Print

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

[Ivan Beaver]

This is a cases of where actual measurement is required, NOT just listening by ear.

I had a case once in which each of the mains and delays were well behaved, but they had different phase responses.

The amplitude and phase response of each loudspeaker were good, but they were different. 

I could either get them to align well above 400Hz or below 400Hz.  But not over the entire range. 

But simply "throwing a highpass filter in" affected the phase response again.

Some quick measurements (assuming proper amplitude and phase measurements) will go a long way towards solving the problem.

Different phase responses will result in dips and peaks in the combined response.  No amount of delay can fix that.  If the phase responses are the same, then delay can help fix the issues.  Notice I said "help fix", not fix. There are a number of issues that also arise with delay speakers (such as what happens BEHIND the loudspeaker), that have to be addressed in different ways, but cannot be eliminated.

The other suggestions presented are good, making sure each speaker is wired up properly, only doing 1 set of speakers at a time etc are all valid if you can't do measurements

[Kevin Maxwell]

I don’t need to repeat what other have already said but I will add that in my opinion if a delayed speaker system is done right people will think that the delayed speaker is off. Because of this I have had clients that don’t want delayed speakers installed because they are tired of having to deal with people telling them that the speakers are off. Even though the room and situation calls for delayed speakers to work the best for what they are trying to do. I have said to people that I think it would be great if you can hide the delayed speakers so the people won’t be listening with their eyes.

Are you possibly trying to run these delayed speakers too loud?

[Luke Geis]

The Meyer probably has a couple of ms of latency and the DXR probably also has a couple of ms of latency. Between the two you may have as much as 5ms or more depending on the latency of the mixer and the Foh system to the subs? So doing a straight conversion of footage between the main hang and the delay speakers can often be off by several feet.

As mentioned before, having a PA with matching units is much easier because they will sound the same. If one speaker has a much different sonic character, trying to hear where things lock-in is hard to do. Also, you are working with two speakers that have totally different crossover topologies. One operates with fairly typical filters ( the Meyer ) and the other utilizes FIR filters ( the DXR ), which can create some weird phase shifts that are frequency-dependent when compared to another conventional product. I.E. you can get it right in one frequency range, but not another. The use of all-pass filters creates a similar effect. In either case, unless you know the processing and have access to processing that can emulate it, it is a fruitless endeavor.

Typically, assuming you have a well-balanced system with products that have similar attributes, you can hear the two parts LOCK-IN when you adjust the delay. Often you can guesstimate the footage and in place delay, and then as you slowly add or subtract delay you can come to a point where you hear it lock in. All of a sudden 1ms is either right or wrong. The delay will just seem to go away and you just hear the FOH sound become more apparent. Your eyes and ears will simply be drawn to the stage as opposed to realizing the delayed speaker. But if the two systems are far enough separated by processing tricks ( FIR, delay, latency, or frequency response ) the task of time aligning becomes very difficult to do by ear; quite literally requiring a DSP and FFT solution to correct.

[Ivan Beaver]

I don’t need to repeat what other have already said but I will add that in my opinion if a delayed speaker system is done right people will think that the delayed speaker is off. Because of this I have had clients that don’t want delayed speakers installed because they are tired of having to deal with people telling them that the speakers are off. Even though the room and situation calls for delayed speakers to work the best for what they are trying to do. I have said to people that I think it would be great if you can hide the delayed speakers so the people won’t be listening with their eyes.

Are you possibly trying to run these delayed speakers too loud?

Exactly.  You should not realize they are on, until you turn them off, and you notice the difference.

I have had jobs in which I HAD to shorten the delay, so the customer could "hear" the speakers, and THEN we would get paid.  They would not pay us until they "heard" them working, even though I had demonstrated that turning them off made a big difference.

Oh well.  The worst thing is that in years afterwards if a true audio guy comes in and finds out that I tuned the system, and then says I didn't know what I was doing because the delay wasn't set right----------

HEY, at least I got paid, and that was the whole purpose of setting it up wrong.

[John Roberts {JR}]

Exactly.  You should not realize they are on, until you turn them off, and you notice the difference.

I have had jobs in which I HAD to shorten the delay, so the customer could "hear" the speakers, and THEN we would get paid.  They would not pay us until they "heard" them working, even though I had demonstrated that turning them off made a big difference.

Oh well.  The worst thing is that in years afterwards if a true audio guy comes in and finds out that I tuned the system, and then says I didn't know what I was doing because the delay wasn't set right----------

HEY, at least I got paid, and that was the whole purpose of setting it up wrong.

The customer (who pays the bills) is always right, even when wrong...

JR

[Miguel Dahl]

The DSP is a Galileo 408, running 3.12 FW,

That's my apporach to tuning delays. They should not be heard, but one can hear the difference when turned off.

I found a soundcard and threw up a condenser, not a measurment reference mic, but a high quality (can't remember which one) condenser, at least, just to look at the relative differences, not important if the mic itself is "off" comparing to what's really happening... Only measured up to 2K looked at the problem area which is lower than that. The SPL-response in REW told me what I wrote earlier what I was hearing, about 200Hz-ish wide bands which either have a dip or peak when I flip the polarity. Can't get the mains and the DXR's to align over that spectrum, it's either this response, or that..
I don't know if using that mic was worth anything to be honest, but even if the mic is "the wrong one", the relative response would be the same. Maybe it was just a coincidence that what I saw in REW matched with what I heard earlier.

Delays are high passed at 120, subs XO mains at 70. The delay settings were in the ballpark after confirming later with the "delay-finder". When doing "fine-tuning" times by ear I usually use a metronome. When I hear one click and sound is the same when flipping polarity I call it good enough.

I'll be able to look at it properly the next week.

We also have some UP-Juniors of some kind, maybe that would be better to hang instead of the DXR's as delays to get more of the same sonic signature.

[Riley Casey]

For voice centric shows in large spaces ( convention centers and arenas ) I generally look to delay speakers for improvements in articulation only and high pass them accordingly sometimes as high as 400-500. There are certainly applications where delays are more akin to separate zones and wider bandwidth makes sense but if your listeners are in direct line with the main speakers try a high cut off and see if that cleans things up.

...

Delays are high passed at 120, subs XO mains at 70. The delay settings were in the ballpark after confirming later with the "delay-finder". When doing "fine-tuning" times by ear I usually use a metronome. When I hear one click and sound is the same when flipping polarity I call it good enough.

...

[Miguel Dahl]

For voice centric shows in large spaces ( convention centers and arenas ) I generally look to delay speakers for improvements in articulation only and high pass them accordingly sometimes as high as 400-500. There are certainly applications where delays are more akin to separate zones and wider bandwidth makes sense but if your listeners are in direct line with the main speakers try a high cut off and see if that cleans things up.

That's a good point. Thanks.

[Ivan Beaver]

For voice centric shows in large spaces ( convention centers and arenas ) I generally look to delay speakers for improvements in articulation only and high pass them accordingly sometimes as high as 400-500. There are certainly applications where delays are more akin to separate zones and wider bandwidth makes sense but if your listeners are in direct line with the main speakers try a high cut off and see if that cleans things up.

Exactly.

The delays should bereproducing "what is missing" from the main coverage.

Sometimes I have highpassed around 800Hz or so.  But yeah, 300-500Hz is VERY common for delay highpass

The one thing I do to keep the delays from "feeling thin", is to use a more gradual highpass than normal.  Usually a 1st or 2nd order filter.

I would start by measuring in the main coverage area, then move to the delay coverage area and see what is missing.  You will see the mid - highs start to roll off.

Start by putting a HP where they start to trail off, and go from there.  The quality of the mic doesn't matter, as long as the same mic is used for both locations, and all you are trying to do is match the other response.  NOT to provide a flat response.

[Miguel Dahl]

Exactly.

The delays should bereproducing "what is missing" from the main coverage.

Sometimes I have highpassed around 800Hz or so.  But yeah, 300-500Hz is VERY common for delay highpass

The one thing I do to keep the delays from "feeling thin", is to use a more gradual highpass than normal.  Usually a 1st or 2nd order filter.

I would start by measuring in the main coverage area, then move to the delay coverage area and see what is missing.  You will see the mid - highs start to roll off.

Start by putting a HP where they start to trail off, and go from there.  The quality of the mic doesn't matter, as long as the same mic is used for both locations, and all you are trying to do is match the other response.  NOT to provide a flat response.

It's a not very long throw from the mains, but there can be some lamps in the way which casts a shadow (heh), and the mains are some degrees downward and inward, to cover the closer seats. But that would mainly only affect the highs.. So high passing higher, and maybe just do it so high that the troublesome area is out of the equation is a great tip. I'll try that. But still I feel like I'd want to try the UP-Juniors for the tonality.

[ProSoundWeb Community]