One of the biggest mistakes I see with beginner producers is that they don’t make good use of the stereo field. Taking sounds that sit dead centre in the mix and widening them can add interest, make them feel bigger and give a more professional sound, it’s also a useful tool for making space for other sounds that should be entered, like your kick, snare, vocals and leads.
How can I make my sounds wide?
In this article I am going to go over 6 simple but highly effective techniques for making sounds wide:
1. Haas Delay
2. Short Reverbs
3. Dimension Expanders
7. Stereo Imagers
What is the Haas Effect? The Haas Effect was discovered by Dr. Helmut Haas in 1949. It is a psycho acoustic phenomenon which states that when one sound is delayed approximately 40 ms or less from another, the two are then perceived as a single sound. It also states that the location of the sound is perceived by the location of the first-arriving sound – so if you have two sounds, one panned left and one panned right and the left sound hits your ear before the right, the location of the whole sound will be perceived as left.
Haas is simply a method of taking a sound and offsetting the left and right signals by a few milliseconds, this causes the sound waves to hit your ears at different times, giving the illusion of a really wide sound.
We can achieve this by using a delay plugin that will allow you to set your delay times in MS (instead of tempo synced) and you need to be able to set different times for the left and right signals. You ideally want to delay the left about 20-40ms away from the right (or vice versa). If you move closer to 5-10ms you will hear the sound starts to phase. If you move too far away it will begin to sound like the audio repeats. You want to find a good balance where no phasing occurs and the audio still feels like one cohesive sound – so just experiment by slowly moving your delay times 20-40ms apart. Make sure to set your feedback parameter to 0%.
Here is an example of a completely mono sound with a haas delay on it:
And here’s how the delay was configured:
Reverb is a great tool for positioning sounds in a mix. You can use it to move sounds backwards, to wash them out, create movement, or to make them feel huge and wide. The great thing about reverb is you are almost always safe from phasing issues.
You can use almost any make of reverb for this but my favourite is Valhalla Room Reverb. The idea is to make a really really short reverb. You want to set your reverb decay to around 20ms, make sure there is no filter on the reverb so you are capturing the whole sound, set the mix level to taste – usually around 20-40%. If your reverb has a “Size” parameter, set it to a smaller size. Depth should be set to around 50%. These settings aren’t gospel so try to experiment until you find a sound you like.
Here is an example of a completely mono sound with a really short reverb on it:
And here’s how the reverb was configured:
Dimension Expander is essentially a four-voice a stereo spreader that deals with the audio in a way that doesn’t cause phasing issues. To put it simply, dimension expanders create four copies of the signal, delay them at different times, and flip the phase of two of the voices so that they are out of phase with the other two. It’s great for making sounds feel HUGE without having to worry about phasing issues.
The best dimension expander I have found is the one made by Steve Duda of Xfer Records, the best thing about it is it’s free to download and use! It has only two knobs, size, and mix. These should be experimented with but my favourite setting is to have a really small size (almost at 0) and then mix it in fairly high – around 50-60%.
Here is an example of a completely mono sound with Dimension Expander on it:
And here’s how it was configured:
Chorus is probably one of the OG ways to widen sounds. They come in many different flavours – from analogue pedals, to VST plugins. A chorus effect works by creating two copies of the sound and offsetting or delaying them slightly, kind of like the Haas Effect delay we spoke about earlier. The thing that sets chorus apart from dimension expanders and Haas delays is that a chorus will also include pitch modulation. The pitch modulation is usually tied to an LFO that will cause the pitch of each voice to move up and down at different times. You can often control the depth and rate of the LFO and setting these too high will begin to make the audio sound out of tune, but mix it in lightly and it can create a very nice wide effect.
Here is an example of a completely mono sound with a chorus plugin on it:
And here’s how the Chorus was configured:
If the sound you are working on is coming from a synth, then there’s usually an option to increase the unison on it. Some samplers will also have a unison parameter. Unison works by making extra copies of the sound and then you can detune each voice slightly from one another. The more you detune the voices, the less likely you are to get phasing issues. This also can add a nice pulsating effect.
Most commonly you will be able to create 2 – 8 voices of unison within your synth, but some more modern synths like serum allow for you to create up to 16 voices of unison. You can also control the volume of the mid voice and the side voices independently, which is great if you are looking for a more subtle effect, or if you don’t want the side signal to be as loud as the mid signal.
Here is an example of a completely mono sound with 7 voices of unison and some generous detune on it:
And here’s how the unison and detune were configured:
Doubling is a technique often used on vocals. One way to achieve this is to have the singer sing two takes of the same line. Due to the imperfect nature of the human voice, this will cause each take to have slightly different tuning and timing. When we pan these two takes left and right it creates the illusion of a very wide vocal and can sound like a “chorus” of singers.
You can quite easily fake this effect with any sound simply by duplicating the sound and offsetting the timing and tuning very slightly. I find this effect can very easily introduce phasing issues so make sure to listen back in mono while you tweak it to make sure you don’t have any problems. If you do then you may need to detune them more or offset the timing a bit more.
You can also introduce more differences by adding more effects to one of the duplicates, anything that will change the waveform of the sound will help with phasing issues. You can try things like distortion, formant shifting, filtering and EQ.
Here is an example of a completely mono sound, duplicated, detuned, offset and formant shifted ever so slightly:
And here’s how I configured this:
Stereo Imagers are great tools but should be used with caution as they can VERY quickly cause more harm then good if you don’t pay attention to phase of your sounds. They work in a similar way to all other widening plugins which is to essentially mess with the phase of all signals using delay, detune, EQ and other techniques, and then pan them out.
My favourite stereo imager is the iZotope Stereo Imager. I love this plugin because it allows you to create multiple bands and to control the width of each band separately. You can therefore set your first band to control everything below 100hz (your sub) and make sure that is perfectly in phase by bringing the width to mono or close to mono. You can then create a few more bands for the mids, and highs and control how wide you want those. Another nice feature is this plugin includes a Vectorscope which gives you live visual feedback on the phase correlation of your mix, so you can immediately see with your eyes if your mix is going out of phase. The trick to using this is to make sure your Phase Correlation meter never goes under 0, -1 means sounds are out of phase and will cause ugly issues, and +1 means they are in phase. A completely mono sound will sit at +1.
I find stereo imagers work best with sounds that already have stereo information.
Here is an example of a mix with the iZotope imager applied to it (pay close attention to the mids):
And here’s how the imager was configured:
Panning is the safest and easiest way to widen your mix. But isn’t so useful for widening individual sounds. You may have noticed that almost all widening and stereo tools use panning to some degree but I am talking about panning individual sounds to create a nice wide stereo image.
This applies more to a group of sounds than to individual sounds.
Let’s take your all your hi-hats for example. Let’s say you have an open hat, closed hat, and 2-3 other hats in your beat. The simplest way to make this nice and wide is to have the main hat hitting dead center and then to pan each of the other ones nice and wide in opposite directions, you might even automate the pan of one of the hats to slowly move from left to right to create movement. Ableton has a nice built-in Auto Pan effect for this but I’m sure you can find other VSTs that accomplish the same thing, or you could map the pan knob of your insert to an LFO to achieve the same result.
Here is an example of a hat pattern with some panning + auto panning applied to it:
As you can tell from all these effects, for the most part they work on the same basic principle. Make copies of the sound and differentiate them enough to prevent phasing issues and to trick the ear into perceiving the location of the sounds as coming from the left and right speakers.
What are some techniques you use to gets sounds nice and wide?