My hybrid MTM project

[Earl K]

Hi Roland,

Rather than one shopping list , I think I'll do an item by item installation of observations/suggestions .
You can either respond or just skip an item this way. It's also better for reference.


My setup; I'm going to make all the necessary measurements & commit them to CAD sometime in the next few days. In the meantime here are a few thoughts .

(i) My speakers are 1650 mm apart ( center of horn to the same ). This 65" spacing is about as wide as I can go in my place. It does works quite well .

(ii) Both speakers are focussed at a window. That point is @ 4300 mm away from each horn. Under that focal point is a table & lamp / no chair . In other words, if this is the sweet spot, then nobody gets to sit there.

(iii) My usual listening chair is to one side of this focal point. There is a duplicate chair on the other side ( of the center table /lamp ).
Both chairs hear ( or see the horns ) at an angle that has yet to be accurately measured ( this is where the CAD comes in ). It looks to be 10° to 15° off-axis
These 2 chairs also have a center to center postion of about 1650mm ( or center of a head to a head ).

(iv) There is a couch along the long wall facing across the room ( TV viewing ). This couch is also @ 15° off axis ( wait for an accurate measurement ) to the closest speaker . This couch , because of placement is "outside" of the imaginary rectangle created by the speakers & 2 chairs. At the far end of this couch is another location of great imaging ( making the couch a great place to stretch out & listen - ears are effectievly turned towards the speakers ) .

- A drawing will follow .

[Earl K]

Hi

I don't use tweeters since I can't seem to integrate them without having them altering the imaging. But you have different gear . If anyone can integrate them you should be able to. I'm always just trying pretty adhoc approaches when tweeters are involved .

(i) I'd start with engaging your 2496(s)' most extreme slopes on the transition between the horn and the tweeters. I'd even try choosing 2 slightly different points ( & creating a small response hole ) but also minimizing any overlap ( comb-filtering ) in response .

(ii) I forget your HF xover points. I wouldn't personally take a tweeter lower than 9 or 10K and would try higher a bit higher just for the heck of it. This may mean creating a HF lift circuit for your horn/driver combo to passively flatten its resposne. A simple "inline" resistor/capacitor combo are all that is required.

- ( Values ?? 16ohms & 1.5uf to 3 uf for the cap / resistor & cap are paralleled to each other & then placed in series with the compression driver ).

- Any DC blocking cap that may already exist on the compression driver line is now in series with the new cap.

- This DC blocking cap "changes" the effective capacitance of the cap that is acting as the HF lift/bypass.

- The net capacitance is always lower than the smaller of the two caps ( at the HF lift circuit position / the DC blocking cap still "sees" the resistor & driver impedances below the crossover point of the "lift" cap ).

- If you don't like the math - ( & who does ) - just buy a quantity of caps, ( .5 uf, 1.0 uf. & 1.5uf are all good values to own ). Observe your UltraCurve EQ ( in RTA mode ) as you add different values ontop of a base cap value of 1.5uf. Add as many as required to create the flattest response above 5K .

(iii) Try adding extra delay to the tweeter. Add in a factor of .125 to .25 ( extra delay ) on top of the existing delay ( which the unit chose - as perfect alignment ). This physco-acoustic trick has no garantee of working. But there are horn advocates out there who will tell you that the higher frequencies are best listened to with a dab of extra delay ( when coming from a compression driver / horn combo ) .

(iv) If you could put your tweeters to the inside of the horn on the baffle ( I know you can't ) then you would need to add extra delay to move the acoustic center of the tweeter to be "in behind the 2441". From my listening position , there is a convergence point ( when looking at the directly opposing speaker ) that may be helpfull in the imaging department. This was the original reason that I thought that a tweeter position to the inside ( & maybe up 45° ) could be useful.



- Again, this tweeter thing is all speculation , since I haven't done it.
And, I also don't see me purchasing the required equipment - any time soon .

[Earl K]

Hi Roland,

- It's possible that you are expecting just too much from a 45° ( 6 db down point ) pattern mid-horn.

- I've always wished for a 60° - though I get by with a 45° in my tight setup.

- You may actually require a 60° for your arrangement . ( I need to see a drawing of your setup. )

- I don't know of any off-the-shelf 60° round horn ( 2" ) entry, so it's a bit of a moot point. And none of the round tractrix stuff that are around publish specs. - so they're not a help .

[Earl K]

Hi

Believe or not, I consider this part quite important to my imaging,,, because -
if the upper mids are at all too rambunctious then the lower (horn) mids suffer . This gives an overly bright presentation that isn't really directly related to EQ. It's a form of dynamic EQ that's at work here. My ears can't relax enough to listen for imaging cues if they are bothered by the dynamics .

IME, the easiest remedy in this area is to use a "MPP in Oil , AC motor run capacitor". MPP = Metallized PolyPropylene. The current cap that has the most "industry buzz" behind it for this category is the ASC type. ( model X386S, in fact ) . Now I haven't heard these since I use a comparable Mallory product ( this Mallory cap line of motor-runs is now owned by CDE ). These things definately rearrange the "dynamic curve" of the typical compression driver by "calming" the upper 2 octaves ( say 2.5K to 10K ) while seemingly releasing dynamics in the lower frequencies.

These frequency-dependant, dynamic rearrangements, will alter the sonic presentation / and / can lead to better imaging . ( It can be a cheap tweak to try out ) . Look here for a review on the ASC caps , as well as many others .

BTW: Solens' used in this same DC blocking "role" offer a somewhat "pinched at the waist" sound - that I'm avoiding with the Mallorys. "Waist = lowermidrange warmth"

[pangea]

Hi Earl!

Wow, many thanks for all the info and your comments!!!!

It'll take me a while comprehend and digest the content, so in the meanwhile I begin by posting the picture you asked me to, hoping it covers what you were talking about.


I have also moved the speakers around a bit and also reduced the distance between the speakers somewhat and also reduced the toe-in angle to 10 degrees. After that I changed the slopes to 48dB/oct. B/W on the 075 and the 2441J. I also added the 48dB slopes on the 2123/2215, but only where they meet the 2441. The rest I left at 18dB/oct. B/W. I'm not sure though if this is OK, or if I should change all the slopes, or if I still should go for the 6dB HP slope even on the 2123, as you mentioned earlier.
Then I also added a small "hole" between the 2441 and the 075, like you suggested. (9,02k and 9,26)

All these things together has made a significant difference in the soundstage, so now the sweet-spot is wide enough for at least two people!

I'll go on with your other suggestions bit by bit! Is this fun or what?

BR
Roland

[pangea]

I've been a bit lazy lately! It has been so much fun listening to my hybrids that I haven't come around to finish the cabs.

But now I have placed the fake leather on to the baffles and I was surprised over the difference it made.

In comparison, the over all sound has become less harsh, cleaner and smoother somehow. It's difficult to describe in a foreign language.
But what surprised me most was that I feel the biggest improvement was in the bass region.

The bass is more tight and easier to focus.

Don't get me wrong, the sound was GREAT before, but now it's even better still. Nothing dramatic, but still noticeably better.

BR
Roland

[pangea]

Hi everyone!
I have made a few alterations on my MTM hybrids, that I would like to share, if anyone is still interested.

I've been experimenting off and on the past six months. Some things have worked, some have not.

A few months ago, I replaced the 2215's, with a very nice pair of 2235's I found at a decent price, so I simply had to buy them.
I'm not sure tho if it made any big difference, perhaps the 2235's go slightly deeper.

The 2215 went back to my old speakers, which are now making my daughter quite happy. Yes she likes them very much!

I have also been experimenting a bit on the x-overs (two Behringer Ultradrive DCX2496)

Going with the MTM concept and having the 2235's and the 2123's blend from about 80Hz to 1000Hz, it has become obvious the 2nd order slopes won't work as well as a 1st or 3rd order slope, due to phase problems as the Ultradrive wants to invert the phase on the 2235's only and I still wasn't quite happy with the soundstage there.
While it is possible to manually alter the phasing on the Ultradrive, it felt as if it would be better to go for 3rd or 1st order slopes instead.
Thanks "scottyj" for your input there!

The 3rd order Butterworth was a big improvement, but still the soundstage was a little bit fuzzy and I wanted more.
So I went for the 1st order Butterworth and at the same time I wanted to get rid of a small dip I had at 80Hz, so, I was thinking perhaps I could let the 2123 go as deep as to where it naturally rolled of, without any HP filtering and at the same time I thought I'd try an old 075 trick, which meant that I let the 2441 go as high as it can and at the same time x-over the 075 at 20.000Hz LP with a 1st order slope.

The entire x-over looks much more simplistic now and let me tell you it sounds absolutely beautiful all the way.
The soundstage is firm as a rock, as well as 3-D holographic and the speakers somehow disappear completely. It's so easy now, to pinpoint the instruments, that my mind and eyes are focusing entirely on where the musicians "are".
It's magical, that's what it is!

BR
Roland

[pos]

Hi Pangea,

Nice setup !
What is the size of your 2123 box? Is it ported?