First real speaker build, MTM troubles..Need answers

[MotoMan_Yz400]

I'm wanting to build a solid set of wall mounted speakers for my TV in the living room. It's a 42" LCD, that sits about 27" tall, so I would like to build these to match the height of the TV (WAF factor here..haha). So looks mean alot with this build, and I have a styling I would like to conform too. I also want a clean response with maybe a little embellishment on the low end for the "piazza" and still be able to get some SPLs for loud music or movies (xmax will be my friend)

My first thought was to use the Dayton ND91-8 full range drivers, and have a line array of six per side. I recently bought four to do a 2 per side demo to test out the speakers in a desktop format before I invest in all 12 of the drivers. Those ND91's are incredible! I did run into a problem. with 2 drives line arrays, they were EXTREMELY beamy! At 3ft they has a listening height of about 4-6". No good. And when I make them 6 per side, this will only get worse. So I was thinking about throwing in a Morel CAT378 in the middle of the group hoping the highs could come from a single source to reduce it's beaminess. But modeling say it would still become beamy in the mids, and I would have to cross over to the tweeter as low as 2kHz, which might be a bit much for the tweeter to handle. To fix the convergence of the mids due to frequency wave length becoming smaller than the driver spacing, I was think of some complicated multi xover design. It all started to become more than I can chew. So I think I might get things simpler.

So I thought maybe I just do a BMTMB. Using the Dayton ND140-8, Dayton ND91-8, and the Dorel CAT378. I figured I could build it with the ND140's spaced 19.5" (695Hz wave length), the ND91 spaced 7.5" (1806Hz) and the CAT378 on center. Make the xover simpler, and provides a great system with ample low end.

I would like to get thoughts on these designs. And see if Im worrying about the speakers spacing vs wave length a bit too much. These are high Q drivers and the enclosure will be around the Vas. I would also like to hear what xover points would you recommend? First real build, but been in the game for a while (my father is a HiFi junky, and is a member of this board). Design considerations extremely welcomed

MotoMan_YZ400

[mdocod]

Hello MotoMan,

I like the direction you have decided to bring the project (WMTMW). I think that will sound much better, and give you much more appreciable off-axis listening performance.

The response charts for the ND140 and ND91 are painting a picture that is several dB worse than their rated sensitivity. I'm not liking the looks of that when there are other drivers to pick from that are either more efficient or have more Xmax or or flatter response charts in the ranges you would be intending to use them in. Keeping in mind here, that the Dayton charts are probably a little optimistic, if they look that "badish" and they are "optimistic" then I wonder??? A build based on those drivers would have to be flattened down to ~80dB@1W@1M efficiency, which is a scary thought (to me) for speakers that are expected to carry out the dynamics of movies and such. Though... maybe that's not a problem, I guess it depends on your listening preferences.

I'm not saying that the drivers necessarily need to cost more, just that I would personally use different drivers. I would also encourage you to consider a standard MTM (technically> WTW) using larger woofers (6-7"), a beefy tweeter, and a relatively low steep x-over point.

Oh.. and the CAT378... well, I guess that would depend on personal preferences but the horn loading makes it less flat than many other domes at similar and lower costs, and you likely aren't going to need the extra sensitivity when paired with such small drivers anyways.

What are your box size limitations?

[MotoMan_Yz400]

Well the boxes I was thinking would be 10" wide, 4" deep (keep them flush with the tv) and roughly 27" tall. So Im fairly constrained on the box volume. With that value I can run them at their Vas. My test setup with x2 ND91s per side at a fairly loud volume, I did notice the drivers dancing pretty good, So I need the xmax they provide (another reason why I'm thinking about using the ND140s, they have x2.8 greater Sd). I've chosen the ND series because of it's black aluminum cone is much better esthetically then say a paper cone or other color cone (plus I've already bought 4 of the ND91s). I'm not trying to be 100% audiophile here, I just didnt like the steep high end drop off when listening off axis. So the SMTMS was in hope to xover before the wave length reaches the driver spacing. I thought about a simple 2 way with just 2 drivers (TW), but it seems too simple looking to me. I want something much better than the .2watt TV speakers im using now, but something that also has a "manly" look to it. Something that looks and sounds like it means business. I was also going to find a x3 - 100watt/2ch kit amps for them. Was thinking putting the 5.5" in series, the 3.5" in series, and the tweeter by itself. making 16 Ohm;16 Ohm; and 8 Ohm respectively. Hopefully that will begin to help balance the drivers. I know I'll need alot of xover compensation for the tweeter due to it's higher efficiency.

My other thought was to replace the tweeter with a single ND91. I know those can pick up around 1.5kHz easy. So it will appear as if I have 3 ND91's in the center, but 2 will be running say 600Hz to 1.5kHz, then the center ND91 will pick up from there. Im testing just a single ND91-8 tonight to see if the off axis is acceptable.

[mdocod]

Anything over ~4-5K is probably going to be pretty on-axis limited on the ND91s.

These: Tang Band W5-1138SMF 5-1/4" Paper Cone Subwoofer Speaker 264-917 ... will work in a smaller box better than the ND140 and have even more of that manly Xmax you are looking for. Also, based on the response charts, it appears to me to be very similar in efficiency to the ND140 (not great but at least not any worse).

Out of curiosity, are you dead set on utilizing the ND91s in the build or are you willing to go another route? These: Tang Band W3-1053SC 3" Full Range Driver 264-880 ... are less expensive and have a vastly better looking efficiancy. There's really no reason to have 4mm of Xmax on a 3.5" driver that you are going to use as a mid-range. The long coil winding just kills efficiency on such a small motor structure and piston size.

I would opt for this tweeter personally: Dayton Audio RS28F-4 1-1/8" Silk Dome Tweeter 275-140 .... nice and flat, can be crossed nice and low. However, there are many even less expensive options that would pair up very well into this system IMO.

[mdocod]

You've definitely got some box size restrictions there! How much clean amplification do you have to work with for this? Have you thought about skipping the heavy xmax and bass production approach here in favor of a powered sub hidden in a corner somewhere?

With so little box size, a pair of 5-7" drivers, will likely have to be installed in a sealed box, though vented is technically possible it may not provide much benefit. In either box type, the amplification costs to play deep bass are going to be pushing the limits of most home amps and pushing the thermal limits of those poor little voice coils. Lots of EQ will be required down low to bring up the bottom end... I like the look of a single bass driver per box a lot better personally, because then you can vent and tune the box to provide a relatively flat performance out much deeper, while reducing power requirements to achieve the same listening levels.

[mdocod]

Here's a ND91, ND140, and this tweeter: Tang Band 25-1933S 1" Fabric Dome Shielded Tweeter 264-900 (chosen for lower cost and low profile faceplate for tighter driver spacing).

I tinkered with all sorts of arrangements and came up with nothing that looked all that good. This is just one of many iterations of my experimentation with those drivers to see if they could be made to play nice together.

I was shooting for lowish x-over points, ideally ~600hz and ~1800hz. Was having a hard time making that happen while keeping things reasonably flat. Please understand that I am a rank amateur at this and just using your project as a chance to practice simulations. With any luck, perhaps other more experienced members will come along and critique the build ideas.

(FYI: I have the sim set up as a ~4Ohm speaker, the sets of 8 ohm mids and lows are in parallel).


Eric

[MotoMan_Yz400]

Ok I've been trying much harder than I should to get these Dayton ND series to play nice with each other in the line array configuration I wanted. Well I have abandoned that ship, and will re-purpose the x4 ND91's I have and use them as computer speakers later. For now, i think 'mdocod' was on the money with those 5.25" Tang Band woofers. So here is my new plan.

I was thinking using x4 of the TB W5's (overkill I know, but theres a reason for this) and using a single 3" or 4" for the mid/highs. In a WW(MT)WW configuration. This will make it a simple 2 way, and I can cross over around ~1kHz or so with no problem, and I dont have to worry about the wave length vs driver spacing issue as much.

I was looking at theseTang Band W4-1320SIF for the single mid/tweet.

My current concern is if I ran the x4 W5's in a series/parallel config, I get a net gain of 6dB with a resulting impedance of 4 Ohms meaning it will have a net result of ~88dB 1w/1m. The mid/tweet is also ~88dB 1w/1m so they would match nicely. But I can only find the W4's in 8 Ohms.

So my question is, have the woofers @ 4 ohm, and the mid at 8 ohm, what issues will I run into? I'm assume I'll loose 3dB in the mid/hi due to the higher impedance, and some say that having an impedance shift like that is bad for an amp. But I was wondering if anyone would help explain any potential issue I might have.

P.S. - single amp configuration. So no bi-amping here

[tinitus]

Originally Posted by MotoMan_Yz400

But I can only find the W4's in 8 Ohms.

So my question is, have the woofers @ 4 ohm, and the mid at 8 ohm, what issues will I run into?

1. mid driver Re is usually 5-6ohm

2. doesnt matter much
what matter is impedance rise, and the 'bumps'

but a design like this will still be tricky
and every time I have had the urge to do a similar thing, budget and cost of drivers mostly point at other ways to achieve the same, or better

[mdocod]

I believe (hopefully someone can correct if I am wrong on this) that a WWTWW configuration with 5" and 4" drivers would need a very low and hard to manage X-over (around or under 500hz and ideally closer to 250hz). A 2.5 way might help resolve the issue here but be prepared for some large inductor values and costs.

A nice low x-over in the 500-1500hz range is really good for a "regular" MTM configuration with 5-7" woofers because you can often get your x-over point an octave or 2 below driver spacing.

On the subject of woofer selection, I must retract my driver recommendation if you intend to use 4 of them. Such a configuration of drivers in that small sealed box would be leaving a lot of Xmax on the table un-used (The drivers would be thermally limited before ever moving those pistons anywhere near the full capability of the motor system). In fact, they are technically thermally limited (rated) to 40W RMS/80W "max" per driver, but in order to use the full 9.3mm Xmax would require nearly 200W per driver (yes, 800W per channel). Assuming you have a clean ~300-400WPC amplification, the drivers could probably technically handle transients in that realm (~75-100W), but you would still only be using ~6mm of the available Xmax.

Any time you are leaving Xmax of a driver unusable due to thermal limitations, you are effectively giving up potential efficiency because a driver with lower Xmax and higher efficiency could have been chosen instead.

The following chart shows 2 drivers vs 4 drivers in the same 600in^3 box space, sealed with equal input power. Notice that once you get down into the deep bass, they have exactly the same performance for equal power input. Once you EQ that system flat, the 2 drivers prove to be just as good as 4 drivers. The only time the 4 drivers would be any "louder" would be with more amplification power above any beyond what the 2 drivers can handle. I should note that with 2 drivers in that box, you can use the available Xmax of the drivers, at ~160W and 20hz.
(yellow = 4 drivers, orange = 2 drivers)

[mdocod]

So with that said;
How much clean amplification do you have to work with?

What are your expectations as far as listening levels?

How deep did you want the system to be able to play "flat to" (<will probably require EQ with your box size restrictions)?

----

edit in: also keep in mind that if you use a 4" wide driver to cover all the way up to 20k, the system will be sensitive to listening position. The off-axis performance of a 4" driver being used as a full range will be limited. For systems intended to fill a room that may have many listeners at many positions (TV room), I would tend to caution against a system that only favors those sitting on axis. Granted in normal home environments, the naturally reflective surfaces will provide some spacial averaging to compensate, but, well, I'd still be leary of trying it.