Purifi PTT6.5W04-01A 6.5" midwoofer

[dwkdnvr]

Those looking for an off the shelf option may want to look at the wavecor tweeter with a waveguide faceplate. Very robust and good response. Would be my choice vs the DXT or an AMT. Also the peerless corundum 32mm would be an interesting choice if you don't want a waveguide.

Yeah, I'm finding the pull of these drivers hard to resist, but the tweeter choice is giving me pause. Having tried my Kef R3s in the living room, I know that I want broad dispersion for this application, so something 'obvious' (and expensive) like the Satori BE w/waveguide as used in the Helios is too narrow. I have the DA25Tx on hand, but with the PTT6.5 you probably ideally want a 1.5kHz xover without a waveguide and that might be a bit much for it. The DA32Tx should handle it fine, though.
The DXT or Wavecor are interesting ideas too - basically recreating the ER18DXT with an upgraded woofer. A mild waveguide on the DA25Tx might also work, but as the HiFiCompass article showed that can involve a lot of trial and error.

 

[617]

Yeah, I'm finding the pull of these drivers hard to resist, but the tweeter choice is giving me pause. Having tried my Kef R3s in the living room, I know that I want broad dispersion for this application, so something 'obvious' (and expensive) like the Satori BE w/waveguide as used in the Helios is too narrow. I have the DA25Tx on hand, but with the PTT6.5 you probably ideally want a 1.5kHz xover without a waveguide and that might be a bit much for it. The DA32Tx should handle it fine, though.
The DXT or Wavecor are interesting ideas too - basically recreating the ER18DXT with an upgraded woofer. A mild waveguide on the DA25Tx might also work, but as the HiFiCompass article showed that can involve a lot of trial and error.

To put it in perspective, I used an off the shelf waveguide and fitted 3 or 4 tweeters to it, and only one really worked well. I would leave waveguide design up the professionals.

Also regarding dispersion, don't worry about the top octave dispersion. A 2 way with a 6" midwoofer is not what I would call an intrinsically wide dispersion design - a 4" or 3" driver doing midrange duties would fulfull that definition. I would either utilize a small midrange (such as a morel dome) or use a big tweeter in a waveguide. Wide dispersion in the top octave is less important than wide dispersion in the upper midrange.

A 3 way with a 6" woofer is normally not justifiable, but the bass performance of this woofer makes it worthwhile.

 

[Peas]

Those looking for an off the shelf option may want to look at the wavecor tweeter with a waveguide faceplate. Very robust and good response. Would be my choice vs the DXT or an AMT. Also the peerless corundum 32mm would be an interesting choice if you don't want a waveguide.

Yes, actually I am working on something with a Wavecor tweeter, but in combination with another midwoofer. Nothing is fixed for life, so maybe I will also try the Purifi later on.

 

[617]

Yes, actually I am working on something with a Wavecor tweeter, but in combination with another midwoofer. Nothing is fixed for life, so maybe I will also try the Purifi later on.

What woofer is that? And how are you liking the wavecor?

 

[Peas]

That one:
https://www.oaudio.de/lautsprecher-...omnes-audio-exclusive-6-tiefmitteltoener.html

Hexacone, long throw, rigid construction.

The Wavecor is great, but I have a feeling that it reacts sensitively on the baffle design. It might be crossed as low as 1 kHz, but then directivity is somewhat compromised. 2 kHz is better, and then it still has advantages compared to other tweeters.

I have heard other DIY speakers with Wavecor tweeters and also woofers, and they all sound great.

 

[617]

That one:
https://www.oaudio.de/lautsprecher-...omnes-audio-exclusive-6-tiefmitteltoener.html

Hexacone, long throw, rigid construction.

The Wavecor is great, but I have a feeling that it reacts sensitively on the baffle design. It might be crossed as low as 1 kHz, but then directivity is somewhat compromised. 2 kHz is better, and then it still has advantages compared to other tweeters.

I have heard other DIY speakers with Wavecor tweeters and also woofers, and they all sound great.

Nice woofer, wish I could get those in the USA. I bought a europe-only visaton waveguide and it was a nightmare getting it.

 

[dwkdnvr]

To put it in perspective, I used an off the shelf waveguide and fitted 3 or 4 tweeters to it, and only one really worked well. I would leave waveguide design up the professionals.

Also regarding dispersion, don't worry about the top octave dispersion. A 2 way with a 6" midwoofer is not what I would call an intrinsically wide dispersion design - a 4" or 3" driver doing midrange duties would fulfull that definition. I would either utilize a small midrange (such as a morel dome) or use a big tweeter in a waveguide. Wide dispersion in the top octave is less important than wide dispersion in the upper midrange.

A 3 way with a 6" woofer is normally not justifiable, but the bass performance of this woofer makes it worthwhile.

Well, my point of reference is the current speakers in the space - a classic Scan Speak lineup of 8545 / 9700 (I think - might be the 9500). So yes - not 'wide dispersion' like the BMR would be. These work well in the room, but swapping over to the R3s results in a significant loss of ambient energy. Sweet spot listening is great with the R3s, but since this is the living room / kitchen open area, most of our listening is casual around-the-space and I find the Scans much better than the Kefs for this. So, this would actually match a Purifi / DA32Tx combo pretty closely.

I think I have bass taken care of, and honestly I'm not sure I can justify the cost of the PTT6.5 for bass duties - I'm after it's low IMD through the midrange. Even there though, as discussed in the other thread the SB15NBAC is an awfully good looking driver if you can supplement the low end. Bass will probably be RSS265 + dual RSS265PR - flat to the mid-20's in 1 cu ft. This is viable since I'll cross actively with the SHD - otherwise efficiency would kill that idea.

As for waveguides, I'll probably do some experimentation, but if I go SB15NBAC + DA25Tx it probably isn't essential. ATH4 + cnc router means that I can iterate fairly quickly at least to sanity check. (in theory - having problems with my spindle at the moment)

 

[Peas]

Nice woofer, wish I could get those in the USA. I bought a europe-only visaton waveguide and it was a nightmare getting it.

They ship to the USA.

 

[Peas]

We were off-topic and talked about the Omnes Audio Exclusive. Sorry about that.

 

[GelbeMusik]

Yeah, I'm finding the pull of these drivers hard to resist, but ... a lot of trial and error.

Does anybody have the Purify driver at hand? My experience with the somehow comparable Adire Audio Extremis 6.8:

The excursion provided with the XBL^2 can hardly be used fully.

The port in a small reflexed box would need to be wide, as to prevent from chuffing or compression. The increased length provides geometrical problems. It quickly becomes huge.

A passive radiator needs large surface area, and subsequently high mass, so two have to be used at opposing walls to prevent from shaking. The surface would leak midrange content.

An effective compromise has to be found, which always limits the excursion. In this case not by the motor, but by a bearable box volume for instance.

The surround deforms with high excursion. It becomes the main root cause for intermodulation. A too high excursion would induce Doppler shift. I suspect, that the cone movement copies into the sound field, as, with its movement it takes its directivity pattern with it.

All these are unusual considerations. The motor is really competent, so that other, unexpected problems show up. On the plus side the intermodulation at quite elevated excursion, say a massive +/- 6mm, behaves very kind. It is kept narrow, doesn’t spread out like with the SB Acoustics 17???NBC, which is claimed to do +/-5mm. The latter shows a hard limit, nasty noises, somehow piercing, shrill, while the AA 6.8 only slowly drowns in a humming background. To have more than 10% IM, which is kind of an unreasonable standard in the industry, would need to drive it to +/- 10mm.

Point is: the IM is there, but it is by far less distracting compared to even most modern drivers. And there is always some reserve.

The compromise I decided for is to use two on each side, which justifies a larger box of around 40 liters of 20x30x90 centimeters^3 external dimensions. In this configuration they easily replace, as far as my measurements hold, a single 12” JBL2206. Taken on its own, that is impressive. The claimed +/-14mm of Xmax would promise even more. You lose 3dB, but at that levels, at home, I don’t care further.

 

[boXem]

Does anybody have the Purify driver at hand? My experience with the somehow comparable Adire Audio Extremis 6.8:

The excursion provided with the XBL^2 can hardly be used fully.

The port in a small reflexed box would need to be wide, as to prevent from chuffing or compression. The increased length provides geometrical problems. It quickly becomes huge.

A passive radiator needs large surface area, and subsequently high mass, so two have to be used at opposing walls to prevent from shaking. The surface would leak midrange content.

An effective compromise has to be found, which always limits the excursion. In this case not by the motor, but by a bearable box volume for instance.

The surround deforms with high excursion. It becomes the main root cause for intermodulation. A too high excursion would induce Doppler shift. I suspect, that the cone movement copies into the sound field, as, with its movement it takes its directivity pattern with it.

All these are unusual considerations. The motor is really competent, so that other, unexpected problems show up. On the plus side the intermodulation at quite elevated excursion, say a massive +/- 6mm, behaves very kind. It is kept narrow, doesn’t spread out like with the SB Acoustics 17???NBC, which is claimed to do +/-5mm. The latter shows a hard limit, nasty noises, somehow piercing, shrill, while the AA 6.8 only slowly drowns in a humming background. To have more than 10% IM, which is kind of an unreasonable standard in the industry, would need to drive it to +/- 10mm.

Point is: the IM is there, but it is by far less distracting compared to even most modern drivers. And there is always some reserve.

The compromise I decided for is to use two on each side, which justifies a larger box of around 40 liters of 20x30x90 centimeters^3 external dimensions. In this configuration they easily replace, as far as my measurements hold, a single 12” JBL2206. Taken on its own, that is impressive. The claimed +/-14mm of Xmax would promise even more. You lose 3dB, but at that levels, at home, I don’t care further.

There are a few articles on the Purifi website addressing your concerns (surround deformation, Doppler effect). For sure they try to sell their stuff, it would be interesting to have the results from your own testing once you get these in hand.

OT and only curiosity (no criticism at all): while reading you, i have the feeling that your mother tongue is German. Am I right? Are you using an auto translator?

 

[GelbeMusik]

There are a few articles on the Purifi website addressing your concerns (surround deformation, Doppler effect). For sure they try to sell their stuff, it would be interesting to have the results from your own testing once you get these in hand.

OT and only curiosity (no criticism at all): while reading you, i have the feeling that your mother tongue is German. Am I right? Are you using an auto translator?

My mothers tongue is Czech. I trained English with Germans, though. My daily praxis is with Indian people now.

I don't plan to purchase the Purify drivers. I was wondering what the discussion is about. They don't address Doppler, the surround is mentioned, but the claims are not supported by any numbers, and not the least, the crackling from magnet hysteresis, I should believe that?

 

[boXem]

My mothers tongue is Czech. I trained English with Germans, though. My daily praxis is with Indian people now.

I don't plan to purchase the Purify drivers. I was wondering what the discussion is about. They don't address Doppler, the surround is mentioned, but the claims are not supported by any numbers, and not the least, the crackling from magnet hysteresis, I should believe that?

I am not asking you to believe anything
Both Kms linearity and Doppler are addressed in an experimental way in some of their blog articles:
Doppler: https://purifi-audio.com/2019/12/07/doppler-distortion-vs-imd/
Kms linearity: https://purifi-audio.com/2019/12/12...so-whats-the-point-of-low-distortion-drivers/
I agree with you that effective IMD numbers seem to be missing

 

[dwkdnvr]

I am not asking you to believe anything
Both Kms linearity and Doppler are addressed in an experimental way in some of their blog articles:
Doppler: https://purifi-audio.com/2019/12/07/doppler-distortion-vs-imd/
Kms linearity: https://purifi-audio.com/2019/12/12...so-whats-the-point-of-low-distortion-drivers/
I agree with you that effective IMD numbers seem to be missing

It would be nice to see a more comprehensive IMD test, but the IMD tests reported by HiFiCompass were fairly impressive - the PTT6.5 is clearly better than other 'top shelf' drivers. But, this does strike straight at the heart of the question - if you don't actually NEED a 2-way design, is the cost of the Purifi justified? The tests seem to show that even if you ignore the bass capability, the distortion performance through the midrange is pretty much as good as anything out there, and the freedom from the type of cone-edge resonance you see on the Satori drivers is impressive.
But, is it better than say a Satori WO24P + SB15NBAC - each of which have impressively low distortion through their respective passbands, and together still cost significantly less than a single PTT6.5.

Furthermore, since this is ASR one has to raise the question as to how much any of these factors is audible, let alone significant. The Toole work wasn't nearly as deep as I'd like to see, but their conclusion was clearly that if distortion is audible in well-designed speakers, it's certainly far less significant than on and off-axis frequency response behavior.

 

[dwkdnvr]

Does anybody have the Purify driver at hand? My experience with the somehow comparable Adire Audio Extremis 6.8:

The excursion provided with the XBL^2 can hardly be used fully.

The port in a small reflexed box would need to be wide, as to prevent from chuffing or compression. The increased length provides geometrical problems. It quickly becomes huge.

A passive radiator needs large surface area, and subsequently high mass, so two have to be used at opposing walls to prevent from shaking. The surface would leak midrange content.

An effective compromise has to be found, which always limits the excursion. In this case not by the motor, but by a bearable box volume for instance.

.

Interesting observations. I've spent a lot of time looking at the Anarchy drivers, which are the successor to the Extremis 6.8. I found that two Anarchy drivers per side in 1.5 cu ft with two Dayton RSS265PR passive radiators is an almost ideal bass module. You're correct that the weight of the PRs is a problem - 375g added mass per PR, so an opposed arrangement would be ideal, but not necessarily the easiest to execute in all cabinet designs

Note that the Anarchy was Klippel tested to have an effective xmax of only about 8.5mm - suspension limited, if I remember correctly.


I think the biggest differences between the Anarchy and the Purifi would be
a) Motor is even more linear, and has been designed specifically to minimize IMD. (the Anarchy is not a particularly low distortion driver from an HD perspective, but the XBL^2 does seem to help IMD)
b) cone is much better - no edge resonance and no hard-cone breakup, so usable bandwidth is extended much higher
c) HD is much lower, to the extent that it matters.

IMHO the Anarchy is a woofer ideally used in a 3-way, but can be used in a 2-way with care. In that role (3-way), it's the most impressive small driver I've seen. The Purifi is clearly designed to be used in high-output 2-way designs, and if used in a 3-way many of it's advantages become a bit less compelling.

 

[boXem]

Furthermore, since this is ASR one has to raise the question as to how much any of these factors is audible, let alone significant. The Toole work wasn't nearly as deep as I'd like to see, but their conclusion was clearly that if distortion is audible in well-designed speakers, it's certainly far less significant than on and off-axis frequency response behavior.

Concerning the audibility of the various types of distortion, the experiments in the two blog posts I mentioned earlier are quite interesting, even if obviously exaggerated.

 

[GelbeMusik]

Concerning the audibility of the various types of distortion, the experiments in the two blog posts I mentioned earlier are quite interesting, even if obviously exaggerated.

It's funny. First I auditioned the examples regarding the Doppler on headphones. Big difference between AM and FM, true. Then I used the speakers. So and so. Then I added the test signal it to some music using a mixer. Level was adjusted to not overload the mix. Result: barely a difference. How come?

 

[boXem]

It's funny. First I auditioned the examples regarding the Doppler on headphones. Big difference between AM and IM, true. Then I used the speakers. So and so. Then I added the test signal it to some music using a mixer. Level was adjusted to not overload the mix. Result: barely a difference. How come?

You made me want to try with speakers.
Kids are sleeping, so I did it at very low volume with my ear at 50 cm from the speaker and could clearly hear the difference, even if not so obvious compared to headphones. I wonder if low volume was not the key since the speakers themselves had low IMD at this level, thus no masking phenomenom. Before anybody asks, it was not a dbt, but I did the experiment and then checked which one was supposed to sound "wrong".
When mixing the distorted tone with music, I suppose that the -20 dB harmonics of the tone become barely audible, everything below being hidden by the music itself?

 

[GelbeMusik]

When mixing the distorted tone with music, I suppose that the -20 dB harmonics of the tone become barely audible, everything below being hidden by the music itself?

Seems to be the opposite. With speakers the quite complex sound field in-room is affected. Different frequency, different amplitude. All this super fine interference of everything with each other, frequency dependent. With other sounds mixed in, which have a meaning, like speech, the relations change with FM--I don't know!

It is all speculative and for sure no final verdict.

 

[Selah Audio]

Are there any Klippel tests for this driver?