Sound Design Live

Build Your Career As A Sound Engineer

Can you match speakers through space and time? (🍎to🍎)

By

What if I want to compare my speaker to your speaker to see if they are compatible? Is there any way to actually do an apples to apples comparison?

Hypothesis: Yes!

Let’s look at an example.

Is a PRX615M compatible with a PRX618S? I don’t have either. How will I find out?

My first idea is to check the loudspeaker manufacturer’s website to see if they have anything that could help. Maybe they have their own modeling software or GLL files I can compare.

Nope. No software and the EASE data offered is only for standard EASE ($2,330), not EASE Focus ($0). Time to start making some calls.

Good news. I found both, but they are in different locations. My friend Amy in Austin has a PRX615M and my friend Burt in Berkeley has a PRX618S. I ask them to take measurements and send them to me.

So far, they don’t seem compatible. The magnitude and phase don’t match.

Magnitude

I thought about asking them to match their preamp settings, but Amy has an OctaCapture and Burt has a UMC404, which has a totally smooth preamp knob. We could probably figure out a way to match their preamps with a meter, but that would still leave the measurement mic out of the equation. Luckily, I remember that Amy and Burt both have microphone calibrators.

I have them do two things:

  1. Calibrate the MIC channel with the calibrator set to 110dB and the microphone preamp set to -12dBFS on the audio analyzer’s input meters.
  2. Adjust the REF loop so that when the signal generator plays a 1kHz sine wave at -12dB, the audio analyzer input meter also reads -12dBFS.
matching levels
level offset

The magnitude data looks like a closer match now. Things are looking up.

Phase

Let’s figure out what’s going on with these phase traces. Are they really that far off, or is there something I’m missing.

I ask Amy and Burt to send me photos of their measurement setups and realized that their measurement mics are at different distances from the speakers. Amy’s measurement is at 0.83m and Burt’s is at 1.6m.

I could ask them to redo the measurements, but maybe I can fix the timing offset manually. If the sub (PRX618S) was measured 0.77m closer than sub (PRX615M), then its measurement needs to be delayed by 0.77m to be in synch. I’ll use the Phase Invaders to add 0.77ms of pre-delay to the sub.

first try

Closer, but they still don’t line up. That’s when I realize that there’s an important feature of the 2-channel analyzer that could also be out of synch. I compare the delay locator setting used with each measurement. They don’t match.

The delay locator for the main measurement was set at 3.94ms and Speaker B was set at…0ms. 0ms? I asked Burt why he didn’t set the delay locator and he said that he tried it a few times and it didn’t work.

Right! If forgot that the delay locator often doesn’t work on LF drivers. Maybe we can figure it out.

I had already added 2.24ms (0.77m) of pre-delay to the Sub because of the distance offset. If I add 3.94 more, that will make 6.18ms.

How do you add pre-delay in the field? It doesn’t exist, but there is a workaround. Try adding 10ms of delay to all of your outputs before the alignment process. Then you’ll be able to add or subtract delay from any channel. Remove any excess delay at the end of tuning.

second try

They are still not aligned. Further investigation reveals that 3.52ms more pre-delay and a polarity inversion or 8.56ms more pre-delay and no polarity inversion give me two workable alignments. If I needed to set these up fast in the field, I would start with those two presets and modify them using their final distance offset.

aligned

Note: The proposed procedure does not test max SPL, but does compare relative sensitivity. For max SPL testing, see m-noise.

Have you tried comparing measurements through space and time? What were your results?

Questions

Will there be enough headroom?

Maybe not to measure max SPL, but we’re just comparing sensitivity. Once the inputs are calibrated you can adjust the signal generator level all day and the measurement will stay the same.

Is it better to face your subs at the wall?

By

I worked on a show last month where I decided to face the subwoofers at the wall. People were looking at me funny and asking questions…and that was before they even saw the subs. 🥁

So I made a video to investigate the effects of path length differences in this binary scenario. One result is not necessarily better than another, but it’s good to make an informed choice.

You can use this formula to estimate the frequency of the first dip of the comb filter.

c / d / 2 = 1st dip

speed of sound / distance (aka total path length difference) / 2 = first dip of comb filter

Subs facing audience
1130ft/s / 13ft / 2 = 43.5Hz
345m/s / 3.96m / 2 = 43.5Hz

Subs facing wall
1130ft/s / 6ft / 2 = 94.2Hz
345m/s / 1.83m / 2 = 94.2Hz

Pick your poison.

sub wall comparison

Taking a look at the field measurements, it looks like the subs were probably closer to 5ft from the wall because of the dip at 112.8hz, indicating a 10ft round trip distance for the reflection.

reflection comb filter

One side benefit that I forgot to include in the video is that the QSC subs were polarity inverted compared to the Kara mains. Facing them away from the audience meant that I didn’t need the polarity inversion. Tiny victory. 👏

Check out Merlijn’s Low-frequency Acoustic Center Calculator.

Have you tried facing your subs at the wall? What were your results?

Does every output in the signal chain need a high-pass filter?

By

For the first 8 years of my career as a live sound engineer I thought that the more high-pass filters I used, the better protected I would be. More is better, right?

I would put a high-pass filter on my console outputs, on my DSP, and on the speakers if they had built in processing.

I didn’t realize it at the time, but the result was a super steep roll-off into a gap crossover. All I knew was that it sounded strange, but at least I was playing it safe.

gap crossover

Why is this happening?

The filters you insert at each point in the signal chain get burned into the signal. They are permanent. Once you reduce a portion of the frequency response so low that it goes into the noise floor, you can’t get it back.

Witness the effect of a high-pass filter at 120Hz.

HPF 100Hz

Then when I try to reverse the effects with a low shelf.

HPF with low shelf

Sad face.

Here’s the result of two matching high-pass filters in line with each other.

two high pass filters in a row

And now three high-pass filters in a row.

three high pass filters in a row

And what if they are asymmetrical, as is probably the case between your console, DSP, and speaker from different brands?

Check out the result of a 24dB/oct Linkwitz-Riley into a 12dB/oct Butterworth into a 48dB/oct Chebyshev.

three asymmetrical filters

And that’s when they all have matching frequency settings. Many times in the past I had thought, for one reason or another, that one filter should be lower and another higher.

3 filters at different frequencies

For this reason it’s best to leave system calibration filters until the end of the signal chain, right before the speaker. That way you can get the job done without a bunch of unnecessary steepness and phase delay.

Some exceptions:

  • Input channels. I should make it clear here that I’m focusing on output channels. When it comes to input channels I am much more liberal with the HPF. Put’em on every channel and crank’em up as high as possible.
  • Either you don’t have access to the DSP or the filters are baked into the speaker (as they often are), but you want to change the filters. You can’t put sound back that’s been taken away, but you can make the filters steeper and/or raise the frequency in the example of a HPF. Proceed with caution and measure the acoustic results.
  • Either you are very experienced or you know what you’re doing, as with any rule.

Case Study

My friend David uses a common setup: DSP into a powered speaker with more onboard DSP. In this case it is the DBX Driverack into a Turbo IQ12 and Yamaha DXS15XLF.

Here’s how he has the crossover filters set up:

  • Out1: HPF 110Hz LR48
  • IQ12: HPF 120Hz LR24
  • Out2: LPF 100Hz LR48
  • DXS15: LPF 110Hz LR24

I didn’t measure each step in the chain individually. Let’s look at the result electrically instead.

asymmetrical stacked crossover filters

If a gap crossover is your goal, you acheived it, but in my case I was often doing this on accident.

I know the acoustical result is really what we care about here, which is what you saw at the beginning of this article.

After David and I had a chat about what he wanted to achieve he decided to remove some filters and move their crossover frequencies. Here’s the result. There’s still a gap, but now it’s less violent.

For more on how to identify crossover slopes, please read Spot Crossover Slopes in Smaart® and Avoid Falling Sharply.

Do ground stacked subwoofers really give you 6dB for free?

By

  • Any measurements made at ground plane will always be 6dB louder, despite subwoofer distance.
  • Flown subwoofers give you 6dB for free just like ground stacked as long as a maximum height is respected.

I have always thought that ground-stacked subs would produce 6dB more SPL because they are coupling with the ground. You’ve probably heard people say things like:

  • “Having subs next to a boundary gets you 6dB of additional sensitivity”
  • “The ground is like a mirror, doubling sub energy. “

I was just running a couple of tests for myself to confirm the reliability of this number, but I couldn’t. In fact, all of the models that I tested actually returned lower average level when the subwoofers were on the ground compared to flown in the air.

50ft Room

50ft room model
50ft room measurements

In the low range the flown sub dominates, but later the ground sub takes over.

150ft Room

150ft room model

I also thought that a corner placement would give you even more SPL, but I couldn’t prove that either.

Maybe the room gain from the walls is making the phenonenon harder to verify. Let’s try a bigger room.

300ft Room

300ft room model
300ft room measurements

Rats.

150ft – Outside

Surely if I turn off all of the walls, except for the floor, we’ll get that free 6dB.

150ft outside

Damn it. I want my free money back.

Why is this happening?

I have a two ideas:

  1. I put the mics in the wrong place.
  2. Flown subs get 6dB for free, too, baby!

The Mirror Effect

mirror effect

I had been taking ground plane measurements in the first models to remove the floor bounce, which made sense to me since I assumed an audience filled with people would have the same effect. There are two problems with this:

  1. Below 100 Hz human bodies have little absorption, which is mostly where our subwoofers live. Plus, it’s hard to predict exactly how the bodies will be distributed.
  2. If my measurement is coupled with the floor, it effectively shows half-space loading at any distance due to the mirror effect.

If the listener is located at the boundary he will hear a 6 dB louder direct signal than he would have heard if there was no floor regardless of where the subwoofer is located.

Comments On Half Space, David Gunness

If we zoom in on on the y-axis, it’s hard to tell which is which because any comb filtering is eliminated. No matter how far away the sub gets, measuring at the ground will show coupling.

ground plane only
Ground Sub
flown at ground plane
Flown Sub

If that’s the case then I should try measuring at head height.

Head Height

Let’s simplify the test by removing the walls and ceiling and using a single microphone position so that I can actually get this article done this year. I’ll move the mics up to head height, since that wasn’t working earlier.

Here’s the result from the same 300ft room, this time without walls, ceiling, or a 3-mic average.

Flown subs get 6dB for free, too, baby!

It is shown that a flown subwoofer [will] have a similar far-field efficiency to that of a ground-stacked subwoofer when a maximum subwoofer height is respected. This maximal height is linked to the venue depth via the DHER criterion, and depends on the listening height. For a standing audience, SPL efficiency is recovered for listening distances that are 5 times or more the subwoofer height. The audience benefits from a more homogeneous SPL distribution and an important SPL reduction close to the stage.

AES Convention Paper 10051, On the efficiency of flown vs. ground-stacked subwoofer configurations, Etienne Corteel, Hugo Coste-Dombre, Christophe Combet, Yoachim Horyn, and François Montignies

Pretty cool, right?

So if we want to recover SPL efficiency at ¾ audience depth, then the sub can be as high as 36.75ft and we get the added benefit of an improved front-to-back ratio.

5x rule for sub height

The real benefit of ground stacking has to with the fact that listeners’ ears are not typically on the floor, but four or five feet above it. If the speakers are elevated above the floor, 45 degrees above horizontal from the listeners perspective, the ground bounce will produce its first comb effect notch at about 80 Hz. If the elevation angle is 30 degrees, the first notch moves up to 113 Hz. If the subwoofers are on the floor, then propagation is parallel to the floor and there is no ground bounce. Hence, there is no comb effect.

Comments On Half Space, David Gunness

It’s interesting that our measurement position is less than 10º, putting the first dip from the comb filter at 297Hz, well out of the operational range of this subwoofer. If you wanted to create a null at 125Hz, you would measure at 72ft depth at 23.8º with the sub.

9.5º

The number isn’t black and white, of course. Even with the first null at 297Hz there is a 3dB drop at 140Hz.

floor bounce calculator
Merlijn Van Veen – Floor Bounce Calculator

So we can see that as we move closer to the sub, the difference in distance between direct and reflected sound will affect more of the operational range of the sub. At this time it is my understanding that with flown subwoofers we accept some amount of comb filtering in the front portion of the audience (anywhere before height of sub multiplied by five for standing head height) in exchange for improved front-to-back ratio, coupling with mains, and SPL efficiency in the rear portion of the audience.

What are your experiences on half-space loading? Comment below.

How to take fast impulse response measurements in Smaart© without pissing people off

By

Use these settings to efficiently investigate room modes with the impulse response module in Smaart©.

  • FFT: 128k
  • Avg: None
  • Delay: 0 (doesn’t matter)
  • Options > IR > Overlap: 0%
  • Signal Generator: Pink noise, Pseudorandom, Drop IR Data Window
    • Level: +10dB above noise floor

Measuring room modes sucks. It takes too long, pisses people off, and the ROI is debatable.

The last one is user dependent. Let’s work on the first two.

Why does it take too long?

This was user error on my part. I thought that multiple averages were required for sufficient dynamic range and actionable data. This meant that all of my measurements took twice as long because I had the averages set to 2.

While increasing averages does improve dynamic range, it is less important in researching standing waves because you’re going to average together at least 6 of them in the end.

Set averages to 0. Fixed.

Why does it piss people off?

  1. Wrong stimulus
  2. Unnecessarily high level

Use period-matched pseudo-random pink noise

The Pink Sweep stimulus pisses people off. People can get used to music and pink noise, but the sweep is surprising and painful. It’s the most common stimulus to enrage civilians.

While the Pink Sweep does improve dynamic range, the results I observed don’t justify the risk of someone telling you to stop. If you’ve got the room to yourself, though, do it.

From what I can see, using pseudo-random period-matched pink noise will get the job done at the same signal generator level and generate less complaints.

Set level +10dB above noise floor

I was using +20dB because according to the Smaart user manual you need to be at least 10dB above the noise floor, but 20dB is OK, and 30dB is preferred.

noise floor comparison

I ran through six tests at six different measurement positions:

  • A: Pink Sweep at +20dB relative to noise floor
  • B: Pink Sweep at +10dB
  • C: Pink Noise at +20dB
  • D: Pink Noise at +10dB
  • E: Pink Sweep at 0dB
  • F: Pink Noise at 0dB

Then I imported the batches of six measurements into Room EQ Wizard to be level matched at 100Hz, time aligned, and vector averaged. Below are the spectrogram graphs for each of the averages.

pink sweep vs noise at three levels
Spectrogram compare

Even though dynamic range drops as the stimulus in the room is lowered, I don’t see any difficulty in identifying the first four peaks in the first four measurements. Only when the stimulus gets to 0dB compared to the noise floor do I start to have doubts.

Obviously the results will be different in larger rooms compared to my living room, but going through all of these tests gives me the confidence to try a -10dB level moving forward.

Extra Credit

Level

Verify signal generator level by comparing it to the noise floor. Switch over to an RTA graph. Measure the ambient noise floor. Offset it +20dB. Turn on the signal generator and match it at every frequency.

While you’re there, make sure that the input levels (Mic/Ref) are matched and within the yellow target zone.

What about wind?

In any scenario where there might be a possibility of any significant time variance during the measurement period, you would probably be better off increasing the measurement time window and/or using a period-matched stimulus signal rather than upping the number of averages.

Rational Acoustics Smaart v8 User Guide

Signal Generator

Pink Noise – Pseudorando

Use a 2-channel measurement with period-matched signal generator.

At first you might wonder, why would I ever use a single channel measurement? IR measurements go back a long way and can be completed with a balloon and stopwatch. This direct method may deliver mixed results, though, since it might not have an instantaneous envelope or uniform spectral content and it doesn’t tell you anything about the sound system.

The newer indirect method uses a 2-channel transfer function to mathematically estimate the response and combined with a test signal of matched length can simplify our goal of actionable data.

feed the DFT what it really wants to eat: a test signal that either fits completely within the measurement time window or cycles with periodicity equal to the length of the DFT time constant. Signals that meet these criteria can produce deterministic, highly repeatable measurements in a fraction of the time it takes to get comparable results using random signals.

Rational Acoustics Smaart v8 User Guide

Unlike the Pink Sweep, pseudorandom cannot be triggered by the play button so you’ll need to start the signal generator before you press play or Smaart may crash, which is what happened to me the first time I tried it.

Pink Sweep

I was curious to see how the results of a pink sweep might differ so I tried that as well.

Using SATlive here to show IR overlay

Ld = level of direct sound

Ln = level of noise

Ld – Ln = dynamic range of the measurement

The dynamic range of the measurement using pseudorandom noise is -44.73dB. The dynamic range of the measurement using pink sweep is -65dB. That’s an improvement of 20.27dB.

FFT

Unless you are in a very large or noisy room, 128k (2730ms = 2.7s) or 240k (5000ms = 5s) should be plenty. I normally use 128k.

Averages

If we want better dynamic range of our measurements we can either turn up the signal generator in the room, increase the averages, or increase the measurement time. Since my goal is speed, I’m looking for the lowest number of averages and measurement time.

When measuring with period-matched noise or sweeps, averaging is normally set to “None” or 2, although it is still possible that a higher setting could prove helpful if measuring in an extremely noisy environment.

Rational Acoustics Smaart v8 User Guide
averages
AveragesDynamic rangeΔ
0-51.31
2-54.643.33
4-59.995.35

Overlap

Normally this is set to 0% for maximum noise reduction, but can be set higher to save time. Let’s see if there’s any benefit to using more overlap in terms of saving time or improving SNR.

This image compares an IR measurement with 0 averages and 0% overlap in yellow with 2 averages and 99% overlap in blue, which took about the same amount of time to measure.

Doesn’t look like much of an improvement.

Delay Locator

You only need to set the delay locator if you are using random stimulus.

The first is to delay the reference signal to match the timing of the measurement signal, so that the data windows line up. You should always do this when measuring with random signals.

Rational Acoustics Smaart v8 User Guide

We are going to use a period-matched stimulus so we can ignore this step. Yay!

Gratitude

Big thanks to Rational Acoustics support. My laptop crashed a lot during my first tests while “Processing frame”, which really sucked when the signal generator was stuck on. I emailed support. John emailed me back and I found out that an update with a fix was coming out that day. It worked!