Subwoofer - Wotcha got? How d'you set it up?

iqi616

iqi616

Yet another Mike
Subject says it all really. What subwoofer/monitor combination do you use and how do you set it up so that you can be sure that the cross-over with the nearfields isn't overlapping or leaving a gap?

Obviously I'm more interested in maintaining the accuracy and translatability of the mixes created with the monitor system than rattling all the crockery in the kitchen.
 
I have is a DIY sonotube sub that my brother built for me. To attain the best phase coherance, I also had my little bro build a 24db/oct crossover (at 70hz HP and LP). I currently use this set up for my home theater and 2 channel music listening. I will be integrating the sub into my new nearfield monitoring system in the same fashion (should be picking up the JBL LSR25P's in the next couple weeks). Should be a killer combo and won't suffer from any gaps, bumps or phase problems.

If you want to see more info on my set up, go here.

Cheers.
 
I use an audio research home theatre sub..

It's usually off, but the tiny amount of time I use it, it is just turned up enough to fill in the lows... I never have it thumping the floors
 
Tannoy PSB110... setup just enough to extend the low-end response of the monitors it's mated with (my secondary monitor system).

As Voxvender said - it doesn't (and shouldn't!) be thumping the floors.

Bruce
 
So, do you set up the subwoofer by ear or just adjust its setting to match the lower -3dB point of the monitors or do you do something more scientific?
 
I used a Klark-Teknik DN60 spectrum analyzer to help, but as I pointed out in another thread - that's of limited use. I made for any limitations by using my ears!


Bruce
 
That's the way to do it!

Trust your ears...or think about quitting the recording world...

Seriously, you can use Analyzers as Bruce said, but it's you who will be mixing, so you should set it for YOUR ears. IMHO

Peace...

PC
 
Ears are what I'd planned to use - if nothing else I've already got a pair of those so it won't cost me anything. :) My ears are actually quite sensitive to low frequencies. I can usually hear the "thump" of a helicopter well before its "chop" becomes audible and I can hear the port on all but the best ported speakers. I suppose that's why I took up the bass.

What sort of technique should I use to adjust the subwoofer? Should I set the crossover frequency a bit below the monitors' bass roll-off, adjust the subwoofer volume until it seems to be right and then gradually raise the roll-off until a "lump" appears, and then back off the roll-off until it seems smooth?

I suppose for fine tuning of subwoofer level, it's just good-old translatability tests of my mixes to big speakers and PA systems?
 
I haven't looked in over a year but I think mine is at 40 hz...

I Basically get the bass drum and the low-lows of the bass...

I find that at 60 hz, I get to much highbass, that just cuts through and rattles everything including my brain....
 
Without proper test equipment or at least detailed response and phase data for both the subs and monitors, DIY subwoofer setup is really just a shot in the dark.

Imagine if your monitors came with no crossover between the woofers and tweeters and it was left to you to set it up. How much luck do you think you would have? Considering manufacturers will sometimes spend hundreds of engineer hours developing these crossovers, my guess is not much.

This is not a completely analogous situation because you typically have more response overlap between a sub and woofer than you do between a woofer and tweeter. Still, other issues make the problem very complex. I've attached an overlay of half space magnitude, phase, and group delay responses for a typical nearfield and sub combo. First of all, can you even identify which plots are which? Knowing all this important information, how would you proceed in designing a crossover? How will proximity and room effects likely impact the situation?

Starting to get a feel for the level of difficulty in properly setting up a subwoofer? Think of all the time, learning and thought required in order to come up with a decent mix. Well, people like me spend as much or more effort trying to figure out solutions to problems like this. Of course you can Do It Yourself, but doing it right is as hard as anything you will encounter in recording.

barefoot

13KB attacment
 

Attachments

  • sat-sub.gif
    sat-sub.gif
    12.5 KB · Views: 48
Interesting, but how does it help me? Should I pay someone to do the setting up? Should I not bother with a subwoofer but get some main monitors?

I just want to hear if something nasty is happening below 50 Hz in case the music gets played over a PA or some other system capable of rendering deep bass.

Come to that, is a shot in the dark all I need? If I'm working on a paying project I'd probably go mix at a proper studio anyway. I just need to hear if there's anything overly dodgy happening. I don't do extreme EQ so it's just a safety check.
 
Well now, even though subwoofer integration is a highly complex subject, I wouldn't say it is beyond the capability of anyone with a good ear!

Barefoot's analogy comparing a sub crossover to the crossover between a woofer and tweeter is interesting, but I would add that it is easier to get a subwoofer and woofer to cross over properly together than it is with a woofer and tweeter, especially considering we're using an active crossover for most subs. The amplitude and phase response of a woofer and subwoofer is much easier to deal with.

Is it a shot in the dark? Well yes and no. Even if you design a subwoofer that crosses over to the woofer with little to no error in phase and amplitude, you still have to contend with room modes, standing waves, boundary effects, etc etc etc. All these issues tend to overpower the subtle adjustments you may make to your crossover.

So the best course of action I think is to experiment! Just keep trying different placements, crossover points, slopes, driver polarity etc until it sounds best!

As a good start, I design my own active crossovers to have a 24db/oct Linkwitz-Riley slope on the highpass and lowpass. That way, there is excellent electrical linearity between the hi and lo pass. Once you hook that up to your sub and mains though, a lot of that gets shot to hell, but with proper placement, excellent results can be achieved with not too much difficulty. Also, the steep 24db slope minimizes the overlapping region, which is by far the most troublesome.

iqi616, I definitely think you should be mixing with a subwoofer! You really aren't getting the whole picture if you don't use one. I have heard far too many recordings that have some kind of nasty low rumble going on that the engineer didn't even know was there!
There are very few main speakers that can actually go flat to 20Hz. VERY few, no matter what the brochure says...
 
I'm always astonished by how cavalier people are when it comes to bass reproduction. The consensus seems to be "As long as I can hear something and it's not boomy, I'm happy". Well, I guess it shouldn't surprise me considering how few people have actually ever experienced accurate bass reproduction. Even relatively high quality 10" or 12" mono subwoofers are producing 5% or 10% distortion at average monitoring levels. Sound reinforcement subs are almost always being pushed close to their limit, so the most you can say for them is "Yes, there's bass alright".

Phloodpant,
Active crossovers eliminate impedence response concerns, but they don't negate other alignment issues. And your L-R filters may have fairly steep asymptotic roll-offs, but they also have fairly soft knees. In any case you're still looking at about an octave of significant crossover concerns. You did an excellent job in building your brother's home theater, as I'm sure you do with many things. However, when posed with the question of how to setup an accurate system for recording monitoring, I don't think these sort of hand waving arguments are adequate.

If you're not going to go all the way and properly align your system's bass response (and by "system" I mean the subs, satellites, and the room), then you might as well just throw on a set of good closed cup headphones from time to time to check the low bass. It's certainly not as pleasurable as listening to the real thing and you loose the visceral sensation , but as long as you stick to running the lows straight down the center it's quite "learnable" and will be more useful than a shot in the dark subwoofer setup.

barefoot
 
I already check my mixes through headphones (I have a pair of really nice Beyerdynamics) and sometimes I play the mix through my Trace Elliot 1215SMX combo via the FX return but that's more just for the hell of experiencing the bottom end because it's a 1x15 with no tweeter so there's not much upper mid and almost no HF, and it's designed to make a bass guitar sound good, not to give accurate reproduction.

I must admit that I'm starting to wonder if I should go for the HR824s instead of the 624s and forget about getting the subwoofer. But then again, in view of a recent rant about 8" speakers that no-one refuted...

Another question: if I don't have a spectrum analyzer, can I use an SPL meter and sweep a sine wave all the way up from 10Hz and plot it on a graph or are SPL meters (and I'm sure you all know which one I'm talking about) non linear in frequency response?
 
Have a listen to the 824's. If you like them better than the 624s, then by all means get them! There is no rule saying you can't make a good 8" 2-way. True, 8" speakers in GENERAL have a harder time producing detailed midrange, but like all generalities, there are exceptions. One way of overcoming the 8"problem is to use a tweeter that can be crossed over lower than most. When you move the crossover point lower, you can overcome the beaming problem as well as allowing the tweeter to do some of the detailed midrange work.

That said, I don't think any 8" monitor is a substitute for a good subwoofer. It's just not there.

Headphones will also not necessarily help you determine the subsonic bass content in a recording. As wonderful as a good set of cans is, there is almost no tactile feel from them.

Now, I'm not advocating a "shot in the dark" subwoofer setup. Integrating a sub is a lot of brain and legwork. Test equipment can be helpful, but even then things don't always sound how they measure.

But it can be done, and I think the results are very much worth it.

The "shot in the dark" comes when you mix without a sub at all!
 
Oh, and about the sine wave sweep...

Unfortunately, that's not a very good method. Room resonances build up while you're taking the measurement, and the graph such a method would produce will be much worse looking than actual transient performance.

The preffered method is an MLS measurement, which requires some sophisticated software. There is a freeware package called Speaker Workshop that will actually do this for you, but it's not easy to do! There's a learning curve involved. Also, you'll need a super flat omni measurement mic. Fortunately a Behringer ECM 8000 is just the ticket!

Basically, an MLS measurement consists of a short burst of pink noise. The computer records the response from the speaker and only analyzes to the tiny moment of sound before the first reflection hits the microphone. That way, only the direct sound is anayzed with no reflections, and therefore no room effects. Effectively, this is as good as using an anechoic chamber.

While this is the preferred method, you can see how this can be problematic for a sub! A sub operates at very long wavelengths, and it's response relies heavily on boundary reflections, room modes, standing waves, etc etc... So even if you have a sub that measures flat using any method you can name, it may not sound that way!

Basically, what I do is design the sub to be flat by design, and the crossover to the mains the be similarly flat. Then play with room placement, level, crossover frequencies and slopes. Take some measurements, but most of all, listen to it!

You're never going to get "repeatability" or "customer translatability" in the sub-bass region like you can get with the main speakers, but you can at least work with a system that is world class.... and that's a lot better than just forgetting about the low bass entirely!
 
Oh, I guess the original title was asking "Wotcha got??" So here's what I gots... :)

4HPtube.jpg


This sub uses a driver made by TC Sounds called the 4HP. It is also sold in Germany as a car audio driver although it's not specifically designed as such.

The GZ Plutonium...
http://www.magma-audio.de/english/gz_plutonium.htm

This is a 15" driver, although in the picture it doesn't look that big because the magnet is so damn huge! You really have to see one of these things to believe it. It's just ridiculous! it weighs 75 pounds!

The driver does an honest 1.5" of linear X-Max. and displaces more than 6 liters of air. That's a lot more than a typical 18" driver.

Now, a lot of 15" drivers have a reputation of being sloppy, but I assure you, with a motor structure that could easily break your leg, cone control is no problem!

The enclosure is a 16cuFt sonotube. Porting is at 16Hz with six 4" ports about 48" long exiting out the top. It stands about 7' tall.

Crossover is a hand built 24db/oct L-R with a 6th order highpass transform circuit.

At about 12Hz, I can literally move furniture!
 
Phloodpants said:
True, 8" speakers in GENERAL have a harder time producing detailed midrange, but like all generalities, there are exceptions. One way of overcoming the 8"problem is to use a tweeter that can be crossed over lower than most. When you move the crossover point lower, you can overcome the beaming problem as well as allowing the tweeter to do some of the detailed midrange work.
Click to expand...
Have you heard the expression "Robbing Peter to Pay Paul"?

In my 8" rant I think I mentioned that I prefer 5" midbass drivers. An average dome tweeter has about 1/10th the surface area of a 5" midbass. When you try and extend the low range of a tweeter you're asking it to travel 10 times the linear excursion required of a 5" in order to achieve the same volume. Dome tweeters aren't particularly well suited for long excursions, especially near their resonance frequencies, no matter how well they are made (btw, I believe the Mackie uses a variation on the Vifa D25AG which is good, but not outstanding). This is a recipe for distortion.

Mackie and others try to reduce the distortion associated with their low crossover frequencies by horn loading the tweeters. This helps to some extent, but then of course the increased directionality runs you back into the same old problem of power response irregularity. The step from a 1" driver to an 8" driver is just too big IMHO.

A Radio Shack type sound level meter will not likely give you reliable information. As far as gated pulse measurements, these are not suitable for low frequency analysis in a normal sized room.

Anyhow, I could go on forever. Maybe more later.
barefoot
 
You must log in or register to reply here.
Back
Top