Exploring Purifi Woofer Speaker Builds

That's a dome with a WG inside a torus, right?

1738906891840.png


This looks like a standard coax driver with a centrally mounted VC for the MF, no outer surround.

That would be tough to market-

"look darling- the speakers are looking at us!"

It would take a heck of an engineering feat to squeeze an convex HF unit within a convex MF unit... but was high neodymium-content magnets even available when ATC made the SM75-150S nearly 50 year ago? If not, it's no had to hang 20lbs (9kg) of motor behind a tiny 3" dome.

Today we can talk about neo and iron nitride...
 
Last edited:
How would the PTT6.5X04-NAA-08 perform in a 17 liters (0.6ft³) sealed enclosure? I don't have time for simulations or even building new enclosures at the moment, but I'm really itching to buy some Purifis and temporarily install them in some old boxes.
PURIFI PTT6.5W04-01 Six-pack.jpg

I also lack time so here is a sim simply opening CAD and setting box to closed with 17 liters. Looks like you get F3 at 78Hz F6 at 50 Hz Qtc 0.48. My guess is with room gain (and if available a bit of bass EQ +2dB below 50 Hz) this would be quite enjoyable for "really itching to buy some Purifis and temporarily install them in some old boxes."
 
That's a dome with a WG inside a torus, right?

View attachment 1418216

This looks like a standard coax driver with a centrally mounted VC for the MF, no outer surround.

IRIS is the picture your words brought to my mind. I’ve also played with those - they’re fun little drivers!

You could also use examples from, e.g. Cabasse or Thiel that have suspended rigid ring radiators with concentric tweeters. That’s my point - unless the “midrange dome” has a hole in it I don’t see it. Though I guess what you’re saying is put a tweeter on the “coherer” (phase plug) of the midrange?
 
Hi,

conduct a fast thought experiment to come up that these things are never going to be ideal, as long as they are physical objects with static proportions.

-Sound wavelength changes greatly over the audible bandwidth.
-Physical objects affect acoustic sound per their size and shape relation to wavelength.
-thus covering whole bandwidth from single point in space would need a physical construct that is both big and small at the same time to be "ideal" for whole bandwidth, which is physically impossible.
=> split the bandwidth to two or more transducer covering different bandwidth are different size and shape to be effective and "ideal" over some smaller bandwidth, concentric or not.

Now, two physical objects cannot occupy same physical location at the same time, so they would have to mold together somehow, big and small, like a concentric driver. But, both transducers are immediate acoustic environment for the other, optimized either for their own bandwidth or for the other, but not both at the same time or you could use just one transducer with reduced bandwidth.

That said all we can do is whine about it because it's physics it doesn't align to our wants and wishes, but we must adapt to physics. Some solutions are working better than others, better implemented compromises with various trade-offs are available, so just pick one suitable.

It's fun to wonder about this stuff so, keep on 🙂
 
  • Like
Reactions: Groundloops
How would the PTT6.5X04-NAA-08 perform in a 17 liters (0.6ft³) sealed enclosure? I don't have time for simulations or even building new enclosures at the moment, but I'm really itching to buy some Purifis and temporarily install them in some old boxes.
Here's 12L sealed near field, then far field 1m merged to nearfield with baffle step correction, 1/12 smth

ptt6.5x04naa hor 0 NF 17L.jpg


ptt6.5x04naa hor 0 NF w BSC mrg to 1m.jpg
 
Thanks! Although my problem is I'm not experienced enough yet to know how to interpret this stuff. <1mm excursion. No idea if that's good or bad, given the THD rises fairly rapidly below 100hz with this driver.
 

Attachments

  • Screenshot 2025-03-06 194218.png
    Screenshot 2025-03-06 194218.png
    194.7 KB · Views: 33
The thd at low frequencies is largely dependent on excursion. So if you have thd measurements they should also say how they were measured and from that you can guestimate what happens if excursion is of say 1mm .
Assume measurement at 2,83 Vrms and full bandwidth withbdriver in closed box or infinite baffle, you can sim that. It gives you an indication of the excursion in that test.
As it is a midrange it has a short voicecoil, thus limited excursion.
Sort of combining these gives an good indication.
 
  • Like
Reactions: maFrodite
Right, got it. It's from here. 16v, 80hz 2nd order, no mention of filter type. Regardless both BW and LR filters put the excursion just above and just below the 3mm maximum, respectively. Ok I'm going to call this one solved. Thanks again for your help!

What a fantastic little driver...