Driver questions / considerations for high sensitivity 3-way tower designs

[KEtheredge87]

Hi Everyone,

I have questions regarding the next project I'd like to undertake, and I'm hoping you all can shed some light on them. I want to design a 5.1 HT setup for my living room to replace the Aperion Audio Intimus 5T towers and 5C center channel that I have currently. There's nothing inherently wrong with these speakers, except that I didn't build them! Of course, I can't resist the urge to make these dual purpose and make sure the L and R towers can sing in 2 channel stereo listening too!

My early questions are around arrangements of drivers and choices of driver sensitivity. My goal is to have > 90 dB of system sensitivity for these towers, and I'd like to use a MTMWW array to accomplish this (with a matching MTM... or WMTMW arranged center channel for the dialogue/home theater aspect). Assumed baffle width is 7 or 8 inches at this point. Looking for a modest size system that won't overwhelm the living room when non-speaker savvy friends / family visit!

The biggest point of confusion is this. How should I arrange my driver sensitivities for the easiest XO design and baffle step compensation? Based on early research, I should aim for the WW section to have around 3 dB of efficiency more than my mid and tweeter. The logic here being that baffle step losses will "pad" the bass response down to match the mids and highs better, resulting in LESS required padding on the mids and highs to reach an even response. Does that make sense?

I've gone through a preliminary round of sensitivity studies with the RS series of drivers from Dayton Audio. I looked at various combinations of 4 ohm series and 8 ohm parallel for the woofer and mid sections. The tweeter section is generally settled, as I picked up 3 of the newly released RST28F tweeters that I'd like to use. My general selection matrix is attached below. I am not 100% committed to using the RS drivers only, but these were the easiest entry point to the thought exercise. I also like some of the SB acoustics stuff available elsewhere, and some of the Tang Band W4 drivers available from Parts Express for mids.

So... how would you guys choose the starting sensitivity ranges for your 3-way tower designs? Am I on the right track by choosing a more efficient woofer section and counting on baffle step from a ~7-8 inch wide baffle to knock me down a few dB?

Thanks!
Keith

[Dirk]

Of this project, it's the center channel that I think will present the biggest challenge. I've designed home theater setups for a couple of friends in addition to my own (a couple of times), so its a design area where I'm happy to say I have some experience. The the center channel was always the most interesting part of the system build.

The crossover you cook up for the mains probably won't be a perfect fit for the center, mostly because the boundary conditions will be different. You'll need less baffle step, for one. If the same crossover is used, vocals come across as boxy and obviously coming from the speaker instead of presenting a seamless soundstage on 5.1 movie mixes.

My advice would be to complete the mains and then measure them, in room. Then take measurements of the center channel drivers in their final enclosure and on/in the TV stand. Use the measurements of your mains as the target frequency response for the center. Don't be afraid to measure a little lower than you would otherwise. It's going to be ugly, and I wouldn't let it drive my decision, but it's also relevant information.

I'd also pay close attention to baffle layout, with a focus on trying to maintain good horizontal off-axis response. How tall of a center can you tolerate? Could you get away with a W(T/M)W arrangement?

[jhollander]

The short answer is yes. The woofer, assuming you can get a bit of floor gain, should start about 3 dB higher than the mtm. My only other comment would be that the MTM is probably sufficient for the center channel as there's likely to be little baffle step and not much content below 80 Hz from a typical avr.

[KEtheredge87]

Hey Derek,

Thanks for sharing your experience! I've heard similar accounts of XO differences between L/R and center channels for HT. That makes sense to me, and I fully expect to design two XOs (well... 3 if you count some basic TM or full-range surrounds) for this project. On the center channel, that Intimus 5C center I'm staring at is 7.5" tall, 8" deep, and 19.5" wide. It's also a W(T/M)W arrangement, and I think this makes good sense for a 3-way center channel. If the box grows a bit for this project, I'm ok with that. My L and R are already going to be somewhat bigger than the Intimus 5T speakers (6.25 wide x 8 deep x 37 tall)

What makes this Intimus speaker setup so odd... the more I look at it, the more I'm sure that Aperion Audio is using the same china OEM that Peerless/Tymphany is using for their truncated frame woofers. 

Anyway... do you have any advice on "sensitivity structure", I'll call it? I'm still wondering if I should try to match sensitivity on the woofer, mid, and tweet section at the beginning of the design, or if I should aim for a more efficient woofer section with the expectation it will appear to decrease due to baffle step losses? I'm trying to decrease the amount of eventual padding that will be needed on the mid and tweeter sections.

- Keith

[KEtheredge87]

The short answer is yes. The woofer, assuming you can get a bit of floor gain, should start about 3 dB higher than the mtm. My only other comment would be that the MTM is probably sufficient for the center channel as there's likely to be little baffle step and not much content below 80 Hz from a typical avr.

Thanks John! I had starting thinking about the center only really needing an MTM as well. Nice to know I'm on the right track here. Honestly, I'm surprised to find myself struggling to best Navy Guy 's Boomsticks in the F3 department. Kerry's write-up says he got an F3 around 48 Hz from a set of four RS100-8's. I was thinking two of the 6" or 7" RS drivers should have enough cone area to at least match that, but BassBox is really twisting my arm with an EBS alignment to get in the low 50's.

I like the Dayton RS drivers for this project since it will be relatively easy to timbre match all the speakers for HT purposes, but I may have to look outside of the PE umbrella at some SB acoustics stuff (even without the upcoming 20% off MWAF sale!)

[Navy Guy]

[Navy Guy]

Thanks John! I had starting thinking about the center only really needing an MTM as well. Nice to know I'm on the right track here. Honestly, I'm surprised to find myself struggling to best Navy Guy

The RS180 will do close 40 Hz flat. The RS125 models better than the RS150. Are you crossing these to a sub? If you are, any of these should be fine for what you are doing.

[jhollander]

That's a good point. The old RS28A liked a contour filter that would be difficult with higher sensitivity designs. RS225 and you are in the 30's. If you want to cross below 200, you could do a side woofer.

[KEtheredge87]

Thanks Kerry! since the RST28F tweeter is published at 93.5 dB sensitivity, it sounds like I should be targeting almost exactly 90 dB sensitivity for the woofers instead of aiming high and assuming I'll average out to around 90 after baffle step losses. I agree with your statement that woofers dictate the sensitivity, although already having decided on this tweeter (because it's new, I bought 3, ... and why not?! ) has kind of turned the logic train around. Now I'm looking for woofer combos that are slightly less efficient than my expected tweeter level!

The RS180 will do close 40 Hz flat. The RS125 models better than the RS150. Are you crossing these to a sub? If you are, any of these should be fine for what you are doing.

I also assumed a pair of the RS180 should get close to 40Hz in a vented alignment since it's a 7" driver. Bassbox keeps telling me that I'm wrong, and the F3 keeps coming in around 60/70 Hz depending on whether I model the aluminum version or the paper version. Maybe I'm doing it wrong... I'll try modeling in WinISD later tonight as a sanity check.

Ultimately, I will have these crossed to a modestly sized subwoofer or two in my living room. I'm not entirely concerned about the bottom end extension of the final application, but I would like these towers to hold their own in a two channel stereo setup... or competition

Of course... my design goals have to compete with the ever important WAF too! Cabinet size probably shouldn't be as big as a full grown man just to satisfy my bass goals!

That's a good point. The old RS28A liked a contour filter that would be difficult with higher sensitivity designs. RS225 and you are in the 30's. If you want to cross below 200, you could do a side woofer

.

John, I had thought a bit about the side firing woofer option after seeing Geoff and Chris's Tundras and Kerry's Juggernauts at InDIYana two weeks back. I thought both of those sounded cohesive, and not like the woofer was pointing a different direction. I think that's a combination of room interactions and crossing below 200 Hz as you mention. Man... going this way almost made me think "active 4-way"... almost!

I'll have to think about the side firing woofer idea some more (as a passive 3 way). Thanks for bringing it back to my attention!

[marvin]

 My goal is to have > 90 dB of system sensitivity for these towers, and I'd like to use a MTMWW array to accomplish this....Looking for a modest size system that won't overwhelm the living room when non-speaker savvy friends / family visit!

Keith, how can you use MTMWW and "modest size" in the same sentence? I am interested in your high efficiency 3 way modest size concept but an MTM will automatically require roughly 6-8" more height than a TMWW would require. Does the MTM give you some advantage over the TM configuration that is worth the extra height? Do you need two mids to achieve the efficiency you want?
 

[KEtheredge87]

Hi Marvin! Yeah... I suppose "modest" may have different subjective associations in different people's minds. When considering a new project to replace an existing setup, I tend to think in terms of a baseline and subsequent deviations from the norm. In this case, that means I'm designing cabinets vs. the Aperion Intimus 5T and 5C setup that's in my field of view while on my couch! If the final cabinet size for the new project is only a little bigger than those (I know... another subjective measurement!) then I'll consider the "modest" box checked.

There's a few thoughts in my head regarding MTMWW's vs. TMWW's:
1) I like the "more drivers" kind of look. Makes me feel like Tim Allen's character, Tim "The Tool Man" Taylor from the old Home Improvement TV sitcom (insert grunting noise here )

2) I wanted to explore the MTM concept because of reading about home theater speakers I've done. I saw several suggestions that using an MTM / D'Appolito arrangement helps with "controlled directivity" or controlling the lobing behavior from the tweeter in these situations. This *should* result in clearer dialogue during movies and TV shows, as well as good music listening when sitting on-axis with the tweeter. Without that extra mid above the tweeter, the high frequency lobe tends to expand/angle upward, which isn't necessarily bad, but simply a different approach to the sound.

3) As you suggested, using a second mid will give me a sensitivity boost that I can use to my advantage when padding the response to level match, and when going for strong power handling. This of course might come at the expense of an emptier wallet when all is said and done! This boost to sensitivity may not be required to meet my >90 dB efficiency goal depending on the choice of driver... but see #1 above! (insert more "Tool Man" grunts here!)

Lastly... the overall tower height. At the moment my 5T towers are ~37 inches tall, including the mounting base. I'm working off my design having an equivalent tweeter height (about ~34" off the floor) when all is said and done. Adding another 4-5" mid above this tweeter height (and cramming the mids as close as possible to the tweeter axis - per best practices on MTM designs) I should be able to get these towers right around 40 inches tall. That feels OK to me just thinking about it. I'll make a rough cardboard mockup and stick it in my living room for a day or so before cutting any wood just to get the idea. Of course, any wood cutting is a ways off right now. I'd like to use the MWAF tent sale to my full advantage this year and save some money!

Thanks for your questions! I think I started to ramble there... go figure. Let me know if I missed a question!

Keith

[jhollander]

Some of the DSA's on the bottom might work well.

[KEtheredge87]

Some of the DSA's on the bottom might work well.

OoOoOoh... Hadn't considered those before. The 5 bolt pattern is different. Maybe good different, maybe not. I'll let that ruminate in my brain. Also considering the ND series now for grins. The "problem" of choice is a good one to have! Maybe a beauty board could be used. This is so much more fun to think about than my actual job (... tell no one!)

[Kevin K.]

Keith, I'm not having any troubles reaching 40's in BB with the RS180's. Are you inputting the latest driver parameters or using the existing library? Most of BB's suggested EB alignments are tuned too low and this causes a higher F3. Try raising the tuning. Make sure your damping is set to "Minimal", that should be fine for woofers in a three way. Also, you don't need a perfectly flat response. Take into consideration where you would be crossing these in a three way. A little bit of rise isn't usually an issue since they'll get crossed as quickly as they will.

Another driver that isn't on your list is the RS180S-8. It's the shielded Aluminum version (not that you need the shielding). It might be a good fit for your application, models really well.

[Millstonemike]

Some of the DSA's on the bottom might work well.

And there's a deal to be had in this thread for a pair of new DSA315-8s. WinISD shows an F3 of 30 Hz in a 3.0 ft³