Haas Effect

Goal

Understand Haas effect.

Haas Effect

In 1949 Dr. Helmut Haas was studying the effect on our ability to locate sound when delayed copies of the sound are involved. What he discovered became known as the Haas effect and also has the name Precedence effect from reasons that will become clear.

The Haas effect (also called the precedence effect) is a time delay related psychoacoustic effect in which delays of a sound occur in a small window of time after each other such that that we cannot tell them apart from direct sound. This small window of time can be considered the "echo threshold" of humans and it varies depending on the content of the sound. We will give "precedence" to the first wave to reach us and use the other waves that hit us in the echo threshold to determine information about the space. Because of this it is also called "The Law of the First Wavefront." Importantly we do not hear these other delays sounds as discrete echos but rather as part of the direct sound, once we start to hear discrete echo we have left the Haas time window or the echo threshold.

This delay is commonly shown has a delay between the left and right speaker. Most Haas plugins do this. This however doesn't need to be the case, it can be a dual mono delay as well. Dual mono delays in plugins however come across as phasers and so to get a clear stereo image plugins focus on the delay between speakers.

Here is a sound with an echo, as it gets brought closer and closer there is a point where it appears to merge with the direct sound instead.

Eventually an audio example will be here.

Haas Window Breakdown

The window the Haas effects takes place in varies with material. Material that is transient heavy such as drums will have a smaller window to work with, as low as 20ms, while sounds that are smoother and take longer to get going will have a longer window to work with, as high as 80ms, such as a brass instrument swelling up or a choir.

Spaciousness and width

Reflections off of walls cause the Haas effect to be a concern when designing rooms. Echos down to about 20dB SPL contribute to our sense of "space" within the Haas zone. If the echo is much louder, which is only possible via a second source, the it may be perceived still as an echo. If the reflection is much softer then is will be inaudible and not matter anyway.

Haas effect


The Fusion Zone

The fusion zone is defined as the distance from the source that will not give discrete echos or in other words is within the Haas effect. Since it varies with material we often just have an amount of time that we assume will be in this region, such as 50ms or 40ms. We may wish to setup a space to be within this zone. In general 50ms is a good target for speech and 80ms for music playback.

To design in this zone it is useful to recall the sound at a distance equation:

d=rt d=rt

We desire to talk about the time delay between the direct sound and the reflected sound. We therefore need to know the difference in distance between the two paths:

distance different =ΔD= Reflection DistanceDirect Distance distance\ different\ =\Delta D=\ Reflection\ Distance-Direct\ Distance

Knowing the difference in distance directly we can compute the time delay that goes with this difference:

Δt=ΔDr \Delta t=\frac{\Delta D}{r}

Where the time given is the difference in time or the time delay between the two sounds, and r is given the be the speed of sound.

Since the loudness of a reflection affects how it perceived we may desire to check various distances to see of the reflections will be problematic. For this we can use the distance dB SPL definition we have used before. Recall without the minus sign this equation gives us the decrease in level.

dBSPL=20log(distancemeasureddistancereference) dB_{SPL} =20\log\left(\frac{distance_{measured}}{distance_{reference}}\right)

Example

Consider a direct sound with path length of 50 ft and an early reflection with path length of 75 ft. What will the time delay be? If we want to be within 50ms what is the maximum path length difference we can have?

Using the distance equation we can find the time delay:

Δt=75ft50ft1130fts.0221 or 22.1ms \Delta t=\frac{75ft-50ft}{1130\frac{ft}{s}} \approx \boxed{.0221\ or\ 22.1ms}

We can solve for the difference in distance in order to find the max difference in distance that remains in the fusion zone.

ΔD=Δt×r \Delta D=\Delta t\times r

Now plugging in we find:

ΔD=.5s×1130fts=56.5ft \Delta D=.5s\times 1130\frac{ft}{s} =\boxed{56.5ft}

Thus to remain in the fusion zone we must design our space such that no early reflections of significant amplitude are within 56.5ft.

Question

You are designing a room in which the direct path length is 25ft. Assuming a fusion zone of 80ms before discrete echos appear what is the maximum length an early reflection can be?

If the reflection is 45 feet, what level decrease will the reflection experience at the listening position?

Show Answer

We are given the Haas requirement and one of the distances. Knowing only the Haas requirement we can determine the max distance allowed:

.08=ΔD1130ΔD=90.4 ft .08=\frac{\Delta D}{1130} \Longrightarrow \Delta D=90.4\ ft

Now we know the max difference in distance allowed and one of the distances. We know the max distance is defined as:

ΔD= Reflection DistanceDirect Distance \Delta D=\ Reflection\ Distance-Direct\ Distance

Plugging in we obtain:

90.4ft= Reflection Distance25ft 90.4 ft=\ Reflection\ Distance-25 ft

Thus:

Reflection Distance=65.4ft Reflection\ Distance=\boxed{65.4 ft}

Now given a reflection at 45ft and using the dB SPL equation for distance we obtain:

Reflection dBSPL at listening position = 20log(45ft25ft)5.1 dB SPL Reflection\ dB_{SPL} \ at\ listening\ position\ =\ 20\log\left(\frac{45ft}{25ft}\right) \approx 5.1\ dB\ SPL

This assumes no losses which will not be the case in reality. 5.1 dB SPL lower than the direct sound will be loud enough to affect our sense of space. Remember we have set up the equation to give us decreases in level.

Anechoic Chambers

I wanted to briefly mention anechoic chambers. Anechoic means "without echos". There is a similar notion in studio design called "once past" where a singular spot is designed so that sound only passes it once. Anechoic chambers do this for the entire chamber. They are useful for taking measurements. The noise of a computer fan for example.

Microsoft currently has the record for the softest room in the world at -20.1 dBA. Sounds below the threshold of hearing are given as negative decibels.

Saxophone in an Anechoic Chamber.

Busting the myth that you go mad if you spend to much time in such a chamber.

Some more fun examples.

To support this series please consider donating via

paypal

or joining the

patreon.