PAP Trio 15 and DIY - Discussion

Can you post pics of the 2 SEN/Neo variants?

 

Exactly where I am with the Voxatives. Very narrow vertical dispersion. The metal stands are drilled to allow placement of the tweeter on top, thought I might try that as well.

 

I've got the old "U" style metal PAP stands, and I did a quick look once and it seemed like the frame holes wouldn't line up if I put the Heil on top. But I need to look again.

 

Ok I was wrong. The old "U" stands are drilled to put the tweeter baffle in the center or on top. I changed mine to top. But I'm still waiting for a Xover from PAP before I play these again. For now, Maggies are still in the system temporarily.

 

Got in some time with the PAP 15 Voxative this weekend. Primary rig was Sonic Frontiers SFCD1 into a Hattor Audio passive pre, into Classe Omega dual monoblocks (450 watts @ 8 ohms). Also doing an amp shootout with gear that has been accumulating, so an Akitika GT-102, FirstWatt F3 , dual Odyssey Kismets, and Benchmark AHB2 got into the mix.

Reiterating what has already been said here, placement is critical. Bass disappeared when closer to 4 feet from the back wall. Imaging was best with the cones of the Voxatives pointed at the middle of your head like a laser, on both the vertical and horizontal axis. Toed out further, parts of the mids were coming from the edges of the image. With the mids laser-pointed, the image cohered but treble got sharper on lesser amps & on the ones highlighting treble (Benchmark, F3) The Omega, with its slightly rolled off top end, was perfect. Toeing the speakers out a slight bit more got rid of the sharpness with the other amps but at the cost of flattening the image.

As has been mentioned, vertical height of the mids is lower than the seated position of your ears. Experimenting, I moved the voxative to the top of the stand for a MWW configuration. The uprights are pre-drilled for this and the woofers run in parallel in this version, so what the heck, Ill try it. This resulted in the treble becoming too hot, a bit honky, and created a suck out in the mids. The crossover is clearly not designed for this: I wonder why the holes are there on the frame?

Restored to WMW configuration, I ended up with the speakers 7 feet apart on center and tilted back, and my chair 14 feet away to get the angled-up cones pointing at my ears. At that point, on the Omega they opened into a very wide sound stage with good imaging. Dynamic; textured; extended highs a bit rolled off; very good and low bass response but not bass heavy; various other superlatives. The Akitika struggled a bit to control the woofers at high volumes, and the Kismet's pushed the bass quite a bit in a way that at first sounded massively wooley compared to the Omegas. The Benchmark handled everything well but pushed the top end too hard and got too sharp at times. The F3 took a dump on everything I put through it.

So: These are sounding great in a big, long room with diffusion on the back end. I wouldn't recommend them in a small room. They are finicky for placement, wouldn't recommend them closer than 6 feet together or less than 4 feet in front of a reflective front wall with minimal absorption on it. As for the height I am going to build stands this week to get the mids to exactly 36 inches high and the baffles vertical, to see if I can get the chair a little closer. The moar power of the Omega works very well in controlling both the low and top end on these, but a bright amp is going to hurt you without modifying the crossover.

Next weekend: Testing out the trio on any tube amps I can get my hands on, and testing the new speaker stands.

 

I have them in TMWW configuration and I noticed if I stand up the bass is tighter. I am 6 feet tall so standing up my ears are around 5.5 ft. quite a bit above the top of the speakers.

with 2 woofers there might be some comb filtering going on.

for WMW configuration, the comb filtering pattern will be different. while on axis (horizontal between the 2 W, or directly in front of M in theory should be best. That's why I think some ppl tilt the whole thing upwards as M is pretty low.

 

Built and tested stands to raise the PAP 15 tweeters to ear height with the speakers completely vertical (no tilt). With the tweeters aimed directly at my head the top end became a bit piercing and the bass was severely diminished, overall a much worse sound than when they were directly on the floor. Imaging still required my chair to be 12 feet out from the speakers, so the point of raising the speakers to reduce that distance was mooted.

End result of all the experiments is the best sound I get with with the default WTW configuration, speakers 8.5 feet apart on center ( in an 18 foot wide room) with my chair 12 feet from each speaker. Vertical and horizontal placement is critical, the tweeters are extremely directional. The crossover was designed well to work with floor bounce and comb filtering considerations in this config

Given all that, I notice PAP is now offering a 'curved' version of the Trio, which is another experiment I had considered. Wonder what this does to the sound?

 

Okay, so I finally got my crossover parts for the PAP Trio 15 clones. I decided to copy the new 'Thrier' crossover, but with the old Leonidas layout. I attached images to better show what I mean. Right now I have it hooked up configured as the Leonidas XO without the optional Voxativ series coil. Note that I got sent a Jantzen MOX resistor instead of the original Mundorf one, but I doubt it makes a difference here. Otherwise the parts are the same as you get with the Voxativ AC-1.6 version of the PAP Trio 15.

Here's what they look like:


The FR at the listening position in the living room is close enough to what I get with my OBs that I'm fairly confident that I got the crossover right. They're about 5db less sensitive than my OBs and seem to have somewhat less bass extension. They're also a lot more sensitive to the vertical listening angle, in other words the vertical off-axis response is a lot more screwy than with my design.

Quick subjective impressions. Currently they're set up in the living room, so they'll most likely end up sounding better when I get them upstairs into my room. The reason I have them downstairs is so I can more easily move them outdoors (as in the pic) for measurements. They seem to confirm what I suspected about 10 pages ago: There's cancellation in the crossover region between 500Hz and 2kHz with this design. Maybe more so with the new woofers with worse FR than with the old ones.

• The imaging and the whole midrange is a bit weird. Voices seem to come from the top woofer mainly, but the higher upper midrange seems to come from the Voxativ beneath it. To me the crossover region is not coherent at all. It feels disjointed. In the beginning it made me slightly nauseous, but you get used to it.
  The imaging lacks 3D-depth, too. Hard to say why.
• The voicing is closer to neutral than I expected. It's still slightly bassy with a one-note tone to the bass, but it's not too bad. I'm sure a lot of people will be happy with the overall voicing. Little bit of a suckout at 1kHz and bump at 1.5-2kHz, otherwise similar to my OBs.
• Bass sounds soft and one-note. For the most part it's from the PAP woofers, though. I think the other 30% or so come from the woofer coil.
• The Voxativ seems to suffer a great deal from the crossover parts. It's much less dynamic sounding and generally just flatter, less involving. Sounds less alive, almost dead. Most of the plankton is gone. I'll have to try with the caps I have parallel to my woofer. Those might be higher quality than what's in series with the Voxativ here.
• I'm not sure the Ragnarok likes the lower efficiency of these speakers. About 5db less sensitive and I bet it's also 4db less efficient (in-room). From my measurements the sensitivity is in the 95db/2.83Vrms region. 87db at 1kHz, 96db at 70Hz anechoically.
• The lower treble sounds a bit rougher than what I'm used to, which I'm not sure I can attribute to the XO parts.
• Similarly, I expected more treble extension from having a resistor in series effectively boosting the upper treble. It sounds the opposite.
• Overall so far I'm not sure I prefer this design to the Ampeggio. I think the loss in plankton is too much for me. Better crossover parts could remedy that partly, which is why I want to biamp it with the parallel XO option, so I can eliminate the Voxativ crossover parts. But the crossover is still screwy. The anechoic FR and the step response both aren't pretty, I'll tell you as much. For now I'll just keep listening.

... To be continued.

 

Did a couple experiments with the PAP clones today:

TMM config
It sounded a bit less coherent in both the imaging and frequency response, but overall I ended up preferring the TMM config for a more believable sounding upper midrange and treble from having the Voxativ driver closer to an optimal height. The overall height this way is too high, though. 107cm for the TMM vs 61cm for the MTM, which is without stands for both configs. I haven't built stands, yet. I have my speakers configured for 89cm height, which is my preferred height. FR at the listening position was a bit different, but neither worse nor better. FR in anechoic conditions and the off-axis response are worse I bet. I kept the TMM config for the other two experiments.

BTW: I tried earlier to listen close to the Voxativ axis in the regular MTM config (essentially lying on the couch). Overall I preferred sitting upright, though. FR gets worse and while you listen to the Voxativs more and imaging gets more focused, the overall midrange tonality gets worse, with a noticeable suckout at ca 500Hz. Some voices trigger it and make it sound really hollow, which is honestly how I remember the Quintet 15 that I heard at High End.

Thrier XO configured as parallel instead of Series/Leonidas XO
Now this was a bit of a shocker, since I thought it would mainly change the FR for the woofer and not much else. Nope, the Voxativs significantly opened up and overall transparency and resolution was improved markedly. Even the treble extension not only got better subjectively, it actually improved in the measurements. It still sounds a bit rolled off here in the living room compared to my room, but overall the sound was more focused and dynamic. FR was much worse with a wide suckout in the midrange, though. The whole 300Hz-1kHz crossover region was low by 3.5db in the room. The bass also subjectively got a bit cleaner and more defined sounding. The series XO seems to help in making the sound more even at the cost of overall transparency.
As a sidenote: I don't really get why there's an option for the series Voxativ coil with the Leonidas XO. It doesn't sound brighter than the Voxativs with no crossover, so you'd have to sit close to on-axis for the coil to make sense. But sitting on-axis with a coil in series means you get less upper treble than by sitting off-axis. I also don't prefer the imaging on-axis.

Biamped: Class D for the woofers, Ragnarok for the Voxativs, Voxativ with no XO
Now this should obviously be the best sounding option, but it wasn't even close. I was lucky that my dad's 400W Class D amp has about 6db more gain than the Ragnarok at full level in high gain, which is just about how much the XO pads it down. We just hooked up both amps up to the Freya. This way the woofer just has the coil in series and the Voxativs just have a cable connecting them to the amp as in my speaker, no crossover parts.
The FR was essentially the same as the 2nd config, but the sound was on another level. Finally the 2kHz+ region had almost the same clarity and resolution as in my OB. The crossover parts really hurt the sound to the point where I'd consider a normal multi-way speaker over the PAPs. There's not much of a point in getting Voxativs and then hooking up cheapo resistors and caps to them. The cap is probably unnecessary anyway (I'll test that, too). The Class D amp here had a quicker, "grippier" sound to the woofers, which doesn't really happen with my OBs. I think the lower efficiency of the PAPs is partly responsible. I still don't think Class D is the best solution for the woofers, a high power Class AB amp would most likely be a better choice, integrating better with the Voxativs, but it seems better suited than the Rag in this case.
The tonality of voices was even more off than the regular MTM PAP already is, but the more dynamic, more resolving nature of this config still made it sound more believable overall. If you're fine with a weird FR and imaging where everything 700Hz and below sounds like it's coming from below you and everything 2kHz and up is coming from above, this could be for you. Overall I liked this config the best.

After that I swapped the Voxativs into my OBs again and let's just say that after initially being disappointed by the sound in the living room I was relieved when I heard them again. They still sound too lean in this room, which the PAPs in the regular MTM config with the Leonidas XO don't, but they're much more resolving, dynamic and focused. Razor sharp imaging all thet way into the bass region. My OBs make me tap my feet and stay up way too late listening to music, which the PAPs don't really do.

 

PAP OB-15SEN (Gen 2) woofer impressions:

By now I think I've made it clear that I prefer the Beta 15A to the PAP woofers. Keep in mind that I seem to be the odd one here with everyone else preferring the PAP woofers. I tried burning them in, I tried them in my frames and then tried them in the PAP frames. I tried them with my XO (which worked surprisingly well on paper) and I tried them with the Leonidas and Thrier XO. In neither configuration I liked what I heard.

• They are slightly less efficient than the Beta 15A woofers 500Hz and below. In the upper bass region (ca 60-80Hz) the efficiency seems to reach that of the Beta 15A woofers, but they don't extend as deep as the Beta 15As either. But we're not talking huge differences here, maybe 1db less efficient in the 300Hz region.
  They also have roll off from 1.5kHz on, while the Beta 15A goes to 2kHz before rolling off. Overall the Beta 15A has a better FR.
• Bass overall sounds softer and keeps getting softer the higher you crank it up. At high levels it can sound a bit loose or bloomy for a 15" driver. This is even more evident with the PAP frames. Articulation down low just doesn't seem to be there and they struggle with delineating the individual bass drum kicks in NIN - Closer for example.
• I don't hear any improvements in the speed or control of the lower midrange, rather the opposite. They sound a bit smoothed over and slower across the board to me. Double bass, bass guitars and cellos seem more smoothed over to me.
• The driver coloration is a different one. Hard to describe, but the Beta 15 sounds more like paper, the PAP woofers more like cardboard. I don't find one preferable to the other in that regard.
• Integration with my Voxativs wasn't nearly as good. The overall FR wasn't as smooth and the lower efficiency made my speakers sound slightly lean. I didn't try the Beta 15As in the PAP frames, but the PAPs aren't lean at least.
  For some reason the imaging in the lower midrange and bass also wasn't as precise as with the Beta 15As.
• The basket is still more ringy than I'd like. With some songs material you can hear the basket ringing with either of the woofers, kinda like the VPI metal armwand. It's a bit more subdued, but it's still there. Dynamat will most likely get rid of it.
• Even the resolution seemed to take a bit of a hit with the PAP woofers.
• They were supposedly already burnt in when I got them. Yet burning them in did make them somewhat more responsive, but only to a point.

The PAP clones I built sound a bit softer, murkier and less "direct" in the bass than my speakers. They also handle higher volume levels less gracefully in the bass. My frames are 3-6db more efficient in the bass, so the woofers basically need to move twice as much in the PAP frames. You effectively lose one of the woofers with the PAP frames.

Spoiler: Distortion measurements (warning, can't be unseen)

After listening to them for a bit I went through my measurements and realized that the distortion from the PAP woofers in the bass region was much higher than for my Betas. I asked @sphinxvc who managed to compare the 1st and 2nd generation of the PAP woofers. Both seem to measure with the same high amounts of distortion in the bass.

First, a very important disclaimer about these measurements:
They were done about a year apart and my UMIK-1 seems to have trouble measuring lower amounts of distortion, especially at higher levels. I suspect that the Beta 15A 2nd order distortion is lower than the graphs suggest. On an FFT the difference between the two woofers is even greater than in this graph. Finally the spikes around 90Hz with the PAP graph was some rattling that I fixed later on. I was too lazy to repeat the measurement as I'd have to swap the drivers again. Even with no baffle the difference is consistent, though.

Beta 15:

PAP woofer:

At 90db (equivalent to ca 85db in the PAP frames) we get D2 at -27db or roughly 4.5% throughout the bass range for the PAP woofers. D2 and D3 seem to be slightly lower from 100Hz to 400Hz for the PAP woofers, presumably from the shorting rings in the magnet gap.

FFT results for 36Hz:
80db: PAP OB-A15SEN: 2.2% D2, Beta 15A: too low to measure (0.1-0.2%)
95db: PAP OB-A15SEN: 8% D2, Beta 15A: 0.5% D2/D3
on average 24db more D2 with the FFT.

With distortion being this high I can only suspect that it's done intentionally.

Sidenotes: The cones for the PAP OB-A15SEN are straight, while the Beta 15A has a bit of a curvature. The cone itself is more flexible with the PAPs. The terminals aren't too accepting of bigger gauge wires, but overall I like how easy to use they are. Like with my Betas I suspect that mounting them from the back (like in the PAP frames) softens impacts.
I attached the FR graphs for the configurations I described above, but I think I will leave the more measurement-heavy discussion of the PAP design to my thread. I don't think it fits here. Like with most of my FR graphs these are supposed to be referenced to 2.83Vrms/1m. I measured at a level ca 20db lower than that.

EDIT: Posted outdoor measurements of the PAP Trio 15 clones in MTM config here: https://www.superbestaudiofriends.o...al-ob-speaker-project.4083/page-3#post-277638

 

I am new to the forum and have a pair of Trios with the Voxativ AC-X drivers. Has anyone experimented with different isolation feet in place of the stock ones?

 

Obviously inspired by this thread, my work in progress. I hope to finish the stands soon but am still considering a few ideas. I have very little technical understanding so i have just followed the great advice in this thread, and i am still tweaking the crossovers...

 

Tom from PAP just sent me this photo of a PAP quintet that was mounted with custom stands in a barber shop in Tel Aviv. I found it interesting and thought I'd pass it along. (I think that might be Ze'ev in the picture)

 

I finally got a set of official PAP Xovers for my Heil build that were formerly used by a reviewer. So I now technically have a production set of speakers after piecing everything together over the past 9 months.

One of the nice things about these Xovers is that there are jumpers to boost or cut the output of each driver. I spent the last week with the tweeter cut and the others boosted and it sounded pretty balanced and non fatiguing. I was also experimenting with the tweeter mounted on top. I found that with this configuration the speakers sounded best when I angled the stands to be as close to vertical as possible.

Yesterday I moved the tweeter back to the middle and I still got good vocal depth and solid center image, but I felt I needed to set the Xover to high tweeter output. This improved the overall balance, but I could see how some material is going to sound better at the low output setting and some at high. With the tweeter in the middle I still like the sound with the stands at minimal angle. Angled back everything seems bigger and wider, but set vertical, the vocals are more precise and you can hear the layering of the instruments better.

I'm still undecided whether I prefer the tweeter on top or middle. There seems to be advantages and disadvantages both ways.

Eventually I'll probably want a subwoofer or two. The bass is pretty tight and articulate, but you can hear on bass guitar how deep notes have different output when familiarity with the recording tells me they should be the same. It's not as bad as one-note bass, but I want more consistent output in the 20s and 30s. My room sucks too. I'm building a house in 2020 with a dedicated sound room so I'll hold off on subs until then.

 

I found them best with the tweeters pointed as closely at my head as possible, wondering if your placement above is replicating that?

 

I forgot to mention that I originally had them toed in significantly, but then tried them parallel to the wall. I ended up with a slight toe in which seemed to give the best balance of image focus and depth of stage.

I've also been playing around with placement away from the wall. I started with them about 5 ft away from the back wall, but the bass output increased and sounded smoother about 3.5 feet from back wall. I have a rack in the middle. When my sound room is built I'll have the rack on the side wall and nothing between the speakers. Right now I can only get them about 6 feet apart (from inner edges) but I have to sit about 12-15 feet back. This is a problem and theres not much I can do about it in the current room.

 

Update: My room sucks. Looking forward to a new sound room in a new home but that's going to be about a year. I haven't RTA'd the room or anything but I can hear peaks and dips in the bass with music that has descending deep bass notes. I have to sit too far back from optimal due to the room configuration. Mids and highs are revealing but still engaging and not overly analytical. I was starting to wonder how the Heil tweeter would integrate without sounding out of place on these speakers, but the factory Xover was done well enough that the transition between drivers is smooth.

One thing I noticed on the production PAP Trio 15 is that the front baffles are mounted to the frame with a 1/4 thick rubber washer between the two. At highish volume you can feel the front baffles move a little since they are floating on these washers. This isn't an issue with DIY'd versions since the baffles usually get screwed directly into the frame. But I might try replacing these rubber washers with some dynamat, felt, or some high duro sorbothane I have laying around to see if tit makes any difference.

 

Strips of dynamat on the uprights to protect the wood, skip the washers all together?