as ESS originally designed it.

 

[toddalin]

The Heil needs to be in phase with the midrange, and not the woofer. If the midrange is wired out of phase relative to the woofer, the Heil should also be out of phase relative to the woofer and in phase with the midrange.

If you are testing the woofer movement with something like a battery, you cannot do it directly on the drivers because the crossover causes the woofer to move a quarter wave one way while the midrange moves a quarter wave the other for the net sum of a half wave difference.

So the crossover makes them out of phase and it is corrected by reversing the phase at the midrange. But if the midrange signal passes through a capacitor in the crossover, it will not respond to a battery through the crossover which filters out the dc. You may get a quick cone jump, but that's about it.

 

[automojo]

The Heil needs to be in phase with the midrange, and not the woofer. If the midrange is wired out of phase relative to the woofer, the Heil should also be out of phase relative to the woofer and in phase with the midrange.

If you are testing the woofer movement with something like a battery, you cannot do it directly on the drivers because the crossover causes the woofer to move a quarter wave one way while the midrange moves a quarter wave the other for the net sum of a half wave difference.

So the crossover makes them out of phase and it is corrected by reversing the phase at the midrange. But if the midrange signal passes through a capacitor in the crossover, it will not respond to a battery through the crossover which filters out the dc. You may get a quick cone jump, but that's about it.

Ok I appreciate the explanation.
So your explaining what the AMT3 crossover, transducer wiring is doing.
The damper tester also serves as a polarity tester by observing cone movement through the crossover. Weems book page 136 (confirms the wiring is as intended) Designing building and testing your on speaker system assuming everything else has been observed, it works correctly
That quick mid cone jump when repeated confirm with the photos everything is wired correctly from the factory
When I replaced my Heil Diaphragms, I observed the foil side, but consulted with Ricky at ESS to confirm this.
But as mentioned, dealing specifically with AMT3 stock crossover..wire it as designed from the factory.
That was my original point-AMT 3 exclusively. The PS1220 was the only other ESS 3-way.
I have owned both.
If it was not assembled correctly at the factory people would’ve noticed the difference and take an issue with it.
My experiment today confirms that- not only to me.
A very audible difference.
So I would have to assume the other people that wired their mid range the other way probably had some of the other drivers wired differ then the factory had them. I have no proof-just what they posted.
Mine are wired as ESS designed them.
As a side note-when I started working with these years ago-I took some advice from speaker builder GordonW-as these are some of his favorite speakers-just finished working on a set-he never once mentioned reversing how the midrange was connected from the way ESS had it.
Having built a pile of speakers-he would have also taken issue with the sound.