|
Post by Hexspa on Feb 8, 2023 23:12:02 GMT
Regarding your room, face the window.
There are two windows, an oversized one on the western wall and a normal sized one on the nothern wall. Which one should I be facing in your opinion?
The northern. The principles behind it is that you want symmetry from your ears forward and maximum distance from your speakers and ears to any boundary. Actually, speaker placement is a little more complex than that. You either want to have your speakers very close or very far from a wall. Anyway, no other orientation meets these criteria.
|
|
|
Post by Hexspa on Feb 7, 2023 8:19:28 GMT
Let me preface this post by admitting that I did not read your whole post for reasons I go into below. I apologize if I don't answer your question in the following. If it's not in there then please just ask one question at a time. Thank you.
Your post is very detailed. In my opinion, too detailed. Acoustic treatment is easy: your room rings and reflects so you introduce absorption (duh). The geniuses have already digested the complicated stuff to the point where you just grab solutions off the rack.
Listening position: 20, 32 or 45% all axial dimensions. Perform measurements to confirm best performance. Balance that with practicality.
Speakers: equilateral triangle, the closer to your ears, the more accurate the sound (min. 0.5m)
Subwoofer: do a crawl or start between 1/4 wavelength and corner of front wall and apply same caveats as listening position. More subwoofers helps bass.
Absorber functions: early reflections, modal ringing, and speaker-boundary interference. (I just added this for clarity as it's repeated in the next section).
Absorption locations: Primarily, early reflections (walls within 10-20 feet that are between you and the speakers) aka RFZ. Secondarily, modal dampening aka 'bass traps'. Corners are simply convenient places to have an air gap and treat two boundaries at once. Other than that, you just treat as much of the wall as will dampen your target resonance to a sufficient degree. Note that speaker-boundary interference can be mitigated with absorption. This happens most strongly at the 1/4 wavelength of a frequency. So, if your speakers are 2' from a wall then you'll get a cancellation of an 8' wavelength. This is separate from FRZ and modal issues. Bear in mind that your speakers are in a 3D space so front, maybe rear, left and right will all be at cancellation points. When placing your speakers, avoid duplicating these distances (yes you might compromise a little symmetry which is necessary anyway for left-right axial mode).
Absorption type: maximum thicknesses per absorber: 6" rigid, 12" safe n sound, 24" fluffy. Technically, density is not related to absorber performance. It is indirectly a property of resistivity (kind of) which, ultimately, contributes to a panel's resistance (resistivity with a given depth). Let's say that the more dense the material, the thinner it will be before it becomes increasingly reflective (not what you want). If a mouse can't grab chunks out of it then it's rigid and needs to be max 6". Maybe 4lb can be 8" but no guarantees.
Bonus points: 1x air gap behind any panel enhances low end at slight cost in lower mids. FRK enhances low end at the expense of higher frequencies.
Method: empty room, place mic at possible listening position and subwoofer at 1/4 wavelength against front wall or front corner. Identify best spots for both (you'll notice response trends). Install stuff in room including RFZ panels and any panels you'll use for recording. Measure again to identify SPL and ringing issues. Place absorption based on ringing frequencies (which correlate to physical dimensions) until you run out of material or space or achieve your targets*. EQ and balance your subs then integrate with mains. Measure to confirm. Listen to make sure everything sounds right. Revisit any previous steps to apply creative or technical tweaks. Done.
*targets: +-10dB spl with a 1dB/oct slope (or +4dB sub bass relative to 1kHz) and 20dB modal decay within 150 ms above 63Hz with a smooth taper in addition to spectrally balanced early reflections - or better.
Resources: How to predict ringing frequencies: calculate modal harmonics (axial mode): F = n1130/2D (imperial) 1125 ft/s F = n345/2D (metric) speed of sound 343m/s @ 20deg Celcius F is frequency n is harmonic (1-4) D is dimension (distance between walls) first harmonic 16ft length F = 1x1130/32 35.3125Hz C#1
Caveat: you can use pressure absorbers but I haven't and don't think they're required for most home audio applications.
Summary: place absorption where there's a problem. Generally, there're a lot of problems (depending who you ask) so you need a lot of absorption. 15-20% of sufficiently thick (1/4 dimension-thick and low-resistance is ideal) is a good start.
Regarding your room, face the window.
|
|
|
Post by Hexspa on Feb 6, 2023 17:57:04 GMT
Yeah, I've seen 'pix' (haven't seen that abbreviation in awhile) of frameless panels and they look very middle aged. I'm currently recombining my 2" rigid panels into 6" ones. This metal drywall bead is very sharp but it's much stronger than the vinyl stuff I'm using for my safe n sound 12" panels.
|
|
|
Post by Hexspa on Feb 5, 2023 18:36:58 GMT
Hi, welcome. I'm just now seeing this but will not be at my computer to review it. Maybe someone else will join in or I can check it out another time. Thanks.
|
|
|
Post by Hexspa on Feb 2, 2023 12:09:58 GMT
Sorry, I've been away for a while. Hex is correct! But add on and put it another way: (cletuschrist wrote:) "Thanks! I was confused about whether the gap had something to do with waves hitting the wall and bouncing back into the insulation, or if it really only matters when the wave energy goes through the insulation once THEN AGAIN after hitting the wall coming back the other way." I'd answer: It's BOTH!. Sound waves bounce off the wall from both going through the absorber first and then back again... and to a lesser degree from any sound waves reflecting off the wall from other angles. Good to have you back. Any war stories?
|
|
|
Post by Hexspa on Jan 23, 2023 6:30:17 GMT
No problem. Yeah, acoustically anything under 10,000ft3 is considered small. Your space's 4500ft3 is nice but it still gets the regular small room approach. I've been in a tiny reverb chamber that was definitely diffuse so I guess the lines can blur but I still think you're looking at the transition approach rather than true schroeder. But, hey, I don't have a degree in this so your guess might be as good as mine.
Also, like I might have mentioned, 6" is the optimal thickness for rigid panels. Beyond that, you might get some reflectiveness which means you're starting to waste material. Ethan's findings show that more coverage beats less but with greater thickness. In other words, if you can stretch out your material to make more 6" panels to cover more area, you have reason to think you'll get a better result. Be sure to measure and find out which dimensions are ringing so you can put your insulation there unless you plan to treat that with pressure units.
|
|
|
Post by Hexspa on Jan 21, 2023 7:42:21 GMT
A few things: Ethan has depreciated those absorbers, I remember him saying. Last I heard, he's mostly with the standard frictional absorber with or without FRK and air gap. From what I understand, nobody here can really help you with resonant absorbers including me.
Second, small rooms do not have a Schroeder frequency because that requires a diffuse field which small rooms do not have. Small rooms have a 'transition' frequency which is much higher than you'd get from calculating a SF. That said, the small rooms certainly have axial modes and you calculate them thus:
calculate modal harmonics (axial mode): F = n1130/2D (imperial) 1125 ft/s F = n345/2D (metric) speed of sound 343m/s @ 20deg Celcius F is frequency n is harmonic (1-4) D is dimension (distance between walls) first harmonic 16ft length F = 1x1130/32 35.3125Hz C#1
Something to keep in mind is that you will need broadband absorption regardless if you use resonant absorbers. Last I remember of Ethan's advice was get all your broadband in then see where you stand. In practice, sub bass will ring longer and that is ok - particularly below 63Hz so long as it's a tapered lengthening; you don't want a single long ring. Besides, this is for a control room and not a live room. You can reduce sub frequency ringing with multiple subwoofers - I've seen decay times reduce with three subs. I can share my notes with you but, again, this isn't my area of expertise. The point is that you might be able to deal with the sub bass range electromechanically via subwoofers or you might need a combination of techniques.
Everyone has their own approach. I recommend getting everything in the room, getting in your broadband then seeing where you're at with LF resonances. Have your predictions ready via the formula I gave you and determine the right tool for the problems, if any. Obviously, you will need targets - there are many available. You will only see how far you are once you take your first shot. I've never heard an acoustician say that everything on paper translates perfectly into the real world.
|
|
|
Post by Hexspa on Jan 21, 2023 7:24:37 GMT
Afaik, it's all about getting the absorber to the 1/4 wavelength of the lowest frequency. Friction works on the velocity aspect of a wave which is 90deg phase shifted from the pressure aspect. The way that manifests in standing waves is the pressure is always against the boundary and, depending on which harmonic of the mode it is will determine how many velocity phases are in the room. The fundamental has one at the center of the room the second harmonic has two and so on. For frictional absorbers, you're looking to absorb that second harmonic and 1/4 wavelength (boundary distance). Naturally, higher frequencies will have shorter 1/4 wavelengths so the absorber works just fine from then on. Luckily, it's not all-or-nothing: the wave still has velocity at all points other than maximal pressure so you'll get some absorption before totally absorbing 1/4 wavelength. Plus, if you use FRK then you add a pressure element but that doesn't really translate directly to depth in the same way. Here's a (somewhat old) blog post of mine that has a good image that I recreated from somewhere: hexspa.com/velocity-vs-pressure/
|
|
|
Post by Hexspa on Jan 19, 2023 22:55:26 GMT
It's the absorbing material. The frame is reflective so gapping it does nothing. It seems like you know but a 1x gap for a frictional material is optimal.
|
|
|
Post by Hexspa on Jan 19, 2023 22:53:37 GMT
|
|
|
Post by Hexspa on Jan 11, 2023 19:09:33 GMT
For waterfall, you don't really need anything above 300Hz. That ~140Hz ringing is from an 8' (or multiple) boundary. Your width is 8.9' and height 8' so I'm not surprised to see 140Hz (a little below) ringing. Luckily, that's your worst problem since you only need to focus above 63Hz. You can tame that with a 1' safe n sound panel with 1' gap on your width or ceiling boundary. Probably better to do the width. Remember that you can also EQ (down mostly) if you've exhausted your positioning and absorption options. Just be sure to take a few measurements in a reasonable radius to make sure you're not worsening problems elsewhere.
Sorry for the late reply. I must've meant to reply then never did it.
|
|
|
Post by Hexspa on Dec 26, 2022 18:04:38 GMT
Thanks for the reply, Hexspa. Do you know if there is a setting for me to be notified when someone posts to this thread? You are correct to link the prime numbers approach to the harmonics. There person said it was based on a much more involved calculation and that using prime numbers was distilled version. In your examples, is there no 1/7 or 14% harmonic? Just curious. Two developments since my initial post: 1. Eight days ago I added a Kef KC62 sub. In the first 48 hours I couldn’t hear a difference when flipping the phase switch and sub placement was limited by a short RCA. But with a long RCA now and about 200 hours of break-in, this evening I heard a clear difference between the phase settings. 0 degrees is way better than 180 in the current position. I have no idea if that could flip (no pun intended) with the sub in a different location. I will ask my dealer. 2. Even before the sub arrived, I revisited the prime 5 (length of room) after removing some remaining acoustic stuff resting on the floor near the front wall. With that removed, the prime 5 spots sounds extremely good. So I left my speakers there, moved the seat forward in increments and added the sub. I am interested in revisiting the prime 3 spot again to see if I prefer it or prime 5, now that the sub is in play. My Lyngdorf integrated amp has 6 built in subwoofer XO’s to choose from. I’m using butterworth 2nd order and have played with XO at 38, 44, 50. I high pass the mains to see nothing below the XO point and the Low pass the sub to see nothing above it. So no overlap. Now that the phase is set correctly, I have turned the sub volume way down and the perception of LF depth stays just as good while quality of LF (better transparency, reduced bloat) has improved as have mids and highs. Do you have any suggestion for optimizing the sub in the vertical dimension of the room? Room H: 106.5”. My Mains are a 2.5 way. The top LF driver covers from 2900 all the way down whiel the bottom one does only from 300 down. Might it be beneficial to put the sub on a stand such that the core of the sub is at the same H of either the lower woofer or the midpoint between the two woofers? Or, does the prime system apply to room H? 20% of 106.5 is 21.3”. Would putting the center of the sub at 21” above the floor sound good? BTW, that height would be fairly in line with the lower woofer. Profile/edit profile there you'll see notifications for when you log in or via email
Perhaps a bit pedantic but if the sub's toggle is 0 or 180 degrees then you are 'inverting the polarity'. "Flipping the phase" is a colloquialism and also inaccurate, afaik, because phase is frequency-specific whereas polarity is for the whole waveform.
Regardless, I think the listening position locations I listed are for just that purpose: listening position. Your sub is a different animal altogether. An obvious example is that putting your sub in eighth space (wall-wall-floor/trihedral corner) gives 'room gain' which is basically free bass. This may be desirable - however - you would absolutely never have your main listening position in an eighth space corner.
With that out of the way, you will virtually always get better bass with more than one sub. The reason is because your room is also divided into polar zones. For example, any given dimension of your room supports a fundamental mode at double that wavelength. If I remember correctly, a standard construction in the US is 8' ceilings which itself is 140Hz wavelength. But, due to what I'm saying, your fundamental axial mode for the room's height is 70Hz. Half of that dimension will be positive polarity and the other half negative and in the center will be a pressure null. For this reason, if you put a sub in the center (energizing both polarities equally) you will cancel the first and third modal harmonics. I forget why this is but Mr. Anthony Grimani has some good presentations on youtube explaining this. I will note that he is a home theater specialist and, in my opinion, does not offer entirely accurate information for home studios; specifically regarding the amount and type of absorption that he recommends. From his sub-mains integration, he says that the polarity of the sub(s) should be whichever produces the best result at the crossover. This makes sense because both speakers and sub(s) are playing there so the phase coherence (remember - phase is frequency-specific!) is important. In my current room, I did not high pass my mains so the polarity of the sub mattered for more than just the crossover region. Actually this is a good strategy for linear bass since your mains will act like mini subs and possibly give you a better response. I don't hear too many people mention this but it makes sense; it's worth trying anyway.
In summary, I am moving rooms right now and I marked off the 20, 32 and 45% lengths and widths but just my regular standing height and put the sub at 1/4 width against the front wall and on the floor (the previous location I had which produced a linear result). From there, I moved the sub one width of itself at a time until I found the most even response with no significant nulls. Of course, I used a measurement mic and REW because I would never do this by ear. Ordinarily, you do not want any speaker or listening to be equidistant to two or more boundaries since this will compound the cancellation. Indeed, the sub is on the floor but slightly forward of the front sliding glass door and even further from the left wall and, naturally, very far from the right wall. It is not quarter wavelength and I doubt that it is exactly 20% (though maybe it's close to that). It is really more like a corner placement but pulled out a little bit. In the end, you will need to iterate and measure to find the best sub location. If you absolutely must do it by ear, place the sub at your listening position and crawl around to put your head at possible locations. Start in the corner and move along the front wall until you find what you like. Those with experience note that a single sub is more localizable than two but if it weren't then even a side wall placement would technically work. Again, the best bass response comes as a result of multiple subs but you can still get a good result with one.
---
Since you care enough about your audio to spend that much on a sub, here are my notes from the various practical resources I found regarding bass reproduction in small rooms. Be aware that this approach recommends getting your bass right first then adding your mains on top of that. After all, bass quality accounts for 30% of listener preference. I make no guarantees about its accuracy or legibility. It is for my own use and home studio-oriented. I'll answer questions but use at your own risk:
Dimensions Damping Listening positions First find listening position Probably around 20,32,45% Not 50% (1st, 3rd harm) 25% (2nd harm) 16.7% (3rd harm) 12.5, 37.5% (4th harm) Subwoofer Placement First Quarter placement? Half distance cancels 1,3 Corners gives max bass Not more than 60cm from boundary ~2’, ~143Hz null (C#3) Speaker placement <5cm<avoid<1.1m ~78Hz null (D#2) (below crossover) 0.5-1m from ears Experiment with toe in/listening window Aim for unequal boundary distances Crossover Where speakers can play Can’t cross over below speaker response Above modal issues Let multisub handle bass, get speaker out Check at 86-100dB Distortion, power compression, limiter issues -3dB from midband good starting point, consider above factors LPF 4th order Linkwitz-Riley (LR4 24dB/oct) Two 2nd order butterworth (squared) HPF 2nd order butterworth (LR2 12dB/oct) Tuning Four microphone positions Have stands for each mic Nearfield sub levels Inter-subwoofer delays Doesn’t always make a big change Rear/front/side/indiv. 2ms increments Inter-subwoofer levels Experiment indiv. level differences at MLP Back/front/side/indiv. 2dB increments -checkpoint- -have established max level, min variation- EQ subwoofer group, some individual Channel levels and delays Crossover point and slope Check polarity of subwoofer group Also indiv. subs if necessary Should be no dip at crossover Listening Balance by ear 1dB increments EQ Spatial averaging Two ear average Each mic by each ear ~7” apart Use stereo mic clip 1/6th oct mains 1/12th oct subs Parametric ideally Gentle bass slope, minimize peaks About 4dB shelf @ 80Hz vs 1kHz Start with lowest, loudest peak to kill Might need to iterate (General) Send LFE and bass management to sub Don’t use separate subwoofers Low pass LFE to 80-90Hz REW has subwoofer optimization routine Also has psychoacoustic smoothing ~120Hz localization issues start Aim for biggest peaks 20-40Hz w sub Free gain that you can eq down (headroom) Watch out for nulls around crossover LR filters are -6dB @ cutoff, even-order Soft knee Butterworth are -3dB, even or odd Hard knee Sealed boxes are 12dB/oct Ported are 24dB oct So ported sub, sealed mains Basically, what’s important is to have symmetrical filters A sealed box is 12 so adding a 12 is 24 www.avsforum.com/threads/guide-to-subwoofer-calibration-and-bass-preferences.2958528/www.audioholics.com/subwoofer-setup/subwoofer-placement-the-place-for-bass-part-1/subwoofer-placement-the-place-for-bass-part-1-page-4Test tracks: www.svsound.com/blogs/subwoofer-setup-and-tuning/the-essential-subwoofer-bass-music-test-tracks-and-demo-list
|
|
|
Post by Hexspa on Dec 15, 2022 9:05:09 GMT
Sounds interesting. The way I understand positioning is to find a axial non-null spots and those end up being 20, 32, and 45% the dimensions. Famously, Wes Lachot recommends 38% but I never found that precise spot to be right; probably for a number of reasons. These numbers are based off the harmonic series of the modes - what you will see in practice. 1/3 is 33% and 1/5 is 20% and 9/20 is 45% which seems pretty prime to my high-school math level of training. These are pretty much what you want for the room. Exact spots depend in part on the absorptive properties of the boundaries. More absorptive surfaces make the room seem bigger. There are many calculators for divining room modes so whatever math which helps you 'run between the raindrops' will work. Besides modal issues, you still have boundary interference issues - particularly at the 1/4 wavelength. That's less about prime numbers and more about whether you're using bass management (crossed over subwoofer) or not. You also need to contend with critical distance and how 'live' (colored by the room) do you want your listening experience.
|
|
|
Post by Hexspa on Nov 24, 2022 17:07:00 GMT
most likely B for a few reasons:
1. you can add more absorption, as per your plan. With no cloud and stone walls, you're going to need all that you can get. 2. boundaries are further away, sides and rear. That's less reflection strength at LP. Every little bit counts. 3. you can try 20, 32, 38, and 45% lengthwise locations. 38 is good but try those as well in addition to wherever your voice sounds most natural in the room (my hack).
Corners are only good places for bass traps because they're acting on two or more modes at once plus you get an air gap. Most acousticians will tell you that the axial modes are the main problem so really, those are best treated on the flat surfaces. Your ideal absorber for 100Hz is about a 2.8' fluffy absorber. Of course, this is huge but with enough coverage, FRK, and ideal placement you can still get a good decay result. I've mentioned in another thread that the ideal absorber material-thickness brackets are 6" rigid, 12" safe n sound and 24" fluffy. You can gap each up to 1x with minimal penalty. The thickest absorber is best among these options.
At the end of the day, it's about what you want and are willing to do to get there. If you don't know any better then, for a home studio absorb everywhere with the greatest quantity of the above options you can. If you can do 2' with a 2' gap fluffy everywhere but the ceiling, that's your best choice (besides that you'd be smooshed). I see you're already using 6" rockwool of some kind so if that's what you're committed to as a material then so be it.
|
|
|
Post by Hexspa on Nov 6, 2022 9:33:17 GMT
You can upload it here. That host is giving me trouble.
|
|