Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Will Abele]

My high school hired a contractor several years ago to install a new sound system in our 980 seat theater. They replaced the 4 PAS PI15-2.2 speakers we had with 3 EV Xi-1152A-64F and used 2 of the PAS speakers as side fills and 2 as rear fills.

They took over a year and a half to install this system. The main Electrovoice cluster was the first thing they installed. We were unhappy with the way it sounded. I looked through the settings on the Dx38 processor and found that the lows had been delayed 5 ms instead of 562 microseconds. Fixing this setting made it sound significantly better. I then looked at the way that the PAS speakers had been plugged in. They had ignored the PAS processor and run a full way signal to both the high and low frequency speakers. Luckily, I found and fixed this before the speakers were used. In the past year I have redone most of their work.

I am now trying to improve the positioning of the speakers. The main cluster and the side fills are located in a speaker bay in the proscenium above the stage. The speakers were spaced several feet apart from each other. I have moved the EV speakers so that they form an arch. There is no space between the boxes. I am looking for advice as to if this is the best positioning for the EV speakers and how I should go about positioning the PAS speakers to avoid problems with comb filtering. The rear fill speakers are currently aimed directly at the rear wall of the auditorium. I am definitely planning to aim them better than they are now. I also need to equalize the room better than it is now.

I tried using RACE to program the Dx38. The brochure for the Dx38 claims that speaker specifications can be imported into the software and it will display speaker response to help program and position the speakers. I have not been able to find the RACE files for my speakers. Has anyone had success with this software? Are there any manuals available. I have had some success with the demo versions of Smaart, but is there any other software available to help me in this process?

thanks,
Will Abele

[Tom Young]

It appears you have had to deal with a bunch or real knuckleheads. Lucky for the school that you have a brain and the initiative to look closely at what was done.

FYI - there can be benefits from spreading devices apart in a  cluster but usually this means separating them by at least a few feet. This is called an "exploded cluster" and you should be able to google on that and find info. However, it would appear that they were spaced as you found them due to laziness or ease of rigging. Nothing wrong with packing them more tightly but they should be splayed to cover what is needed and not overlap.

Smaart is by far the most widely used and supported complex measurement system out there. Sure, there are cheaper programs but they do "time blind" real time analysis and that has severe limitations.

I woud contact EV directly and ask about RACE and the Dx38.

[Brad Weber]

Have you tried looking at the manufacturer's web site?  The Xi-1152A/64 page http://www.electrovoice.com/products/32.html has direct links to the Dx38 settings and the RACE files for that speaker.  EV's Tech Library http://www.electrovoice.com/tech-library.php?pt=3 also contains such useful information as rigging information for the Xi series, the Xi Arrayshow program, Xi series RACE files, EASE files and even some applications diagrams.

Since it didn't seem to be stated, what is your experience with rigging, installing and tuning sound systems?  Knowing a little more about you and your experience and abilities would be useful.  For example, if you have never used something like Smaart and never tuned a system, then it might make sense to hire someone for this work who would let you work with them and learn from that experience.  The same with moving and rehanging speakers, that is best left to someone with rigging experience.  You may be, it's just not real clear what your role and background are.

Quote:

I then looked at the way that the PAS speakers had been plugged in. They had ignored the PAS processor and run a full way signal to both the high and low frequency speakers. Luckily, I found and fixed this before the speakers were used.

You said the speakers were the PAS PI15-1, which is a full range speaker with an internal passive crossover, therefore no biamping or separate high and low inputs.  So either they are actually the PI15-2.2 or they should be wired full range.

Given that three Xi-1152-64 speakers in an array would be at least a 180 degree horizontal pattern, I'm a little surprised that you need side fills.  You haven't provided any information on the room itself such as dimensions, where the audience is, what height the speaker are at or what the room finishes and acoustical environment are.  So it is a little difficult to provide much of any response regarding coverage and aiming.  In fact, without this type of information I'm not even clear why there are any side or rear fills.

[Will Abele]

They are PI15-2.2 speakers. I have no rigging experience. My experience with Smaart has only been with the version 6 beta for Mac. I am definitely not qualified to do an installation of this size, but it is my only option to improve the sound. I am a student at the school. The theatre director does have rigging experience and will help me move speakers. The school has spent all of the money they are willing to spend on the sound system. So far my changes have significantly improved the system and I would like to learn to do more.

I have attached a drawing of the building. The speaker bay is approximately thirty feet above the stage. It is several feet back from the lip of the stage. The opening is approximately 10' wide by 5' tall. There are no aisles in the middle of the house. The seats take up all the space in the house with three foot wide aisles on the sides. The first row of seats is only about three feet from the stage. Two of the PAS speakers are intended to cover the first several rows of seats on the sides. The other two PAS speakers cover the rear corners of the house.

I would appreciate any more advice you can give me based on this information.

thanks,
Will Abele

[Michael 'Bink' Knowles]

Looks like three-speaker cluster is too wide of a pattern. Two of them will cover ~120 degrees which will prevent some of the late reflections you get when you paint the side walls with sound. I'd get rid of one of the Xi-1152s and re-aim the remaining two.

I'd also get rid of the rear or delay speakers.

Having some front fill action is a good idea for the extreme right and left sides. Be careful about keeping their volume low--just enough to add a skosh of presence.

How is the cluster powered? Are the three EV speakers using two amplifiers (four channels total) or one amplifier with two amp channels? How are they cabled up? Are there four conductors going up there or are there eight? If four channels/eight conductors are used then you have an unusual option where your three main speakers can be amplitude shaded. You would put the amp in bridge mode and hook the center speaker up to both red terminals on the amp so that it would get the lion's share of power. The left and right speakers would get a lot less signal; they would still paint the walls and side aisles but the negative effects would be dramatically reduced. There's more to the method than I just described... Peavey has an article by Marty McCann about it online. Most of the info is on page two. Marty's example is a speaker cluster that is passively crossed over but the method works just as well for bi-amped speakers if you double the speaker cabling and amplifier channels.

-Bink

[Brad Weber]

Is there a balcony?  Like Bink, with a 63' deep room and the speakers at 30', I question the need for any fill or delay speakers unless they are covering a balcony.

I would also question tight packing the Xi speakers.  The angle on the cabinet sides is only 15 degrees (the Xi-1152A-94F also have the same cabinet), so the splay between boxes is only 30 degrees when tight packed, resulting in lots of overlap with the 60 degree horizontal patterns.  You might want to contact EV and ask about the best splay angle to use in a tight array, but it looks like 2 boxes would be expected to provide 100-120 degree coverage.

Admittedly, this is just a complete guess, but it is possible that the intent of the three boxes was to have the outer two aimed down to provide wider coverage of the front part of the main floor and the center box aimed higher and covering the back part, sort of a short throw/long throw arrangement.  Just thinking that this might make sense with the two outside boxes then having a 60 degree splay and as an attempt to minimize the sound off the side walls.  This would make even more sense if you have a crossover aisle that splits the audience into two sections.

It sounds like you might not have a great relationship with the original installer, but it might be worthwhile to try to contact them and simply find out how they envisioned the speakers working.  Maybe if you presented it as you have here, that you are simply wanting to learn and understand more, then they might be willing to explain their ideas on how it was supposed to work.  That might at least give you a better feel for why you have what you have and what was intended.

Smaart does come with a manual, but it may not be available for download and only supplied with a purchase.  The Smaart classes can also be well worth it if there is ever one in your area.

[Will Abele]

There is no balcony. I have been questioning the need for the delay speakers as well. I think they are causing more harm than good. I spoke with the contractor when they were finishing the installation and I do not think that there was much thought put into how the speakers were to be used. The only thing they envisioned was the money they would get for the installation. As I said, most of what they did was wrong. They even managed to swap pins 2 and 3 on the mic cables in some of the snakes which caused very strange problems when I tried to use the snakes for some intelligent lights we rented.

The EV speakers use 4 amplifier channels. The side speakers are run off of the same channels, but the center speaker has its own channels. There are a total of twelve conductors of 12 gauge speaker cable going to the EV speakers. The cables for the two side speakers are spliced together before going into the amplifiers.

My only concern with eliminating a speaker is that the speakers' coverage will overlap in the center of the house. We would obviously prefer for the center seats to be some of the better seats in the house. If I do remove a speaker would you recommend putting a thirty degree spacer between the speakers and having the backs of the boxes touching each other? There are two metal pipes that are about 1.5" in diameter running vertically along the opening of the speaker bay approximately 1.5' off each side of center. Will this have an audible affect on the sound? Two speakers can fit between the pipes, but a spacer would cause the pipes to be in front of part of the speakers.

I unplugged the two side speakers today and listened to just the center speaker. It sounded much better than the system sounds with all three speakers on. I knew that this would be true to some extent, but I was surprised by the difference that it made. I thought I read somewhere that the angle on the cabinet sides was thirty degrees, which was the reason I put them together in a tight cluster. However, after looking at the speakers again today I realized that you are right, it is only a fifteen degree angle.

[Michael 'Bink' Knowles]

I wouldn't want the metal pipes in front of part of the speaker cluster--they'll mess with the pattern. If I were trying to make two speakers work in that space I'd consider adding more steel rigging hardware so that I could scoot the two speakers forward to sneak them out beyond the vertical pipes.

The center seats will sound good if you are able to find the right angle of splay. I doubt it will be exactly 60 degrees from center to center... you would have to try different angles and listen to the results. For me, the goal is to make the most seats sound good, not concentrate on the center seats. Ideally, ALL the seats will sound good.

Going to a 2-box cluster is a big change. You will have be certain of your rigging hardware.

Jumping back into the 3-speaker scenario... your 12 conductor run to the cluster allows you to experiment with Marty McCann's bridge mode idea I linked to earlier. You might find it a happy compromise between one speaker only and all three blasting equally. If you need more guidance with the bridge mode concept as regards bi-amped speakers PM me. Sounds like you could try it with very little effort.

I'm curious; when you fired up only the center EV was it loud enough? Did the intelligibility fall off at the seats near the side aisles?

-Bink

[Will Abele]

I have not had a chance to work on this in a while, but our spring musical is now about a month away and I would like to do something with the sound system before then.

I have made some additional measurements and drawn a lot of triangles since my previous post. Assuming my math is correct, the vertical coverage needed for the house from the speaker bay is approximately 75 degrees. The horizontal angle from the speaker bay to the front row of the house is 112 degrees and the horizontal angle to the back row is 80 degrees.

I would like to have the majority of the house covered by the EV speakers, so I have considered rotating the horns to give me an array with approximately 120 degree horizontal coverage by 60 degree vertical coverage. However, I do not know how to cover the remaining 15 degrees of the house.

The other option that was previously discussed is to have two EVs cover the front half of the house and one cover the back half of the house. This would work well for the front half of the house, but the coverage for the back half of the house would not be so good. The horizontal angle of coverage needed in the front row of the area that would be covered by the single speaker is 95 degrees.

I am having trouble determining a solution that would not result in either too much overlapping of the speakers or too much spill onto the walls.

[Will Abele]

Michael 'Bink' Knowles wrote on Tue, 30 January 2007 22:36

I wouldn't want the metal pipes in front of part of the speaker cluster--they'll mess with the pattern. If I were trying to make two speakers work in that space I'd consider adding more steel rigging hardware so that I could scoot the two speakers forward to sneak them out beyond the vertical pipes.

It is not possible to move the speakers out past the pipes because they are flush with the mesh that encloses the speaker bay.

Michael 'Bink' Knowles wrote on Tue, 30 January 2007 22:36

Jumping back into the 3-speaker scenario... your 12 conductor run to the cluster allows you to experiment with Marty McCann's bridge mode idea I linked to earlier. You might find it a happy compromise between one speaker only and all three blasting equally. If you need more guidance with the bridge mode concept as regards bi-amped speakers PM me. Sounds like you could try it with very little effort.

I may try this if I end up using the speakers in this arrangement.

Michael 'Bink' Knowles wrote on Tue, 30 January 2007 22:36

I'm curious; when you fired up only the center EV was it loud enough? Did the intelligibility fall off at the seats near the side aisles?

It was loud enough, but the coverage did fall off at the front and sides of the house.

[Rob Timmerman]

Remember that the nominal coverage patterns are for the 6dB down points on the horns, and there is usually still some output beyond those points.  This can work to your advantage in some cases.

I'd be inclined to go with the 3-wide, rotated horn approach, aiming for the back of the room, and adding some lip-fill if possible and if needed.  You probably won't be missing more than the first couple of rows anyway.  Lip-fill can also help anchor the image to the stage when the cluster is almost overhead.

Another approach might be to contact EV, and see if you can substitute the horn from a Xi-1152A/94 into one of the boxes and use that box to cover the back of the room.  It looks like a drop-in retrofit, and shouldn't be that expensive.  Just some food for thought.

[Brad Weber]

Rob is absolutely correct about the nominal coverage being the -6dB points.  Along with that keep in mind that this is also only a nominal number and the actual pattern changes with frequency.  Based on the EV datasheets, the Xi-1152A/64 appears to actually be 60x40 only from about 4kHz and higher.  At 2kHz it looks more like 85x55, at 1kHz 90x60 and at 500Hz maybe 105x120 (yep, apparently the pattern flips from around 400-700Hz).  Looking at the directivity and response data, there also seems to be some rather significant issues in the vertical pattern right around the crossover point with the -6dB pattern jumping to 100x100 and lobing that causes some major drops in level from about 15 degrees to 30 degrees off axis.

Anyways, the point is that it isn't really a 60x40 speaker at all frequencies and one thing you may have to watch is the resulting levels on the stage for frequencies where the vertical coverage gets wider.  The front rows may get some direct sound from the live action and you may think about letting the level from the speakers drop down in the front rows in order to keep the levels on the stage down and then counting on the live reinforcement from the stage to offset that for the first few rows.

Another aspect to make sure you considered in your calculations is distance.  The front row is likely quite a bit closer to the speakers, so you may actually want the coverage down quite a bit at that point to offset the higher level from being closer.  It does get rather complex trying to envision the 3-D pattern and distance effects, which is why even quick direct coverage modeling can be so helpful as it does this math for you and then lets you see the predicted results a various frequencies quite quickly.