Porting the midrange in a 3-way system?

[kfr01]

I've noticed that some of the Wilson Audio 3-way units also port the midrange cabinet. Why would someone do this? What effect does this have on transient response and/or overall sound of the unit?

Curiousity has me asking... thanks!

[RJ]

Not familiar with the Wilson speaker.
Let's say you have the woofer in a 4th order single bandpass setup and the upper F3 is 95 hz.
You'd than find a 4 inch midbass that has a lower F3 of 95 hz in a ported enclosure.
No huge inductor involved.

[paulspencer]

I have noticed this before as well. My gut reaction was "but you can't stuff it as much and hence you don't get to kill reflections as much." At first I thought it was a bad idea for this reason. However, you must also consider that venting the design allows it to be more linear over a wider bandwidth. It may give you an extended low end, which you may not use in a 3 way as such, but it is useful to have flat well tamed reponse below the crossover point in order for the crossover to funtion as it should. I have heard that a LR4 should have 2 octaves flat either side of the crossover region. Suppose you have a midrange driver that rolls off below 200 Hz and you wish to set the xo point at 300 Hz. Ideally you would have flat response down to 75 Hz. You may not quite get there, but you will be closer than if you had used a sealed chamber. You might tune it at 100 Hz and be flat to that point.

I believe an alternative is to seal it and combine the elctrical filters with the acoustic rolloff.

[kfr01]

Paul, excellent response. That seems to make perfect sense.

Anyway, I'm trying to replicate a Wilson Watt/Puppy type design, you mention that the alternative is to seal and use electrical filters + acoustic rolloff. Which do you think is better?

Do you think it makes sense to try a ported midrange, or do the potential negatives / risks outweigh the benefits?

[paulspencer]

I don't know which is better but I prefer the idea of crossing well above rolloff of the driver. If you combine acoustic and eletric response then you have to measure acoustic response first to be sure. I don't yet have the means to measure (accurately). I think you will find those who prefer one and those who prefer the other. I doubt there is a definitive answer, otherwise everyone would do it the one way (or at least those who knew what they were doing)

[wintermute]

Originally posted by kfr01
Paul, excellent response. That seems to make perfect sense.

Anyway, I'm trying to replicate a Wilson Watt/Puppy type design, you mention that the alternative is to seal and use electrical filters + acoustic rolloff. Which do you think is better?

Do you think it makes sense to try a ported midrange, or do the potential negatives / risks outweigh the benefits?

You might find if you start modeling that port length is an issue, often the required length is very difficult to obtain in small volume enclosures, it will depend on the driver.

Also something to be carefull about is port resonances. You may get some nasties in places you don't want them. I think from memory you can turn on and off port resonance simulation in unibox to see it's effect. My general approach is to try and size the port so that the port resonance is outside the range the speaker will be operating, however I think this could be very difficult for a midrange.

Tony.

[kfr01]

Excellent issue-spotting post. That is probably not something I would have considered before it was too late!

[paulspencer]

Tony, you are right. This makes it a bit tricky. An example may help.

I had a play around with this idea for the Vifa P13 and it's a juggling match. The box needs to be bigger, the vent small and flared at both ends in order to get the resonance out of the passband, say as high as 3.5k. I can see why the mid chamber is usually sealed!

I believe Sonus Faber also do this with some of their speakers. If they are Wilson Audio do it, I'm sure it can be done. Not saying it will be easy!

[Nat Eddy]

Two thoughts: Is the Wilson ported as in bass reflex, or its it more a transmission line or even simply an unenclosed baffle? If so, the point is to minimise resonances.
But iif the Wilson is ported in the bass reflex way, the idea might be to protect the driver against being overdriven below its resonance point.

[rick57]

An alternative to achieve lower end extension, esp. where port length is an issue, depending on the desired Hz and volume - may be heavy stuffing.

Unibox can model this, though without defining how heavy is “heavy fill”.
Does anyone know the density that Unibox assumes?