Description

I have not updated this in more than a year. Finally sold the Sarastro 2s to a friend. The Sarastros were not a good match to my room. The rear firing config. excited the longitudinal mode significantly. They have some non-linearities that I cannot effectively compensate for given my room limitations. This is not to the speaker's fault as they sound much better at my friend's larger listening space. To say the speakers are too big for the room would only be partial truth.

This is not due to lack of effort. I went thru acoustic consulting with Rives and then local expert Bob Hodus. Read up on Handbook of Room Acoustics and Sound reproduction. After the consultations, I did a lot of experiments to improve the room acoustics. Rm is 22'x14' x 10'. Dimensions are not bad but one side wall has windows occupying >2/3 of surface area. There are also a door & a fire place along the mid axis of the room, two walls were interior walls which are less robust in bass reinforcement. Room is at mid level with living space above and below. Previously, after reading the Get better sound book, I had settled for a listening location at 4ft from backwall. This allowed for boundary bass boost which overcame a bass cancellation problem (mid axis, quater wavelength listener to boundary) at 35-40hz. However, primary length mode at 25hz (7db) muddled up the sound. Despite treating the rearwall, the comb filtering effect from the rearwall markedly reduced transpancy and smoothness.

Rives Audio: I went with level 2 consultation and also bought the room measuring kit to allow for repeated measurements along the way. I was quite weary of the actual translation of mathematical modeling into the field. Richard was very helpful in educating me to use ETF and BARE. He was excellent in recommending RPG products, different type of curtains, blinds for different applications. Rives also had all the contruction methods for bass traps and diffusors. Although I preferred the looks of built in traps and diffusors, I went the way of RPG and GIK. My friends had ripped out acoustic constructions that they didn't like and that's just too painful for my wife. Room tuning is complex and requires much trial and error. Base of my previous experiences, I was also worried about resonances of wood diffusors (ceiling and sidewalls). Rives ultimately set the listening location at 9ft from the rear wall. The result was much improved clarity and smoothness along with reduction of room modes. The suckout at 35-40hz is still present but reduced in bandwidth and severity. Richards felt that is acceptable in my case as there is smoothing in human hearing. Bass was flat on psychoacoustic curve on BARE. (I thought this has something to do with combination of gating and smoothing. I could produce a similar curve on Fuzzmeasure when I played with smoothing and gating time.) Rives also had even handed approach in balancing RT 60, freq response, absorption and diffusion.

Bob Hodus: I seeked the help of local expert Bod Hodus in hope of eliminating the suckout entirely and also to provide a second opinion before engaging in a sofit / ceiling construction proposed by Rives. Bob was a pleasure to work with and very willing to educate. His approach was entirely hands on. He focus mainly on frequency response curve and uses absorption mostly. Using continuous test sweep, we went thru inumerable listening locations, reconfiguration of bass traps and speaker locations. At the end of the day, he got rid of the suckout entirely, low bass 20-40hz was linear without boost or suckouts. It is done by sitting 12 ft from the rear wall. THat left me with only 10ft betw me and the front wall, not a lot of room to position the rear firing Sarastros. I was excited with the new result and invited listening group over. All felt low bass was excellent but trade off was made in clarity of midrange as speakers are too close to the wall.

I thought about this for a while and kept rereading the Handbook of Room Acoustics. "Bass is best thought of as waves or pressure zones and high frequencies are more similar to light rays (hence the term specular reflection)." Although my room is symmetric in dimension, the asymmetric construction allowed for the most even bass pressure distribution at the 12 ft point. What if I were to set up the speakers along the opposite wall and check if bass linearity persist at the same spot. Sure enough, the bass alignment remained the same and I got 12ft betw me and my new front wall. This allowed for more room for the speakers to breath. It also prompted me to choose a speaker with no rear wiring port/woofers. I went with the Isis after auditioning Rockport, YG, Magico. I am quite happy with my third pair of Avalons.

After the bass foundation was set, I proceeded to adorn all surfaces and experimented with a few tweaks.

Frontal wall: Tried RPG skylines, hemifusors, abfusors, wood constructed diffusors, BAD Arcs. The wood constructed diffusors were the worst. They diffused effective but they also sounded. THere was marked resonace in 150-200hz range measurable on both Fuzzmeasure and ETF. Subjectively, they were like adding more speakers in the room. They enriched the wood tone during playback. With more of them, the main speakers disappeared as more of these diffusors made sound. This was pleasing for a short while. I think there are marketed room tweaks that work on similar effect. They extend the midrange resonance to balance out the slap echo in the highs and bass boom in many untreated room. The 7inch skylines were very good and produced smooth midrange and treble (largest effective bandwidth). They brought a lot of focus to the center stage if you put them in the center (suggested by Rives). I almost installed them permanently until I played the XLO test disc. It consisted of Bill Johnson walking around the room while percussing a gong. With skylines at the center, he could not walk away from the center but he could walked far away to the side. THe extruding blocks of the skylines were actually vibrating during playback. It was ruining the subtle spatial clues need for proper localization. I suspect the effect would be less if I was further away from the front wall. The hemifusors were better in this regard as there are no extruded blocks to vibrate. The BAD ARCs were the best as they were combination tools (absorb + diffuse). The soundstage was the most organized and yet able to portray variations in depths. Abfusors were almost as good but why pay more for the same thing. I later constructed a dome with 3 to 4 BAD ARC panels of varying curvature and width to be placed at the center. This dome pushed the centerstage forward and increased focus ( effect can be modified by changing width and curvature). I could see why so many rooms have this. This did not work for me as I was quite close to the speakers (8 ft). I ended up with three panels evenly spread out in front.

Sidewall: Tried aborption panels first. They decreased reflections and created more focused but smaller soundstage in comparison to the diffusors. They decreased the amplitude of the reflections as seen on the ETF impulse response but the reflections were still very focused in the time domain. With combo tools like BAD ARCs, the reflections were decreased in amplitude and also spreaded out over time. It tricked the ears into thinking the boundaries were farther away. Not treating the first reflection point of the ipsilateral speaker created a wide soundstage in a different way. The sound source appeared wider but price was paid on impulse response and soundstage specifity. Treating the 1st reflection of the contralateral speaker was just as important as the ipsilateral reflection point. The crosstalk was reduced. ie, the left ear heard less of the right speaker.

Backwall: I used a few RPG skylines to great effect. The slight resonance is no longer an issue as they are behind me and ten feet away. Later, I tried 5 columns of Ikea expedit shelfs positioned in at an angle. This worked even better with better bandwidth. Depth was increased to 12inch and area of coverage was increased. I had two very narrow band suckout(less than one sixth octave) from only the left speaker at 50 Hz and 110Hz. These completely disappeared after installing the Ikea shelfs. Only part of each shelf was filled with a mix of cds and LPs to maximize the diffusive effects. One drawback, they increased midbass reverb at 60-80hz via combination of blocking the fire place, altering overall room acoustics and there own resonances. I had to increase bass traps at the corner.

Ceilling: Very important yet frequently overlooked area. Before treating the ceiling, the soundstage would rise when orchestra went full tilt (Similar effect to broadening of sound source). Hemifusors and skylines were similar in effect. When they fell off, hemifusors survived much better as there were no extruded blocks to be broken. One of my more embarressing moments when hosting. Lucky it did not land near the tonearm during play. I had considered doing suspended wood panels or maybe acrylic panels. After having the Expedit shelfs increased bass reverbs, I am not going that route on the ceiling. Just imagine if a wood panel came loose from the ceiling.

Bringing it together: All these traps, diffusers, combo tools are used to optimize soundstage, impulse response and frequency response. They frequent have unintended effect on reverb at a particular bandwidth. The soft diffusors were great as they did not make sound of their own. However, they markedly curtailed the high frequencies (>10khz) reverbs as they increased soft surface area of the room significantly. Large hard surface adds resonance of their own and alter room acoustics. Attention needs to be paid to maintain RT60 in an even handed manner. This is at least as important as freq response curve. As the RT60 trends down for a higher resolution sound, it is very easy to kill the high frequencies and quite difficult to bring down mid and low bass reverb.. At one point, my system sounded lean in bass. I had five large removable wood panels installed to block off all windows in the room and installed a 2 inch solid wood door with soundproofing. Surprising, the amplitudes of the room modes and bass alignment are not significantly different but the RT 60 at the bass 40-80hz were drastically prolonged. The noise floor of the room dropped to 40 db but it was like a tomb. The bass was muddy and standing waves were very problematic. In this exercise, I learnt that at least in my situation, the room dimensions governed the amplitude and freq. of the modal response but the degree of reverb is largely controlled by the structure. It is very hard to paaively trap mid and low bass reverb (

I am sure my experiments and observations are flawed. Feel free to point out any deficiencies so I can learn and benefit.

Speakers previously owned: B&W 805, JM lab Diva BE, Avalon Opus Ceramique, Thiels, Harbeths, Quad 2805, Avalon Diamond, Verity Sarastro 2

Amps previously owned: Boulder 1060, Vac phi 300.1, Jadis DA88S, Pass lab X350.5, Naim 250.2, ASL 1009 and ASL hurricane, Mcintosh 275.... etc. Heard in my sytem: ML432, Lamm M2.2, etc/

Preamp previously owned: ARC Ref3, Ref 5, BAt 51se, Naim 282. Heard in my system:,Shindo, Halcro

Previous Phono: AYre px5e; ARC PH7, ASR exclusive 2010. Heard in my system: Lamm, Nagra VPS, Shindo, Halco, Allnic

Previous carts: Zyx Universe, airy; My Sonic lab ultraeminent, airtight PC1. Heard in my system: Goldfinger, PC1 supreme

Previous tonearm: Graham 2.2, Davinci 12", triplanar 7

Previous tables: SME 20/2; Avid Acutus

Here is my opinion and preference about different carts that have gone thru my system. Just opinions and I am no expert. My preference is limited by my ability to setup (sub)optimally and by phonostage matching.

My preference in carts have changed and current favourites are Lyra titan i and dyna v1T.

Zyx Universe: Very refined and detailed but not enough macrodynamics and bass.

Airtight PC1/Supreme: Most amount of bass. A little on the warm side. Played well on all three arms, worked well with tube + ss phono. Sounds good with many different loading. Lacked a little detail and refinement comparing to other top carts. The supreme is slightly more refined comparing to the regular PC1. Slow transients.

My Sonic Lab Ultraeminent: This used to be my favourite cart.. More detailed and energetic than the Airtight line up. Quantity of bass is less but quicker. Excellent bass power carried into the decay. Transient attack is not as fast as XV1T or Titan I and perhaps not as open sounding. In combination with the Davinci, it conveys the most ambient info and float the soundstage best.

Dynavector XV1T: I think this is the most neutral cart.. Faster transient attack than MSL/AT and more open sounding. Bass is impactful with less energy in the decay. I like this one on the phantom best. On the Davinci, the midrange is marginally richer (maybe wood wand) but the lightening transient and focus were attenuated.

Titan I: My number one cart after finding an excellent match with DV 507 arm. Fast and open like the XV1T (both stiff body contruction) but with even better grip on the Bass. The transient changes on double bass is very well portrayed from the attack to the decay. It portrayed wide tonal and macrodynamic contrast and make the performance very exciting. More focused and controlled than PC1. To my surprise, I never find it bright when properly adjusted. In terms of setup, the window of optimal performance may be small but not too hard to arrive at. You know when u get this one wrong. I like this less on the Davinci & phantom. The triplanar and Davinci adds a little warmth and roundedness that it does not need. The Graham could not control resonance as well and had a little top end glare. With the 507mk2, the heavy arm and flux damping dissipate the energy perfectly. Just all the glories, nothing more and nothing less.

Coralstone: I never intended to get this one because of the warm fuzzy reputation of Koetsu. My dealer offer a partial trade deal and I could not resist. Another surprise, it is very detailed with a fluid midrange and extended topend. Bass power is slightly less than Titan/ultraeminent but not too fat behind. Only have this on the Phantom so far. Very difficult to align because the stone block my line of sight. I probably have not optimize this one due to neck spasm. It is also better played on an arm with Azimuth adjustment to minimize crosstalk.

Goldfinger v2: Another excellent cart.. Very limited experience but a friend was kind enough to bring it over for a spin. Similar sound to the MSL. Heaviest cart i have handled.
Read more...

Components Toggle details

    • TW Acustic Raven AC
    3 motor
    • Esoteric X 01 Limited
    CD/SACD
    • Graham Engineering Phantom 2
    Unipivot
    • Einstein The tube Mk 2
    Linestage
    • Einstein TT Choice
    phono
    • Luxman M800a
    Two as monoblocks
    • Stealth Indra
    XLR
    • Studer A810
    Reel to reel
    • Finite Elemente Reference double width
    I will not spend this much on a stand again
    • Grand Prix amp stand
    not bad
    • Avalon Acoustics Isis
    Isis
    • Dynavector 507mk2 SE
    black
    • Lyra Titan i
    stereo
    • Lyra Kleos
    stereo
    • Dynavector XV1 T
    cart
    • Minus K BM 8
    0.5hz
    • RPG skylines, hemifusors, BAD ARC
    diffusors
    • fuzzmeasure 10ft from wall behind listener
    8ft from speakers
    • fuzzmeaure 7-8ft from wall behing listener
    10ft from speakers
    • fuzzmeaure 4ft from wall behind listener
    15-16ft from speakers
    • Spiral Groove SG2
    Centroid

Comments 111

Showing all comments by jaytor.

View all comments

Boy, you've put a lot of effort into room tuning and have a very nice system. I hope you are now happy with the sound. If you are still having issues with room modes, have you considered adding a couple of small subwoofers (to implement a swarm effect)? I haven't tried this personally since my room response is fairly smooth as it is (I have 4-driver open-baffle woofer towers), but I've heard other's systems where this really helped.

jaytor