Speaker Placement Calculator (Link)

[mpearce38]

Not sure if this has been posted before but came across a simple speaker placement calculator for near-field set-ups. 

 

http://noaudiophile.com/speakercalc/

 

I have a question for the more mathematically inclined SNA'ers . If i have. a front ported speaker what is the optimal distance from the rear wall if I want to try and tame a peaker at 145Hz. I calculated the 1/4 wave length at 61cm does this mean I should aim for 61cm from the front of the speaker to the rear wall?

Edited November 6, 2020 by mpearce38

[Neo]

@mpearce38

You may want to include the actual link in your post

Neo

3 minutes ago, mpearce38 said:

Not sure if this has been posted before but came across a simple speaker placement calculator for near-field set-ups. 

 

I have a question for the more mathematically inclined SNA'ers . If i have. a front ported speaker what is the optimal distance from the rear wall if I want to try and tame a peaker at 145Hz. I calculated the 1/4 wave length at 61cm does this mean I should aim for 61cm from the front of the speaker to the rear wall?

 

[mpearce38]

Sorted!

Just now, Neo said:

@mpearce38

You may want to include the actual link in your post

Neo

 

 

[almikel]

On 07/11/2020 at 9:14 AM, mpearce38 said:

came across a simple speaker placement calculator for near-field set-ups. 

(my bold above) - "near field" set-ups typically don't consider the boundary/room response - "near field" means close enough to the speakers that the direct sound swamps the room response

 

On 07/11/2020 at 9:14 AM, mpearce38 said:

I have a question for the more mathematically inclined SNA'ers . If i have. a front ported speaker what is the optimal distance from the rear wall if I want to try and tame a peaker at 145Hz. I calculated the 1/4 wave length at 61cm does this mean I should aim for 61cm from the front of the speaker to the rear wall?

A speaker will have a "Speaker Boundary Interference Response" (SBIR) dip at 1/4 wavelength path difference between the direct sound (speaker to listening position) and the reflected sound from the speaker to a rigid boundary to the listening position.

 

You can experiment with an SBIR calculator created by SNA member Ken Tripp here http://tripp.com.au/sbir.htm.

 

Normally with SBIR you're trying to ameliorate response dips, not peaks, but hey, too much/too little are both issues...Goldilocks was right.

 

Play with your rear wall and sidewall offsets to reduce your 145Hz peak (I assume at your listening position) - stay symmetrical (left/right) to not muck with imaging.

 

If that peak is due to SBIR, then reducing your wall offsets will raise the SBIR frequency, making absorption treatment more effective to ameliorate it. 

 

Checking your maths, I get 1/4 wavelength at 145 Hz as 591mm, but close enough to 610mm (velocity of sound @343m/s).

Assuming direct reflection from the rear wall (no triangles), then with the speaker 591mm from the rear wall, the path difference of 1/2 wavelength would create an SBIR dip at 145 Hz.

 

I've never considered SBIR effects vs room resonance at frequencies below the transition point of the room - SBIR is always considered using "specular" models, ie path differences like rays of light...which become irrelevant below the transition point in the room where room resonant behaviour takes over...

... @davewantsmoore - I'd appreciate your input - is SBIR a consideration below the transition point in a room? does SBIR contribute to the modal response of a room, or is it swamped by the modal response of the room???

 

cheers

Mike

[davewantsmoore]

14 hours ago, almikel said:

... @davewantsmoore - I'd appreciate your input - is SBIR a consideration below the transition point in a room? does SBIR contribute to the modal response of a room, or is it swamped by the modal response of the room???

If you think about it.... they're the same thing.   Just on different scales.

 

So the question is kinda moot.   The sound is always bouncing off the boundaries and interfering with itself - at all frequencies.

 

Depending on what frequency range you're looking at, you have different things which can affect the problem/outcome.   ie. 20hz is very different to 200hz... or to 2000hz.

 

eg. (just as an off the cuff example) where you sit, strongly affects 200hz.... but not so much 20hz or 2khz.

[almikel]

On 09/11/2020 at 7:49 AM, davewantsmoore said:

If you think about it.... they're the same thing.   Just on different scales.

 

So the question is kinda moot.   The sound is always bouncing off the boundaries and interfering with itself - at all frequencies.

but it could make a big difference to when you choose to apply EQ or not - as you say, with EQ, it's important to know when not to apply EQ...

 

...and I'm only referring to a room's response at the listening position below 150Hz or so...

 

...I would regard SBIR as a case where not to apply EQ (as it's never minimum phase at the listening position), and adjust speaker/listening position and/or apply treatment to ameliorate SBIR effects, but never EQ...

...on the other hand, room modal behaviour can be minimum phase at the listening position, and where the room response is minimum phase it would benefit from EQ correction...

 

The below is for the thread, not directed at @davewantsmoore...

 

 

...where you know the response is minimum phase, EQ cut and boost works fine, but generally room peaks are more likely to be "minimum phase" than room dips - sharp room dips will never be minimum phase.

REW can measure where your room response is "minimum phase" - it's in these regions that EQ can help a great deal - for both the amplitude response and the time response...counter intuitive that EQ could effect the time domain response but true.

 

With EQ:

• the human ear/brain doesn't "hear" sharp/high Q dips/peaks in the frequency response - even though they look awful on a graph - the human ear is way more sensitive to "hearing" low Q/broad peaks/dips in the frequency response. Clearly a crossover null is not ideal, but fix the crossover, don't try to EQ boost the dip 
• low Q, broad EQ cut is generally the recommended approach
• where you know the room response is minimum phase, then low Q, broad EQ boost can work also
• never use hi Q narrow EQ cut or boost in domestic scenarios - an example of where high Q narrow EQ cut may be required is in live music where a feedback destroyer circuit sucks out a narrow band problem frequency that's causing feedback...or the mixer does it manually
• never use hi Q/narrow EQ boost - ever

cheers

Mike

[davewantsmoore]

8 hours ago, almikel said:

...I would regard SBIR as a case where not to apply EQ (as it's never minimum phase at the listening position), and adjust speaker/listening position and/or apply treatment to ameliorate SBIR effects, but never EQ...

...on the other hand, room modal behaviour can be minimum phase at the listening position, and where the room response is minimum phase it would benefit from EQ correction...

 

Yes, but this depends on how you measure/window.

 

I prefer to put it like this:   You need to correct the speaker response.... but you don't want to be confused by a big dip/peak.   You may still need to apply EQ in this region..... but not simply EQ "to correct what you see in a typical measurement".

 

I think this is actually one of the most critical parts of a playback sound..... the degree to which the response "flatness" is correct in the 100 to 400Hz range  (ie. where you'd normally see a very rough response).    It's this range where all the tone/richness/warmth, etc. lives.   (and where the majority of the energy in typical program material is).

[mpearce38]

wow thanks for the detailed information a lot to work through  

[mpearce38]

so here is my current room treatment set up currently the speakers are 51cm from the rear wall (to front face of driver)

 

still have the bump at 140hz and a dip at 95hz but more controlled then an untreated room. I'm going to run Rew this weekend and really narrow in on the remanning issue. above 300Hz rooms sound flat to my ears which is great. 

 

Next steps is to add a clound above the listening position and a bass trap on at the wall to ceiling point above the listening position. 

 

I use the room for audio production and I 've done a couple mixes since the treatments went in and they are translating well but bass guitar has been a little hit and miss but I'm learning where it needs to sit with the bump and a little lean forward to get me out of the dip helps

 

With regard to EQ it was generally frowned upon in the pro audio world for monitors. but more and more DSP is being integrated into speaker designs and being accepted.  May consider DSP running only below 300hz to see if that helps. Other option as my speakers are only 5" a sub could benefit my 90hz flatspot and fill out the 40-80 range where I'm missing fundmentals 

Edited November 11, 2020 by mpearce38

[davewantsmoore]

2 minutes ago, mpearce38 said:

With regard to EQ it was generally frowned upon in the pro audio world for monitors.

 

Think of EQ (I mean EQ in addition to what is already natrually part of a speaker) as "medicine" for a "patient".    Is it "generally frowned apon"?.... that's a bit of an over generalisation.

 

Medicine can fix all sorts of problems or even save someones life ..... but the wrong medicine for the wrong situation could be very bad .... and could be bad but also not super obvious/acute.... or bad, like, put you in an ambulance obvious.

 

Like patients ..... speakers in rooms, are quite hard to measure/diagnose in a way which is meaningful for applying EQ..... but it is easy to do some not very meaningful measuremnts, and apply what looks like (but probably isn't) the solution..... it's also very easy to change someone with "medicine" (make you feel "good" or "bad") ..... but that outcome might not be related to the underlying problem  (like prescribing you redbull for a sleeping disorder, for example..... or antidepressants for a diet issue... or whatever).

 

I hope that's too abstract.   

[davewantsmoore]

10 minutes ago, mpearce38 said:

a sub could benefit my 90hz flatspot and fill out the 40-80 range where I'm missing fundmentals 

The "true" solutions to issues under 400Hz:

 

Move the speaker

Move the listener

Add additional sound sources

 

 

EQ is also a valid solution.... but it's difficult to measure speakers in a meaningful way at these sorts of frequencies (eg. 80 to 400)

[mpearce38]

LOL no not too abstract, when I say frowned upon it "was" generally believed to get the room right and use EQ as last resort. but this was back in the day of only have 31 band or paramatric eq.  This was generally tuned in with an RTA to get as close to flat as possible but if you were in a room with standing waves, dips early reflections EQ as you said was not the best medicine. There has been a real movement in the industry driven my manufactures like Genelec to use "room correction" which is being adopted more and more as mainstream. Again however the less you need to compensate via signal processing the better so I'm happy I was able to narrow down to only 2 main issues

 

 

 

[mpearce38]

10 minutes ago, davewantsmoore said:

The "true" solutions to issues under 400Hz:

 

Move the speaker

Move the listener

Add additional sound sources

 

 

EQ is also a valid solution.... but it's difficult to measure speakers in a meaningful way at these sorts of frequencies (eg. 80 to 400)

good advice,

 

 1, I've moved to the best listening position in the room

2, Played with speaker postion best I could but could not pick-up any noticeable difference in response 

3. Will trial a sub over the weekend and see if this helps, have a small MJ acoustics ref 150 which should suite the room nicely

4. Final option will be room correction (any recommendations) I see the dsp units pop up in the classifieds from time to time

 

[almikel]

On 12/11/2020 at 8:17 AM, davewantsmoore said:

The "true" solutions to issues under 400Hz:

 

Move the speaker

Move the listener

Add additional sound sources

 

 

EQ is also a valid solution.... but it's difficult to measure speakers in a meaningful way at these sorts of frequencies (eg. 80 to 400)

For issues 150Hz - 400Hz I would add room absorption acoustic treatment to that list...but absorption is getting big to be working down to 150Hz

For issues < 150Hz I would add room "specialist/tuned" pressure trap acoustic treatment to that list...but these get big also 

 

On 11/11/2020 at 8:30 AM, davewantsmoore said:

Yes, but this depends on how you measure/window.

I agree we're running into the limitations of software/measurement at low frequencies - but let's assume we're trying to measure the room at the listening position (not the speaker), with no "user configured" windowing.

Is it possible to make a room measurement, or a series of room measurements (eg at different listening positions), in order to draw conclusions on where EQ could be appropriately applied to improve the sound at those measured listening positions?...

...not blindly applying "inverse" EQ to a peak/dip, but also reviewing say the REW measurement of "excess group delay", ie whether the measurement is minimum phase or not, to determine if EQ is appropriate...

...or are the REW measurements insufficiently accurate due to internal windowing/measurement inaccuracies for "in room" measurements at low frequencies? 

 

cheers,

Mike

[almikel]

On 12/11/2020 at 8:25 AM, mpearce38 said:

when I say frowned upon it "was" generally believed to get the room right and use EQ as last resort. but this was back in the day of only have 31 band or paramatric eq. 

Agreed that old school 31 band (or fewer band) graphic equalisers have a lot to do with the poor reputation of EQ.

 

IMO, IIR parametric EQ is still the best EQ tool to apply to room bass issues after:

1. Move the speaker

2. Move the listener

3. apply additional sound sources

4. apply room treatment

I'm much more dubious of DSP/FIR solutions where you lose control of what the DSP solution is doing (eg DSP/FIR filters can adjust amplitude and phase independently )

 

I use a DSP solution (DEQX) as a 4 way active crossover and room correction.

Standard DEQX room correction uses DSP/FIR approximations of "minimum phase" IIR parametric EQ - ie phase and amplitude are both adjusted at the same time predictably, as @davewantsmoore says, "a wiggle in amplitude shows as a wiggle in phase", as they're the same thing with a different view...

...start mucking with amplitude and phase independently (which DSP/IIR filters are able to do), and it just does my head in...

 

 

Mike

[petetherock]

Check out this nice speaker layout design tool:

https://slc.cinema-tools.com/#!/

[davewantsmoore]

15 hours ago, almikel said:

Is it possible to make a room measurement, or a series of room measurements (eg at different listening positions), in order to draw conclusions on where EQ could be appropriately applied to improve the sound at those measured listening positions?...

 

It's difficut....  if you take a few measurements you will start to get an idea for what not to correct.....  but it's not usually possible to look at the uncorrupted first arrival part of the sound, which makes things a bit hit and miss.

 

It's where the whole "have a well balanaced speaker, and just plonk it in the room and leave it alone" approach can work well/better....    OTOH applying small amounts of relatively wide EQ in this 100 to 400 range can work "magic" too.

 

 

15 hours ago, almikel said:

...not blindly applying "inverse" EQ to a peak/dip, but also reviewing say the REW measurement of "excess group delay", ie whether the measurement is minimum phase or not, to determine if EQ is appropriate...

...or are the REW measurements insufficiently accurate due to internal windowing/measurement inaccuracies for "in room" measurements at low frequencies? 

 

Is it minimum phase or not.... is a bit of a too coarse binary question.    At much below 500hz in  typical room, there will be reflections starting to mix in with the first arriving sound.     Pays to be careful, is all.