To time align your Main to your Front-fill using Smaart, first find the acoustic crossover point where they have matching level, then match their arrival times using the delay finder.
Setting the delay for the front-fill speakers is a very easy process using Smaart. The challenge is in choosing the correct measurement microphone location. If your mic location is too far into the audience, your front-fills will be too loud, tearing off the faces of the season ticket holders in the front row. If the mic location is too close to the stage, timing errors and comb filtering will worsen.
Here’s an overview of the process:
- Set relative solo level and EQ for Main and Front-fill at on-axis positions.
- Find the acoustic crossover point.
- Set delay.
Crossover (acoustic): The frequency and/or location where two separate sound sources combine together at equal level.
ONAX: A mic position used for calibration located at or near the horizontal and vertical center of a given speaker or array.
Sound Systems: Design and Optimization by Bob McCarthy
A system can only be optimized as well as it is designed. For this example, we are going to assume that the system has been perfectly designed so that it has minimum level, spectral, and ripple variance from vertical top to vertical bottom of the audience area covered by the Main. In laymen’s terms: it sounds the same everywhere we expected it to.
Now, all we need to do is fill in the gap left in the front of the stage with our Front-fills.
Let’s get a little more detailed.
Set relative solo level and EQ for Main and Front-fill at on-axis positions.
Another way to say this would be ONAXL (on-axis with left) = ONAXF (on-axis with front-fill). And it goes like this:
- Measure the Left Main solo @ONAXL. Set EQ. Capture trace.
- Measure a single Front-fill solo @ONAXF. Set EQ and level to match ONAXL.
Before we combine the Left Main and the Front-fill, we need to match their solo on-axis response and level. Otherwise, it will be difficult to create a seamless transition at the acoustic crossover point.
Here are the solo measurements of my Main and Front-fills. Pink is the Left Main array and green is the Front-fill.
Find the acoustic crossover point.
Now that the sub systems are playing nicely alone, let’s schedule a play date. Where L solo = FF solo we will find XLF (crossover between left and right).
- Start at the geometric mid-point between L and FF or wherever your ears find matching level.
- With both arrays playing at the same time, move the mic while watching the Live IR until the peaks match.
The way I used to do this was very slow. I would guess the position of XLF, measure each system solo, resetting the delay each time, look at the magnitude graph, then move the mic, over and over again until I found matching level.
A turning point for me was discovering the option in the Transfer Function Options page called proportional panes. This makes the Live IR graph bigger, which really helps for tracking it on an iPad screen.
Measuring two sources at once can be tricky. If you don’t see two distinct peaks, they may be out of bounds of the graph, already on top of each other, too low in level, or something else. If I zoom out, zoom in, and still don’t see them, I’ll go back to solo measurements to track them down.
Set delay
Now that you have found XLF, the process of setting the delay for the front-fill is simple.
- Measure Main solo. Use the Delay Locator to set the compensation delay in Smaart.
- Solo the Front-fill.
- Click Find in the delay locator. Observe the delta delay value. This is the difference in the current delay and the measured delay. Put this number into the delay line for your Front-fill.
- Click Find again to verify the change. Delta delay should be very close to 0.
- Measure L+F @XLF to verify summation.
You could also just mouse over the second peak in the Live IR window to see the time offset, but you have to get your mouse in the exact right spot. I find using the Delay Locator a bit easier.
Don’t worry if you don’t get a full 6dB of summation at every frequency. Focus on the areas where both traces have high coherence. If you suspect a misalignment, take solo measurements again, observe their phase relationships and adjust.
Have you tried this method for aligning Mains to Front-fills in Smaart? What were your results? Do you use a different method? Let me know in the comments below.
Bonus
If you’re curious, here’s my microphone map and system tuning checklist for the example above.
Smaart® and the Smaart logo are registered trademarks of Rational Acoustics LLC and are not affiliated with Nathan Lively or Sound Design Live.
Hello Nathan,
first of all I’m so grateful for all your tutorials, it brings my college knowledge about sound engineering to the next level. It also gives me a couple of ideas for my own classes in secondary technical school.
I have some questions regarding this article. If the purpose of finding an acoustical crossover is matching the time of sound arrival from both Main and FF in this position, I probably should match the phase response as well – just like I’ve been told when learning about sub alignment.
Then, should I be using only the the same loudspeaker model?
Also, won’t some EQing the FFs mess the phase up?
Moreover, I’m a lil’ bit confused about on-axis solo measurements. Am I supposed to do them in the free field or it doesn’t really matter?
Hope you’re doing well during pandemic. Stay healthy up there!
Best regards,
Łukasz
Hey Łukasz, thanks for checking out the article!
> If the purpose of finding an acoustical crossover is matching the time of sound arrival from both Main and FF in this position, I probably should match the phase response as well
Yes, great idea. Just remember that in the HF, as wavelength gets shorter, the phase will change very rapidly with distance, so don’t obsess about it being off by a few degree at 10kHz. As soon as you move the microphone an inch, it will change. 🙂
> should I be using only the the same loudspeaker model?
Should? Yes. Ideally we would only ever use matching loudspeakers, but that often does not happen in practice. More often, especially for FF, you’ll get whatever was available at the warehouse at the time, so it’s good to practice making it work with whatever you have.
> won’t some EQing the FFs mess the phase up?
Yes, but it might not be as bad as you expect.
1. Often a corrective EQ filter will also correct the phase response, making it more compatible.
2. https://youtu.be/14wF0zGiJuQ
> I’m a lil’ bit confused about on-axis solo measurements. Am I supposed to do them in the free field or it doesn’t really matter?
I’m not clear on this question. What would be the alternative to free field? diffuse field? Ground plane?
Whoa, that’s a lot of useful information, thank you!
Let me bother you a little more:
>Just remember that in the HF, as wavelength gets shorter, the phase will change very rapidly with distance
Is that true even in anechoic / ground plane conditions? I did noticed that, but I though it’s mostly related to reflections causing comb filtering etc. I didn’t saw any loudspeaker phase vs directivity plot in my life though, doubt that they even exists.
>I’m not clear on this question. What would be the alternative to free field? diffuse field? Ground plane?
The opposite would be a diffuse field – practically, when D/R ratio become less than 10:1 and room starts to interfere. Maybe I get something wrong, but comparing MAPP 3D razor-sharp predictions to my own measurements, made with cheap studio monitors in acoustically unprepared room. I think that at home I mostly measure my room response even if my mic is only 1,5m from the loudspeaker. That applies mostly to low frequencies but it’s not limited to them – I can clearly see comb filtering caused by table reflection and so on.
That [bunch of nonsense] being said: should I make “relative solo and EQ” in my target location or do it in some better conditions – just like sub-main relative alignment – to be sure about traces? If the location is just fine, do I need to make 5+ measurements and make some optical-mathematical averaging to get more reliable data? If not, then I don’t get it: where is the exact position where I should measure them both? On-axis seems like infinite number of places to me…
> Is that true even in anechoic / ground plane conditions?
Yes. To confirm, take a measurement or simulate in MAPP, then move the mic 2 inches and compare the phase shift.
> The opposite would be a diffuse field
Ok, I see. On-axis doesn’t say anything about the distance. In the case of measuring a FF, you are often measuring from the first or second row. I should be mostly direct sound.
> should I make “relative solo and EQ” in my target location or do it in some better conditions
At the end of the day you have to do whatever is necessary to get actionable data. If your measurement does not lead you to any clear action then something must be changed until it does. It depends on what’s causing the data corruption. If you are measuring in a small room and you believe that the reflections are a problem, you could try taking several measurements around that location and creating an average. The biggest help, in my experience, is to have a near-anechoic measurement of the speaker to compare it to. You can do this in the field by taking a near-field measurement of the speaker that is mostly reflection-free. Then use it as a comparison with your room measurements and you should be able to start to understand what affect the room is having on the system and make some better decisions.