I enjoy reviewing innovative loudspeakers like the Dutch and Dutch 8c and the Kii THREE. I also spend time on diyAudio’s loudspeaker forums looking at what to build next or modify what I have, essentially looking for the “next big thing.” I noticed Purifi’s woofer getting good reviews on diyAudio and ASR. On Audiophile Style, @DuckToller wrote an article on meeting the folks from Purifi in Munich with @The Computer Audiophile. Reading those really peaked my interest.
HiFiCompass’ review of the PTT6.5 shows the woofer measuring considerably better than its competitors, especially in reduced distortion, both THD and IMD. This puts Purifi’s PTT6.5 woofer into a class on to itself. Now I really wanted to hear this woofer! Before I could track down a pair, I received a note from DuckToller asking if I would be interested in evaluating the woofers in kit form. A short while later I received a demo kit and a stereo IET400A amplifier.
I must say upfront that this technology represents a major audible improvement in mid/bass clarity and low frequency extension. Meaning, I have not heard a 6.5” woofer go this low or sound this clean, at concert level volumes. To my ears, the Purifi PTT6.5 woofer is the “next big thing” in transducer technology. Coupled with Purifi’s 1ET400A amplifier, this is the lowest distortion sound I have heard from any amp/speaker combo of this size in my listening room.
Purifi SPK4 Loudspeaker Demo Kit and 1ET400A stereo amplifier (top view).
Much has been written about the technical aspects of Purifi’s PTT6.5 woofer technology including HiFiCompass’s extensive objective measurements linked above and Amir’s objective measurements of the 1ET400A amplifier.
My review will be a combination of a few objective loudspeaker measurements in my room and subjective listening impressions. My review is aimed at the consumer and DIY audio enthusiasts wondering what these technology advancements are and how they translate into a superior listening experience.
A 6.5” woofer is not big and has real world limitations on how low of a frequency the driver can realistically reproduce, yet still sound clear in the bass and midrange as volume increases.
While our ears are relatively insensitive to low frequency harmonic distortion, they are more sensitive to intermodulation distortion (IMD). How can a small driver output significant low frequency energy yet still sound clean in the lower mids and midrange? This is the crux of the technical dilemma that no loudspeaker designer/manufacturer has been able to crack… until now.
I thought I would ask Purifi’s Lars Risbo, one of the brains behind the PTT6.5 woofer, to describe their design approach on how they were able to obtain such low measured intermodulation distortion numbers:
“Our PTT6.5 woofer is the result of about 4 years of research. It has been a fruitful teamwork between Bruno, Carsten, Morten, and myself. This has resulted in a design that is vastly different from conventional drivers. In short, we have been on a crusade against all sources of intermodulation distortion. Psychoacoustics has correctly shown that our ears are not that sensitive to pure harmonic distortion – especially in the bass. However, our ears are way less forgiving when it comes to IMD when listening to complex music program material. Since it is impossible to only generate harmonic distortion without IMD, it has been necessary to reduce all nonlinear mechanisms in a speaker.
The most known source of IMD is the position dependent force factor Bl(x) which we have made ruler flat over +-6mm excursion (+-10mm at -10% Bl). Less known is the current and frequency dependent modulation of the force factor (aka force factor modulation, flux modulation, reluctance force) which we have reduced by up to 40dB over comparable standard motors. Then we have the position dependent inductance which causes a modulation of the drive current, again reduced by a huge factor. A new (re)discovery is the gain modulation caused by the varying radiating area of conventional surrounds – this can cause an amplitude modulation of the midrange by +-10% easily. This has been reduced by orders of magnitude by our very special surround geometry.
Finally, an extremely annoying distortion mechanism is the hysteretic memory of the iron in the motor. This distortion leads to a grainy non dynamic sound and lack of ‘black canvas’. We have managed to reduce this distortion by up to 50dB compared to industry benchmark drivers. Interestingly, the exact same hysteretic distortion can be found in class D amplifier output coils. Our 1ET400A amplifier embeds the coil in a control loop with over 70dB loop gain all the way to 20 kHz which effectively removes this distortion. We believe that the class D coil’s hysteretic distortion has been a large part of the bad reputation of early class D designs.”
When I asked Lars about how loud I can turn up the demo kit, he had this to say:
“How loud? We always end up playing insanely loud due to the lack of distortion (my ears become the limit). The woofer handles 350W IEC long term and survives bottoming (which of course sounds bad). Actually, the 100 hour power tested units have never had damages to the soft parts which is quite unusual (our special surround works like a dream to stabilize the motion). You have +-14mm of mechanical stroke and with the lowered port tuning to 30Hz it takes quite a bit to bottom it. So just fire it up and ease back if you hear bottoming – mostly because it then suddenly sounds bad, it’s not harmful.”
Wow. Usually I am careful not to blow up review speakers, but this sounds like a challenge!
Kidding aside, very interesting as I have found most bookshelf speakers don’t go very loud before hearing the onset of cone breakup. For example, I own a pair of KEF LS50’s which have a nice voicing, but don’t go very loud (or very low in frequency) before audible distortion is heard as compared to PTT 6.5 (i.e. demo kit). The traditional approach to combat this is to pair satellite speakers with sub(s) to reduce the low frequency load (and IMD) on the small driver. Or in the case of the Dutch and Dutch 8c’s, put two 8” long stroke subwoofers in the back of the stand mount speaker enclosure operating at 100 Hz and below.
The PTT6.5 woofer is unique in being able to reproduce a prodigious amount of low frequency energy (i.e. in the 30 Hz range), yet remain absolutely clean sounding throughout the lower mids and midrange. The only comparison I have is using large, stiff cone, short stroke 15” woofers, which is what I use today in my JBL system, in addition to Rythmik dual 18” subs. The (major) disadvantage is that these cabinets tend to be the size of fridges, look fugly and cost more dollars. Further, the 15” drivers can barely make it to an 800 Hz crossover point, whereas the PTT6.5 is crossed over at 2.5 kHz in the demo kit, but can still hit 32 Hz, all delivered with low distortion. The PTT6.5 is a subwoofer, woofer and midrange to 2.5 kHz all in one unit.
Of course, it is all relative to size and SPL’s. There is no “replacement for displacement” and my large JBL’s with dual Rythmik 18” subs can easily outperform the PTT6.5 in sheer SPL’s. But, listening at “reference” level of 83 dB SPL, the PTT6.5 sounds unbelievably clean with good low frequency extension and outstanding clarity in the lower mids and midrange. A wide bandwidth, clean sounding, stand mount audiophile monitor it is.
Setup and Configuration
You can see the size of the subs (not used) relative to the Purifi SPK4 demo kit. I pushed my large JBL’s out of the way (see “the fridge” far left) and placed the SPK4 in the same location, which is also the same location where I have reviewed other speakers. I set them up in a 9 ft. equilateral triangle, toed in to point right at my ears on axis. The Monoprice 24” steel stands are filled with sand and the speakers are doubly isolated from the floor using Vibrapods. One set under the stands, another set under the speakers.
I have attached the PDF sheet for the SPK4 Demo Kit, a ported box tuned to 30Hz, which is the same frequency as the woofers Thiel/Small parameters resonance frequency (i.e. Fs). Thus tuned for maximum low frequency extension in a 15 liter bookshelf enclosure. I have also attached the datasheet for the PTT6.5” woofer and 1ET400A amplifier. Just before publication I received an update that there is a SPK5 demo speaker, which I have also attached the data sheet.
The passive crossovers are built using premium parts including Jantzen air core inductors. Due to the extremely low distortion of the PTT6.5 woofer, it is important to choose good crossover filter components. Inductors must by air-core.
These are heavy duty crossovers crossing over to the AMT tweeter at 2.5 kHz. The Mundorf Air Motion Transformer (AMT) tweeter is well-known for its excellent transient response and outstanding low distortion. A good match for the PTT6.5 as a demo platform.
Amir measured the 1ET400A amplifier to output 257 watts @ 4 ohms with THD+N of 0.0002% and 131 watts @ 8 ohms with 0.00019% THD+N. Looking at the impedance graph in the SPK4 app note, we see impedance as low as 5 ohms in the low frequencies and lower mids. Given Lars said that the PTT6.5 can handle 350 watts and with SPK4 system sensitivity of ~ 82 dB SPL @ 2.83Vrms at 1 meter, we should be able to hit concert levels at the listening position (i.e. ~ 95 dB SPL C weighting, slow response) no problem.
I am using a Lynx Hilo DAC which measures around 0.001% THD+N. The Hilo’s ADC measures even better and the main reason I use this converter for my DSP loudspeaker measurements and binaural recordings. Suffice to say, all of these distortion levels are well below our audibility threshold in a typical living room listening environment.
I connected the Hilo to the 1ET400A amp via its XLR balanced input connections:
Putting my ear literally against the AMT tweeter I hear the faintest hiss, dead silence on the woofer. For critical listening, I listen at reference level (i.e. 83 dB SPL C weighting at the listening position). Here is an excellent article on calibrating for reference level and why we listen at this level.
With the higher than normal 6.5” woofer sensitivity of 88 dB @2.83v at 1m and 350 watt power handling rating, this makes for one powerful 6.5” woofer. Note that the PTT6.5 data sheet sensitivity is for infinite baffle (2pi) whereas the usual speaker box sensitivity is stated for free space (4pi) - this tradition is a source of much confusion. This means that one should subtract 6 dB from the data sheet to get the free space sensitivity (i.e. the box baffle step is equalized down to the free field response in the passive xover and the tweeter is attenuated accordingly).
I am still amazed how the PTT6.5 can cleanly belt it out without any strain every time I listen. The only other way to get that kind of kick is using what I have now, which are short stroke, large stiff cone area, 15” woofers. The “quality” of bass coming from the PTT6.5 sounds like it should be coming from my large JBL’s. It is quite the audible illusion. Of course this is only up to a certain SPL, but much louder than I expected and certainly louder and cleaner than other similar sized loudspeakers, by a good margin.
To switch it around a bit, I ended up listening to the amp/speaker combo for days before I pulled out the measurement gear. In fact, this is the longest I have procrastinated to get the measurement mic out. I just wanted to continue listening. So, let’s begin with the listening impressions first and then look at the measurements to confirm what I am hearing.
I have listened to this demo amp/speaker combo for numerous hours over days using a wide variety of source material. Not only my reference recordings that I have linked to in previous articles, but tunes that have a combination of good low bass and midrange elements that would stress the amp/speaker combo to “give it up” in the bottom end and stress the midrange.
Driving loudspeakers towards their limits can result in unexpected failures. I have “popped” woofers right out of their gap and instantly “smoked” tweeters into silence with a clipped amp signal. Sometimes it is good to start with material that is dynamically compromised to test the limits.
Yotto’s Radiate (DR6) is a good thumping mix with significant bass centered at 39 Hz and highly compressed. So does it sound good? The PTT6.5 can certainly reproduce the deep bass. I can feel the couch vibrating from the 6.5” woofers. Ridiculous you say! I could not believe it myself.
Sounds clean with good transient punch on the kick drum and the throbbing bass sounds solid. I cranked it up until 96 dB SPL C weighting at the listening position 9 feet away in 30” x 16.5’ x 8’ room. At that SPL, I just started to notice bottoming out of the woofers on peaks. That’s +13 dB more SPL than reference level, which to our ears sounds over twice as loud. It’s concert level loud! Very rarely do I listen at this level, but it is fun for short periods 🙂
Up until that point, the amp/demo speaker combo sounds crystal clear without the bass and kick drum losing any articulation and the midrange syths still sounding clear and well defined in the mix. Most speakers will tend to “break up” with distortion before reaching their full limits. Again, the PTT6.5” just kept getting louder, showing no strain, and then the bottoming of the woofer is very brief at the limit. Turn the level down a bit, and back to sounding crystal clear. A unique property that I have not heard from any other small transducer.
Above on the left is a spectral analysis of the first couple of minutes of Radiate along with a waveform view on the right. Can you say clipped! Significant energy focused in the 40 Hz region.
Madonna’s Power of Goodbye (DR8) has an excellent low frequency bass line with long sustained notes. Literally feels like rolling waves as you can feel each wave that has significant energy down to 3 Hz. With my dual Rythmik F18’s I can make the house, couch, and my body vibrate and pressurize the room with this tune turned up. Makes me smile.
The PTT6.5 could indeed pressurize my room and make the couch and room vibrate with some body sensation. Really impressive coming from a 6.5” (sub)woofer! At that level Madonna’s lower mids and midrange from her voice and syths still sound clear as a bell while the woofer is able to shake the room a bit on the low notes. This is unheard of from a 6.5” woofer. Mind blown.
We can see in the frequency spectrum above, the wave like pulse below 10 Hz. On the right is the waveform which is compressed, but not horrendously clipped like Radiate.
Patricia Barber’s, Regular Pleasures (DR15) is a torture test for amps/loudspeakers with its concussive bass drum and sliding bass line. With the punch of DR15, it is downright dangerous for popping woofer cones. It is another recording with significant low frequency energy focused at 40 Hz and good content down below 10 Hz. On my dual F18’s subs the drop note bass drum shakes the house on its foundations after the initial concussive hit.
Obviously not on a 6.5” woofer but within its operational limits, sounds very deep, solid and in balance with the rest of the mix. I can still feel the bass and drums sitting on the couch. I am amazed at how low, loud, and clean these woofers can go. They have the same transient impact like my 15” JBL’s have (up to a certain SPL, but way louder than I expected) and the lower midrange is crystal clear. I mean the cleanest I have ever heard.
Patricia’s voice is mixed up in level in the overall mix with the drums and bass. The lower midrange and mids from her voice sounds clear with no modulation that I could hear even when cranked right up. I can hear/feel the resonance of the lower mids of her voice unlike I have heard from many a different speaker. On most speakers I have listened to, Patricia’s voice sounds good, but a bit thin and missing the power of her lower midrange register, which I only hear on the large JBL’s… and now the PTT6.5.
I could go on, but the above three examples are representative of the continued enjoyment I get listening to this amp/speaker combo. Even blasting for a period of time only barely warmed up the 1ET400A chassis and case. Not once did I hear any strain from the amplification in any way. The top end sounded transparent with excellent transient response.
Well, now I am curious how they measure, up close, and from the listening position. Given what I have heard, I expect good low frequency extension with a smooth lower mid and midrange frequency response.
Here is a video of an REW measurement of the right speaker. Turn your sound down to a safe level if your web browser is hooked up to your sound system. If you look closely at the beginning, you can see how much cone excursion is going on starting from 10 Hz on up. Perfectly linear and clean sounding.
The first set of measures are 30cm away from the speaker on-axis with the mic positioned between the two drivers. While the room will have its way below Schroeder, (i.e. the room’s transition frequency), I still want to see some representation of bass output. So I don’t gate my measurements and for sure the room is in the picture, but still a very smooth and extended response at both frequency extremes:
My room comes into play from about 300 Hz on down, with some excess room energy at 250 Hz and typical room modes down lower. The -3 dB point measured was about 32 Hz. Wow, this is a small two way with a 32 Hz to 20 kHz full range response. The SPK4 demo speaker has a similar midrange voicing as the KEF LS50 in frequency response, but the LS50 rolls off quickly after 100 Hz.
Here is a quick comparo of frequency response measured in the same location and 30cm away. I have offset the measurement pairs so it is easier to compare. Purifi SPK4 Demo Kit on top, KEF LS50 on bottom:
As you can see, the response on the LS50 drops quickly after 100 Hz and is -10 dB down at 40 Hz. So this confirms for me what I heard during subjective listening as the Purifi PTT6.5 in the demo kit can put out significant output to 32 Hz, again with the lower mids and midrange sounding crystal clear.
Let’s look at the demo speakers step (timing) response. What we are looking for is the “right triangle” shape which denotes an ideal speaker’s minimum phase response:
The doublet spike in front is the AMT tweeter which the passive crossover blends perfectly into the Purifi PTT6.5 woofer and has near perfect right triangle step response. Look how smooth the right triangle is with hardly any “bumps” at all which translates into how smooth the woofers frequency response is.
I did not take any distortion measurements as I am not setup to perform those. No anechoic or semi-anechoic chamber, unknown distortion contributed by the measurement mic and mic preamp. Plus it’s (really!) hard to do correctly ☺ I would rely on HiFiCompass’s excellent IMD measurements to be representative of the PTT6.5 unit.
Here is the frequency response as measured at the listening position:
Of course, there is more room in the picture as we know below Schroeder the room is in control of the low frequency response. I have dips at 80Hz and 90Hz and peaks in the low end – classic room modes. Nothing some DSP can’t fix, but my point here is again to show the wide bandwidth coverage of this small two way system with in-room response down to 32 Hz whether near field or far field in my room. Impressive!
If you drew a straight horizontal line in-between the peaks and dips and average them across the frequency band, it would be exceptionally flat response with the normal high frequency roll-off.
For fun, I hooked up my dual Rythmik F18” subs with a 90 Hz crossover point as you can see in the above chart where I have some major room issues and dialed in some DSP and this is what I got at the listening position:
One of the reasons I did this is because I am used to a particular sonic signature that I could only get with subs and large JBL 15” woofers. So I was curious if I could replicate that with the PTT6.5 and maybe lose “the fridges.” I still can’t believe how clean it sounds. Cleaner than my JBL’s… Hmmmm.
Do you remember the THX intro? Sony’s intro to Spider-Man into the Spider-verse is the new THX intro. While I linked to the YouTube recordings, hearing it on Blu-ray is mind blowing. With a -3 dB point of 6 Hz in my room with the dual Rythmik F18 subs dialed in, it’s mind blowing. As the intro builds with the volume turned way up, both my daughter wife tried to hide behind the couch because it sounded like the room was going to implode at the climax. The SPK4 handled it with aplomb. Love it!!!
I must say, I have heard a lot of amp/speaker combos over 50 years in audio, 10 of which being a pro recording/mixing engineer, and have never heard anything like this. Or should I say the lack of distortion and outstanding low frequency extension from a 6.5” woofer. The clarity is stunning. Congrats, truly in a class upon itself.
Is the PTT6.5 a subwoofer? A woofer? A midrange driver? Yes, it is all three with a measured nearfield and far field response from 32 Hz to 2.5 kHz in the SPK4 demo package. This really got me to thinking of applications for the driver. Literally makes for a perfect driver for a small format, full range 2 way accurate sound reproduction system. This made me wonder about Purifi’s business strategy.
I asked Lar’s about Purifi’s technology roadmap, “Our plan is to become an OEM supplier but also serving the DIY’ers. For larger volumes, we also license our tech like we have done with NAD. The roadmap also calls for DSP and algorithms as well as PSU’s. Our overarching strategy is to go in and overhaul the weakest links in the audio chain and break into new performance territory. Speakers and amplifiers were the natural first points of attack. Exciting to see where all this leads us to.”
Makes perfect sense to me and I wish Purifi all the best. They are certainly off to an auspicious start!
Mitch “Mitchco” Barnett.
I love music and audio. I grew up with music around me, as my mom was a piano player (swing) and my dad was an audiophile (jazz). My hobby is building speakers, amps, preamps, etc., and I still DIY today.
I mixed live sound for a variety of bands, which led to working full-time in multiple 24-track recording studios. Over 10 years, I recorded, mixed, and sometimes produced over 30 albums. I am into Digital Signal Processing (DSP) and run an Accurate Sound Calibration service.