which q? .707 or .577

[slr 5000]

hi all im building a speaker design using an eton 12"
hexacone for bass and a manger transducer for mid/high
it is a b&w 801 copy but is a sealed design

ive biult the bass cabs based on a q of around .577 which gives me 80 odd litres after bracing and the proportions of the cabs just dont look right (too tall)

if i reduce the cab size to 50 litres by cutting 200mm off the bottom of the cabs then the cab proportions will look a lot better and be very similar to the design im copying
but this will mean a cab q closer to .7
every one seems to suggest a q of .5 would be best but this makes the cabs to big

what are the downsides and benefets of different q

the speakers will be active using a dbx260 and hypex amps

 

[Moondog55]

While looks are important, I think you should decide your priority Sound quality or looks and then build your box to suit.
But 80 litres is a small box by my standards (( how-ever not by the wifes HA HA ))
Haven't even looked at specs for that driver so can't help there, but how tall are these bass boxes??

regards
Ted

 

[slr 5000]

hi mate
they were one metre tall and 400 wide
they are the same shape as the b&w 801 with a round back

1 metre doesn,t sound big but if you look at the b&w design with only one bass driver in the bass cab and the mid/tweet on top in its own cab it just doesn,t look right being this tall
it should be more of a wide sqaut looking design

as of this afternoon ive cut the cabs down to 800 high and recalculated the volume for a total of 70litres after bracing (i must have had over 100 litres and a q of .5 before i cut them instead of the 80l i thought i had)

it now looks great and has the same look and proportion as the b&w 801 nautilus and i just have to fitt a new bottom

the q is now around .6 so should still blend with the mangers
(fingers crossed)

 

[Drew Eckhardt]

ive biult the bass cabs based on a q of around .577 which gives me 80 odd litres after bracing and the proportions of the cabs just dont look right (too tall)

if i reduce the cab size to 50 litres by cutting 200mm off the bottom of the cabs then the cab proportions will look a lot better and be very similar to the design im copying
but this will mean a cab q closer to .7

And Fb 20% higher.

the speakers will be active using a dbx260 and hypex amps

Since you're already active, you might look at your power requirements to equalize to the lower Q + Fb in the smaller box at the output level you require and provided it works make the cabinet fit your aesthetics and electronics fit the response you want.

 

[pacificblue]

what are the downsides and benefets of different q

The higher the Q the better the rise time, the worse the fall time. Q above 1,1 is not considered hi-fi by many people. Best fall time is at 0,5, best rise time at 1,0. At 0,707 rise and fall time should be equal.

Another characteristic of Q is the frequency response roll-off. With a Q of 0,707 you get the flattest roll-off and the deepest resonant frequency. A higher Q leads to a peak before roll-off and then the roll-off is steeper. A lower Q leads to an earlier and very smooth roll-off.

For bass you don't want a low Q, if you don't use equalization, because the bass may sound weak. A higher Q gives you more punch. Values between 0,65 and 0,9 are a good compromise between depth and precision without equalization. More can be desirable, if you want hefty punch, and precision is not such an issue.

Around 0,5 is nice for the mids and highs, because it sounds more precise.

Which Q you want depends on many factors. For one thing Q factors cumulate, so a speaker's Q of 0,5 doesn't help much, if your crossover already has a Q of 10, which is more common than one would think.
Then the Q is also a question of your room response. A highly dampened room can compensate for a high Q factor. A room with little damping can easily make any attempt at precision futile, whatever the Q factor.

Literature about filter design usually explains well about Q factors and their properties.

 

[Moondog55]

Hi Pacificblue, what would give a Xover "Q" of 10??
Chebychev with its peak is a (1) n a 10 would surely be unlistenable>> finger slip perhaps?

 

[planet10]

Re: Re: which q? .707 or .577

Another characteristic of Q is the frequency response roll-off. With a Q of 0,707 you get the flattest roll-off and the deepest resonant frequency. A higher Q leads to a peak before roll-off and then the roll-off is steeper. A lower Q leads to an earlier and very smooth roll-off.

For bass you don't want a low Q, if you don't use equalization, because the bass may sound weak. A higher Q gives you more punch. Values between 0,65 and 0,9 are a good compromise between depth and precision without equalization. More can be desirable, if you want hefty punch, and precision is not such an issue.

Actually the above has to be reconsidered once the box is put in a room... room lift will often give a box with a nominal butterworth Q (0.7) a fat, peaky bottom.

One has to consider the room, where in the room, and where the box rolls off.

I typically target between 0.5 and 0.7, at least small changes can be made with just the amount & type of damping.

dave

 

[slr 5000]

hi all
the new cab volume gives me a q of dead on .6 so i think this should work well

the system is in a 6m x 6m room which is just for audio so ive got heaps of freedom regarding speaker placement and the speakers can be placed up to 2m from the back wall if i need to

can someone tell me more about the q of the crossover and how it affects the system and give some examples if possible

once built im planning on taking the speakers outside and measuring them
i will them use the auto eq on the dbx260 to get a flat responce before taking them inside

id prefer to get them as flat as possible outside and then use room treatments to get the in room responce good instead of using the dsp in the dbx to change the in room responce

i hope this will work out ok

 

[planet10]

id prefer to get them as flat as possible outside and then use room treatments to get the in room responce good instead of using the dsp in the dbx to change the in room responce

i hope this will work out ok

The room is the dominant component in any system... and where it dominates in the LF you aren't going to fix with room treatment.

Best to work with it and not against it.

dave

 

[Moondog55]

Bear in mind that I am not an expert on crossovers
As I understand it low "Q" crossovers are better damped and tend to flatter responce ie LR type crossovers are Q= 0.5 where the inductor value is doubled and the capacitor value is halved, Chebychev have value equal of both inductor and cap at "1" and there is quite a peak at the crossover frequency.
I conjecture thatif you used values in the reverse of LR ie double the capacitor and halve the inductor there would be a big peak at resonance and the possibility of "ringing"
Electrical crossovers NOT audio

 

[Moondog55]

In box reponcse "Q's" look to follow the same shape as crossovers, low "Q" 0.5-0.6box no peak, medium "Q" box small "Q" box flat to minimal peak, high "Q" box peaky at box resonance ( think cheap BOOM BOX )

As the box gets smaller ( for the woofer involved ) "Q" increases
I think most of us were so used to fairly high :"Q": boxes that that was the sound we built to before we learnt better.
That said party boxes for dancing probably benefit from Q=1 or higher, the bass is louder at the 100 Hz frequecy of kick drums, and when dancing rythym is everthing, just bounce to the beat.

This probably didn't help much, somebody smarter and more experienced than me please jump in and help

Regards TED:

 

[slr 5000]

hi ted
thanks for your help
any ifo on this matter is much appreciated
cheers

 

[Moondog55]

Where in Melb RU???
See my garage sale listing in Trading post section

 

[slr 5000]

hi mate
im in werribee south
ill check out your listing

 

[richie00boy]

Provided the crossover frequency is a couple of octaves or more away from the bass resonance frequency, the Q of each element does not interact. So for most normal 2 way speakers, the Q of the bass resonance and the Q of the crossover are completely separate entities and won't affect each other.

 

[pacificblue]

Hi Pacificblue, what would give a Xover "Q" of 10??
Chebychev with its peak is a (1) n a 10 would surely be unlistenable>> finger slip perhaps?

In bandpass filters Q = f/B, where B is the bandwidth. Imagine you want to flatten out a small dip of 100Hz bandwidth in the frequency response at 1000Hz. You get a Q of 10. Now even if that bandpass is in series with a Butterworth filter (Q=0,707), the result is a total Q of 7,07. If you want to achieve a total Q below 1 after that filter combination, your loudspeaker would need a Q below 0,14.

You may call that a finger slip, but it happens quite often in cascaded active filters or in equalizers. It is not always unlistenable. In some cases you are lucky and the Qs of all filters and the speaker together add up to a listenable value. And in most cases...

The room is the dominant component in any system

... and determines the audio quality much more than the Q of a single component, whether that be the speakers or the crossover.