Double Stack ESS AMT-1 with Wings--Possible Kit for Heritage

[Rudy81]

22 hours ago, dkalsi said:

 

Correct! You certainly don't want the center (of the AMT) to center (of the mid-bass/bass driver) distance to exceed the wavelength at the crossover frequency. Doing so will likely result in severe lobbing/cancellation. 

 

As an example, if you are crossing at 800hz, you don't want the center to center distance between the AMT and the mid-bass driver to exceed 43 cm, or approximately 17" (wavelength of 800hz). 

 

Once I get my AMTs, I plan of putting them directly on top of a 12" driver and will plan on crossing the drivers somewhere between 1200 - 1600 hz. 

 

Dang it!  Now I will have to look at flipping my bins.  I wonder what kind of paint job I did on the bottoms?  Thanks @dkalsi, you just created more work for me! Current center to center measurement is 25", closer to the 500hz wavelength. 

[Pete H]

3 hours ago, Rudy81 said:

 

Dang it!  Now I will have to look at flipping my bins.  I wonder what kind of paint job I did on the bottoms?  Thanks @dkalsi, you just created more work for me! Current center to center measurement is 25", closer to the 500hz wavelength. 

Is this referring to the longest measurement between the centers of the 2 furthest points?  If I'm placing a stack (2) of the ESS drivers on the top of my dual 12" BB's then I'm assuming the measurement is from the center of the bottom 12" driver (they are installed vertically)  to the center of the top ESS driver.  

[Chris A]

Actually, I believe that you need to be within 1/4 wavelength--which is the reason why MEHs came into being:

 

[Rudy81]

Thanks Chris.  At 1/4 wavelength, no point in flipping the bins.  Lots of work for no or minimal gain.

[rplace]

On ‎1‎/‎15‎/‎2020 at 11:40 AM, Rudy81 said:

I could flip my entire unit upside down and the whole thing would stay together

 

How tall is it with the drivers top/bottom and whatever else  you have there? ie how big would my hold in my baffle need to be if I continue down the PAP DIY avenue?

[dkalsi]

1 hour ago, Chris A said:

Actually, I believe that you need to be within 1/4 wavelength--which is the reason why MEHs came into being:

 

Chris, 

 

I'm not nearly as technically competent as you, but I'm not 100% sure whether the 1/4 wavelength rule is applicable to non-MEH designs. 

 

Take for example the thousands of the typical 1" dome tweeter and 6.5" woofer  2-way bookshelf speakers. Most of these design typically cross at 2.0Kz - 2.5Kz. 

 

The 1/4 wavelength at 2.25Khz is approximately 38mm, or 1.5" ---which would mean its impossible to create 1" tweeter + 6.5" woofer design that does not exhibit sever lobing. 

 

Its not that the 1-wavelength rule does not exhibit lobing, it just that it doesn't suffer as severely as the one whether the c-t-c distance is in excess of the wavelength at the crossover frequency. 

 

See below: 

http://audiojudgement.com/speaker-lobing-polar-response/

http://audiojudgement.com/understand-the-passive-crossover-network/

 

Again, I know absolutely nothing about MEH designs, but if they are achieving less than 1/4 wavelength spacing between drivers, I would imagine the vertical and horizontal polar response must be amazing. 

 

Depending on the crossover slope (if its less than 4th order), I do think it would be good for Rudy to try flipping the bassbins. 

 

 

[Rudy81]

1 hour ago, rplace said:

 

How tall is it with the drivers top/bottom and whatever else  you have there? ie how big would my hold in my baffle need to be if I continue down the PAP DIY avenue?

 

The way I built mine, it is 17" wide by 13" tall.  You can make the bay as narrow as you wish and just let the waveguides float in front of the drivers by attaching them to the baffle.  I think once you have a set of drivers on hand it will be very simple to figure what you need.

[Chris A]

13 minutes ago, dkalsi said:

The 1/4 wavelength at 2.25Khz is approximately 38mm, or 1.5" ---which would mean its impossible to create 1" tweeter + 6.5" woofer design that does not exhibit severe lobing. 

You're right.  The issue is "where are the polar lobes pointing vs. frequency?". 

 

13 minutes ago, dkalsi said:

Its not that the 1-wavelength rule does not exhibit lobing, it just that it doesn't suffer as severely as the one whether the c-t-c distance is in excess of the wavelength at the crossover frequency. 

True.  The design constraint is to put the lobes on-axis at the listener to the degree possible (across the entire crossover interference frequency band covered by the two drivers), and to make that lobe as large as possible. 

 

In MEHs, all the drivers appear as one driver--which is a pretty nice thing to have. 

 

I think some of the thinking about steep slope filters for passive crossovers use (in the past) is making that assumption that lobing is inevitable when using separate tweeters on flat front baffles that are more than a quarter inch from the centerline of the midrange driver axis. (It isn't inevitable, BTW, if you're using drivers all within 1/4 wavelength at the crossover interference bands like MEHs use, and by avoiding all crossovers above 1 kHz.) That kind of thinking dictates that the only way to control lobing is to use steep slope filters--which as a side effect also introduce much greater phase distortion at the crossover bands.

 

Chris

[Pete H]

Does anyone have the exact angle for the top and bottom?  I'm assuming that if the horn could be made including a top and bottom that would continue those angles it would be of some benefit and I'm sure someone will correct me if I'm wrong, but I'm still waiting for the drivers to arrive and spoke with the CEO this morning (he actually answered the phone) and explained that they are behind due to the volume of orders as well as the CES show and my order would ship by the end of the day today.  Just planning ahead so if anyone can chime in, great.

[Rudy81]

Spoke to Ricky?  Very nice guy.  He usually answers the phone. 

 

The side angle is 90 degrees from center, within a couple of degrees.  That is the angle of my wings.  The top and bottom, IIRC, are 120 degrees.  When I tested my 'full' horn prototype, I just make the top and bottom also 90, which is why I likely got some anomalies in the response until I removed the top and bottom. 

 

Even if the top and bottom would be helpful, the whole thing become just too tall for me to use.

[Pete H]

13 minutes ago, Rudy81 said:

Spoke to Ricky?  Very nice guy.  He usually answers the phone. 

 

The side angle is 90 degrees from center, within a couple of degrees.  That is the angle of my wings.  The top and bottom, IIRC, are 120 degrees.  When I tested my 'full' horn prototype, I just make the top and bottom also 90, which is why I likely got some anomalies in the response until I removed the top and bottom. 

 

Even if the top and bottom would be helpful, the whole thing become just too tall for me to use.

Thanks Rudy.  Contemplating a partial extension for the bottom and top trying to compromise, based on the height of my BB's, distance between drivers, placement on top of the BB's (completely forward or back while trying to eliminate the flat bottom  if they are sitting back a little.)  All that said, having zero clue what different length sides versus top and bottom flares might do.  Just mulling things around a bit.  

[Rudy81]

7 minutes ago, Pete H said:

Thanks Rudy.  Contemplating a partial extension for the bottom and top trying to compromise, based on the height of my BB's, distance between drivers, placement on top of the BB's (completely forward or back while trying to eliminate the flat bottom  if they are sitting back a little.)  All that said, having zero clue what different length sides versus top and bottom flares might do.  Just mulling things around a bit.  

 

Totally understand. The best thing we can do is keep experimenting and trying out different designs.  We are close to a good solution, just need to fine tune the results.  The only way to know what a particular design does is to measure it.  That was how I ended up with just the sides. 

[Pete H]

6 minutes ago, Rudy81 said:

 

Totally understand. The best thing we can do is keep experimenting and trying out different designs.  We are close to a good solution, just need to fine tune the results.  The only way to know what a particular design does is to measure it.  That was how I ended up with just the sides. 

Agreed.  I don't however need to duplicate something that has already been tested out that doesn't provide an improvement.  I've been tying to get to a 2 way configuration that I'm satisfied with and hopefully, this will be the key.  If there is something you haven't tried but wish you had, let me know so that we can check something off of that list.

[Rudy81]

4 minutes ago, Pete H said:

Agreed.  I don't however need to duplicate something that has already been tested out that doesn't provide an improvement.  I've been tying to get to a 2 way configuration that I'm satisfied with and hopefully, this will be the key.  If there is something you haven't tried but wish you had, let me know so that we can check something off of that list.

 

Optimizing the length of the wings is the real unknown right now.  I'm not sure if longer wings would produce any worthwhile results.  The is also the issue of curved wings with some sort of proven curve. Other than that, there isn't much else we can do other than keep stacking drivers.  No real point in that I don't think.

 

[CECAA850]

Just now, Rudy81 said:

some sort of proven curve

Tractrix???????

[capo72]

Here is a Tractrix calculator I have used in the past. It might be able to be applied in this situation. Variables for different set ups are going to be target frequency cut-offs for different bass bins.

 

 http://wp.volvotreter.de/dl-section/tools/

[Thaddeus Smith]

Maybe a horn with measles? I think horns with mumps are patented or something..

[No.4]

MiniDSP arrived this weekend, but no progress made due to house painting.

 

just a little side note if anyone does not already know. According to the ESS website the purple wire is the positive connection, blue is negative.

[Chris A]

That wire color thing that you stated above is correct from my measurements.  I just dialed in a pair of single AMT-1s on top of the Khorn clones yesterday, after removing the top hats and salvaging the top of the top hats themselves for covering the bass bins and setting the AMT-1 on top (centered).  I had to unscramble the phase on the bass bins and AMT-1s so that they are in phase at the crossover frequency band (to the measured AMT-1 channel delay using excess group delay), and in-phase left to right stereo channels.  The purple wire is positive from my measurements. 

 

[Also note that the AMT-1 diaphragms take a fair amount of time to break-in.  When I first got this pair of AMT-1s last year, it took about a week of playing them in the surround speakers before they finally loosened up.]

 

Don't try to place anything soft on top of the top plate if you do this, because I found that the decrease in midrange SPL from using one absorption pad on top between the AMT-1 and the top of the bass bin was severe.  The AMT-1 needs the entire top surface of the top hat, along with 8-10 inches of setback from the front of the bass bin top to develop the correct SPL response in the midrange (300-900 Hz). 

 

The final results are still in progress, but I can say one thing:  the sound quality while sitting in the listening chair is much more listenable and smooth, and the soundstage is spectacular due to resolving the driver/horn sound quality improvements.  There is also a definite lightness and increased apparent ambience that wasn't there before.  This is in a room that is 13.5' x 11.5' x 8' (wide dimension front wall).  The Khorn clones with stock top hats just didn't sound smooth and relaxed even though they were tri-amped and time-aligned.  The natural crossover point to the Khorn clone bass bins is 560 Hz--with no crossover filters required--it crosses naturally at almost 24 dB/octave (single-high AMT-1 stack). 

 

The La Scala and Belle bass bins both will respond to over 1 kHz on-axis, so some form of crossover filter or wide-attenuating PEQ(s) will be required when using them. 

 

Chris

[Pete H]

1 hour ago, Chris A said:

The AMT-1 needs the entire top surface of the top hat, along with 8-10 inches of setback from the front of the bass bin top to develop the correct SPL response in the midrange (300-900 Hz).

On 1/14/2020 at 3:20 PM, Rudy81 said:

The only problem with longer wings is that there is definitely a better, more open sound when the drivers don't have any solid surface immediately under the waveguides. 

Is there a conflict between these two statements, or is it based on the lower crossover point that Chris is using?