Pure Audio Project open baffle speakers

I have a set of PAP Trio with Horn 1 midrange. They replaced a set of B&W 803Ds. Here are my impressions:

I am using the Leonides XO. Have very mixed feelings about this XO. It is a first order XO and does little to roll off either the bass drivers or the Horn 1 midrange. The Horn 1 is good down to 300 Hz or so, and the bass drivers go all the way up to 1500 to 2K Hz (where they have a nasty peak). So there is a large area in the lower midrange where both the Horn 1 and the bass drivers are active. I added an RC network to the bass drivers to try to attenuate the upper midrange somewhat, and it works after a fashion but is not perfect. The speaker sounds great on Jazz but is trying for compressed rock. Am in the process of looking at an active XO system so I can better roll off the horn and the bass drivers at about 800 Hz.

Compared to the B&Ws, the imaging and sound stage depth is vastly better. Even with the frequency response issues, these have much better upper midrange (not as harsh and they don't "beam" like the 803s did). Bass is quite good compared to the B&Ws, which surprised me.

I also have the Voxative drivers which I have yet to try, and may swap these for the horn. The horns use Beyma Spanish made drivers and are incredibly efficient, at 108 dB/Watt. May also take a look at the ESS drivers as these are quite inexpensive at the moment.

I do have to comment on the CS from this company. Ze'ev is quite responsive until he has your money, thereafter not so much. His business model is to drop ship the components from all over the world, so the user has to assemble and test the speaker. We had a number of QC problems (the initial shipment of woofers were damaged by the manufacturer in packaging) and many of the parts were quite late in delivery.

So for me these are a work in progress, which I don't mind so much as I am a tweaker. I think they have real potential if the frequency response aberrations can be brought under control.  

douglas

Different resistors and capacitors will not change the basic design, which is that of a first order XO. This means that both the horn and bass driver will be operating (and fighting each other) in the lower mid range below 1K Hz. The Eminence OB-A15neo drivers frequency response continuously rises to its peak at 1200 hz, which is very noticeable particularly because the XO does not roll this driver off well before this. When summed with the Beyma compression horn driver (good for below 300 Hz), the lower mid range emphasis is unmistakable. The best electronics in the world will not change this.

I  am using Parasound JC1 amps, not exactly what I would consider lower end electronics. 

douglas
I considered the C1. I do believe that is the one designed by Nelson Pass for PAP. That gets a lot closer to what I want, and does roll off the drivers properly as opposed to the Leonides passive XO. The C1 and the Leonides are night and day different in how they XO the drivers.

Only issue I have with the C1 is it is not quite flexible enough for me, in that is a single ended design and I used balanced connections and would require buying two C1s. Pass also used 2nd order Linkwitz Riley slopes, which create 180 degree phase shift at the XO point, requiring one of the drivers to be wired out of phase with the other. I would prefer 4rth order LR slopes instead. That said, the C1 is definitely a step in the right direction.

PAP also has a passive 2nd order XO that was used prior to the Leonides version (I think). Have not used that XO, as I was convinced to order the Leonides instead.

 

douglas:

Well I decided to build my own active XO, based on a class A discrete op amp design (a modified Spectral design using a Pass Labs type JFet output stage). Took quite a while to build and test due to the number of amps involved. Also had to procure two more JC-1s. I substituted the Voxative midrange driver for the Horn, and cross the bass drivers over at 400 Hz with 4rth order Linkwitz Riley slopes.

Resulting sound is amazing. Best my system has ever sounded. Considerably better than the Horn 1 with the passive Leonides XO. I have yet to try the horn with the electronic XO (400 Hz is probably a bit too low and would have to alter my XO point to 800 Hz), but the Voxative sounds so good I may not bother with the horn. Gone is all the stridency in the vocals and the symbols and bells are wonderful. 

douglas:

As an addendum, since you mentioned VH Audio. I am using their teflon caps in the critical low pass and high pass filters.

Doug:

I know its been quite a while since your last post to this string but here is my update to your suggestions. I tried the horn again and lowered the XO point to 400 Hz. Still was not that satisfied as there appears to be something missing at the very top end. Sold them (the PAP horns) and the new buyer confirmed the same observation. Many have added a "super tweeter" to the horn to fix this, but since that requires yet another XO point, I decided not to pursue this complexity as my current active XO is two way only. My in room pink noise measurements confirmed a drop off in frequency for the horns at about 10-11K Hz so this explains a lot. The Voxativ AC 1.5 reaches up to 15-16KHz in the same setup. The Voxativ is nice and flat with the main speaker chassis pointed straight ahead (as the off axis response is flat). They have a slight peak at 10KHz on axis, so I listen to them straight on (an not towed in as with the horns).
One additional advantage I have found with my active XO. I can tune the upper frequency response (to the Voxativ) by 0.5 dB increments and this really helps in tweaking certain recordings. I also built in an old school Yamaha (80's era) high frequency "tilt" control which tilts the HF response by variable degrees ( 1 through about 6 dB at 20KHz; hinged at 800 Hz). Don't really care for this much as the fixed shelving response works better for me.

One additional addendum. Looking to perhaps upgrade to the Voxativ PiFe. This is the driver similar to the AC 1.5, but with wood cones and phase plug. Cannot find any comparisons between these two drivers, except from Ines at Voxativ, who claims the wood makes the driver smoother and more detailed. Comes at quite a cost as these are about $5K a pair. 
It would be a plug and play replacement for the AC 1.5 as far as the XO goes, which makes life simple. Another suggested tweak is to replace the Al metal phase plugs in the 1.5 with a wood version from the upper end drivers, but I'll be d'mned if I cannot get a grip on those plugs to remove them.