Beginner Crossover Design - Advice

[thoglette]

 how do you then clean up the job and mount the crossover inside?

 

Hope its not a stupid question 

It's not a stupid question.  The n'est plus ultra approach is to keep the crossover outside in a separate box - capacitors and unpotted inductors are microphonic.    If that doesn't float your boat aesthetically, then you can prepare a box-within-a-box or leave a removable panel section.  Or remove a driver to get access to the inside of the box.  I've used all of these approaches over the years

[davewantsmoore]

Well i have sort of been thinking about sealed, Origianlly i was just going to go ported so the bass would be a little nicer.

 

You will get "more" bass from a ported box.    You might call this "nicer", but it is simply "more".    In a perfect world that additional bass will be of the same quality as bass from another enclosure.... but the world is not perfect, and so whether or not the bass is any good will depend on your ability to design a good ported system   (it's very difficult when keeping the box small).

 

For the 3 or 4 extra dB at 20hz you are geting from the ported box ..... I would just go sealed.   Especially seeing as you have access to EQ, which you can use to set whatever bass output you want from the speaker   (with the only restriction being that boosting the bass, will reduce the overall maximum SPL the speaker is capable of before it reaches it's excursion limits).

[Neo__04]

Yeah possibly poor wording on my part. Lower bass i guess i was getting at.

I do think ill run a seperate sub.

 

So im pretty much set on sealed now. Time to start drawing it all up

[davewantsmoore]

EDIT   Whoops!!!.   SB Acoustics specify their Xmax value as peak-to-peak (not standard), which caught me out.   This means that the Xmax in these charts is actuall half what I though.   So the linear excursion limit line in the charts below should be 5.5mm not 11mm.

 

.... which means the driver won't play as loud as shown.   I should have really realised this when making the charts, as 11mm linear travel on a 6" driver is 'too good to be true'.

 

----------------

 

Good choice for sealed and subwoofer IMO.

 

This really is a nice little midwoofer, with a surprising amount of linear travel.    I've heard speakers with it, but never built with it myself.

 

 

Here some quick screen grabs to show the difference between box sizes, and then how those box sizes behave when EQed to a crossover point of 60Hz and 24dB/octave, which you might use for a subwoofer, for example.

 

Green is a 30L box, and yellow is a 9L box.  These are about as big and as small, as will make a significant difference to the response.   All charts are at 60 watts.

I'd pick somewhere around 15 or so.....   but not a big deal.

Plan to drive it with at _least_ a 100w (4ohm) solid state amplifier.

 

 

The first pic.... you can see the response given by the different box sizes, without EQ.

 

a)  cone excursion in mm  --- larger box lets the driver move more at lower Hz

b)  SPL  ---   cone moving more, means more SPL at lower Hz for the same input power

c)  Impedance ---   box size affects what the speaker looks like to the amplifier

 

The second pic....   you can see that once the two boxes are EQed so they produce the exact same response

 

a)   SPL  ---  once the two different boxes have EQ applied so they produce the same response - that the lines overlap.

b)  Cone excursion --- we can see that when the SPL is the same, the cone movement is the same --- no matter what size box.

c)  Apparent power --- this is how much load the amplifier sees.    Look back at  1a and 1b, the bigger box lets the cone move more .... se we can see here that to produce the same response, the larger box requires less power  below 70Hz.

 

 

#1

 

 

#2

 

Edited March 14, 2016 by davewantsmoore

[Sub Sonic]

@@davewantsmoore which size box would you pick for best driver damping?

[davewantsmoore]

@@davewantsmoore which size box would you pick for best driver damping?

 

The word 'damping' is one of these things which creates so much confusion and misinformation.

 

A lot of this confusion comes from people who have written articles from the perspective that we have nothing else but the box to control the response of the cone (which is the case for a typical traditional speaker with passive filter networks).    But if we have EQ, we can set ANY response we like...  and hence any concept of box "Q"  (ie. Qtc)  or 'damping' becomes irrelevant.

 

 

You can see that in the second picture.....   When EQed to the same response,  the cone movement for the two is identical.   So the box has no effect on the steady state response.

 

Another thing to look at is what happens when the amplifier stops suddenly (ie.  not a steady state signal)   .... and the cone is allowed to fall back to rest on it's own.    What happens is that the cone follows the amplifier (ie. it returns to rest when the signal does) .... except at the region where the mechanical system is in resonance  (this resonant energy makes the cone keep moving).     The box size effects where this region is --- we can see it in the impedance chart.      In this region, the amount of 'damping' provided to the cone is dominated by the amplifiers electrical 'damping factor', and not the box.   This is because the mechanical system is in resonance, and by definition, the mechanical damping => 0

 

 

I can explain whichever bits of this in more detail (or in simpler terms)  ...... but so I don't create any confusion or debate.... perhaps it could help for you to say exactly what you mean by "damping".

[Sub Sonic]

Thanks Dave.

I was wondering about the bass driver's reaction to the input signal, how quickly it comes to rest after a signal stops I.e. impulse response.

[davewantsmoore]

I was wondering about the bass driver's reaction to the input signal, how quickly it comes to rest after a signal stops I.e. impulse response.

 

No worries.

 

Above the impedance peak.... the mass of the cone dominates, the cone comes to rest with the signal.

Below the impedance peak....   the compliance of the suspension (surround, spider, etc.) dominates, the cone comes to rest with the signal.

 

In the region of elevated impedance.... what is happening is that the combination of all the mechanical parts of the system, mass, compliance, air in the box, etc....  reach the point where they resonate .....    all that is left to stop the cone from moving too much is the electrical damping of amplifier.     

 

As mentioned, varying the box size will affect what frequency this region occurs at (slightly) ....  but what is the "best" in this regard?    There isn't one.     If you are designing a speaker which will not have EQ on the low end, then you might want to position this region where it extends the bass    (eg.  in the first picture, where the bigger box has more bass extension).

 

 

 

To cut a very long stroy short.   For this speaker.  If I were going to cross it at 60hz or 80hz to a subwoofer, the I would probably lean towards a larger box to keep this region of resonance away from where I was using the speaker.

[davewantsmoore]

What you hear a lot about this is people talking about the box Q, for a sealed box.     I chose 0.5 and 0.7 in the example, they happen to be common ones.    People talk about the performance of the speaker in relation to the box Q (qtc).

 

.... but what in effect they are talking about is simply different frequency responses for the low end roll off.

 

One you use EQ on the low end, it no longer matters what the box Q (qtc) was.....    aside from its relation to setting the region of resonance.

[Sub Sonic]

It was the Q I had in mind when asking about damping. Not absolute levels as such, but "speed" (he he) of the driver in relation to following the input signal, I.e. impulse response around the region of resonance.

[davewantsmoore]

The Q of the box, dictates the frequency response  (you can see this in the first picture).

 

Obviously the cone never follows the input signal exactly in this region.... otherwise the SPL would be a straight line down to 0 Hz.     The driver rolls off at low frequencies instead....   so the Q simply relates to how the cone doesn't follow the input signal.     Which rolloff is "best"?    That depends on many things....  but the point is that there is no universal 'best'.

 

The big thing to realise, is that we can transform this to any other response using EQ .... and then the box Q is irrelevant.

 

 

 

Sorry for the major (minor?) threadjack.    :-S

Edited March 12, 2016 by davewantsmoore

[Sub Sonic]

Sooo... if we took two of the above drivers, placed one in a 25 litre sealed enlosure and the other in a 15 litre enclosure and EQ'd both to have identical FR curves, then they would sound and measure identical in every way (ceteris paribus etc etc.)?

[davewantsmoore]

Yes.     If they both have the same response.... then they ARE the same.

 

If the sound different, then they have a different response (for whatever reason).

[Neo__04]

Hi again,

 

Jumping back in here as I was going to do my final designs of the speaker in unibox and draw up the enclosure.

I found a discrepancy and am unsure what to do. I need a bit of advice before i start anything.

 

Through all the discussion previously in this thread, we decided on around a 27L box.

 

I modeled the speaker again and i get quite different results based on the original speaker specs and what I measure in DATS v2.

 

Now the big thing that makes the difference is the QES.

Original specs is 0.34, my measured is 0.45,

 

That basically DOUBLES the size of the enclosure.

 

So when im designing this, should i be using the advertised specs, or DATS measured specs?

 

I've attached the unibox screenshots for both lots of measurements.

 

 

Thanks again guys. Such a great help you all are!

[henry218]

I would recommend to stick with factory specs, in general SBA drivers are very consistent.

[Sub Sonic]

Have a look at the link below for some interesting information about speaker burn in/running in, it should largely explain what you are seeing:

http://www.gr-research.com/myths.htm

The manufacturer's specs will have a been taken after a period of run in, so I'd go with the published specs.

Edited March 15, 2016 by Sub Sonic

[Neo__04]

Ok great, Thanks guys, Ill work on the manufacturers specs and build a sealed box to suit.

 

That will be a bit easier as its much smaller

 

Will post updates when the box is done.

 

Thanks again

[henry218]

Ok great, Thanks guys, Ill work on the manufacturers specs and build a sealed box to suit.

 

That will be a bit easier as its much smaller

 

Will post updates when the box is done.

 

Thanks again

with sealed box, just go as big as you can

[Neo__04]

Just to throw another question into the mix.

 

If i consider going an MTM design, would the box size be double what a single driver designed up to be?

[Sub Sonic]

Yes.

You'll need to make sure that the sensitivity of the tweeter is high enough to match the twin woofers though.

Edited March 16, 2016 by Sub Sonic

[henry218]

if you are going to keep the 4ohm mid woofer, MTM in 2way will have very low impedance and will give your amp very hard time. if its 2.5 way, it won't be a problem.

[Cloth Ears]

Sooo... if we took two of the above drivers, placed one in a 25 litre sealed enlosure and the other in a 15 litre enclosure and EQ'd both to have identical FR curves, then they would sound and measure identical in every way (ceteris paribus etc etc.)?

 

 

Yes.     If they both have the same response.... then they ARE the same.

 

If the sound different, then they have a different response (for whatever reason).

Ummm. Yes, you could EQ them to get the frequency responses to theoretically sound the same, but the load seen by the amplifier at the affecting frequencies (i.e. the lower end of the frequency range) would be different. So, to get them to sound the same in the real world would require more EQ - and you would have to find an amplifier that could handle this extra load. Was it Carver who was producing subwoofers using that same theory - requiring 1200W to get the same levels/range of sound from a 12" box that you could have got with a lot less from a larger box (even with the same sized driver).

So the 25l and 15l question - given a powerful enough amplifier, they could sound the same. In the real world the 15l box would put much more load on the amplifier. Possibly resulting in a premature death...

[Cloth Ears]

Have a look at the link below for some interesting information about speaker burn in/running in, it should largely explain what you are seeing:

http://www.gr-research.com/myths.htm

The manufacturer's specs will have a been taken after a period of run in, so I'd go with the published specs.

Hi SS,

I remember reading an article on burn in specs and it turned out, for the drivers that were being tested, that the way the T/S parameters changed meant that the modelling of the enclosure actually remained unchanged. I can't remember what the things were, but I think Fs went down, Vas went up, Qts went down (actually, I think both Qes and Qms went down) and Cms went up. Plus other minor changes as a result (efficiency was up slightly also). But the model enclosure for the same response was the same size as before.

[davewantsmoore]

So, to get them to sound the same in the real world would require more EQ

 

That's correct, the assumption inherent in two speaker having the same frequency response, is that you have an amplifier which can drive each of the to the previously mentioned response.

 

 

Was it Carver who was producing subwoofers using that same theory - requiring 1200W to get the same levels/range of sound from a 12" box that you could have got with a lot less from a larger box (even with the same sized driver)

 

Many do this.  It is very common.

 

 

In the real world the 15l box would put much more load on the amplifier. Possibly resulting in a premature death...

 

This is easy to calculate (using the sort of charts back around post #50) .... so see where the drive approaches its power and excursion limits.

[Sub Sonic]

So would transient response around resonance suffer in the smaller enclosure (assuming eq'd to same FR). Still trying to get my head around this :-/