[Jeffrey Hedback]

Zydeco,

I was in a room last evening with speakers just as close to sidewall and similar room width as yours.  I had BAD samples as well as other devices...bottom line, the BAD Arc will give you the best control possible and really help.  One 2'x4'x4" RPG BAD ARC per side.  Simple stands will be nice and you can play with the location of the ARC.

I fully agree with Nylar on the rear wall and look forward to answers on bass response concerns.

[zydeco]

Jeffrey / Nuance / Nyal

Thanks for the words of encouragement and advice. Much appreciated.

I don't think it's time to walk the ledge of the building.  How does is sound?  I remember that once you adjusted some items on the shelving, the imaging squared away.  But, beyond that...how does it sound?

I asked about the sound. Well, the room is fine for general conversation but is noticeably quiet and, I think, damped at high frequencies. If, for example, I take my portable radio into the room then tone of the voice drops a little. This effect was noticeable when the acoustic foam was added behind the speakers. Note, prior to the addition of the acoustic foam behind the speakers there was, occasionally, a “ringing” around 3kHz but this has gone with a subtle reduction in gain on the tweeter and the foam.

So, right now, with music the main issue relates to the sound-stage which is narrow, doesn’t envelop the listener in a realistic manner, and has indistinct imaging. The interesting observations is that I’m able to pick left / right, up / down, front / back on Keith Johnson’s set-up CD but that isolation of the different instruments / vocals is much harder to real music. It also seems as though the central image is shifted a little to the upper left and, if I had to guess, has an emphasis on mid-range. Bass is present but not impactful – none of the physical impact that one sometimes feels – but more important the connection with the mid-range is somewhat lacking. This effect is noticeable, for example, in some low male vocals (e.g., Lloyd Cole)

As for other speakers, let's see others thoughts.  There are options...planar speakers would negate this sidewall issue tremendously...others designs have the ambience oriented tweeters that could work as well.  However, even a great waveguide designed like AetherAudio would suffer the same time skewed coloration.

Interesting – I’d been thinking of building a pair of planar / open-baffle speakers as my first DIY project so this makes it all the more pertinent. For now, I’ll stick with attempting to solve the issues acoustically with room treatment on the basis that the treatment will remain useful.

I'll admit, that room looks very small, but possibly doable...I dunno.  What are the dimensions of the room ( L W H)?  In my opinion you certainly need more diffusion.  Have you removed the foam from behind the speakers and compared?  With panels placed behind my speakers the sound got worse (objective and subjective).  Since your room is tiny perhaps more reflection will add ambiance.  Give it a try, then compare by listening and measuring.

The room is 16 x 12 x 10ft (LxWxH) but as seen in the photograph the effective width is a little less due to the built in furniture. Ok, I’ll report back after I’ve removed the acoustic foam from behind the speakers. Note, as per the earlier measurements, the foam seems to have an impact of reducing RT60 across the full frequency range.

 

I still think some BAD panels would be effective, and as Jeffrey pointed out it might be worth going with the curved panels (BAD arc) for the sidewall due to the high angle of incidence. I see now from the photo which I didn't appreciate before that your bookshelves do not extend all the way to the floor and in fact the area around the acoustical center of the speakers (between the midrange and tweeter) is mostly reflective. In this case it would be definitely worthwhile mounting the RPG panels to stands so that they cover an area 1ft below and 2ft above the acoustical center of the speakers.

Thanks. What, to be specific, is the “acoustic centre of my speakers? Is it correct that think that this is the tweeter so that the set-up is that shown below in which the panel doesn’t extend down to the bass drivers? And is a single panel width (2ft) enough or should I set-up two adjacent panels so that a 4ft length of the sidewalls are treated?

On the back wall I might be tempted to leave a curved piece of MDF or plywood in place if it is working out for you rather than introducing any more absorption into the room.

Yes, I’ve kept this in place as it did help with the realism of the sound. What type of diffusers would you recommend as a long-term solution for the back-wall (where the listening position is 6ft from the back-wall? It seems as though RPG Omniffuser is a good option but I’m not sure if this is more applicable, for example, to Skyline or 1D QRD Diffusor.

You also mentioned that there are still some bass issues? Can you explain further what you are hearing?

I’ve noticed that the bass isn’t that impactful and mid-bass a little lacking with low male voices and some music. One interesting point is that I did attempt to flatten the response in 4o to 60Hz via using a Room EQ software on a computer audio source but found that the results weren’t audible. The real thinking is to replace the acoustic foam with proper bass traps that are more effective – hopefully smoothing mid-bass response – while not killing the HF response.

I was in a room last evening with speakers just as close to sidewall and similar room width as yours.  I had BAD samples as well as other devices...bottom line, the BAD Arc will give you the best control possible and really help.  One 2'x4'x4" RPG BAD ARC per side.  Simple stands will be nice and you can play with the location of the ARC.
 

Thanks again. Just saw this post after I drafted the response to the questions.

Regards,
Zydeco

[Nuance]

The room is 16 x 12 x 10ft (LxWxH) but as seen in the photograph the effective width is a little less due to the built in furniture. Ok, I’ll report back after I’ve removed the acoustic foam from behind the speakers. Note, as per the earlier measurements, the foam seems to have an impact of reducing RT60 across the full frequency range.

Regards,
Zydeco

I follow ya.  But, and correct me if I am wrong, acoustic foam doesn't absorb the midrange and lower frequencies very well.  Isn't it more geared toward HF absorption?  If so, you certainly do not need any of that, at least based on your measurements.  Also, according to Floyd Toole's books some reflections in smaller rooms can be good, as they add ambiance in the timbre, spaciousness and a better depiction of the venue the recording was made in.   Of course, this all depends on the room and speakers used...

Lowering RT60 is good, but isn't that more applicable to large rooms?  I'd be looking more at the FR, spectral decay and waterfall plots.  If it were me, I'd ditch the foam and instead pick up some bass tri-taps from someone like Real Traps or Gik Acoustics.  That foam just won't cut it, in my opinion, as its aborption properties are nothing in comparison to rockboard or 703/704 material.  Tri traps behind the speakers in the corners, floor to ceiling will greatly help with the lower frequency ringing, and probably smooth out nulls and peaks.  Another set in the corners behind the LP will help too.  After that I'd listen and re-measure.  If still not happy, put some first reflection side-wall panels up and a panel on the ceiling to handle the floor/ceiling bounce.  Listen and measure again, perhaps adding some diffusion panels.  This won't be easy, but its IS doable. 

Just my $0.02.  Good luck to you.

[Nuance]

So, right now, with music the main issue relates to the sound-stage which is narrow, doesn’t envelop the listener in a realistic manner, and has indistinct imaging. The interesting observations is that I’m able to pick left / right, up / down, front / back on Keith Johnson’s set-up CD but that isolation of the different instruments / vocals is much harder to real music. It also seems as though the central image is shifted a little to the upper left and, if I had to guess, has an emphasis on mid-range. Bass is present but not impactful – none of the physical impact that one sometimes feels – but more important the connection with the mid-range is somewhat lacking. This effect is noticeable, for example, in some low male vocals (e.g., Lloyd Cole)
Interesting – I’d been thinking of building a pair of planar / open-baffle speakers as my first DIY project so this makes it all the more pertinent. For now, I’ll stick with attempting to solve the issues acoustically with room treatment on the basis that the treatment will remain useful.

This might sound silly, so forgive me if you've already done this...

1) Sometimes in an attempt to resolve an annoying situation we overlook the easy stuff.  Are you measuring your speakers one at a time, then averaging the two measurements to get an idea of what you're actually hearing?  You cannot measure both at the same time and get an accurate depiction.  Also, what microphone are you using again?  Is it calibrated, and are you using the cal file for it?

2) Have you used REW to level match your speakers?  If one speaker is even 1dB louder than the other, the sound stage can collapse and imaging will suffer greatly.  As mentioned above, measure each speaker separately, then overlay them to see if they are matched (use the measured tab in REW).

[Nyal Mellor]

Thanks. What, to be specific, is the “acoustic centre of my speakers? Is it correct that think that this is the tweeter so that the set-up is that shown below in which the panel doesn’t extend down to the bass drivers? And is a single panel width (2ft) enough or should I set-up two adjacent panels so that a 4ft length of the sidewalls are treated?

U got it! Remember when dealing with room acoustics that the frequency range can be broken down into low frequency below around 300Hz and high frequency, above about 300Hz. It is not a hard rule, more a guideline. The panels are there to assist with the sound quality metrics of soundstage focus and envelopment which are mostly related to high frequencies.

These panels should help with the issues with imaging precision and fix any problems of images not being centrally located. Your complaint about soundstage width I don't think you are going to resolve in that room (it is physically narrow). To increase envelopment you will need to further work on getting lateral sound energy into the portion of the room behind your head.

And re increasing bass sound quality - Nuance is right - foam isn't as effective as fiberglass in this application. The general rule of thumb being that you need twice as much thickness of foam to have the same effect as fiberglass. Take a look at the absorption coefficients of the two materials.

[JoshK]

WTH is BAD?  Treat first reflections.  Check out realtraps.com for a good tutorial on how to determine first reflections.  Nothing else will work as well as treating first reflections, period.

[bpape]

BAD is an abbreviation for Binary Array Diffuser. 

Bryan

[Nuance]

WTH is BAD?  Treat first reflections.  Check out realtraps.com for a good tutorial on how to determine first reflections.  Nothing else will work as well as treating first reflections, period.

Really?  In his small room I bet he yields better results from using bass traps.  Check out Floyd Toole's book, by the way; some first reflections are good, depending on the speakers, of course.  Saying "nothing else will work as well as treating the first reflections, period" is untrue, as its highly situation based. 

Bryan from Gik would know best, but I'd start with four tri-traps (bass traps), floor to ceiling in each corner.  If you fix the bass, other frequencies will be effected (less midrange smear caused by said bass issues).  A panel on the ceiling to absorb ceiling bounce would be beneficial as well IMO.

Oh, that coffee table probably isn't helping...  Have you tried moving it, zydeco, just for kicks?  If you do, take some measurements too.  Also, do you still have the acoustic panel behind the listening position, as is illustrated in the picture in the OP?  If so, is it foam?

[Nyal Mellor]

Nuance / JoshK

There are 'good' first reflections and 'bad' first reflections. The 'good' ones are those that increase what Toole classifies as ASW (apparent source width). These are the lateral reflections from the left and right side walls. 'Bad' reflections are those that don't have any beneficial effects, but rather tend to decrease sound stage focus and clarity. These are back wall, ceiling and floor. The other reflection is from the front wall, which for straight monopole speakers (i.e. boxes) is considered bad but for dipoles some recent work by Linkwitz has suggested that these are actually 'good' (see the discussion on the wall effect on his site).

What to do with left/right side wall reflections in particular is to some extent personal preference related. Those who like pin point imaging will like absorptive sidewalls, at the expense of sound stage width and the 'spaciousness' of the soundstage. Those that like the opposite will prefer reflective sidewalls. Then there is the use of combination absorptive / diffusive products which absorb some of the incident sound whilst diffusing some back into the room. This sidewall treatment will suit those who go for a more of a balance between focus and envelopment.

Hope that helps to clarify! If not then sorry for confusing things more!

[Jeffrey Hedback]

Nylar...BEAUTIFULLY stated!

Looking at the "Toole" beneficial lateral reflection concept in relation to Zydeco's room.  It is not possible to leave the sidewalls un-attended and not have tremendous smearing, image shift and phase coloration.  Thus the suggested combo panel, RPG's BAD ARC, which offers: absorption in upper bass and lower midrange, diffusion in midrange and due to the polycylindrical arc...scattering of energy that might still be specular (angle of reflection= angle of incident) due to proximity of drivers.  Not that the BAD ARC is a single cure-all to all issues at hand...but it is the best approach for this major issue and then remaining concerns can be much better identified and addressed.

[Nuance]

Nuance / JoshK

There are 'good' first reflections and 'bad' first reflections. The 'good' ones are those that increase what Toole classifies as ASW (apparent source width). These are the lateral reflections from the left and right side walls. 'Bad' reflections are those that don't have any beneficial effects, but rather tend to decrease sound stage focus and clarity. These are back wall, ceiling and floor. The other reflection is from the front wall, which for straight monopole speakers (i.e. boxes) is considered bad but for dipoles some recent work by Linkwitz has suggested that these are actually 'good' (see the discussion on the wall effect on his site).

What to do with left/right side wall reflections in particular is to some extent personal preference related. Those who like pin point imaging will like absorptive sidewalls, at the expense of sound stage width and the 'spaciousness' of the soundstage. Those that like the opposite will prefer reflective sidewalls. Then there is the use of combination absorptive / diffusive products which absorb some of the incident sound whilst diffusing some back into the room. This sidewall treatment will suit those who go for a more of a balance between focus and envelopment.

Hope that helps to clarify! If not then sorry for confusing things more!

Not confusing at all.  What you said is exactly what my understanding was on the subject, but I am no expert so I wasn't 100% positive.  Thank you very much for the explanation.    It is also my understanding that Toole's explanation of the ASW and "good" lateral reflections only applies if the speakers have very good off-axis measurements/dispersion. 

By the way, I was not saying zydeco didn't need side-wall first reflection panels at all.  I was simply responding to the statement (below) in which JoshK stated his opinion as fact.  I have had a sore spot for people who do that lately.    In fact, he may greatly benefit from such panels because, since his room is not wide enough to experience an extremely large sound stage, he might as well not worry about the negatives and instead use them to improve other aspects (midrange smearing, imaging, etc).  Correct me if I am misunderstanding. 

Nothing else will work as well as treating first reflections, period.

Edit: oops - I see Jeffrey already explained that using the BAD panel would probably benefit zydeco; sorry. 

[zydeco]

Nuance / Nyal / Jeffery

The last couple of months have been hectic with work and 6-weeks of house guests. Over the past few weeks, though, I've had a little more time to listen and tinker with the set-up. The issue has been the lack of ambience, as well as some image distortion, associated with narrow room / short-wall set-up. Funds have been a little tight so, for a while, I've persisted with some crude diffusers on the left / right and rear wall first reflection points.

A week or so ago I decided to trial a near-field set-up. This set-up saw the speakers about 5ft apart angled in to fact a listening position located at the apex of an equilateral triangle. Such a set-up gave me a about 1.5ft gap to the sidewalls along with a 10ft gap to the rear wall. This approach centred the image and made it more precise but the soundstage was a lot smaller while the lack of envelopment continued for all but the best recordings. I'm not sure that I optimised the set-up but, all in all, it didn't recreate the natural soundstage that I'm after and sitting on top of the speakers wasn't relaxing.

So, one evening last week, I threw caution to the wind and set-up the speakers along the long-wall. This set-up allows the speakers to be about 7.5ft apart with a substantial gap to the sidewall with the downside being that the listening position is about 1ft from the wall / book-case. The irony is that half-way through this process one of the shelves / bookcases on the front portion of the sidewalls was requisitioned for another room.








I'd not expected this arrangement to work but all of a sudden it seems as if the soundstage is both more precise and larger I've not had a chance to address bass response but hope to make some measurements over the weekend. The strange thing is that at times the soundstage almost snaps into focus - with a three-dimensional feel - but at other times it collapses back to the two-dimensional (but large) soundstage. I'm not sure of what needs to be done (either in speaker set-up and/or room treatment) to keep the three-dimensional image.

Regards
Zydeco

[zydeco]

Some more listening and tinkering with the room / speaker set-up. The sound-stage width / height seems good but the image is two-dimensional - lacking real depth except on the best recordings. What does one do to gain a deeper sound-stage?

Regards,
Zydeco

[richidoo]

To optimize soundstaging the front wall should be empty. The furnishings are reflecting and confusing the image. Don't put any soft stuff behind the front baffles.  Pull the speakers out as far as practical to increase depth illusion. Make sure tweeters are exactly the same distance from your ears. All this stuff will simplify the reflections coming to you and increase image coherence.  Add absorbtive treatment as desired, while listening to each piece as you add it. Each speaker is different, some can tolerate a lot of treatment (they are poor imagers anyway) and some cannot.

Try the Master Set placement technique to optimize flat EQ, it works very well but takes some effort and practice - well worth it. Not related to soundstaging, but it will make them sound as good as possible while relatively still close to the wall. Search AC for Master Set. There are several threads on AC. 

Make sure your left and right walls are treated the same, whether hard or soft. This prevents the image center from skewing to the right or left. It will move toward the softer side.

The tweeter and mid drivers are kinda far apart, but nothing you can do about that. Just make sure the tweeters are exactly at ear height, that will place all mids at equal distance to ears too. The passive crossovers have phase error inducing parts. Again,  nothing you can do about it now, but think about it when you upgrade, consider active crossovers. Rounded corners reduce edge diffraction. It all adds up, everything is audible.

Use a low distortion amp with very low phase error across the frequency band. Minimal feedback, or well designed feedback loop are important to keep the feeling of life in the music and preserve the subtlest spatial cues. Only the best designers can do feedback well. The others just diss it. Poorly designed feedback adds phase error while lowering THD, as with most cheap pro audio SS amps. Tubes use less feedback but have higher distortion in other areas, but soundstaging is often better (among budget amps) if quality transformers are used.

Good luck!!
Rich

[zydeco]

Another month, another set of changes and observations and another post. Still persisting with the long-wall set-up. Current set-up sees the speakers 2.20m apart and listening position 1.85m from each tweeter - as per the Audio Physic model for long-wall speaker set-up. The other major change is that I've moved the main speakers from sitting upon the woofers to sitting upon small stands - which has improved WAF as well as allowed for optimal placement of the two units.

Starting point for the main speakers has been to set the toe-in so that the speakers face directly at the listening position. I've found that this creates a good centre image but I can hear the speakers . Yes, the soundstage width is good but it seems as if there are three sources - centre image, left / right speakers -  and feels rather un-natural. Reducing the toe-in (so that the speakers cross-over behind the listening position) but I'm wondering if moving speakers closer together to form an equilateral triangle is useful.

Your advice, Richidoo, on the set-up for sound-stage depth is much appreciated. I've moved some large acoustic foam bass traps to line up across the front wall this Another week, another set of changes and another post. Your advice, Richidoo, on the elements that impact sound-stage is much appreciated. I've put some large foam bass traps that were unused across the front wall and this does seem to have helped with sound-stage depth. Plan is to adjust speaker height with some new stands to get it closer to that of the listening position.

Zydeco

[zydeco]

So, I'm still using the long-wall set-up and, for now, focused on getting the speaker / chair locations set to optimise bass performance. The main finding has been that moving the speakers out from the front wall acts to reduce a major null centred at 185Hz. Figure below shows REW sweeps when the speakers are progressively moved from near the back wall (purple) to middle of the room (blue).





The first step was to see if adding some bass traps would remove / reduce this issue so I've added some DIY super-chunks in four room corners. These traps are made of a local product similar to 703 with the traps cut into an isosceles triangle with lengths 2' - 2' - 2'8" with each corner having a 3' high stack off the floor and 2' high stake at the ceiling (tri-corner).

Overall result is that bass is sounding better than it has in this room and the extra room around speakers from the long-wall set-up is creating a less distorted sound-stage. Solo acoustic / instruments sound great with a strong, large, centre image, good depth and tonality. Major issue, though, is that in more complex music one starts to hear the left / right speakers. Let me give an example.

A solo vocal sounds good and I'm not able to hear the speakers. If, then, a guitar is added then all of a sudden it seems as though the guitar is at the left and/or right speakers. Most disconcerting is that it can seem as though different strings are being played on the left speaker and right speaker. It's almost as if this big, deep, soundstage disintegrates due to left / right speakers.

Your help on what might be causing this issue would be much appreciated. I've wondered if it's a result of the near-field set-up (its 12' from front to rear wall so with speakers at middle of room and chair 2ft from rear wall it's just 5' from ears to speaker). Another thought was that it might be due to the highs (>4.5kHz) running a little hot so I've turned down the gain to the tweeter amplifier which seems to have helped but not eliminated the issue.

Any and all advice appreciated

Regards
Zydeco

[bpape]

If pulling the speakers farther out to get smoother bass response is hurting the imaging or cramping seating position, you can likely achieve the same null reduction by treating behind the main speakers. 

Bryan

[Nyal Mellor]

I wonder if the null you are seeing is speaker boundary interference (nulling due to phase cancellation of the direct wave from the speakers and a reflected wave from a boundary). You need to check if there are any boundaries that are causing that null and place treatment at the reflection (mirror) point to absorb the reflected wave. I have some material on my site that explains this phenomenon.

[richidoo]

Check to see if your speakers are connected with correct phase L and R. Out of phase is common cause of your problem, although I think it is not severe enough in your case to be caused by out of phase wiring.

Also, you may be sitting too close ot the speakers now after they were pulled out from wall.  Multi driver speakers are designed for certain minimum listening distance. That blue curve looks awesome, btw... Check it again sometime with the curve smoothing turned off. I think REW always has some smoothing, but it can be turned down. No smoothing is what you really want to see. But moving the speakers really made a nice difference. Maybe you need to move them back a bit to regain your listening distance.  Compromise! UGH!!

Sound stuck to speakers is often the amplifiers fault, with high group delay or significant distortions that are different between the channels. The measure of a really great amp is that is can float an image right close to the speaker without getting sucked into the speaker like a magnet.
Rich

[zydeco]

Thanks, I'm appreciative of all the time and effort that's gone into the responses. I've attempted to go through all the suggestions and, better still, can hear things getting better as the various suggestions are put in place.

If pulling the speakers farther out to get smoother bass response is hurting the imaging or cramping seating position, you can likely achieve the same null reduction by treating behind the main speakers. 

Bryan

So, I kept the super-chunks in the front corners but used those from the left / right rear corners to treat the area behind the speakers. I tried a range of trap configurations including treating the entire floor-wall boundary as well as building 2' x 2' footprint towers that were 3' high. The basic finding from the REQ measurements was that none of the front wall configuration treatment options had a material impact on the frequency plot - in all cases the shape of the curve (including the location of the null) was more or less the same.

I wonder if the null you are seeing is speaker boundary interference (nulling due to phase cancellation of the direct wave from the speakers and a reflected wave from a boundary). You need to check if there are any boundaries that are causing that null and place treatment at the reflection (mirror) point to absorb the reflected wave. I have some material on my site that explains this phenomenon.

Thanks. Your web-site is a fabulous resource. The null that I'm seeing is located at 165Hz which implies a path difference of 1.04m or 3'5"; i.e., 165Hz = 343m/s / (2 x 1.04m). Now, the speakers are about 6ft from the front wall, 4ft from sidewalls and the room has a 10ft tall ceiling so it doesn't appear as though these boundaries are relevant. One option is the reflection off the rear wall was the current listening position is about 21" in front of the back wall which is about right. This idea would seem to make sense as the specific frequency didn't change when I moved the speakers towards the front wall.  Have we made a break-through in understanding the issue? I guess that next set of measurements will involve a) putting some treatment behind the listening position and b) moving the microphone away from the rear wall to see if the frequency descreases.

Check to see if your speakers are connected with correct phase L and R. Out of phase is common cause of your problem, although I think it is not severe enough in your case to be caused by out of phase wiring.
Rich

The speakers are a 3-way active configuration. So, then, I've checked - for each pair of drivers - the left / right channel and phase using Keith Johnson's test CD. This confirmed that the set-up is correct although, to be honest, it was hard to hear the difference on the in-/out-phase options for the tweeters.

Also, you may be sitting too close ot the speakers now after they were pulled out from wall.  Multi driver speakers are designed for certain minimum listening distance. That blue curve looks awesome, btw... Check it again sometime with the curve smoothing turned off. I think REW always has some smoothing, but it can be turned down. No smoothing is what you really want to see. But moving the speakers really made a nice difference. Maybe you need to move them back a bit to regain your listening distance.  Compromise! UGH!!
Rich

Yes, I've wondered if I'm too close (now about 6') to the speakers. One interesting observation is that using a laser measurement tool to get set identical speaker to listener distances for the left / right speakers has provided a noticeable improvement. I'm not adverse to replacing the MTM speaker (that sits upon woofers) with a unit that has a single full-range driver but, first, I want to see what can be gained from getting the speaker placement / room treatment right.

Regards,
Zydeco