Stereo / Mono / Distributed Systems
|This article applies to using multiple L1® s in various ways: Stereo, Mono, Distributed Systems.|
- The Compact looks really attractive as a lightweight solution for running stereo. Are there any special considerations for doing this? I am looking for differences from the way we might have approached this with the other L1® models
Ever since the L1 approach was introduced in 2003, we've had a vibrant discussion here about the subject of instruments that work in stereo -- predominantly keyboards and occasionally guitar (Adrian Belew, Pat Metheny for example).
Here a few things to consider from my perspective:
- There are many reasons a player would want to hear his or her instrument in stereo. Having different sounds come from different directions can be very exciting sonically and I doubt there's much controversy about that.
- An L1 system can project a wide stereo field on stage and into the audience -- particularly the Model I and the Model II. Because the I and II are essentially 14' line arrays (7' speaker with a 7' mirror image source created by the floor reflection) they project sound at much less loss with distance than a conventional speaker. This has the effect of greatly enhancing the size of the stereo field.
- It's a mistake to place stereo speakers too closely together. Not only does the area over which listeners hear stereo sound shrink but you can get undesirable interference effects caused when the same (or very nearly the same) signal is played simultaneously through two or more like speakers. (Multi-source interference can cause audible dropouts in tone.)
- For these reasons, I'd recommend spacing L1 systems used for stereo at least 10' apart. There's competing interests here. A player located too close to one of the L1 systems spaced too far apart won't hear stereo — but his or her audience will. On the other hand, space the L1 systems closer so that the player gets stereo and you won't project stereo to the audience.
- A player located roughly between two spaced L1 systems will hear stereo well and many in the audience will too. 10-20' feels good to me for this kind of setup although I don't want to have this come across as more than a rule of thumb.
- The L1® Compact uses a significantly shorter array than the L1® Model I or L1® Model I. As a result, it doesn't project sound as far. Said another way, the level falls off more rapidly with distance on the Compact. As a consequence, the stereo field won't be quite as large. This will translate I think to a smaller stereo sweet area for the player and for the audience. On the other hand, as has recently been pointed out on this forum, a player can buy two L1 Compacts for the price of one L1 Model I and can buy three for the price of a Model II and a T1 ToneMatch engine. For those that value a stereo instrumental sound, experimentation with the various L1 models is encourage.
- I noticed in the L1® Compact Owner's Guide (pdf format) (page 7), there are instructions on how to use more than one Compact to provide additional coverage in larger spaces. This is new. It hasn't been easy to do this with the L1® Classic / L1® Model I / L1® Model II.
- There were recommendations that the minimum distance between two L1®s running the same source was 20-50 feet. (see: Multiple Source Interference). We know that the L1® Compact performs differently than the other models, so does this mean that this recommendation is no longer applicable?
- Whenever you feed an identical signal to two or more like sources you must ask the question: will I get undesirable interference? Will I get unwanted dropouts in tone?
- The most important application for an "identical signal feeding two or more like speakers" is when covering a large area. A good example would be an audience of 1,000 or more seated for a graduation ceremony. A distributed system of several L1® systems all fed with the same signal is perfectly appropriate. We make it very easy in the L1® Compact to do this with the line output jacks on the back. You can easily daisy chain them.
- I can't think of a situation like this where you would want to space the L1® Compacts (or any other L1® system) closer than 20' and thus I see no issue with undesirable multi-source interference. I think that spacing closer to the 50' mark would be more typical in terms of covering large audiences. Said another way, I don't think an L1® Compact has trouble throwing up to 50'.
- As said above "The L1® Compact uses a significantly shorter array than the L1® Model I or L1® Model II. As a result, it doesn't project sound as far." Thus, spacing of multiple L1® Compacts would be somewhat less than the spacing of multiple Model I or Model II systems.
Two L1®s in Mono
How much gain do you get when you run two L1® with the same source.
- The most you can ever ever get from putting two sources fed by the same signal together is 6 dB. The reason? sound waves add in space through the principle of superposition. If the waves from two sources perfectly add, you get twice the amplitude, which is 6 dB. Now, I've been proven wrong on a number of occasions, but if that happens on this occasion, Alan, the sound engineer you've been talking to is going to get a Nobel Prize.
- The problem is that to get the full 6 dB the two sources have to be spaced very closely together relative to a wavelength of sound. at 50 Hz, where the wavelength of sound is about 20' (about 7 meters) that's pretty easy to do. Massing your B1 bass modules is done for this reason -- to get the 6 dB per doubling of B1 systems. But at 1,000 Hz, where the wavelengths are 1' (about 30 cm) you just can't do get L1 speakers close enough. As a result, you get something less than 6 dB because at some angles instead of the sound waves constructively adding they will destructively cancel (one will be partly or completely out of phase with the other.
- Once you separate the sources by several or more wavelengths, the summations and cancellations are so numerous and so complex that we say the two sources are essentially uncorrelated (not strictly true, but for the purposes of this argument, a very close approximation.) When two sources are uncorrelated the only thing you can count on is that the total amount of acoustic power going into the room will be doubled. Doubling the power is 3 dB.
- So what we've said is
- put B1s together if you can because they'll give you an honest 6 dB per doubling of B1 systems;
- if you are sending a mono or stereo signal to spaced L1 systems, try to space them 20' (7m) or more if you can because this will minimize any audible cancellations.
- Alan, you've got enough spacing of your L1 systems and so I really don't see any problem with destructive interference. On the other hand, given the maximum benefit in level is only 3 dB, and you are somewhat reducing the clarity and accuracy of localization, I question the benefit. Your willingness to experiment is a good ingredient.
- I'm going to work next week on a Wiki article on this subject, someplace where I can try to put these thoughts together where they can be found time and time a again.
Distributed Systems for Covering a Large Area
Summarizing from above:
A good example of an application for a distributed system would be an audience of 1,000 or more seated for a graduation ceremony. A distributed system of several L1® systems all fed with the same signal is perfectly appropriate.
Place the L1®s at least 20 feet apart to avoid undesirable multi-source interference. Spacing closer to 50 feet is more typical in terms of covering large audiences.
Bass Power Alley
- The Compact line array is significantly shorter than the Model I or II they could be used closer together than the 20ft mark.
Yes, my instincts and experience tell me that this should be true -- say down to 10' -- but I haven't done a clean experiment to prove it to myself, so some caution is advised.
And I'll point out that below say 200 Hz, all the L1 systems are essentially omnidirectional and the spacing of 10, 20, or even 30' can create an audible "power alley". Once you get beyond these distances, the balloon animal has so many closely spaced bulges and creases -- even at lower frequencies -- that our ear doesn't hear them individually but rather hears the average. The balloon animal turns into something more like a Japanese fan.
Delay and Mono
- Would a few ms delay be enough to avoid the "interference" from dual mono?
Do you have a value for the latency when daisy chaining Compacts?
Pure delay does nothing to fix interference "globally" (I'll explain) but can be used "locally". In this case (L1 systems, spaced, all getting the same signal) I believe delay will turn out to be not useful.
Imagine a balloon animal as a representation of the radiating characteristics of two like sources fed the same signal. The bulges are high SPL and the creases are low SPL, caused by interference.
Say the balloon animal is a dachshund.
Point the nose at the audience.
Pure delay on one of the signals is like rotating the dog.
It doesn't change the bulges or creases (that's the global part) but does change their direction (that's the local part).
If you wanted to change the direction of "power alley" (explained above) caused by two subs on either side of a stage, because the mix engineer was getting blasted by bass, you could do it by applying delay to one of the subs.
All that does, however, is change the aiming of the howitzer to someone else.
This technique can be extremely useful in optimizing the radiation pattern of a speaker in the crossover region, and is used all the time in big-venue boxes.