View RSS Feed

mitchco

Hear music the way it was intended to be reproduced - part 2

Rating: 1 votes, 5.00 average.
Part 1 is here. Thanks for your comments. Before we can measure the frequency response of your sound system at the listening position, we need to configure the speakers to the listening room. These set up steps are required in the quest to hear music the way it was intended to be reproduced – i.e. best effort timbre. This is the first part of a three part process. The three parts are setup, measure, and adjust. Then we iterate, sometimes a few times, sometimes more. It will cost you nothing but a few hours or more of your time moving your speakers and perhaps listening position around your listening room. A tape measure is required.

I was going to dive into acoustics, like in this article: http://www.nonoise.org/quietnet/tcaa/smallrooms.pdf But I thought it would be better to explain a few quick wins that you can easily achieve in your own listening room using a bit of muscle and a tape measure. Of course, these are setup calibration steps in order to establish a baseline for the series of frequency response measurements we are going to perform.

Have a look at this listening room. Actually it is a control room in a recording studio. In fact, the vast majority of control rooms in the world will be a variation of this set up:


blogs/mitchco/attachments/364-hear-music-way-it-was-intended-be-reproduced-part-2-examplecrroom.jpg


You can find many examples of these setups on the internet, including specifications, plans, golden room ratios, etc. We will come back to that when we look at acoustics. What’s important at this point is:

1. The speakers and the listening position form an equilateral triangle.
2. A best effort attempt at passive room treatments to calibrate the damping, reflections, and RT60 of the room. http://en.wikipedia.org/wiki/Reverberation
3. Best effort attempt at using the tape measure for all measurements, symmetrical or otherwise.

Let’s walk though through this in some detail, but we are going to leave step 2 out for a while.

But first a comment. I take no credit for any of this. These are all public specifications, tried and true, and are generally accepted as industry standard in the pro audio biz. If this is “old hat” to you, just think of it as a quick review of how important it is in the quest to hear music the way it was intended to be reproduced.

An equilateral triangle of the speakers to the listening position is required for proper decoding of the stereo mix. Look at the diagrams above for both the control room and critical listening room. Both are equilateral triangles. In my listening room, my speakers are 9ft apart (center to center) and each speaker is 9ft away from where my ears are located at the listening position. The speakers should be toed in so that they are on axis to your ears in the listening position.

Get out your tape measure. Make sure that whatever equilateral triangle that you end up with that the distance between the speakers and that each speaker to your ears are as exact as possible, down to a ¼ inch tolerance or less if you can do it. This is absolutely critical to ensure you are getting the exact sound stage that was mixed in the control room in the recording studio.

Here is an analogy with respect to sound waves. Ever throw a rock in water and watch the waves it produces? Now throw two rocks in the water, spaced apart (like 3 to 6ft for example) and try to do it so they land in the water at exactly the same time. Really hard to do, but look at the waves produced and when they meet – beautiful symmetry. If the rocks land at different times, then observe the waves produced. The one that landed first will produce a wave sooner than the rock that landed second and when the waveforms mix, it will look distorted, (i.e. loss of symmetry) that is because it is. Quick rule of thumb, sound waves travel 1 foot per millisecond.

Short story. When I was working in a about to be built LEDE studio on the West Coast, Chips Davis used a professional laser distance meter, levels, and transits to layout the design of the studio including measuring the equilateral triangle down to 1/16” tolerance. While I am old school with the tape measure, I see the prices of some of these laser distance meters are down in the $100 to $200 range. So if you spend time measuring, moving the speaker ever so slightly and re-measuring, over and over again – it’s perfectly normal to increment and iterate.

I can’t stress enough how it important that everything is measured and as symmetrical as possible in your listening room. That includes measuring the toe-in of the speakers from the back wall for example so that they are as near a perfect mirror of each other. This is critical to attaining proper timbre, especially related to the perceived depth of the sound stage.

The end result is that your speaker system is calibrated to properly reproduce (i.e. decode) stereo sound. From a listening perspective, each speaker’s sound will arrive at your ears at the same time. This will result in a perfect, none distorted (from a time perspective) representation of the stereo signal. You will hear pinpoint imaging, dead center phantom image, and now in a position to move to the next step of the calibration process.

Before we continue, I know some will ask, how far do I move the speakers into the room from the back wall? Great question and one we will measure, but for a starting point, and avoiding an acoustics conversation, hear is a quick way to “voice” the speaker position in your room and train your ears at the same time.

Play music that has good bass content. If you have sound level meter, like the infamous Radio Shack http://www.maxim-ic.com/images/appno...DI127Fig04.jpg meter, then select C weighting and slow response and crank up the music to average 85 to 90db (note we will come back to this sound level and why it is important in another post). If you don’t have a sound level meter, no worries, just crank up the sound a bit, but not really loud, we just want to load the room with sound.

Turn your balance control to either left our right so only one speaker is playing. Now go stand beside the speaker and listen to the bass sound. Listen to how even the bass sounds as the notes go from high to low and vice versa. Does the bass sound louder on some notes and less on others? If so, start moving the speaker slowly forward while listening. For really trained ears, this is like blowing air into a Coke bottle and hearing the resonance. That’s what we are doing. We are trying to find the sweet spot where all of the bass notes sound even up and down the scale.

If you can’t hear the difference, no worries. Try moving the speaker against or as close to the back wall as possible. It is likely to sound boxy, or too much bass. Now move the speaker several feet from the back wall and listen again – the bass response should be considerably different. It may be that most of the bass seems to have disappeared. Somewhere between the two positions is the best position for the speaker based on your specific room ratios. Patience and practice will assist in finding the sweet spot.

Now turn the balance control to the other speaker and move the speaker the same distance from the rear wall that you had moved the other speaker from. It should sound the same in the bass region. Use a tape measure to get it exact. Now turn the balance control to the center and listen again. Bass notes sound even through the scale? Does the balance of bass to mids, to highs sound ok? Use your ears, they are wonderful measuring devices. Congrats, you just voiced your speaker to room interface without having a PHD in acoustics or breaking out the measuring equipment. Remember this is a starting point or baseline in order to continue the calibration process.

Now that you have located the sweet spot, make the measurements exact using the tape measure to form that equalateral triangle. Take the time to get it within a ¼” tolerance.

I am a bit reluctant to get into room treatments and acoustics until we take some measurements. The reality is that you have the listening room you have. You could work out the room modes with a room mode calculator like http://www.mcsquared.com/metricmodes.htm and you would do well to read the reference links at the bottom of: http://en.wikipedia.org/wiki/Resonant_room_modes You could also check to see if your room falls into the gold room ratios that are in the slides I referenced earlier. You will notice in the article that there are different types of rooms from the LEDE to RFZ to ESS. My own room falls into the latter ESS category. You can also look at this speaker set up guide: http://www.cardas.com/pdf/roomsetup.pdf

We are at a point where we have a best effort speaker to room setup and calibrated to a well-known standard (i.e. equilateral triangle). Now we have a baseline in which we can start taking frequency response measurements. In my next post, I will start taking measurements of this setup and we will see how close I voiced my speaker setup in the bass frequencies – remember they should be as evenly distributed as possible. I will get into a bit of room acoustics and basic room treatments if the measurements warrant it.

Happy Listening!

Mitch

Updated 05-09-2012 at 01:48 AM by mitchco

Categories
Personal Blogs

Comments

  1. barito7's Avatar
    Thanks for this basic advice!



    I was wondering if you have a common song that you like to test with that has good bass?



    Thanks again Charles
  2. AudioExplorations's Avatar
    Fantastic blog! Looking forward to following entries!
  3. AudioExplorations's Avatar
    Scuba - You Got Me





    Gil Scott-Heron and Jamie XX - I'm New Here (has a nice bass sweep)

  4. mitchco's Avatar
    Hey Charles, I am glad you find it useful. I don't really have a common song I use. It's kind of whatever I feel like at the time.



    However, the criteria is that it is well compressed, has a wide range of bass notes, and a fairly fast beat. The reason for this is the analogy back to blowing air into a Coke bottle to hear the resonance. We are trying to get the speaker to "load the room" with constant sound to excite the room modes so it makes it easier for our ears to hear the resonance/standing wave differences as we move the speakers around the room.



    Hope that helps and happy listening!



    Mitch
  5. barito7's Avatar
    Thank you all!
  6. barito7's Avatar
    I was also wondering what height your speakers should be at? I have some bookshelf speakers and I am assuming that line of sight should be directly inline with my ears??



    Cheers
  7. mitchco's Avatar
    Yes, if you can, the speakers should be directly inline with your ears.



    Cheers, Mitch
  8. bleedink's Avatar
    Hi Mitchco...I sort of stumbled on the triangular shape of the front 2 speakers somewhat by accident. It's great to see it confirmed. I'd done the same as you (ie take measurements from the walls behind the speakers to get the toe in correct and make sure they are equidistant.) Now that we have the stereo out of the way (I was surprised just how much it adds to the imaging and particularly the anchoring of the phantom center channel in stereo recordings) what relation would this have to a 5.1 set up. Can we go from stereo to 5.1 or 5.0 (or even 4.0 for the real purists out there) with this info? Im particularly interested in placement as there seems to be a few different set ups for sacd, dvd-a, and home theater proper. Thanks. Anything you might add I am sure would be helpful. BTW my speakers are all from the same manufacturer and are close in sonic attributes though my rears are bookshelfs. Psychoacoustically this shouldn't matter though right? As I assume that the larger sound coming from the theoretically closer rears would obscure the fact that they are actually smaller as the info in the rears tends to be more atmospheric. Or am I way off base? (Some have insisted the only way to do 5 channel sound right is to have 5 of the same speaker...I have found the practicalities of getting 5 towers spread equidistant in the room to be a no go! Hence the question. Like I said earlier it seems the psychoacoustics of placement would eliminate the need for 5 full size speakerz)
  9. mitchco's Avatar
    Hi bleedink, yes, there is an industry standard specification for 5.1 music mixing from Dolby Labs that builds on top of our 2 channel specification: http://www.dolby.com/uploadedFiles/z...sic_Mixing.pdf



    On page 20, there is a specification of the reference loudspeaker placement:







    It is an excellent read. It goes into detail on how 5.1 music is mixed, calibrations, speaker placement, etc. Basically, a "how it's made" kind of read. All of your q's will be answered. And yes, you can mix and match speaker pairs, different center speaker and different sub.



    If you don't have the space for a proper 5.1 set up, there are a couple of alternatives. Depending on what 5.1 system you have, some come with a mic that you place at the listening position and automatically balances the 5.1 levels for you. Or you could do it manually with a mic and some software - adjusting the levels to make up distance differences. If you can, try and keep the angles the same, but rear channels likely be closer, so the rear channel level should be lowered appropriately.



    Hope that helps.
    Updated 05-09-2012 at 10:53 AM by mitchco
  10. bleedink's Avatar
    I've been on that dolby site a zillion times and never saw this information. Great stuff!