Myths and facts about studio acoustics II: Monitors in a room
Italian acoustician DONATO MASCI from Studio Sound Service continues his series of articles on studio acoustics with a look at the pros and cons of main and nearfield monitoring.
2.1 Damned big in-wall monitors!
These days, the choice of studio monitors is very influenced by fashions and trends and against this background the issue of big monitors is definitely the most debated maybe because they are usually sold only once a control room is built. As in many similar situations, if you ask around, those who have them say they are necessary while those who don't tend to say they are useless.
From a technical point of view, most popular recording studio designs (Non-Environment, LEDE, RFZ) base their entire theory on these on big monitors and especially on their in-wall mounting. For nearfield and free-standing monitors you generally accept various reproduction compromises.
From my research many engineers are unhappy with the big monitors either because the are accustomed to working only with nearfield monitoring positioned a maximum 1.70 m from the listening position or because they have been "burned" by the incorrect positioning and mounting of big monitors, which, unfortunately, happens more often than you might think. The first reason might be generational as it is common for younger engineers who debuted in their home studio (with a compromised listening environment) to not feel the need to change. On the other hand, we should bear in mind that in the 70s and 80s, big monitors struggled to provide good reproduction quality at low SPLs so many engineers preferred small monitors for working at lower levels for many hours without fatigue. This could be one of the reasons why nearfields got a foothold together with the fact that rooms have got smaller and nearfields are the only possible solution.
Nowadays big monitors (or at least the best ones) sound very good at low SPLs too. Obviously, if you need a full-range system with nearfields it should be mandatory to add a subwoofer. It is very interesting to see how many engineers hate this configuration, and in many cases I would agree with them, because if the phase and the level of subwoofers are not perfectly calibrated with the system it is definitely better to avoid the subwoofer. (I talked about this topic in the last article V13.2 and I will discuss it further in the next one). Clearly, if you always use a high pass filter at 45-50 Hz, as many do, surely you avoid a lot of listening problems even with a nearfield system. However, many of my clients working in mastering studios tell me that it common to find mixing errors in the range between 20÷50 Hz, due to the excessive use of bass enhancement plug-ins because these frequencies are not correctly monitored with aforementioned configuration.
There is also a technical aspect that is related to the difficulty in installing a monitor that could reproduce lower frequencies. From me it is hard, if not impossible, to have a satisfying listening experience with big monitors if they are not flush-mounted into a wall. I have also noticed a lot of differences in the frequency response and the perception of the phantom centre image when mounting monitors in a big box made with chipboard, MDF, plasterboard etc., rather than mounting them in a masonry or concrete cavity. If the box is not really massive there will be some problems, in particular related to the low frequency resonances of the lighter layers. In addition, the interaction with the other walls (particularly the one at the back of the monitors) results in cancellations due to phase shifts of the direct and reflected sound. You should never forget that monitors are almost omnidirectional below 200 Hz.
Many engineers who don’t analyze it from a scientific point of view might disagree because in-wall flush mounting emphasizes low frequencies. This is normal because the monitor is usually factory-calibrated in an anechoic chamber. It is easily understood that the energy of the low frequencies radiated from the back of the monitor are reflected forward by a solid and massive infinite baffle. However, this would also be the case with walls placed 1 m or more behind the monitors (you must remember that usually big monitors are 50-80cm deep and with the usual 30° horizontal angle and the vertical tilt, they can easily be placed several cm from the wall behind them).
When the reflected sound interacts with the direct one, it creates cancellations (comb filters). The wavelength related to the cancellation’s lower frequency (which is also the stronger one) is strictly related with the distance from the monitor to the walls behind them. The only way to avoid this cancellation is to reduce this distance to zero – to flush mount. With this phase issue, a simple equalisation will never solve the problems, whereas for flush-mounted and very close-to-the-wall monitors, the bass boost could be controlled with equalisation.
2.2 ...and for free-standing monitors?
For the same reason, it is hard to obtain good results with free-standing monitors that are not close to the wall because on the back wall you should have total absorption (of the frequency related to the cancellation) to retrieve a flat frequency response. This is very difficult to achieve from a practical point of view because we are talking about low frequencies. For instance, if a monitor is placed 1 m from the front wall, the cancellation will be at about 87 Hz and you would have to use a sound-absorbing trap thicker than 1 m (where to put it?) or some resonant device that would ensure a precise absorption with the same Q as the cancellation, which is challenging to design.
As Genelec reported with its simple AcoustiTape (a truly useful tool to predict precisely this cancellation) "sound travels 1/4 wavelength to the wall, bounces off, then travels 1/4 wavelength back from the wall, thus and is 180 degrees out of phase (half wavelength) with the direct sound". For this reasons, I am always puzzled to see photos of control rooms with free-standing monitors set far from the front wall. I rarely change my mind after listening or measurement.
2.3 Case study: big monitor control room pre and post treatment (Dunastudio)
A few months ago I was called to analyze a studio (Dunastudio) in the province of Ravenna (Italy), whose control room had several acoustic problems. In particular, the sound absorption was unbalanced (too low absorption at LF and too high at HF) and the listening point was exactly halfway between the front and the back wall. The control room had a pair of Dynaudio M4s the acoustic axis of which were not directed towards the listening position and the monitors were not properly flush-mounted into the wall.
The correcting intervention consisted of putting adequate absorption on the back of the room, on the side and on the ceiling, a slight repositioning of the listening point, and a complete redesign of the monitors flush-mounting.
In Figure 1 you can see all the pretreatment frequency responses problems and in particular resonances and cancellations (respectively at 60-70 and 90-120 Hz) at low frequency due to incorrect construction of the baffle and to the room modal resonances that were not controlled. You can also see the loss of high frequencies, mainly due to the tweeter not aimed towards the listening position in a room that had too much high frequencies absorption. In the posttreatment frequency response these effects have been substantially improved, moreover the responses of the two monitors are very similar to each other.
The pre-treatment reverberation times (Figure 2) were too unbalanced between the mid-high (0.15 s) and low frequencies (0.55 s). In addition a resonance peak at 125 Hz was very noticeable (it is not a coincidence that in this band there was also a dip in the frequency response).
The posttreatment readings show how the relationship between the reverberation times in the mid and low frequencies were optimized, the modal resonance at 125 Hz had been controlled and the control room gained a little extra brightness on the midrange with a light diffusion.
The pretreatment Center Times (Figure 3) were too long on LF and too short on HF, the correction led the values perfectly on the average that we discussed in the last article (Resolution V13.2).
2.4 Dispelled myths
Back to our myths then (outlined in the last issue) and the process of dispelling them.
5. Big monitors are good for clients but not for mixing. They're just too big for that, they lack “definition” – On the basis of the results shown and the explanations proposed, I can say that flush-mounted big monitors are truly very challenging to install in a control room, but personally give me more satisfaction, because when they sound good, it is almost as if all the acoustic design is brought to life.
6. I don't want my monitors in the wall, it's not necessary and I can change them easily if I ever want to. Furthermore, they can be moved about in case I need to perform a “fine tuning” – On this point I would say that you should not think of a monitor as an object of trend to be changed on a whim but as a real part of a control room. It is good to choose it when you decide what kind of control room to build.
7. Nearfield monitors have much more “definition” than far-field ones – Wrong. Far-fields introduce more energy into the room so they excite its modal resonances more than the nearfields, but the effect of the first reflections is surely worse for a free standing nearfield than for a flush-mounted big monitor.
In addition, nearfields have a lot of other problems, such as those related to the directivity. Indeed, the dimensions of a monitor cabinet influence the sound radiation. This effect exists when the wavelength that is generated by the monitor is identical (or proportional) to one of the dimension. The enclosure starts to be directive for those frequencies and the related harmonics. So, for a small monitor, the dimensions correspond to midrange which is already directive. This way it reinforces the phenomena and degrades the off-axis response. In a large enclosure the box dimensions correspond to lower frequencies where the energy is much less directive, and hence the effect becomes negligible. This is why the same drivers in two different enclosure sizes will generate different sound characteristics. The smaller the box the more effect it has on the directivity of higher frequencies. Furthermore, a small enclosure is very affected by a large mixing desk in its radiation load at LF, by the wall behind and the induced comb filtering, and the console reflection. It is an effect of the proximity of large objects that are close to the drivers. For a big monitor in-wall there are no objects close to the drivers in the case of the mid and tweeter at least. In a small enclosure it is often the case, and that has an effect on the response.
To conclude, my advice is to think about big monitors for your control room if you have the opportunity. Because of the necessary phase coherence between individual drivers, bear in mind that the minimum listening distance cannot be less than 1.5 m, so it would definitely be useless to place them in rooms that don't have a suitable size. In large control rooms the common typical listening distance is about 2.5 meters. In smaller control rooms, however, it is possible to flush-mount mid-fields with excellent results and shorter listening distance. The mounting and positioning of a monitor is fundamental in a control room; incorrect positioning may cause dips in the frequency response of up to 15-20 dB! In the next article, I’ll look at (auto)calibrations, more myths to be dispelled and some conclusions.
Donato would like to thank Christophe Anet and Valentina Cardinali, for useful discussions and thoughts, and his friend Christopher Martinuzzi, for reviewing the text.