Celuaris SPK5 Purifi Review (speaker)

[ROOSKIE]

The port resonance can act as a cancelation to the rear wave as it is rear ported, thus causing a cardioid high Q effect at that frequency

Yup that big 1st resonance wave is trying wrap itself around that box from the rear.
A front port can also cause cancelations. All port placements can cause either a peak, null, both or ideally neither.
This port has such a low untamed1st resonance that it is hitting the system hard.
If this must have a long port to meet response goals, then Purifi can try moving the inlet around in box and find a spot where the resonace is in more of a null state. They can also try placing the outlet on the bottom, top, side(s), front. Any of these option can affect the response in a way meaningful to a SOTA type project.

Again this is a reason to really invest in the port design, right now it just seems tuned for the box tuning without addressing the additional issues a state of the art design would ideally address. It is more like a port I designed and built vs a very experienced engineer. Not the end of the world but under the microscope it is not possitive.

However, those who value optimal directivity should wait until angular frequency response measurements are available for the kit before buying.

Due to the fact that the crossover frequency is at 1.3 kHz, there should still be significant influences of edge diffraction on the tweeter despite the waveguide.

It would always be great to get a Kipple test for a Helios DIY, but I fear that design is not in the Q.
The designer is well known and respected and known to be extremely thorough.
That said a test is test and that is what counts here. He did public some basic off axis measurements though I do not know how much smoothing was applied here. o,22,45 degrees. What is shown here is exceptional - again not near complete as a full Polar or Kipple test.

With the low X-Over frequency are you saying that the smallish/medium sized waveguide on a semi-wide baffle can no longer responsibly direct the sound and the waveguide has given up it's primary influence to front baffle?

I ask why the 1300hrz crossover is suspicious for the waveguided tweet in the Helios example as I am actually designing a DIY active using a similar design of my own. In my case I would be willing to place the tweeter in a separate box sitting on top on the woofer box. This is both for the ability to mix and match drivers easily and to optimize the case size for each driver.

Not a pot shot at all. I meant it. The MODDERS and DIY crowd like to build themselves and they enjoy that and buy tools and all sorts of stuff for the tasks. I've been a DIY MODDER in the past and it's fun.

Yes, at this point it is hard to say how much more I have spent on audio due to DIY.
Sometimes I wish a could just buy a set of the JBL 708p's a couple subs and be done.... nah! Really this is the only fun thing I do that regularly cost me $, most of my other loves are very inexpensive or free activities - which is good as I am deff not flush in $USD or BTC or Gamestop or whatever. (I am rich in enjoyment )

 

[ctrl]

OFF-TOPIC

What is shown here is exceptional ....

This can be an excellent sounding speaker. The directivity does not have to be perfect for this.
It's just that this speaker will most likely not show exceptional directivity.

It looks like the measurements are unfortunately ungated, with all the room reflections. We also don't know if the measurement distance was large enough to fully account for the baffle influence.

Nevertheless, we can see that in the range around 2kHz the edge diffraction has the strongest effect and "left and right" of it the speaker radiates less broadly.

It can be seen that the waveguide dominates the radiation up to about 5 kHz and below this frequency the influence of the baffle increases more and more.

With the low X-Over frequency are you saying that the smallish/medium sized waveguide on a semi-wide baffle can no longer responsibly direct the sound and the waveguide has given up it's primary influence to front baffle?

Yep. Again, don't get me wrong, you don't need perfect directivity for a good sounding speaker and a speaker with perfect directivity can sound terrible with a mediocre crossover tuning.

I ask why the 1300hrz crossover is suspicious for the waveguided tweet in the Helios example as I am actually designing a DIY active using a similar design of my own.

If a similar situation arises there, simply simulate a higher crossover frequency, if the woofer allows it. If the horizontal directivity improves, listen to the crossover tuning - perhaps it's tonally advantageous.
The trade is then, better horizontal directivity, against worse vertical - this is a matter of consideration and personal design philosophy.

Especially if you develop an active loudspeaker, a LR crossover of second order can also sound very good (then of course with crossover >=2kHz)
.

 

[PeteL]

This get me curious about Kii, Putzeys's active speakers company. The hype is quite big and not only from some semi rigourous reviewers but it seams to legitimately pop up on the short lists for serious mastering rooms. They do rely heavily on 6.5 inch woofers, in interesting configurations. they do look somehow similar but still quite not the same, but seeing the lukewarm performance, especially in the bass, wondering if he does better stuff, just implemented better, or rely on an other manufacturers, it's not clear, but I would think that his thing is showcasing his own drivers. So basically, i have to say i'm a bit surprised that we don't get much of a wow here, at least for the mid woofer, but never heard neither (purifi or Kii) myself I have to say.

 

[Massimo]

This get me curious about Kii, Putzeys's active speakers company. The hype is quite big and not only from some semi rigourous reviewers but it seams to legitimately pop up on the short lists for serious mastering rooms. They do rely heavily on 6.5 inch woofers, in interesting configurations. they do look somehow similar but still quite not the same, but seeing the lukewarm performance, especially in the bass, wondering if he does better stuff, just implemented better, or rely on an other manufacturers, it's not clear, but I would think that his thing is showcasing his own drivers. So basically, i have to say i'm a bit surprised that we don't get much of a wow here, at least for the mid woofer, but never heard neither (purifi or Kii) myself I have to say.

There seems to be a disconnect between the principles of Purifi Audio and Kii Audio. On the one hand, Purifi is pushing very expensive components as a necessary means to achieving high performance while on the other hand Kii Audio deploys very low cost components to achieve high performance.

 

[tktran303]

There seems to be a disconnect between the principles of Purifi Audio and Kii Audio. On the one hand, Purifi is pushing very expensive components as a necessary means to achieving high performance while on the other hand Kii Audio deploys very low cost components to achieve high performance.

It might be hard to connect the dots if you don't know the history of the 2 entities.

Even though Bruno Putzeys been involved in both (and a few other companies) Kii Audio came first, and relied on drivers and amplifiers made by other companies eg. SEAS, Peerless, Hypex.

Purifi Tranducer Technology is a subsidiary of Purifi, and design/build their own drivers... Bruno and Lars Risbo started by trying to fix non-linear distortion of drivers by electronic/software/DSP means, but that stalled after many years of R&D. Together they, and Carsten Tinggaard, Morten Halvorsen found more success in improving electromechanical aspects (suspension, motor, soft and hard parts) of the driver... which eventuated in the release of the PTT6.5W-01.

Anyway this is getting OT.

But Kii are NOT Purifi.

 

[BYRTT]

Could be wrong here but in general it looks there some nayers that we didn't really get improvements then please study below detailed overlaid animation of Amir's acoustic data (thanks a ton) that shows tiny bit smoother verticals plus bit smother overall responses and extended low end reach, not to speak of it scores 5.8 which is not bad..

Regarding low end extension was sensed be weak and worse than Revel M106 into listening test session then suggest if @amirm have the time plus find below comparison interesting to research that he rerun a listening test with Celuaris SPK5 positioned in exactly same position as for Revel F328Be, also added a dirty quick 7 times PEQ that could be used in test for Celuaris SPK5..

 

[Tks]

I'm a bit confused why even Amir thinks these are good measurements. Of course the disotrtion metrics are going to be good (not great in my view, simply due to the treble in both output levels tested). I imagine many woofers would perform this good if you had such rolloffs both at the top end and low end of the frequencies.

These drivers don't seem cheap coming from Purifi, so idk... Not great.

 

[Francis Vaughan]

The distortion numbers of the bass driver are the standout. Really very good. But as implemented in this speaker, even once the damage that had occurred to the first sample has been rectified the weaknesses in the tweeter chosen and use of a ported enclosure let things down. Not a lot of surprise here.
I’m starting to get interested in thoughts of a three way of some form again. But probably with the higher efficiency lower X-max version.
I just can’t see this one particular speaker design as justifying itself.

 

[amirm]

Regarding low end extension was sensed be weak and worse than Revel M106 into listening test session then suggest if @amirm have the time plus find below comparison interesting to research that he rerun a listening test with Celuaris SPK5 positioned in exactly same position as for Revel F328Be, also added a dirty quick 7 times PEQ that could be used in test for Celuaris SPK5..

The F328Be was tested in a different part of the house. And Klippel measurements underestimated its low-end by fair bit due to its low frequency complexity. So it is not a good comparison.

 

[mitchco]

@mitchco reviewed the SPK4 and was duly impressed by its performance incl bass. Where did Purifi go wrong with the updated/improved SPK5 version?

I wondered what happened for SPK5? The SPK4 I had in my room just smoked my KEF LS50's in every way possible, lower distortion, much higher SPL and absolute clean bass response down to 32 Hz. I am just speaking to the Purifi driver. See post https://www.audiosciencereview.com/...ker-review-prototype.17806/page-2#post-578454 for some in room measurements.

 

[Robbo99999]

This is a second review and detailed measurements of Purifi woofer reference design implementation by Celuaris called SPK5. The original sample had some enclosure leakage issues which have been resolved in this unit due to collaboration between our DIY expert, @Rick Sykora, and builder, @sgoldwin.

I am going to borrow the picture from the last review since nothing is changed in that regard:

The crossover is now built inside the unit rather using the external one in the previous reference design.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber. In a nutshell, the measurements show the actual sound coming out of the speaker independent of the room.

I performed over 1000 measurement which resulted in error rate of around 1%.

Temperature was 58 degrees F at sea level. I kept the speaker indoor at 70 degrees prior to starting the measurements.

Measurements are compliant with latest speaker research into what can predict the speaker preference and is standardized in CEA/CTA-2034 ANSI specifications. Likewise listening tests are performed per research that shows mono listening is much more revealing of differences between speakers than stereo or multichannel.

Reference axis was the tweeter center which in case of these AMT tweeters is a bit inexact.

Celuaris SPK5 Purifi Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 110820

Compared to the first sample, bass extends lower but it starts to droop starting at 100 Hz. We have a dip around 350 Hz which is caused by port resonance:

View attachment 110819

I am disappointed in the uneven response of the tweeter which has at least one resonance which showed up in the spin graph above.

Early window is fine other than showing that the tweeter due to being narrow width-wise, is starting to "beam" (narrow its radiation angle:

View attachment 110821

We see that just the same in our predicted in-room response:

View attachment 110822

The main claim to fame of Purifi driver is low distortion -- something we did not see in the first sample I tested. Fortunately here it becomes the star with strong showing:

View attachment 110823

View attachment 110824

It is so good that it makes the tweeter really look bad.

Horizontal directivity now is much smoother:
View attachment 110825

View attachment 110826

Here is our vertical:

View attachment 110827

Finally, the impedance and phase:
View attachment 110828

Celuaris SPK5 Purifi Speaker Listening Tests
The sample speaker I have has SpeakOn connector and the only cord I had for it did not extend long enough to test the unit where I normally do. So I had to move the speaker to the right side of my room where there is no bass enhancement. Whether it was that or something else, the results were quite unsatisfying. There was not a whole lot of bass at normal playback levels. Due to low efficiency, I had to push the speaker hard to get bass out of it but then the winding would suddenly jump the gap and cause nasty static. Tracks with deep bass didn't distort the speaker but they were hardly played down low either. I played around a bit with EQ but didn't get far so gave up.

To make sure the new location or my mood was not at fault, I replace the SPK5 with Revel M106 and boy what joy there was with that speaker. Thundering and clean bass was back as was warm tonality. I could detect no more distortion there than I did in SPK5.

Conclusions
If this were a test of distortion for the Purifi driver, mission is accomplished. This is one low distortion driver. Alas, the reference design that Purifi has created is not optimal. The tweeter I feel is not the right match nor is the enclosure to get proper bass and efficiency here. I am not sure of the purpose of a fancy driver if you don't have much bass and you have low efficiency to boot. Objectively the results look better than my impression so perhaps I am being too harsh. I don't know.

As is, I can't recommend SPK5. Feel bad saying that despite the heroic efforts of its builder who went through some five (5) samples to get us to this point with transatlantic shipping issues, samples getting lost, etc.

------------
As always, questions, comments, recommendations, etc. are welcome.

Appreciate any donations using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

I was gonna say I like the look of that woofer in terms of low distortion, but then I saw your listening tests re disappointing bass. Judging by the low distortion of the woofer and the roll off only really starting to accelerate at 35Hz I would say this speaker is begging for a +5dB Low Shelf filter at 100Hz which would blend in very nicely with the slope of the rest of the spinorama, and would take useful bass down to 30Hz....so at moderate listening levels I think that would be pretty useful. I don't know if this would have helped with your listening tests, but you made a comment about static, so I'm not sure. Judging by the spinorama though just putting in the Low Shelf Bass Boost would make the speaker even duller sounding due to the wide deficiencies around 3kHz and above 10kHz, so I'd put in a wide +2dB Peak Filter of Q0.7 at 3kHz to lift that area, a sharper peak filter at 8kHz cutting the resonance, and a High Shelf Filter at 10kHz of about +2 or 3 dB to lift that far right roll off of the treble. I don't know if you tried that combination during your EQ testing, but it looks fixable from that point of view.

 

[ZestClub]

The PTT6.5 feels like it has a lot of potential - but when I ran my tests it felt like standing in front of a sheet of tissue paper - you can see it moving, you can hear it moving, but you can't feel it!

 

[Massimo]

I wondered what happened for SPK5? The SPK4 I had in my room just smoked my KEF LS50's in every way possible, lower distortion, much higher SPL and absolute clean bass response down to 32 Hz. I am just speaking to the Purifi driver. See post https://www.audiosciencereview.com/...ker-review-prototype.17806/page-2#post-578454 for some in room measurements.

The difference between the SPK4 and SPK5 is quite minimal. Unfortunately, the SPK5 tested here is not built as the Purifi blueprints/specs. There are numerous key differences which may (or may not) influence the end result.

 

[Rick Sykora]

I wonder if @Rick Sykora 's experiment with a passive radiator with provide some interesting results that would deal with that port cancellation.

Yes, the port resonance is more tame, but so is the overall output (purple trace is port and red is pr)...

EDIT: per @ctrl post below, cannot compare levels here

PLEASE NOTE: the passive radiator used above is NOT the Purifi one. I had the SB 5x8 on hand. Am obtaining the Purifi one for comparison.

 

[TNT]

Since the port resonance is out of phase with the woofer, it is causing cancellations in different places around the speaker. It is not beaming per-se.

An an other note - you speaker testig system, does it scan the driver-cones and calculate distorsion or does it measure acustically with mic in the "good ol' way"?

//

 

[PeteL]

It might be hard to connect the dots if you don't know the history of the 2 entities.

Even though Bruno Putzeys been involved in both (and a few other companies) Kii Audio came first, and relied on drivers and amplifiers made by other companies eg. SEAS, Peerless, Hypex.

Purifi Tranducer Technology is a subsidiary of Purifi, and design/build their own drivers... Bruno and Lars Risbo started by trying to fix non-linear distortion of drivers by electronic/software/DSP means, but that stalled after many years of R&D. Together they, and Carsten Tinggaard, Morten Halvorsen found more success in improving electromechanical aspects (suspension, motor, soft and hard parts) of the driver... which eventuated in the release of the PTT6.5W-01.

Anyway this is getting OT.

But Kii are NOT Purifi.

Thanks, I didn't think they where the same entity, but did not have the information on who came first and who made the driver. so those 6.5 inch woofers in Kii serie are Pearless? Separate entities, sure, the point is if you have an involvment, financially and as an engineer with a speaker company, and in an other area you have researched and developped expertise on driver design, it's not far fetch that eventually you are trying to use it, unless Pearless is a better solution for Kii's design. The same way they use hypex amps. If I read between the lines of what you said, what they could do in dsp with Kii "stalled" and found "more success" with making drivers. If this is true, and I'm not sure based on this review, we should see at some point Purifi drivers in Kii speakers no?

 

[ctrl]

Yes, the port resonance is more tame, but so is the overall output (purple trace is port and red is pr)...

This way you can't compare the effect of BR port and PR because the "membrane areas" are different.
You would have to correct one of the two values in the sound level.

It is best to always determine the total frequency response from the corrected BR port / PR and the woofer. This will simplify the comparison.

In the Arta handbook you will find from page 127 described how this is done.

 

[ctrl]

OFF-TOPIC

This way you can't compare the effect of BR port and PR because the "membrane areas" are different.

Near-field measurements must be corrected in level depending on the measurement distance and the sound source dimension.

If the measuring distance for BR port / PR and woofer was the same during the near field measurement, for example 1cm (0.4''), then only the different sound surfaces to each other must be corrected.

If the PR is 16cm in diameter but the BR port is only 4cm, then the level correction would be 36dB - 27dB = 9dB (relative to 1m distance)

Source: Arta handbook page 95

 

[Rick Sykora]

This way you can't compare the effect of BR port and PR because the "membrane areas" are different.
You would have to correct one of the two values in the sound level.

It is best to always determine the total frequency response from the corrected BR port / PR and the woofer. This will simplify the comparison.

In the Arta manual you will find from page 127 described how this is done.

Thanks, but understand these are the respective outputs with the woofer output level set to 75 dB for each alignment.

So, why are they not comparable?

 

[Rick Sykora]

OFF-TOPIC



Near-field measurements must be corrected in level depending on the measurement distance and the sound source dimension.

If the measuring distance for BR port / PR and woofer was the same during the near field measurement, for example 1cm (0.4''), then only the different sound surfaces to each other must be corrected.

If the PR is 16cm in diameter but the BR port is only 4cm, then the level correction would be 36dB - 27dB = 9dB (relative to 1m distance)

View attachment 111044

Source: Arta handbook page 95

Got it. Will note on earlier post that levels are not comparable. Thanks!