Can a sub be too big for the room?

[andyr]

The fly in the ointment is any latency in a modern sub due to dsp.

As my miniDSP 'nanoDIGI' provides all the XOs - ie. sub-mid/bass & mid/bass-tweeter - I suspect the same latency is imposed on all 3 drivers?

What I want to do is be able to prove this using REW; I just don't know which REW graph will show me.

It’s good to get to get the delays correct, although I’ve found there is a certain amount of wiggle room before errors become audible. In practice

I'm sure there is - however, there's no harm in getting it absolutely right!

I’ve found it is better to err on the side of the sub being delayed relative to the mains.

In my case subs are about 5' further away from my ears than the 'mains' - so they certainly do not want to be delayed further.

REW is one’s friend though, although I can’t remember, without firing it up, exactly how to do it. Not too difficult though.

Yes - I'm sure it certainly is 'not too difficult'! (As long as one knows which graph and which settings to use! )

Are you just looking for a speed of sound calculation, so the waveform arrives at the same time?

I've done the maths, Mike - I'd simply like to be able to confirm from a graph that my calculations were correct.

 

[camverton]

As my miniDSP 'nanoDIGI' provides all the XOs - ie. sub-mid/bass & mid/bass-tweeter - I suspect the same latency is imposed on all 3 drivers?

What I want to do is be able to prove this using REW; I just don't know which REW graph will show me.



I'm sure there is - however, there's no harm in getting it absolutely right!



In my case subs are about 5' further away from my ears than the 'mains' - so they certainly do not want to be delayed further.



Yes - I'm sure it certainly is 'not too difficult'! (As long as one knows which graph and which settings to use! )



I've done the maths, Mike - I'd simply like to be able to confirm from a graph that my calculations were correct.

REW phase Have a look in here. It’s bit complex but about half way down there is an example which shows the delay of a sub. You might find easier instructions with a bit of searching of the jolly old internet. Hope this helps.

 

[darrenyeats]

Which is not something I'm interested in doing, unfortunately.



That is not what I am trying to find out how to do - as I know how to do it.

I am simply trying to find out how to get the delay right, between my 'mains' and my subs..



Thank you - I'll have a look.

Although I suspect he will be talking about sub placement and phase setting - rather than delay.

IME phase knobs on subs are really delay knobs. The phase value is only valid for one frequency i.e. it's really a delay.
Usually, you would want the mains to be delayed (positioning + many subs have DSP) - most AV receivers provide such a function. But the phase (delay) knob on a sub does the opposite.

This is different to polarity which I would leave untouched.

 

[andyr]

IME phase knobs on subs are really delay knobs. The phase value is only valid for one frequency i.e. it's really a delay.

Not sure I follow your argument?

I use a miniDSP unit to set the XO between my (stereo) subs and the mains; 48dB L-R slopes @ 110Hz. Because they're 48dB slopes ... the subs need to be in phase with the woofers - so the phase knob on each sub is set to '0 deg'.

The miniDSP unit enables me to invert the sub driver 'channel' - which is the same as turning the sub's phase knob round to 180 deg.

The result, after running an REW sweep is the same in either case; if the sub is not in phase with the woofer ... there is a huge dip at the XO frequ.

 

[darrenyeats]

Andy, I'm wondering if could look into using your miniDSP to set an exact time delay (probably on the mains) which is usually the optimal thing for a sub.

In the following I'll cover why I don't like reversed polarity, even though you aren't reversing polarity!

First point, the thing that needs "fixing" is a time lag, usually due to DSP in the sub itself and/or positioning of the sub+mains. This time lag will correspond to X degrees of phase at the crossover frequency - X could be 0 or 180, but just as likely X could be 30 or 175 degrees etc. Second point, the same time difference would correspond to a different number of degrees than X at other frequencies, for a non-zero lag.

In contrast to a delay:
(a) considering the crossover frequency, normal/reversed polarity acts like a 0/180 degree phase shift that's not likely to give exact phase alignment (but clearly one setting could give a much better result than the other - this is what you're reporting with normal polarity)
(b) let us suppose reversing polarity gives optimal phase alignment at the crossover frequency - then it won't do so away from the crossover frequency - don't expect optimal phase alignment across a whole crossover region simply by reversing polarity (you can with a delay). This is why reversing polarity is bad, so good that you're not doing that.

For a 48db/octave crossover (b) might be less of an issue though.

Right so about sub knobs ...

If the sub has a switchable two-position 0/180 degree phase knob, then yeah, that'll really be polarity yes. Sounds like yours have this Andy?

IME when it's a variable analogue 0-180 degree phase knob, that acts as a delay (good news) albeit delaying the sub (which is not usually where ideally you'd want to add the delay, somewhat less good news) - and in this scenario there will probably be a separate dedicated control for polarity on the sub, labelled such.

The good news, you have the means (miniDSP) to introduce an exact delay to the right place, so circling back to my first sentence.

 

[Neil P]

I'd like to add a sub to my AV/stereo system, simultaneously supporting both front channels and providing LFE. Can subs ever "overpower" the room, or is it just a case of turning them down a bit? The model I've got my eye on is 15" 1000w... My room is 6x4m. My previous sub was 12" 500w but couldn't simultaneously do life and hi-level input, so I sold that to fund the upgrade.

I'd forgotten that my room actually opens into 2 further rooms! Both doorways are "door-less". The red doorway has bookshelves (with backs) both sides, the yellow one is open. I wonder how much of the total space is "seen" by the sub ... I have only ever considered the 6x4m room.

 

[Sonority]

Erm, their are some really big subs out their - and crazy people.

 

[daytona600]

 

[andyr]

Darren, it seems my setup is confusing for you?

As I said, I use a miniDSP unit (the all-digital 'nanoDIGI') to set the XOs between my (stereo) subs and the 'mains' drivers:
* 48dB L-R slopes @ 110Hz for the sub-to-mid/bass XO, and
* 24dB L-R slopes @ 2800Hz for the mid/bass-to-tweeter XO.

REW can show you whether you have adjacent drivers connected with the correct polarity. Theory says ... 24dB & 48dB slopes should have the drivers connected with the same polarity; if you reverse the polarity of one of the drivers in a pair - you should see a dip in the FR at their XO point.

REW confirmed that all 3 drivers in each channel of my system -ie:
* sub
* mid/bass, and
* tweeter

... need to be connected with the same polarity.

So that is "driver polarity" out of the way!

My subs are situated outboard of my mains - hence I need to delay the mains, to time-align them with the subs. My calculations of the difference in the distance from my ears, of the mains and the subs ... produced a required delay for the mains of 3.13ms - so that is what is set, in the nanoDIGI.

All I am interested in knowing is ... which of the many REW graphs will show me that my theoretical calculations are correct - and the mid-bass drivers are time-aligned with my subs?

 

[camverton]

Darren, it seems my setup is confusing for you?

As I said, I use a miniDSP unit (the all-digital 'nanoDIGI') to set the XOs between my (stereo) subs and the 'mains' drivers:
* 48dB L-R slopes @ 110Hz for the sub-to-mid/bass XO, and
* 24dB L-R slopes @ 2800Hz for the mid/bass-to-tweeter XO.

REW can show you whether you have adjacent drivers connected with the correct polarity. Theory says ... 24dB & 48dB slopes should have the drivers connected with the same polarity; if you reverse the polarity of one of the drivers in a pair - you should see a dip in the FR at their XO point.

REW confirmed that all 3 drivers in each channel of my system -ie:
* sub
* mid/bass, and
* tweeter

... need to be connected with the same polarity.

So that is "driver polarity" out of the way!

My subs are situated outboard of my mains - hence I need to delay the mains, to time-align them with the subs. My calculations of the difference in the distance from my ears, of the mains and the subs ... produced a required delay for the mains of 3.13ms - so that is what is set, in the nanoDIGI.

All I am interested in knowing is ... which of the many REW graphs will show me that my theoretical calculations are correct - and the mid-bass drivers are time-aligned with my subs?

You’ll find the answer to your question in the link in post #22. Given the very high crossover frequency and steep slope you are using from your sub you are wise to want to check the delay.

 

[gez]

Probably also helps if you have a large detached house in the country a very long way from the nearest neighbours!

The problem with big subs in small rooms often isn't so much their performance in the room as outside it.

That's a problem with subs (and speakers that have bass extension down below 40hz in any meaningful way) period. They're likely to make your neighbours every unhappy. Pretty much every house built since 1900 (in the UK and I'm betting most of the world) is essentially acoustically transparent to those frequencies. (ok not literally but in comparison to mid range frequencies - by orders of magnitude more).

Whilst it's great to have a system with decent output low down (my own speakers have audible bass at 25hz when the overall listening level is 60dBA), most neighbours will be very unsympathetic. Thankfully my neighbours are half deaf in their late 70s/mid 80's and have never complained or even mentioned my film/music viewing to date, and I'm on good speaking terms with them so I know they'd say something if they cared.

Anyway, as others have said. There's no such thing as "too large" a sub, except that it'll possibly visually dominate the room. That said size isn't a guarantee of more volume. It's not even a guarantee of more bass extention. As with everything a lot comes down to design and implementation. I'd buy a well designed high quality 12" sub over a cheap 15" or even 18" every day of the week.

The main advantage of multiple subs over a single is bass evenness, and the potential (if you place them correctly) to reduce suck outs at certain frequencies or in certain locations in the room. But if you have a single listening position, aren't building your system for other people to listen to at the same time as you, one sub is sufficient. Just for volume for a given cost, you're likely to be better off buying a single more powerful sub, than multiple subs (of course if you have multiple of any given sub you'll get 3dB more bass output, but you'll be oaying twice as much or more for it). Bass isn't directional in the slightest in the sub 40hz region, so there's no "stereo/surround sound" advantage with multiple subs.

 

[Colin L]

Totally agree with all the above, except twin subs does enhance the soundfield (airiness and imaging) versus one sub IMO.
Oh, and living in a detached house helps when running 2 x REL Carbon Specials

 

[gez]

Darren, it seems my setup is confusing for you?

As I said, I use a miniDSP unit (the all-digital 'nanoDIGI') to set the XOs between my (stereo) subs and the 'mains' drivers:
* 48dB L-R slopes @ 110Hz for the sub-to-mid/bass XO, and
* 24dB L-R slopes @ 2800Hz for the mid/bass-to-tweeter XO.

REW can show you whether you have adjacent drivers connected with the correct polarity. Theory says ... 24dB & 48dB slopes should have the drivers connected with the same polarity; if you reverse the polarity of one of the drivers in a pair - you should see a dip in the FR at their XO point.

REW confirmed that all 3 drivers in each channel of my system -ie:
* sub
* mid/bass, and
* tweeter

... need to be connected with the same polarity.

So that is "driver polarity" out of the way!

My subs are situated outboard of my mains - hence I need to delay the mains, to time-align them with the subs. My calculations of the difference in the distance from my ears, of the mains and the subs ... produced a required delay for the mains of 3.13ms - so that is what is set, in the nanoDIGI.

All I am interested in knowing is ... which of the many REW graphs will show me that my theoretical calculations are correct - and the mid-bass drivers are time-aligned with my subs?

Frequency response.

Delay affects relative phase between the mains and the subs. If the delay is aligned then phase is also aligned. The only place it matters is at the XO frequency (you can't control how phase varies with frequency within the subs output itself or the mains or how the room impacts any of it, so the XO region is the only place you can affect the integration). The frequency response of the XO region will have maximal output when the sub and mains are aligned in phase (i.e. the delay is zero - because zero delay = 0 phase difference at the XO frequency). So you need to adjust delay and the delay that gives the highest spl at the XO frequency is the correct phase, as you can't get more spl than perfectly aligned phase will give you. That may not leave you with a flat response at the XO point (or region) by the way, that'll be determined by the relative outputs of the sub and mains at the XO frequency and the steepness of the XO curves (as well as the XO frequency you've chosen on the sub obviously).

 

[ToTo Man]

Whilst it's great to have a system with decent output low down (my own speakers have audible bass at 25hz when the overall listening level is 60dBA)

IIRC the threshold of audibility of 20Hz is 75dB SPL so I'm very impressed that you can hear 25Hz when listening at 60dB SPL.

 

[gez]

Have you seen these videos by the way?

 

[gez]

IIRC the threshold of audibility of 20Hz is 75dB SPL so I'm very impressed that you can hear 25Hz when listening at 60dB SPL.

Well I'm only going by an app on my tablet, which likely as not isn't very accurate. But according to the tablet the absolute dBA volume when playing the 25hz tone was around 40-45dBA. Off my head I have no idea how that relates to the human perception curves in terms of SPL (I believe those curves are unweighted). Wouldnt dBA (which is perceptual volume) give a figure significantly different from unweighted SPL?


Just checked, yes those human perception curves are linear SPL or Z "weighting" (i.e none). So given the app I used is dBA, the level it will show in the bass frequencies is going to be much lower than the human perception curves indicate.

This site (if it's to be believed) states for example that at 63hz dBA is 26dB lower than unweighted, the difference at 25hz is going to be a lot bigger (meaning that if my dBA meter was showing 40dB at 63hz then the unweighted measurement would actually be 66dB - so a 40-45dbA at 25 hz is very likely more than 70-75dB unweighted) :

Understanding A-C-Z noise frequency weightings

Understanding A-C-Z noise frequency weightings , what you need to know and why you need to know it.

pulsarinstruments.com

Those human perception curves are averages too, they're not absolute. They don't mean that no human can hear better than that (we know that some exceptional people can hear above 20khz for example, where the curves would imply that's impossible). I'm not claiming that the variation is going to be huge but there is likely to be some variance between individuals (eg I wouldn't be surprised if one person could hear 20hz at 5-10db lower levels than another person).

I'm not claiming any special hearing ability by the way (even though recent youtube tests would appear to indicate I have the high frequency hearing ablity of someone 7-10 years younger than me )

 

[stevec67]

Go for as big as your partner/finances/foundations/walls will allow

You will be far better off with a good monster(s) who is just ticking over flapping your trousers than a smaller sub that has to operate near limits at lower frequencys to give you the same `effect`
Tuning the beast small or large with good DSP or eq is key, lowering output rather than boosting frequencys is best.
Headroom, allways good and allows blowing the socks off visitors!
Infrasonics, once experienced you want more

Kids' stuff. You want a vane subwoofer, then you can build it into a side wall. 80"? Pfft. I want something that makes me feel nauseous when I turn it up. 7Hz at least.

 

[gez]

Kids' stuff. You want a vane subwoofer, then you can build it into a side wall. 80"? Pfft. I want something that makes me feel nauseous when I turn it up. 7Hz at least.

I once went to a concert where the bass guitarist was using a 5 string bass, with the extra string being a low string (don't know what note/frequency that would be off the top of my head but a standard Bass E string is 41hz). I'd drunk and smoked a bit too much that evening and the low bass from the bassists guitar was messing with my stomach. I ended up having to go to the loo to throw up it made me that queezy.

Just looked it up a low B string is 31hz

 

[darrenyeats]

Darren, it seems my setup is confusing for you?

As I said, I use a miniDSP unit (the all-digital 'nanoDIGI') to set the XOs between my (stereo) subs and the 'mains' drivers:

I'm sorry, I should have read all your other posts, where you state you're indeed trying to set a correct delay.

I hope someone will still get some value out of my prattlings.

 

[stevec67]

In my case subs are about 5' further away from my ears than the 'mains' - so they certainly do not want to be delayed further.

So, they're not just smaller, then?

I once went to a concert where the bass guitarist was using a 5 string bass, with the extra string being a low string (don't know what note/frequency that would be off the top of my head but a standard Bass E string is 41hz). I'd drunk and smoked a bit too much that evening and the low bass from the bassists guitar was messing with my stomach. I ended up having to go to the loo to throw up it made me that queezy.

Good excuse. Better than the dodgy goujon one.
I have heard that frequencies below 15Hz can make you nauseous, with 7Hz being a resonant frequency that can cause physical harm if it's loud enough. There was of course an infamous sonic weapon that attempted to exploit this, but the difficulty was that the sound wasn't directional enough and the weapon operators were being harmed instead.
Years ago I made a sub and installed it behind the listening sofa. I had a mate round, played some music and dialled the sub up without saying while I "made a cup of tea". A few seconds later he was giggling away and complaining that the sofa was trying to assault him.