Print

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

[Scott Carneval]

You stated 'starting' off with the limiters. we're going to continue into the xover and delaying parts of dsp correct?

It's going to be near impossible to recommend crossover and delay settings without measuring your system. The crossover will be influenced by your K12's. The TH118 can go up to 100-125hz if it has to. But if your speakers can reproduce this, then it usually sounds better to let them. Your personal preference will also determine xover. I personally don't like to run the subs much about 70-80hz, but that's assuming I have a top that can fill in that region.

If you want to phase-align the system, you can't set the delay until you've determined the crossover. If you just want to time align, then it really just depends on the physical location of your tops and your subs. I believe you said in an earlier post that you center cluster the subs. This usually provides more output from the subs, but it's impossible to align the system properly if the tops and subs aren't co-located.

In your case it might just be easier to invert the polarity of the subs and send a sine wave at the crossover region. Then delay the tops until you get maximum cancellation. Once you flip the polarity back, you should get maximum summation, at that frequency, as that point in the room. This won't get it perfect, but unless you stack the tops on the subs, or fly the subs with the tops, there's not a whole lot you can do to get it properly aligned.

[Chris Hindle]

I could use the trial of SMAART to do it properly I think?

Not to dampen the parade, but get the trial to START learning how it all works.
You have to learn a shitload before you can turn out meaningful data.
This is NOT an afternoon project......
Best way ?  Bring in someone VERY familiar with one of the analysis programs, and let him loose while you watch.
Chris.

[Nathan Riddle]

It's going to be near impossible to recommend crossover and delay settings without measuring your system. The crossover will be influenced by your K12's. The TH118 can go up to 100-125hz if it has to. But if your speakers can reproduce this, then it usually sounds better to let them. Your personal preference will also determine xover. I personally don't like to run the subs much about 70-80hz, but that's assuming I have a top that can fill in that region.

If you want to phase-align the system, you can't set the delay until you've determined the crossover. If you just want to time align, then it really just depends on the physical location of your tops and your subs. I believe you said in an earlier post that you center cluster the subs. This usually provides more output from the subs, but it's impossible to align the system properly if the tops and subs aren't co-located.

In your case it might just be easier to invert the polarity of the subs and send a sine wave at the crossover region. Then delay the tops until you get maximum cancellation. Once you flip the polarity back, you should get maximum summation, at that frequency, as that point in the room. This won't get it perfect, but unless you stack the tops on the subs, or fly the subs with the tops, there's not a whole lot you can do to get it properly aligned.

I know that time aligning is impossible with subs/speakers spaced out, but I was curious if phase aligning was too? probably not though. I guess what i'm after maybe isn't perfect system setting up because I can't fly the TH118s with the K12s on top of our truss towers, but maybe i'm looking for the best compromise?

I'm curious which is better center cluster or ground stacked with K's ontop? I see issues with both.
If the subs are center clustered then they align properly and don't cancel eachother out, but then they can't be aligned properly into a semi-point source with the mains.
If the subs are co-located with the mains then the subs cause cancellation (i've looked at the modeling and it isn't pretty imo).

Which is more important? For me it is the low end being tight and punchy; Which method helps with that? co-locating for phase alignment? or center clustering?

Not to dampen the parade, but get the trial to START learning how it all works.
You have to learn a shitload before you can turn out meaningful data.
This is NOT an afternoon project......
Best way ?  Bring in someone VERY familiar with one of the analysis programs, and let him loose while you watch.
Chris.

I want to say "you underestimate my power" (insert pic of anakin & obiwan on the lava planet) in my pride haha. But I shall listen to teh wiser and not attempt in one afternoon. perhaps many afternoons But I do want to eventually learn so why not start tinkering right?

Here is what I would "start with" on the DBX.

Turn the threshold on the DBX all the way up.

Turn up the level until you read around 95V.

Decrease the threshold on the DBX until the volt meter reads 80V.

Turn off the sine wave and put in some music and see if the amp clips.  Speakers still disconnected.

Drive it hard.  If the clip lights on the amps are just barely lighting, you may be fine.

If they are flashing pretty good, then back down on the threshold until they stop flashing or just barely flash.

Got the gear and have been playing with it some.

In an effort to understand how you got those numbers and this method. I'm assuming you're just applying ohms law with 4ohms and 1700w for continuous and then 2300w for peaks?

60.1hz output from signal gen on our QU-24 freq reads 60.12 on meter
~95v output from amp
dbx limiter @ 15db with over easy @ 5 (half) = ~80v

when run hard with music the clip light on amp just lighting full red max voltage = 114v
when burried does the same with just slightly longer max voltage = 156.8v

So I setup the K12 on top of the TH118 and set the xover at 100hz 24db BW for both tops and subs. Then I sent a 100hz sine into the system and delayed the top at first and found it summed better at 0 and fully at ~8ms. I figured those were extremes and so I switched to delaying the sub; delayed 1.60ms which summed very well.

Turned up the music and I thought it sounded very nice. Good impact in the xover region that i've been missing I think. Also, with the over-easy at 5 and limiter turned on i felt like the sub was a little punchier and well enough protected when i'm running things. I might turn it down 6db or so when I let the system out of my sight or 3db when i'm there and not running it.

Overall I feel like the system is tuned better. Though it could be confirmation bias kicking in. Now I want to try with the mains non co-located (simulate subs in middle) Try some RTA traces with SMAART. Move on to the xilica since a friend loaned me one for the foreseeable future. Eventually get sc48 and new amps.

I'll keep updating my progress encase anyone else is where i'm at and wants some "starting settings" or just to learn.
Thanks for the help everyone, Ivan, Scott.

[Nathan Riddle]

Hey Ivan,

A friend is letting me borrow his Xilica nearly indefinitely, so the SC48 will have to come next year since it's not a wise investment at the moment for my companies' immediate future since the Xilica will help us limp along for now.

What are you recommended limiter settings for that one? I liked how the dbx was limiting once I set it up the way you told me to, but now that I have attack time release time and threshold things might have changed on your recommendation. I'm going to sort of assume, because I have tried to understand so that I know why certain values are chosen, that I should still use 95 and 80v and set the attack time to max and release time to max.

[Scott Carneval]

I setup the K12 on top of the TH118 and set the xover at 100hz 24db BW for both tops and subs. Then I sent a 100hz sine into the system and delayed the top at first and found it summed better at 0 and fully at ~8ms. I figured those were extremes and so I switched to delaying the sub; delayed 1.60ms which summed very well.

Unless you run the tops at the same level (acoustical output) as the subs, you'll need to 'underlap' the crossover some. There's sometimes a big difference between the acoustical crossover (the frequency at which the subs and the tops are equal) and the electrical crossover (the numbers in the DSP). For example, with a 24dB/octave slope, and assuming for the sake of simple math that you're running your subs 12dB hotter than the tops, your acoustical crossover will be about half an octave higher than the electrical crossover. This assumes that both the sub and the top are perfectly flat in magnitude, which they definitely are not. This is why it's important to measure with something like SMAART. But the purpose of saying all of that is to point out that you would need to send a sine wave at the acoustical crossover point, not the electrical crossover.

Also, the subs are almost certainly going to be lagging the tops because they are horn loaded. I'm sure Ivan can provide the exact dimension, but I believe the TH118 has something like a 13' path length of the horn. So in essence your sub driver is 13' behind the tops. BUT, your tops have a built-in DSP which is probably causing a few milliseconds of delay. The 8ms of delay you found the first time may actually be correct.

[Nathan Riddle]

Unless you run the tops at the same level (acoustical output) as the subs, you'll need to 'underlap' the crossover some. There's sometimes a big difference between the acoustical crossover (the frequency at which the subs and the tops are equal) and the electrical crossover (the numbers in the DSP). For example, with a 24dB/octave slope, and assuming for the sake of simple math that you're running your subs 12dB hotter than the tops, your acoustical crossover will be about half an octave higher than the electrical crossover. This assumes that both the sub and the top are perfectly flat in magnitude, which they definitely are not. This is why it's important to measure with something like SMAART. But the purpose of saying all of that is to point out that you would need to send a sine wave at the acoustical crossover point, not the electrical crossover.

Also, the subs are almost certainly going to be lagging the tops because they are horn loaded. I'm sure Ivan can provide the exact dimension, but I believe the TH118 has something like a 13' path length of the horn. So in essence your sub driver is 13' behind the tops. BUT, your tops have a built-in DSP which is probably causing a few milliseconds of delay. The 8ms of delay you found the first time may actually be correct.

Scott,

Thanks for the response. That's the direction I had originally desired to go. Doing things the correct way. But I felt what was expressed throughout the prior conversation was that correctly MEASURING things was a huge learning curve and I would not be able to accomplish my desired goals anytime soon.

At this point I feel like there are differing suggestions and don't have an accurate picture of all of the data in order to understand and make a proper decision.

I have read that although usually tuning the system so it sounds right with music is a good idea that it is actually easier to mix on a system that is flat and to boost the low end on the channels for ipod/music. I haven't tested this idea yet, but I do desire to do so. That said, I did at one point have the subs xovered at 90 so that the acoustic crossover point was closer to the desired electronic crossover of 100hz and that is what got me on the train of thought of obtaining SMAART and measuring the response to get the system response to be correct. But alas I was dissuaded because of prior mentioned reason.

I too thought about that when I originally obtained the TH118s, upon reading the Delay section in the FQT section on Tapped Horns on Danley's  website It was determined that delaying to the path length of the horn wasn't a very good idea. If I remember correctly the path length was dependent on frequency and the nominal path length was somewhere in the middle. But effectively the system behaved as if it wasn't horn loaded at all. Never mind, re-read the FAQ question.
http://www.danleysoundlabs.com/support/frequently-asked-questions/#question_10

All of that said, I have new mains (SRX835p's) which will need to be tested and tuned.

I have all the hardware for a SMAART trace and downloaded SMAART to try playing with it. If someone could shoot me a quick rundown on how to get started - by all means I'm willing to try to utilize that for a little simple tuning in the xover region.

One of my main holes in my understanding currently, is what happens when I tune the system with the main on the sub (effectively a point source) and then move the mains to flying. Will I have to re-do all of my DSP tuning or is it good enough?

[Mark Wilkinson]

In addition to the Smaart demo, you may want to try the freeware REW.  It has very good help files....I read them all time to help learn Smaart.

[Scott Carneval]

I too thought about that when I originally obtained the TH118s, upon reading the Delay section in the FQT section on Tapped Horns on Danley's  website It was determined that delaying to the path length of the horn wasn't a very good idea. If I remember correctly the path length was dependent on frequency and the nominal path length was somewhere in the middle. But effectively the system behaved as if it wasn't horn loaded at all. Never mind, re-read the FAQ question.
http://www.danleysoundlabs.com/support/frequently-asked-questions/#question_10

I'm not suggesting to just calculate the delay based on 13'. I'm saying that the inherent delay built into the subs will almost certainly dictate that you will need to add your delay to the tops. You mentioned somewhere above that you felt you had the best results by adding delay to the subs.

I have all the hardware for a SMAART trace and downloaded SMAART to try playing with it. If someone could shoot me a quick rundown on how to get started - by all means I'm willing to try to utilize that for a little simple tuning in the xover region.

Unfortunately, it's really not that simple. The SMAART training class is 3 days long. Learning to navigate the software is the easy part. Learning how to take accurate measurements, and then learning how to interpret those measurements, is what the other 2.5 days are for. There are some decent Youtube videos you can watch, and there is finally a SMAART operators manual available. I would suggest you start there. Please understand that it's not for lack of desire that we can't help you. It would be akin to teaching someone how to drive and all of the rules of the road, in a matter of an hour or so. It's just not possible.

One of my main holes in my understanding currently, is what happens when I tune the system with the main on the sub (effectively a point source) and then move the mains to flying. Will I have to re-do all of my DSP tuning or is it good enough?

If you move either of the speakers by more than a couple inches (or less, depending on the frequency of interest) you will change the results. If you fly your mains than all of your tuning goes out the window. Any time you change the differential of the arrival times between the sub(s) and the main(s) your delay needs to change.

[Ivan Beaver]

Hey Ivan,

A friend is letting me borrow his Xilica nearly indefinitely, so the SC48 will have to come next year since it's not a wise investment at the moment for my companies' immediate future since the Xilica will help us limp along for now.

What are you recommended limiter settings for that one? I liked how the dbx was limiting once I set it up the way you told me to, but now that I have attack time release time and threshold things might have changed on your recommendation.

As usual there are a number of "it depends".

And in this case it depends on the gain and the output capacity of the amplifier used, does it have its own limiting and type of music etc.

Different types of answers to the above question would result in different answers.

One size does NOT fit all.

[Nathan Riddle]

As usual there are a number of "it depends".

And in this case it depends on the gain and the output capacity of the amplifier used, does it have its own limiting and type of music etc.

Different types of answers to the above question would result in different answers.

One size does NOT fit all.

Lets say 2x presets; one for EDM one for more rock. Same amps (34dB gain), clip limiting only [Though i'm not sure what it's doing as I can lite it up red all day long and the voltage just keeps going up which makes me believe it is becoming distorted...?]

[ProSoundWeb Community]