[ebag4]

Between Mike's recent thread on REW and Danny's 'Room Room Room' thread, I was inspired to measure my system, something I have been wanting to do for a few months.  I am just getting started, first time using REW with this equipment, Dayton UMM-6 USB mic, MacBook Pro using it's optical out into my Vinnie Rossi LIO DAC 2.0 optical input.

The room is 16'x28'x cathedral ceiling up to 10'.  The room is open to two other spaces; 8'x'11' and 7.5'x6.5' both with 8' ceilings.  The only sound treatment is a pair of PiAudio Diffusers, one behind each speaker. I sit about 8' from each speaker.  The speakers are 49" from the front of the baffle to the front wall.

I located the microphone at the listening seat and just worked with the right speaker this go around.  Most graphs are with both the Otica and the sub running, I did take one with the Otica only.

Here are the readings with various changes made to phase settings, moving the speaker and changing toe in:



R Otica Only


R Combined, Phase 0


R Combined, Phase 0, no toe in


R Combined, Ph 90


R Combined , Ph 180


R Combined, Ph 180, Speaker moved front baffle 43" from front wall


If anything occurs to you how I might tame the 40 Hz peak or the dip at 63 Hz please throw it at me. Right now it appears that thing got worse when I moved the speaker from it's original location (2nd graph).

Best,
Ed

[HAL]

You should be able to use the parametric EQ on the servo amps to drop the 40Hz peak down by adjusting the controls.   You can adjust the controls and take measurements to find the correct settings.  Do one channel at a time as you did here.

I would just turn on the subs without the mains to adjust controls first then add in the mains for a check. 

The 63Hz dip appears in both the Otica and combined response, so that looks like a room cancellation.  See if it is still there with just the sub on.

[JLM]

Be careful not to try to boost that 63 Hz dip.  A 15 dB dip equates to 15 fold power increase which could clip the amp and blow out your woofers.  Note that the ear "covers over" narrow dips anyway.  If you are hell bent on eliminating such dips, reduce the overall output signal by the same volume (15 dB).  Your overall maximum volume will be reduced, but you'll save your amp/speakers. 

DSP can easily used to reduce the 40 Hz peak.

Most recommend limiting DSP to under 300 Hz (room effects). 

[artur9]

Since the dip is always there it must be a room induced null.  You can move the sub so that the null is not where you listen.  Maybe you can move it to where the SO sits

[Jonathon Janusz]

Since the dip is always there it must be a room induced null.  You can move the sub so that the null is not where you listen.  Maybe you can move it to where the SO sits

Although an entertaining thought at least for a minute as a playful joke on one's significant other, in this case, that isn't quite practical as the "sub" is actually the OB servo woofers that make up the lower frequency producing part of each of the main speakers - an NX-Otica MTM upper section atop a pair of OB servo woofers, one set of all the above for each channel.  With as high up in frequency as the OB servos play in this speaker design as well as the fact that the enclosure for these is the stand for the MTM, separating them from the MTM section in the room won't really work.

DSP can easily used to reduce the 40 Hz peak.

Most recommend limiting DSP to under 300 Hz (room effects). 

As I've read you basically shoot from the hip with this solution frequently across multiple threads recently (ad nauseam?), I have to ask if you have any thoughts on a solution that wouldn't involve: demolishing and rebuilding an entire room/house/building to "ideal" parameters, selling the entire system and replacing it with headphones, selling the entire system and replacing it with a "proper" active speaker design, or not buying a pile of electronics/software to DSP the source signal into an unrecognizable shadow of what it once was?

I'm not arguing that in this case a little judicious use of the single band parametric EQ already built in to the servo plate amps to shave down a single peak a bit (maybe I state my case above a little dramatically) is out of line, but although a hammer is a fine and good tool, for sure, not every problem is, can be, or should be a nail.

[nickd]

It is not trendy in the hobby yet, but DSP well implemented is getting less expensive as technology evolves. Once you have expierenced it, you will either rejoice in the improvements or get sick realizing the wasted cash over the years changing speakers and gear looking to make things sound better.

I still have my turntable, I’m not all digital. But I don’t miss tubes or class A solid state amps.
Technology is not all bad.

[Jonathon Janusz]

I'm neither saying it is nor am I against the (strategic) use of "technology".  All I was trying to (politely, diplomatically, or at least constructively) say is (nearly) always suggesting scrapping/replacing/changing (even by implication) everything already in place simply to use "technology" or suggesting "technology" as the first solution to all problems - regardless of practicality, applicability, or relevance of implementing such a solution (no matter if it would work in the end or not) perhaps demonstrates - giving the benefit of the doubt in offering constructive insight - at best enthusiasm without even a few moments pause to understand a situation - as it currently exists - before evangelizing the "one true path into the future" [my words].

Using the specific challenges presented in this thread as an example, taking a few minutes looking at the system under review as a whole and as it actually is, in no part of that system (please correct me if I'm wrong) is there anywhere to currently implement a source side (or even mid stream at the servo amps, as if I remember correctly the adjustment functions are in the analog domian?) DSP based solution, nor any inference that making such a (relative) drastic change to the gear being used is either desirable or under consideration.

My tongue in cheek (mostly) rhetorical question was really intended to point out how metaphorically suggesting to use a pool cue to sink a golf ball on a putting green, although clearly functionally effective and arguably advantageous, isn't as helpful a suggestion as starting with solutions that acknowledge that one is playing golf in the first place and thus accepting the basic rules and limitations within the game of golf when discussing how to best sink a putt.

[note that this isn't saying that I can either confirm or deny that I may or may not have in the past used the handle end of a putter to bank a (few...) putts on a mini golf course... ]

My apologies for the hopefully slight aside as I kindly direct the conversation back to our scheduled program.

[mlundy57]

Ed,

In my case a phase control position of about 10:30-11:00 o’clock, rather than 90 degrees (12:00 c’clock) had the best results at bringing up my 60 Hz dip in both channels.

For the 40Hz peak I used the PEQ. For the left chanel Gain is at 1 o’clock, Bandwidth is at minimum, and Frequency is one click past 12:00  o’clock ( about 12:01).

For the left channel, Gain is at 3:00 o’clock, Babdwidth is at minimum and Frequency is at 12:01 o’clock.

That doesn’t mean these settings will work in your room, just what worked in mine

Mike

[Captainhemo]

Ed,    as  rich stated, use the PEQ (maybe start with settings close to what Mike states to get you close as yo are both trying to  knock down a  40hz peak)  and try  moving the subs further  as well as   closer to the front wall to try and  deal with the dip. 

Moving the subs   can   make   signifigant differences....

jay

[ebag4]

Thanks for the responses guys.  I wasn't able to get back to this until this morning.  After moving the speakers and adjusting the sub amps PEQ and other settings, this is the best I can come up with.  In each of these I have the Otica and sub section playing.  I worked on the right speaker, then the left, and finished by measuring them together.  The speakers ending up several inches closer to the front wall than when I started, they are now about 38" from the front wall.  Here are the response graphs:



Right Speaker


Left Speaker


Left and Right Speaker Responses overlay


Left and Right Speakers Measured Together

The left speaker has a wall to the left with a leather love seat sitting there whereas the right speaker has an opening to another room to the right of it, I don't know if that could be causing the suckout around 70 hz or not, unfortunately there won't be much I can do with treatments in this space.

Best,
Ed

[mlundy57]

Ed,

That's what the numbers say. What do your ears say?

Mike

[ebag4]

Ed,

That's what the numbers say. What do your ears say?

Mike

Hey Mike, my initial knee jerk was that I lost some soundstage depth, but as I listen to my go to tracks I don’t believe that’s true.  This system has sounded pretty great in this space but imaging has improved focus, the music seems better balanced.  These are all initial impressions.  One thing I changed after measurements seemed to help and may be the reason for the better overall clarity, I moved the diffusors in between the speakers (in front of a window).  I will take some measurements with the diffusors in place to see what changes they make.

Best,
Ed

[ebag4]

I made a few more measurements today with the diffusers in different locations, I had one unexpected response, it surprised me enough I ran it again, here they are:



Left Speaker without Diffuser


Left Speaker with Diffuser at Window


Left Speaker in Front of Rack


Left Speaker Overlay

The reduction of the 40Hz peak is what surprised me here, I didn't think the diffuser would have any impact on these bass frequencies.  I measured it again to verify, same result.



Right Speaker no Diffuser


Right Speaker with Diffuser at Window


Right Speaker with Diffuser in front of Rack


Right Speaker Overlay

I also tried moving the mic from my current listening seat to several locations in front and behind, it became obvious you would have to tweak the setup for each location as the suckouts and peaks changed with each position change, regardless, I am limited within a couple of feet where my listening seat can be located from a practical aspect. 

I also moved the mic to about 3' from each speaker, here I noted a difference at high frequencies with the mic pointed at the speaker vs having it point up.  For the rest of the measurements the speaker was pointing straight up.

Best,
Ed

[Tyson]

Believe it or not, that's not terrible response.  Can you set it to 1/8th and then 1/3rd octave and show those smoothed graphs?  That's generally what you see in magazine and other professional reviews.

[Captainhemo]

Believe it or not, that's not terrible response.  Can you set it to 1/8th and then 1/3rd octave and show those smoothed graphs?  That's generally what you see in magazine and other professional reviews.

Yeah, and more often than not...  they don't use a  5db scale .    graphs tend to get  prety darn smooth when they use a  10 or  sometimes even higher db scale 

[ebag4]

Thanks guys, for troubleshooting the less smoothing the better I would think, but for your viewing pleasure, find the 1/6th and 1/3rd smoothed graphs below:



R Speaker w Diffuser 1/6th Smoothing


L Speaker w Diffuser 1/6th Smoothing



R Speaker w Diffuser 1/3rd Smoothing


L Speaker w Diffuser 1/3rd Smoothing

I tried to recreate the peak reduction I saw earlier in the left speaker, I could not.  Apparently there was some other anomaly with the earlier reading.

Best,
Ed

[mlundy57]

That's when it really gets irritating. You stop for the day then when you start back up you get different readings than when you left off. Grrrrrrr!!

[Danny Richie]

I also tried moving the mic from my current listening seat to several locations in front and behind, it became obvious you would have to tweak the setup for each location as the suckouts and peaks changed with each position change, regardless, I am limited within a couple of feet where my listening seat can be located from a practical aspect. 

This is why digital room correction does not work. After corrections are made, just move the mic a little bit and you have to start all over.

That diffuser might make more of a notable difference above that 400Hz range that you are limiting your measurements to.