Subwoofer crossover with full range mains

100Hz was my preferred crossover frequency. It took enough out of the mains that they seemed happy with getting pushed hard, but didn't make the subs go too high.
 
And if you really want to get into things on that level, get SMAART and a reference mic, set the system up, and test it rather than guess.

I have a UMIK-1 and a notebook computer that I plan to install REW on. I have already tuned dual subwoofers in my home theater, so I know something about it.

Never used SMAART. Doubtful if there is any funding for it in this group, but I have used REW. In the plan, just not there yet.
 
I run a system exactly like this.

The front of house graphics on the full range: I slope the eq from about 100hz down to 0 at 20hz, you need to experiment with this both starting point and the shape of the curve some of the guys that use the system do a full cut from 100 hz down but I prefer a slope. If you do the old one two testing through a mic with the subs off it works for me, I hate reference mics, white noise and all that, you can tell if it's not right using your ears.

Then I have the subs set to a crossover point around 80 Hz.

I use a post fader Aux send to the subs (this by passes all the FOH EQ) I then balance the instruments going to the subs with the aux send. Usually Kick, Bass, Floor tom/low toms, sometimes Keys.

There is nothing below the 80hz (or even about 120hz) that guitars need to have, and especially vocals.

The full range set up like this sounds much cleaner as do the subs not having to filter out stuff they don't want via the crossover. Tell the drummer that doing it this way will actually get more power due to the system trying to reproduce frequencies in places they should not be, I.E 12" or 15" speakers trying to reproduce subs and subs trying to reproduce highs.

Alan.

The drummer originally had something similar, except he did the lowpass on the sub in graphic also. We reverted to this for the last gig for lack of time to develop a better solution, then I proposed the 60Hz 6dB/octave crossover as a compromise solution and he seems OK with this.
 
The thing that smart systems do that you can't do by ear is fix the timing between the tops and subs. Once you do that, the system comes into focus and you don't have so many issues around the crossover frequency.

I was out of the live sound gig for a long time, when did everyone stop using proper crossovers? I still use a driverack for smaller systems. The big line arrays have a system to address each cabinet in the array, but that's a giant concert system.

If the mixer were less capable we would not be having this discussion because the subs would be crossed at their built-in 120Hz. Master Fader and the DL series opened a can of worms when they put more capability into the system than most people need. I think that is the core of the issue here, but I am willing to run with an alternate solution that increases overall headroom provided it sounds OK. (not that we need more headroom, but that is another story for another day when the band is playing larger venues regularly)
 
I have a UMIK-1 and a notebook computer that I plan to install REW on. I have already tuned dual subwoofers in my home theater, so I know something about it.

Never used SMAART. Doubtful if there is any funding for it in this group, but I have used REW. In the plan, just not there yet.

Then you've got a head start. I also use REW and it's more than adequate for this. I tend to rely about equally on measurement and listening.
 
Remember I am only talking about filtering the subs out of the full range, the full range part of the system is capable of receiving the sub frequencies, I just remove them to make the system cleaner and to let the full range run easier. The sub side already has a crossover. I am therefore using the graphic as a filter not a crossover.

I also forgot to mention that on my own system (as opposed to the in house system I use) my power amps have inbuilt adjustable crossovers. The in house I use most of the time does not have this which is why I use the graphic, some operators don't use the graphic filtering on the full range but that up to them, each to their own.

Alan.

By the way I hope we have not bamboozled the OP.

I am still trying to decide how to approach this.

If I run with the built-in crossover, everything is simple and phase-coherent (provided I set the polarity on the sub properly) but the crossover is at 120Hz and the drummer will think he is leaving some SPL on the table. Me, not so sure it matters, and I would be happier knowing that the crossover was designed by an engineer with some experience in this.

If I run with a 60Hz crossover at 6dB/octave, the drummer is OK with it, and the headroom may be maximized, but I may still be creating phase issues that will muck up the sound. The phase is 90 degrees out at crossover and that may create vertical beaming. Not sure if that matters for bass. Also, below 60Hz the mains may have large phase shift that aggravates the existing phase issue in the crossover.

If I run with 80Hz crossover at 12dB/octave, everything will probably be (actually) optimized, per my prior experience with home theater, though not sure why the built-in crossover would be 120Hz if it is not already optimal.

Then I am unsure about the Linkwitz-Riley aspect, whether the HPF/LPF in the mixer is cascaded from lower order stages or not, and if that may create bump at crossover?
 
If vocal/guitar/etc. mics are adding too much rumble to the sound even when you high pass them

I have not noticed any issues, but then again we have been using aux fed sub...

I would suggest setting the system up and testing it when you don't have a show to put on, especially if you can do it outdoors away from buildings.

That was my first suggestion to the drummer. This is just a hobby for us and he seems reluctant/too busy to give me time on the system, plus with gigs only once per month or so and half of them using the house PA my time on the system is limited.

I think I might suggest to him that we try the 60Hz crossover and hopefully I can also get a minute to run some sweeps and check the phase/mag for aberrations. We might get lucky, or it might be off. Who knows?
 
The thing that smart systems do that you can't do by ear is fix the timing between the tops and subs.

The DL1608 has adjustable delay on the outputs, so I should be able to accomplish this, but it seems like too much work during setup. I need to do this offline unless the venue demands alternate subwoofer placement. We have one small bar that seems to want the subwoofers placed for modal response and I am working on a solution for that, including stands for the mains so the subs can move, but it is speculative at this point. That solution would require a timing measurement to do it right.
 
Speakers themselves have phase response. Just looking at the crossover phase response isn't enough, you have to take the total phase response of the whole system into account.

The only real way to do that is to set the system up and measure. Run a tone at the crossover frequency, level match it in the tops and subs, combine them to see the change in output. If they sum with +3dB output then they're in phase (or some multiple of 360° out of phase). If there's no change in output then they're 90° out of phase (or 90° plus some multiple of 360°). If they cancel almost entirely then they're 180° out of phase (or 180° plus some multiple of 360°).
 
One nice thing about low frequencies, they're really forgiving about top vs. sub placement. Moving the subs a few feet either way isn't going to matter much given the wavelengths. What matters more is their placement relative to boundaries (walls and ceilings) and the other sub.

Since you're already thinking about this stuff, do some research on "power alley" and "cardioid subs".
 
The DL1608 has adjustable delay on the outputs, so I should be able to accomplish this, but it seems like too much work during setup. I need to do this offline unless the venue demands alternate subwoofer placement. We have one small bar that seems to want the subwoofers placed for modal response and I am working on a solution for that, including stands for the mains so the subs can move, but it is speculative at this point. That solution would require a timing measurement to do it right.

Yes, the smaart system (and others like it) integrate the timing measurement.

You have a good handle on the science of speaker placement and the effects of crossovers, however you are missing the part about the speakers themselves having odd phase relationships to each other (even in an anechoic chamber). Moving the subs away from underneath the mains will make that a big issue.

Your mixer will be able to adjust the delays of the outputs, but you will need to do the measurement to have a hope in hell of getting it right.
 
One nice thing about low frequencies, they're really forgiving about top vs. sub placement. Moving the subs a few feet either way isn't going to matter much given the wavelengths. What matters more is their placement relative to boundaries (walls and ceilings) and the other sub.

Since you're already thinking about this stuff, do some research on "power alley" and "cardioid subs".

Thanks for the tip.

The power alley is probably not an issue for most of what they do given that most of the stages are only about 20' wide max, but there is the possibility of outdoor gigs. I will need to keep this in mind for later.

Presumably cardioid sub only applies in larger venue. We are only using the 2 subs and I doubt I could set it up reliably in the time we have to prepare. Anyway the feedback at low frequencies has not been an issue so far (cross fingers) though there is at least one venue where we might relocate the subs for modal response and that could cause problems, though not solvable with a phased array.
 
Yes, the smaart system (and others like it) integrate the timing measurement.

You have a good handle on the science of speaker placement and the effects of crossovers, however you are missing the part about the speakers themselves having odd phase relationships to each other (even in an anechoic chamber). Moving the subs away from underneath the mains will make that a big issue.

Your mixer will be able to adjust the delays of the outputs, but you will need to do the measurement to have a hope in hell of getting it right.

I am aware that the phase shift is inherent to the speakers also. I was just hoping it was not such a big issue as to negate the benefit of driving through both subs and mains simultaneously. Obviously that can only be determined though experimentation and careful measurement.

Well here is hoping that I get some dedicated time on the system to calibrate the sub integration. Maybe some vague manual polar response data will convince both of us to choose one solution over the others for venues where the subs are under the mains and the positioning is known in advance.

For venues where the subs are separated from the mains due to modal issues, well... modal response is complicated. Who knows what will work? I had to tune my home theater by trial and error because in irregular space modal response is difficult to predict.

His yard is urban with homes nearby so outdoors there is not such a good option either. I need to think this through and find a location to calibrate in.

After this conversation I am less optimistic about sharing the bandwidth. I looked over the response plots again and the mains have sharp rolloff whereas the subs have less steep rolloff. They probably fight lots after the mains start rolling off since the slope of the rolloff is a fairly good indicator of the phase shift. The mismatched rolloff is evidence that the mains not only exhibit phase shift beginning at a higher frequency but also that there is more shift from them per octave as well in the LFE.

Some phase measurements with REW would go a long way toward figuring this out. I guess it is time for me to figure out how to set up a reference sweep on an alternate channel so I get an absolute timing reference. Someone told me I cannot do this with a USB microphone, but my impression is that the timing reference is generated by a separate speaker so should not have anything to do with the USB microphone.

Well I have another gig to go to this evening so ciao. The competition wants access to my skill set now. This is getting out of hand. Before I know it, every wayward amateur band is going to be asking me to do sound for them. Old dog, meet new tricks.
 
The reason not to use a USB mic is that you need to loop the output back to an input, going through the same A/D conversion, for the time measurement to be meaningful.
 
That is the part I am not getting.

Why is electrically looping back through the sound card meaningful? Does that occur once, or once with every new timing measurement?
 
The software has to measure the timing of the signal coming from the speakers against the signal it sent. In order to take latency or any other effects the interface has on the signal, it needs to see what it sent and when it comes back through the interface.
 
That is the part I am not getting.

Why is electrically looping back through the sound card meaningful? Does that occur once, or once with every new timing measurement?

What farview said. The electronic latency must be subtracted from the measurement to get an accurate result. It happens with every measurement. A USB mic would have to have an integrated output to drive the playback system to be able to do this. Using a different output device, like the stock sound card, means that the software can't know how much of the delay is acoustic and how much is electronic.

It might not be as important when aligning drivers, but when measuring acoustic spaces it's essential so the software can better distinguish between the direct and reflected sound. It makes it possible to measure absolute acoustic delay between source and mic.
 
Almost forgot, looping back also accounts for any phase and frequency response deviations of the analog circuitry and converters.
 
The drummer upgraded the subs from the SRM1801 to the KW181.:D

I had the model number wrong and misled you all, sorry. The SRM1801 has 500W output and a strong peak at 90Hz, as opposed to the SWA1801z I thought it was, that has 800W and a more natural straight-line bass-boosted response. The SRM is definitely a lower performing sub and the drummer wanted to run it with the mains full range to help cover the LFE, and based upon the power rating and frequency response of those SRM subs it seems that was probably the right thing to do.

At 1000W the new KW181 should keep up with the 1200W SA1530z mains better, so the drummer is OK with crossing over the mains now instead of running them full range. Details in the next reply.:)
 
Back
Top