High frequency range / temporal distribution
LEDE - Non Environment Methods
Another important issue for a control room is to avoid early reflections occurring in the median plane which could give rise to a series of regularly distanced spikes in the frequency response (comb filter). Those components can be eliminated by using an absorbing or scattering treatment of the surfaces responsible for the early reflections.The general principle is however to prevent the early reflections from the front and side walls from reaching the mixing position by absorption or reflection and take care of the reflections from the back by either diffusion or absorption. For a few years now, methods like the LEDE (Live End Dead End), the RFZ (Reflection Free Zone) and the Non-environment methods are used in order to design the room with regard to those problems.
In this lab we investigated mainly the LEDE and Non-environment methods.
All our measurements are now recorded in the mix-position (center of the room in the width dimension and close to the front wall), and the frequency range is 20 - 2000 Hz (200 - 20000 Hz in Room-Capture).
The first task in this kind of pre-shaped room is to locate the first reflections from the wall. In order to see clearly the reflections we put absorbers on the ceiling and the floor.
Figure 1: Studio covered by absorbers on the ceiling, floor and short walls
The position of the microphone is chosen that way because the mix position is the position were the recorded sound should be the "best". Therefore, the first reflections must be far enough from this position.
Figure 2: Impulse response of the room with absorbers on the ceiling and the floor
As we can see from our design the first reflection should come from the side walls. We put some absorbers on the floor and the ceiling to make the wall reflections clearly visible. As we expected the first wall reflection from sides is in 1ms (corresponds to the distance between the headphone and the microphone via the side wall). We can clearly see the other reflections of the sound from the other side wall of our studio (the sound comes back and forth between the two side walls).
In order to locate the first reflection in a more precise way, we put absorbers also on the short walls (see figure 1). This can tell us if the first reflection comes from one of the short walls or one of the side walls.
Figure 3: Comparison of the impulse responses with (blue) and without absorbers (red) on the short walls
From figure 3 one can clearly see that the first reflection comes at the same time for both cases (around 1 ms), therefore one can say that the first reflection doesn't come from the short wall but from one of the side walls. The effect of the short walls (rear and front wall) is visible after 6ms, which is perfectly understandable when one looks at the "flatness" of the blue curve after 6ms. That is why the LEDE and Non-environment methods are used, in order to avoid this first reflection from this wall which is in most of the cases the closest one to the mixing console.
One can also notice here that the red curve (without absorbers on the short walls) shows sharp peaks in the first reflections which indicates wide frequency band components mainly (high frequencies). The reflections coming from the short walls show only small fluctuations (likely smaller frequency components).
On the figure 4 below one can see the time and frequency domain of the configurations studied above.
Figure 4: Frequency responses and impulse responses of the room with absorbers on the short walls (blue) and without absorbers (red)
The flutter echo effect can be improved by adding :
We can clearly see that the first reflection stays like before when there was no absorbers on the short wall, but there is a high attenuation for the other reflections from the other walls after the first reflection. So by adding absorbers to the short wall, we can reduce the flutter echoes coming from the short wall.
On the figure 5 below, some diffusers have been added inside the room in order to see if those can avoid the flutter echoes.
Figure 5: Configuration of the room with absorbers everywhere and some diffusers on the side walls.
Figure 6 : Impulse response and frequency response of the room using diffusers on the side walls (blue) compared to the ones of the room without diffusers (red).
In this case we added some diffusers and some reflectors at the side walls to make a better reflection and a flatter frequency response. We can see that the impulse response curve is flatter than before but the frequency responses of the room still look similar to each other in both cases up to 1500Hz. After this frequency the improvements change the frequency response clearly.
Having a look at the impulse response one can see that the diffusers create modulation in the impulse response (which could be expected). At least except for the first one there is no clear reflection... One has to be careful while using the diffusers and one could think about coupling the diffusers with absorbers in order to have better impulse responses.
In order to find out what could be the best design for this room, we investigated two different ways of covering the room: the LEDE and the Non-Environment.
Firstly the LEDE design was investigated by placing absorbers on one half of the room (front half) to create a reflexion-free zone around the listeners. This section is acoustically dead to prevent any room effect from returning from the frontal direction. The Live end is obtained by placing diffusers on the other half (rear half) to create a "live" impression. This section is highly acoustically diffusive to simulate a sense of ambiance within the room that does not obscure the general clarity of the monitoring.
The dead end is used to eliminate secondary paths to the listening position. The live end, created with a set of diffusers to avoid sharp reflection, provides a set of late reflections to the listening position that do not color the sound.
Figure 7: LEDE configuration. The picture has been taken before placing the absorbers on the front wall and the ceiling
Since this method is basically used to avoid the first reflection, and to have a "good" room in low frequencies also, we added low frequency absorbers, i.e. porous absorbers in order to absorb a wider frequency range.
The second method investigated is the Non-Environment. This design consists of applying absorbers on every wall except the front wall and the floor. Non Environment control rooms are basically “hemi-anechoic chambers”. The floor and front wall are not damped in order to be able to work in this room. An anechoic chamber would be really uncomfortable for the listeners to work in. The main drawback of this kind of room is that they do not translate a typical living room where people listen to music. But here, we are supposed to design a studio room, intended for recording music, not for listening, so this drawback is not relevant here.
Figure 8: Non-Environment method applied to the room. The ceiling has also absorbers on it.
Having an important absorbing surface and a low quantity of reflexions, the ratio of the direct sound to the reflected sound is increased and hence the level of coloration is reduced.
On the figure 9 below is displayed the comparison between the two methods under study. The frequency response and the time response are plotted.
Figure 9: Comparison between LEDE method (blue) and Non-Environment method (green)
As expected both methods give improved frequency responses and impulse responses compared to the unmodified room. Unless both methods give quite similar results overall (especially in low frequencies) one can detect tiny differences. Regarding the impulse response on can see that the LEDE shifts a bit the first reflection in smaller time, but this can also be due to the fact that the room is smaller when using the LEDE method than the NE method. Having a deeper thought about the differences one can notice that the impulse response is smeard out faster using the LEDE method than the NE one. One can then think of the LEDE as a better method. But one has to be careful about it because it is a bit tricky to conclude about how good or bad a method is, since the headphone, used as a source, influences a lot the responses.
Considering now the high frequencies (between 1500 Hz and 5 kHz), even if the LEDE method attenuates more the frequency response, there is only a small difference between both. But at higher frequencies (above 5 kHz), the LEDE method starts to drop a lot, whereas the Non-Environment method remains more stable. So, as an overall tendency, we can say that the Non-Environment method gives a more even and flat response of the room for a large frequency band.