Surround Speaker Project Concept, need feedback

[dlneubec]

I had been thinking about doing a surround speaker similar to one Dickason did in the Loudspeaker Design Cookbook. His contention is that surround speakers present a completely different design problem than the typical LCR speaker. They should have a more diffuse sound field. His design had one front firing woofer with two side firing tweeters.

My current surrounds are RJB Seas Microbe’s, which are a single RS125 with a Seas 22TAF/G tweeter. They sound good, but are low efficiency and aren’t really able to keep up with my mains at times. They are mounted on the wall and the SOF demands a small, wall mountable solution, not a floor or stand solution. I was thinking of needing to go to a MTM or other dual driver solution, hopefully with a little larger cone than the RS125 as well.

I got to thinking about the special PE has going on the RS150-4ohm fro $23 and wondered if I could use it for this purpose. Recently someone posted this link to the surround speaker pictured below (They are selling them for $599/pair BTW). Axiom QS8v2

Based on that design and the RS150-4, I decided to try and quickly design a box to see what size it would have to be (see pdf plans below). The box is pretty small at about 7" d X 11" X 10.5" tall and designed to match the looks of my mains. They would have and up and downfiring RS150-4 in series and two front firing tweeters arranged at a 45º angle and 90º to one another. I have not given too much thought to the tweeter, but might be an RS28a or one of the Seas 27 series. I was thinking of a crossover around 1800hz with some pretty steep slopes. Below is a quick crossover design with FR, impedance and phase plots using two RS150-4’s, and two RS28a’s (both are in series and the tweeters are reverse polarity). These were done using the Clio files from PE in SoundEasy.

Enclosures are about 5.5l or so. In a quick Unibox model, they have an f3 of 96hz sealed, fb about 100hz, qtc of .738. They will play up to about 107db with 30watts before reaching xmax at about 80hz, about 7db higher than my current surrounds.

It would seem to meet the needs of a wide, diffuse sound field and increased spl. One question I have is if I go with this, what would the proper measurement procedure be? Would I make a test box, mount the drivers and measure them in place hanging on the wall? If so, would on axis be considered 45º off from the center of the speaker, since that is how the tweeters are oriented. I assume there will be a null in the tweeter response if measured 90º to the front of the speaker as one would usually do. Anyone have any thoughts on this?

My goal is to be able to make a decision in the next day or two whether it is a go with the RS150-4’s since the sale is over on the 18th. I’d love to get any feedback that folks could offer regarding this design. It is only my second complete design (crossover design and all), so I still have lots to learn. I’d sure like some feedback to help me make a decision.

Thanks in advance!

[DougyD19]

RS150 sounds good except for when you try to make them play bass and play vocals. The more excursion the RS150 has, the more distortion at higher volume levels. A 100 Hz crossover will allow you to crank the volume a good bit higher vs the 80 Hz. That's my experience in a .25 sealed box. Not sure if my cheap two way 12db crossover at 3K helped to hurt the distortion levels.

[Dennis H]

Dang Dan, I've gotta love your spirit! You don't even have the finish on the last one and you're already planning the next one. Pretty soon you'll be like JonMarsh with 4 projects going at once. Sorry, I don't have any experience with diffuse surrounds but have fun with it.

[augerpro]

I like the idea Dan. I've thought about this style too. I would measure them onwall so the BSC is correct. I would take several measurements at different angles, including one at the listening position. Then see what they look like and what makes sense to use in the XO. I don't know if there is any defined method for this style speaker.

Did you just copy the angles of the QS8? I've also considered this: http://www.av123.com/products_produc...rs&product=7.1 as a model. But the Axiom has got really great reviews so that's probably a godd starting point.

I'd be curious how you modeled that in SE.

[dlneubec]

RS150 sounds good except for when you try to make them play bass and play vocals. The more excursion the RS150 has, the more distortion at higher volume levels. A 100 Hz crossover will allow you to crank the volume a good bit higher vs the 80 Hz. That's my experience in a .25 sealed box. Not sure if my cheap two way 12db crossover at 3K helped to hurt the distortion levels.

I may have been a bit confusing in my explanation. I would be crossing these over at 100hz. The 80hz figure was just where the reached xmax at 30w, which gives a little margin for error. I would think pushing the RS150 all the way up to 3kHz with a 12db slope would probably have some distotion problems.

[dlneubec]

Dang Dan, I've gotta love your spirit! You don't even have the finish on the last one and you're already planning the next one. Pretty soon you'll be like JonMarsh with 4 projects going at once. Sorry, I don't have any experience with diffuse surrounds but have fun with it.

Hi Dennis,

You know, the challenge is half or more of the fun :E

Really though, I wouldn't be considering this at this time if it weren't for the sale urging me on! I will finish the Mentor omni's before I do anything but buy the RS150's.

[dlneubec]

I like the idea Dan. I've thought about this style too. I would measure them onwall so the BSC is correct. I would take several measurements at different angles, including one at the listening position. Then see what they look like and what makes sense to use in the XO. I don't know if there is any defined method for this style speaker.

Did you just copy the angles of the QS8? I've also considered this: http://www.av123.com/products_produc...rs&product=7.1 as a model. But the Axiom has got really great reviews so that's probably a godd starting point.

I'd be curious how you modeled that in SE.

I simply loaded up the clio files from PE's website. Since the RS150's and RS28's are 4ohm, I figured I had to connect them in series, so that's what I started with. The reverse polarity connection simply gave the best response. Of course this is all assuming these were all playing on axis, so it probably won't work like this in reality. I'm not smart enough to know how to predict the response in this situation.

I do know from my Mentor omni project that the up and downfiring woofers will most likely drop off in their responce sooner than they would in a front firing arrangement. I suppose one could measure the tweeters on axis to their baffle and then on axis to the front of the speaker, maybe at 22.5º in between. I'm not sure if one would try and design for a flat response on the tweeter axis or from the front of the speaker or somehwere in between. I hope someone has some good ideas on this.

[wmilas]

Don't know if it helps any but I actually own a pair of those Qs8v2's I'll help you out however I can.

[dlneubec]

Don't know if it helps any but I actually own a pair of those Qs8v2's I'll help you out however I can.

Thanks for the offer. I don't know yet if it will help or not. :B

What's your impression of them? Do you like the diffuse sound approach for surrounds?

[indygreg]

this is pretty much what i am doing but i decided to do an IB design to get the really flat to the wall. we finished framing my theater today and i have 2 26" wide x 15" tall openings in each side wall with unfinished space behind them. i have a stack of the rs180's and seas tweets i am about to start testing with. i plan to build a flattened wedge shaped baffle almost identical to the av123 that was posted and let the drivers vent into the unfinished space. i ran 4 speaker wire pairs to each speaker and plan to biamp them with a behringer i have sitting here. this is the VERY low effort surround project. no crossovers to build and no boxes to build. just a proper front baffle with the drivers in it. i plan to experiment with having the drivers in phase (like the axioms) and the tweets out of phase (like the av123). if one works better for music and one for movies i will wire a relay to switch them since i home ran all 4 pairs to the equipment room.

greg

[wmilas]

Thanks for the offer. I don't know yet if it will help or not. :B

What's your impression of them? Do you like the diffuse sound approach for surrounds?

Well I'm not sure. I'm still getting used to them. I used to have a pair of Klpisch Chorus II's for fronts. Academy center, Forte II's as surrounds. These were in my listening/theater room. We moved and the new house doesn't have a basement so I went with an axiom setup up in the living room mostly because WAF, but also because even I couldn't take the monster boxes in the family room.

Going for a large monopole horn speaker to a diffuse small speaker is a bit wierd. I don't have them on the wall but instead have them on the qs surround stands axiom sells. Btw the stands look great online but are only so so. If I stay with the axioms long term I'll either hang em (not likely) or build better stands.

The sound is indeed diffuse. I don't have a back wall and I'm thinking that its too diffuse because of this. One side doesn't have the reflections that the speaker was probably moddled with. The speakers are extremely forgiving in placement. Listening to something like a football game you can hear the fans around you.

The difference I think between these and a direct radiator is that the non sweet spot seats have a marginally better immersion for surround. Its marginal at best. I think I still prefer a direct radiator for the sweet spot though.

For 5 channel sound the direct wins hands down. The axiom doesn't do a bad job at all, its just that a quality direct sounds better.

Something to consider when hanging the axioms is they are alot heavier than they look. They feel like the internal volume is solid magnet.

Overall for the size and WAF and price I'm pleased.

[dlneubec]

this is pretty much what i am doing but i decided to do an IB design to get the really flat to the wall. we finished framing my theater today and i have 2 26" wide x 15" tall openings in each side wall with unfinished space behind them. i have a stack of the rs180's and seas tweets i am about to start testing with. i plan to build a flattened wedge shaped baffle almost identical to the av123 that was posted and let the drivers vent into the unfinished space. i ran 4 speaker wire pairs to each speaker and plan to biamp them with a behringer i have sitting here. this is the VERY low effort surround project. no crossovers to build and no boxes to build. just a proper front baffle with the drivers in it. i plan to experiment with having the drivers in phase (like the axioms) and the tweets out of phase (like the av123). if one works better for music and one for movies i will wire a relay to switch them since i home ran all 4 pairs to the equipment room.

greg

Hi Greg,

Do you mean you plan to have the woofers in phase with each other and the tweeters in phase with each other, but the tweeters out of phase with the woofers, or do you mean you plan to have to tweeters out of phase with each other?

[indygreg]

hey dan. i will test with everything in phase (like the qs8) and reversing 1 tweeter in each speaker (like the av123). if i find that one is not the clear winner i might rig a relay to switch them but i suspect i will find a winner. the thx manual says to run one tweet in each row out of phase and alternate the rows.

greg

greg

[dlneubec]

OK, I've went so far as purchasing the RS150-4's to do a surround project with, probably with the Seas 27tdfc, since there will probably be 2 tweeters per speaker to obtain the diffuse sound I'm looking for.

I've drawn up 3 options for the design and would like to get others thoughts, critisizms, compliments, etc. on them. :W

Attached are the plan options 1 through 3. My plan would be for them to be wall mounted, probably high on the wall above ear level, perhaps as close as 2-3' from the ceiling to the top of the speaker.

Marzen pointed me to an interesting ripple study that Dave Brown did in regards to surround design:
DBrown Ripple Study
The tool he used for this is fascinating and available online here:
Ripple Tank Applet
If you haven't looked at this site, I encourage you to do so. Very cool simulation stuff there.

I'm not sure I agree with Dave's conclusions, but it is good stuff. It seems like he primarily weighs horizontal dispersion performance in drawing his conclusions and when check the ripple on them model them mounted vertically on a wall, other problems seem to show up.

At any rate, any feedback/ideas would be greatly appreciated!

[cjd]

What are the intended goals here sound-wise?

Horizontal dispersion patterns ARE the issue for surround duty, and the general premise is that sound should be, on average, diffuse.

Dave R's conclusions are generally spot on as far as they go, but they investigate lobing patterns almost exclusively and they don't seem to look at how adding a crossover changes things.

I'm looking at surround construction options right now as well and am planning to do some experimentation and measurements. Will see how it goes.

C

[dlneubec]

What are the intended goals here sound-wise?

Horizontal dispersion patterns ARE the issue for surround duty, and the general premise is that sound should be, on average, diffuse.

Dave R's conclusions are generally spot on as far as they go, but they investigate lobing patterns almost exclusively and they don't seem to look at how adding a crossover changes things.

I'm looking at surround construction options right now as well and am planning to do some experimentation and measurements. Will see how it goes.

C

Goals I would say are as follows, not necessarily in order:

1) Small and wall mountable
2) Increased spl and sensitivity vs my current Microbe surrounds. The Microbes can't keep up with the mains for HT.
3) diffuse sound, not just horizonatally, but vertically, since this is in my living room, not an HT room and there are couches, chairs, etc. directly under where these need to hang
4) Good, clean sound quality at reasonably high levels for HT only, not used for music. Plan to drive CC and surrounds with HK AVR240 HT reciever only.
5) Reasonably flat response, maybe a bit of valley in the crossover area and not a hot top end that could cause listening fatigue over the length of a movie.
6) Drivers that are a decent tonal match to my mains, combo of stereo subs and RS180x2/RS28. (See the omni project thread)
7) Meet the wifes approval, the main reason for #1 above!

My plan is to pick a design, build a test mule and see what I can learn from in room measurements and subsequent crossover modeling. I'd like to pick the design that has the best potential to meet those goals before I build anything and I'm feedback from the forum can help me do that (or go back to the drawing board if need be).

[Dennis H]

'A' is most like many commercial designs I've see. The woofers should be pretty omni and most seem to wire them in phase. The tweeters can be in phase or out of phase with each other. I think they usually wire the front one in phase with the woofers and reverse the rear one.

[cjd]

I'm not really sure why multiple tweeters are desirable for diffuse sound unless you're trying for a dipole, in which case you get dipole radiation patterning and nulls that may well line up with the listening spot(s). This MAY not be a bad thing, but may be as easily resolved with other mounting techniques.

One of the things I want to mess with is a single center mounted tweeter and mid-woofers mounted at 90 degrees to each-other (take that classic "surround" box, put the mids where the tweeters are, and a tweeter on the side facing out). Cross below the frequency at which combing becomes an issue. It seems that should give quite broad horizontal dispersion characteristics. But then I wonder, why not just do a simple MTM completely in-wall.

C

[dlneubec]

Here's another idea, sort of a combo of direct firing and diffuse firing. Call it an MTMT, sort of, but spanning 3 sides along the veritcal axis. This way you have a direct firing M&T and a diffuse firing M&T. :E

[indygreg]

i am moving ahead with mine. i believe these will be more forgiving of design than traditional speakers. in short i think it will be easier to get a diffuse sound than it is to get good imaging. my driver is that i want something in wall that is not your typical inwall and i don't want to pay an arm and a leg. i will be designing for the way many theaters today are built with 2 layers of sheetrock and 1" of sound treatment on top behind GOM fabric. this means i have 5.5 inches of depth to work with. i framed in openings in my walls that are 26" wide x 15" tall and i am thinking to do a layout a little like the kef

http://www.platinumhome.ie/shop/product_info.php?name=KEF+Ci130DS+In+Wall+Speaker+(Pair)&products_id=644

i plan to use 2 rs180's and 2 seas tweeters in each one. i originally planned to do an IB but i am finding that i have too much furnace noise from the unfinished space getting in so i will go ahead and seal up the hole. i calculate that i can only get about 12 l of volume behind the rs180s but for surround i am hoping this will do. as i mentioned earlier i ran 4 pairs to each speaker so i can switch from bipole to dipole from the equipment room.

any input on my ideas is welcome. i am a newbie here.

greg

[cjd]

I think that what is happening with "diffuse" sound is a classic dipole figure 8 radiation pattern with a null for the listener so it's not quite localisable from the listening position. Some of the other variants may play on keeping nulls for the listener position without going dipole. I suspect that is the point.

Tweeter up on top isn't going to do a whole lot though you'll get some reflected sound and thus delay off the wall/ceiling.

C

[indygreg]

several of the designs that get great reviews are not dipoles. for instance the axiom qs8 runs the drivers in phase.

g

[dlneubec]

I think that what is happening with "diffuse" sound is a classic dipole figure 8 radiation pattern with a null for the listener so it's not quite localisable from the listening position. Some of the other variants may play on keeping nulls for the listener position without going dipole. I suspect that is the point.

Tweeter up on top isn't going to do a whole lot though you'll get some reflected sound and thus delay off the wall/ceiling
C

You can get a decent tweeter response up to a point when it is upfiring. Below is a meaurement 90º to the axis of an upfiring RS28 that I took during testing for my omni project. You can see that it looks ok up to about 5500hz. Then it rolls off about 10db quickly to about 8kHz where it levels out until about 17khz, where it drops out completely. This was a 4ms window measurement, so there is little, if any, reflected sound in the measurement. This was also with the tweeter out in the room, so there is no reinforcement from a wall immediately adjacent, so the tweeter was operating in 4pi space. I would think you might get a little more in the upper registers in this case since the adjacent wall should be making perform in 2pi space. It might not be too bad.

The same should be true of the downfiring woofer. It will be interesting to see how it measures if I decide to build a test mule for this one. Allison made a couple of speakers back in the early 80's that had upfiring woofers, desinged to be placed right against the wall.

I'm leaning toward trying option 4 first at this point.

[kgveteran]

Is it too late to say discrete isn't meant to be diffused, but rather located.I'm positive mixing is all done with direct radiating mono poles.

I know this is late in this post, but have you given direct radiators a chance ?

Just my 2cents.

KG

[dlneubec]

Is it too late to say discrete isn't meant to be diffused, but rather located.I'm positive mixing is all done with direct radiating mono poles.

I know this is late in this post, but have you given direct radiators a chance ?

Just my 2cents.

KG

FWIW, the following discussion was posted in another forum in regards to surround channels:


> Until program material mixing is
> standardized there will be no
> "correct" type of surround speaker
> configuration to use. Early Dolby Pro Logic
> with "surround sound" generated
> from a matrixed source used diffuse rear
> channels. At the other end of the spectrum
> we have digital channels, each of which is a
> discrete sound source (not extracted from a
> matrix), for which point source speakers are
> preferred. No matter which type of speaker
> you design/use it will only be the correct
> one some of the time. So, surround speaker
> designs are going to be a compromise
> solution most of the time. There is no
> "correct" answer.

Agreed that there is no correct answer, but the only reason the early dolby suround and similar schemes recommended diffuse sound is because the matrixed surround information (l-r, r-l) was mono which did not lend itself to complex pans through the soundfield. Today’s discrete surround-channel formats allow directors to be more creative in that regard. Even so, most of what you hear during movie playback in the x.1-format surround channels is ambient sound which is required to create sense of space and/or venue.

[cjd]

LOTR makes very good use of localisable surrounds...