• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Purifi Audio PTT6.5W04-01A 6.5 Inch Woofer Review

hardisj

Major Contributor
Joined
Jul 18, 2019
Messages
2,299
Likes
10,171
Location
North Alabama
#1
Here's the link to my original review because some of the tables and such won't easily translate here:
https://www.erinsaudiocorner.com/driveunits/purifi_ptt65w04/



Purifi Audio PTT6.5W04-01A 6.5 Inch Woofer Review
  • Sunday, Aug 16, 2020





Information
Per the manufacturer:
The PTT6.5W04 is a 6.5” driver that truly cracks the long-stroke code. PURIFI’s research has identified the parameters that have so far prevented long stroke drivers from breaking through in truly high performance audio. The distortion of the acoustic output of a driver is a combination of several separate distortion mechanisms in the motor, cone and suspension. When testing a complete driver with sine waves, a situation often arises where a small tweak seemingly improves harmonic distortion (HD) by letting two mechanisms counteract each other. This always leads to a clear worsening of intermodulation distortion (IMD) that becomes obvious when testing the unit with a more complex signal. This is why traditional HD tests fail to predict subjective sound quality. PURIFI takes care to optimize the various distortion mechanisms separately, thus guaranteeing that any measured improvement truly reflects a real audible improvement no matter what the signal is. Accurate mathematical models are developed for the motor, suspension and the vibroacoustics of the dome and surround. These models clearly explain several distortion mechanisms in each domain, which provides fundamental insights into better ways of constructing the motor, cone and suspension. Finally, the same models are used to numerically fine-tune the geometry. This way of working reliably reduces multiple distortions mechanisms which secures low IMD for complex signals.​
Retail price is about $740/pair from Madisound.com.

Product Spec Sheet can be found here.

Test Results
Foreword: Subjective Analysis vs Objective Data of Drive Units (click here)

Small Signal Testing (Thiele-Small Results)
Using Klippel’s Distortion Analyzer 2, Linear Lumped Parameter Measurement Module, Pro Driver Stand and provided Panasonic ANR12821 Laser along with Klippel’s Training 1 - Linear Lumped Parameter Measurement tutorial, I measured this drive unit’s impedance and small-signal parameters. Below are the results.
Note: The impedance graph can be found in the Frequency Response Linearity graphic later in this review.
Electrical Parameters
Re3.69Ohmelectrical voice coil resistance at DCLe0.218mHfrequency independent part of voice coil inductanceL20.385mHpara-inductance of voice coilR22.06Ohmelectrical resistance due to eddy current lossesCmes440.29µFelectrical capacitance representing moving massLces62.52mHelectrical inductance representing driver complianceRes84.51Ohmresistance due to mechanical lossesfs30.3Hzdriver resonance frequencyMechanical Parameters(using laser)Mms26.119gmechanical mass of driver diaphragm assembly including air load and voice coilMmd (Sd)24.391gmechanical mass of voice coil and diaphragm without air loadRms0.702kg/smechanical resistance of total-driver lossesCms1.054mm/Nmechanical compliance of driver suspensionKms0.95N/mmmechanical stiffness of driver suspensionBl7.702N/Aforce factor (Bl product)Lambda s0.007suspension creep factorLoss factorsQtp0.297total Q-factor considering all lossesQms7.092mechanical Q-factor of driver in free air considering Rms onlyQes0.309electrical Q-factor of driver in free air considering Re onlyQts0.296total Q-factor considering Re and Rms onlyOther ParametersVas26.2642lequivalent air volume of suspensionn00.228%reference efficiency (2 pi-radiation using Re)Lm85.78dBcharacteristic sound pressure level (SPL at 1m for 1W @ Re)Lnom86.13dBnominal sensitivity (SPL at 1m for 1W @ Zn)Sd132.70cm²diaphragm area


Large Signal Modeling (Linear Xmax Results)
Using Klippel’s Distortion Analyzer 2, Large Signal Identification Module, Pro Driver Stand and provided Panasonic ANR12821 Laser along with Klippel’s Training 3 - Loudspeaker Nonlinearities tutorial, I measured the linear, nonlinear and thermal parameters of this drive unit.
Nonlinearities
Traditionally, Xmax has been defined in one of the following ways:
  • the physical overhang of the voice coil (height of the voice coil relative to height of the gap)
  • 115% times the physical overhang above
  • the point where displacement limit(s) is/are exceeded
The third option is where the Klippel LSI module comes in to play. It permits a more “apples to apples” approach of defining the displacement (Xmax) limits based on the XBL, XC, XL and Xd. The displacement limits XBL, XC, XL and Xd describe the limiting effect for the force factor Bl(x), compliance Cms(x), inductance Le(x) and Doppler effect, respectively, according to the threshold values Blmin, Cmin, Zmax and d2 used by the operator.
There are one of two sets of thresholds which can be used to define linear excursion:
  1. Non-Subwoofer Drivers: The thresholds Blmin= 82 %, Cmin=75 %, Zmax=10 % and d2=10% generate for a two-tone-signal (f1=fs, f2=8.5fs) 10 % total harmonic distortion and 10 % intermodulation distortion.
  2. Subwoofer Drivers: The thresholds Blmin= 70 %, Cmin=50 %, Zmax=17 % create 20 % total harmonic distortion which is becoming the standard for acceptable subwoofer distortion thresholds.
These parameters are defined in more detail in the (Klippel) papers:
  • “AN04 – Measurement of Peak Displacement Xmax”
  • “AN05 - Displacement Limits due to Driver Nonlinearities”
  • “AN17 - Credibility of Nonlinear Parameters”
  • “Prediction of Speaker Performance at High Amplitudes”
  • “Assessment of Voice Coil Peak Displacement Xmax”
  • “Assessing Large Signal Performance of Loudspeakers”


Below are the displacement limits’ results for this drive unit obtained from Klippel’s LSI module:
Displacement Limits
Code:
X Bl @ Bl min=82%    9.1    mm    Displacement limit due to force factor variation
X C @ C min=75%    6.9    mm    Displacement limit due to compliance variation
X L @ Z max=10 %    >11.2    mm    Displacement limit due to inductance variation
X d @ d2=10%    42.1    mm    Displacement limit due to IM distortion (Doppler)


Per the above table, this drive unit's linear excursion is limited to 6.9mm due to exceeding the Cms displacement limit of 75% for the distortion limit of 10%.

We can break the above information down further. The below text is written by Patrick Turnmire of Red Rock Acoustics and used with his permission, substituting data from this drive unit’s test results where applicable.
Large Signal Modeling
At higher amplitudes, loudspeakers produce substantial distortion in the output signal, generated by nonlinear ties inherent in the transducer. The dominant nonlinear distortions are predictable and are closely related with the general principle, particular design, material properties and assembling techniques of the loudspeaker. The Klippel Distortion Analyzer combines nonlinear measurement techniques with computer simulation to explain the generation of the nonlinear distortions, to identify their physical causes and to give suggestion for constructional improvements. Better insight into the nonlinear mechanisms makes it possible to further optimize the transducer in respect with sound quality, weight, size and cost.
Nonlinear Characteristics
The dominant nonlinearities are modelled by variable parameters such as
Bl(x)instantaneous electro-dynamic coupling factor (force factor of the motor) defined by the integral of the magnetic flux density B over voice coil length l as a function of displacementKMS(x)mechanical stiffness of driver suspension a function of displacementLE(i)voice coil inductance as a function of input current (describes nonlinear permeability of the iron path)LE(x)voice coil inductance as a function of displacement

Nonlinear Parameters
Bl change with excursion



The electrodynamic coupling factor, also called Bl-product or force factor Bl(x), is defined by the integral of the magnetic flux density B over voice coil length l and translates current into force. In traditional modeling this parameter is assumed to be constant. The force factor Bl(0) at the rest position corresponds with the Bl-product used in linear modeling. The red curve displays Bl over the entire displacement range covered during the measurement. You see the typical decay of Bl when the voice coil moves out of the gap. At the end of the measurement, the black curve shows the confidential range (interval where the voice coil displacement in this range occurred 99% of the measurement time). During the measurement, the black curve shows the current working range. The dashed curve displays Bl(x) mirrored at the rest position of the voice coil – this way, asymmetries can be quickly identified. Since a laser was connected during the measurement, a coil in / coil out marker is displayed on the bottom left / bottom right.

Suspension Stiffness change with excursion


The stiffness KMS(x) describes the mechanical properties of the suspension. Its inverse is the compliance CMS(x).

Inductance change with excursion





The inductance components Le (x) and Bl(i) of most drivers have a strong asymmetric characteristic. If the voice coil moves towards the back plate the inductance usually increases since the magnetic field generated by the current in the voice coil has a lower magnetic resistance due to the shorter air path. The nonlinear inductance Le(x) has two nonlinear effects. First the variation of the electrical impedance with voice coil displacement x affects the input current of the driver. Here the nonlinear source of distortion is the multiplication of displacement and current. The second effect is the generation of a reluctance force which may be interpreted as an electromagnetic motor force proportional to the squared input current.
The flux modulation Bl(i) has two effects too. On the electrical side the back EMF Bl(i)*v produces nonlinear distortion due to the multiplication of current i and velocity v. On the mechanical side the driving force F = Bl(i)*i comprises a nonlinear term due to the squared current i. This force produces similar effects as the variable term Le(x).

Fs shift with excursion


Qts change with excursion



Asymmetrical Nonlinearities
Asymmetrical nonlinearities produce not only second- and higher-order distortions but also a dc-part in the displacement by rectifying low frequency components. For an asymmetric stiffness characteristic the dc-components moves the voice coil for any excitation signal in the direction of the stiffness minimum. For an asymmetric force factor characteristic the dc-component depends on the frequency of the excitation signal. A sinusoidal tone below resonance (f<fS) would generate or force moving the voice coil always in the force factor maximum. This effect is most welcome for stabilizing voice coil position. However, above the resonance frequency (f>fS) would generate a dc-component moving the voice coil in the force factor minimum and may cause severe stability problems. For an asymmetric inductance characteristic the dc-component moves the voice coil for any excitation signal in the direction of the inductance maximum. Please note that the dynamically generated DC-components cause interactions between the driver nonlinearities. An optimal rest position of the coil in the gap may be destroyed by an asymmetric compliance or inductance characteristic at higher amplitudes. The module Large Signal Simulation (SIM) allows systematic investigation of the complicated behavior.
Bl symmetry xb(x)
This curve shows the symmetry point in the nonlinear Bl-curve where a negative and positive displacement x=xpeak will produce the same force factor Bl(xb(x) + x) = Bl(xb(x) – x).
If the shift xb(x) is independent on the displacement amplitude x then the force factor asymmetry is caused by an offset of the voice coil position and can be simply compensated.
If the optimal shift xb(x) varies with the displacement amplitude x then the force factor asymmetry is caused by an asymmetrical geometry of the magnetic field and cannot completely be compensated by coil shifting.



Kms Symmetry xc(x)
This curve shows the symmetry point in the nonlinear compliance curve where a negative and positive displacement x=xpeak will produce the same compliance value kms(xc(x) + x) = kms(xc(x) – x).
A high value of the symmetry point xc(x) at small displacement amplitudes x » 0 indicates that the rest position does not agree with the minimum of the stiffness characteristic. This may be caused by an asymmetry in the geometry of the spider (cup form) or surround (half wave). A high value of the symmetry point xc(x) at maximal displacement x» xmax may be caused by asymmetric limiting of the surround.


Parameters at the Rest Position
For accurate system modelling “Large + Cold” parameters are preferable to “Small Signal” parameters because they more closely reflect the parameters in their typical operating range.
Symbol Large + Warm Large + Cold Small Signal Unit Comment Note:for accurate small signal parameters, use LPM module———————–———————————————————————————————————————————————–Delta Tv [referenced]79.70Kincrease of voice coil temperature during the measurement referenced to imported Re(Delta Tv=0K)Delta Tv = Tv-Ta52.700Kincrease of voice coil temperature during the measurementXprot12.712.71.9mmmaximal voice coil excursion (limited by protection system)Re (Tv)4.804.004.00Ohmvoice coil resistance considering increase of voice coil temperature TvLe (X=0)0.220.220.20mHvoice coil inductance at the rest position of the voice coilL2 (X=0)1.161.160.47mHpara-inductance at the rest position due to the effect of eddy currentR2 (X=0)0.550.550.77Ohmresistance at the rest position due to eddy currentsCmes (X=0)544544490µFelectrical capacitance representing moving massLces (X=0)100.37100.3780.02mHelectrical inductance at the rest position representing driver complianceRes (X=0)95.3695.3637.27Ohmresistance at the rest position due to mechanical lossesQms (X=0, Tv)7.027.022.92mechanical Q-factor considering the mechanical system onlyQes (Tv)0.290.250.31electrical Q-factor considering Re (Tv) onlyQts (X=0, Tv)0.280.240.28total Q-factor considering Re (Tv) and Rms onlyfs21.521.525.4Hzdriver resonance frequencyMms26.87526.875g(calculated from imported Bl) mechanical mass of driver diaphragm assembly including voice-coil and air loadRms (X=0)0.5180.5181.591kg/smechanical resistance of total-driver lossesCms (X=0)2.032.031.35mm/Nmechanical compliance of driver suspension at the rest positionKms (X=0)0.490.490.74N/mmmechanical stiffness of driver suspension at the rest positionBl (X=0)7.707.707.70N/A(imported) force factor at the rest position (Bl product)Vas50.426750.426733.4917lequivalent air volume of suspensionN00.1640.1970.168%reference efficiency (2Pi-sr radiation using Re)Lm84.385.184.4dBcharacteristic sound pressure levelSd132.70132.70132.70cm²diaphragm area

Frequency Response
Frequency Response data is generated using Klippel’s Transfer Function Mueasurement module in conjunction with Klippel’s In-Situ Compensation module. This pairing enables me to generate anechoic-quality measurements in the farfield with a resolution of response down to 20Hz increments; far better than splicing nearfield with gated farfield data (the latter of which is often limited to 200Hz or 300Hz resolution and will not show high-Q resonance in the midrange and bass frequencies). Data is represented at 2.83v/1m.

On-Axis Linearity and Impedance




On and Off-Axis Frequency Response






Total Harmonic Distortion (THD) and Compression:
Klippel’s 3D-DISTORTION MEASUREMENT (DIS) module is used to calculate the Total Harmonic Distortion and Compression for this drive unit.
Distortion and Compression measurements were completed in the nearfield (approximately 0.3 meters). However, SPL provided is relative to 1 meter distance.
Harmonic Distortion and Compression: What does this data mean? (click me for info)











Maximum Long Term SPL (Multitone Distortion Testing)
Klippel’s Multi-Tone Measurement (MTON) module is used to calculate the maximum SPL for this drive unit.
The below data provides the metrics for how Maximum Long Term SPL is determined. This measurement follows the IEC 60268-21 Long Term SPL protocol, per Klippel’s template, as such:
  • Rated maximum sound pressure according IEC 60268-21 §18.4
  • Using broadband multi-tone stimulus according §8.4
  • Stimulus time = 60 s Excitation time + Preloops according §18.4.1
Each voltage test is 1 minute long (hence, the “Long Term” nomenclature).
The thresholds to determine the maximum SPL are:
  • -20dB Distortion relative to the fundamental
  • -3dB compression relative to the reference (1V) measurement
When the speaker has reached either or both above thresholds, the test is terminated and the SPL of the last test is the maximum SPL. In the below results I provide the summarized table as well as the data showing how/why this SPL was deemed to be the maximum.
This measurement is conducted twice:
  • First with a 20Hz to 20kHz multitone signal
  • Second with a limited bandwidth multitone signal
The reason for the two measurements is because it is unfair to expect a small bookshelf speaker to extend low in frequency. Applying both will provide a good idea of the limitations if you were to want to run a speaker full range vs using one with a typical 80Hz HPF. And you will have a way to compare various speakers’ SPL limitations with each other. However, note: the 80Hz signal is a “brick wall” and does not emulate a typical 80Hz HPF slope of 24dB/octave. But… it’s close enough.
You can watch a demonstration of this testing via my YouTube channel:
Test 1: 20Hz to 20kHz
Multitone compression testing. The red line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Multitone distortion testing. The dashed blue line represents the -20dB (10% distortion) threshold for failure. The dashed red line is for reference and shows the 1% distortion mark (but has no bearing on pass/fail). The green line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Test 2: 80Hz to 5kHz
Multitone compression testing. The red line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Multitone distortion testing. The dashed blue line represents the -20dB (10% distortion) threshold for failure. The dashed red line is for reference and shows the 1% distortion mark (but has no bearing on pass/fail). The green line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


The above data can be summed up by looking at the tables above but is provided here again:
  • Max SPL for 20Hz to 20kHz is approximately 97dB @ 1 meter. The compression threshold was exceeded above this SPL.
  • Max SPL for 80Hz to 5kHz is approximately 105dB @ 1 meter. The compression threshold was exceeded above this SPL.


Bottom Line
T/S Parameters and Linear Excursion:
  • My measured linear excursion is limited to about 6.9mm one-way. This is based on 10% distortion limits. This jives with the spec provided if one were to take the graphics and apply the displacement limits (75% for Cms). The product spec sheet states a linear xmax of 10mm one-way linear likely defined by the physical overhang of the voice coil [VC length (23.7mm) - gap height (4mm)].
  • This is good linear excursion for a speaker this size. However, the Scan Illuminator 18wu I tested in October 2011 still takes the cake in the 6-7 inch midwoofer linearity category with a Klippel verified 9.1mm one-way excursion (also limited by compliance).
  • My specs fall well within line of the manufacturer’s spec. Not much else to here.
Frequency Response:
  • The average sensitivity is measured at about 87dB from 300Hz to 1000Hz.
  • On-axis response linearity is ±1.5dB within 100-1kHz with some mild bumps here and there.
  • On-axis response linearity is ±3.0dB within 60-2.5kHz.
  • Breakup is kept to a very respectable +5dB above the mean SPL at 3kHz.
  • Off-axis response shows nice linearity until about 4kHz (evidenced by the 60° measurement showing a +4dB resonance at ~3.9kHz).
Distortion and Compression:
  • This is the best speaker I have tested (I don’t have the same distortion data for the Scan 18wu I mentioned above). An incredibly low 2% THD at 80Hz with 12vRMS (making for about 101dB @ 1 meter) and well below 1% THD above 100Hz.
  • The dominant mode of distortion is 2nd-order. But, really, it is so low… who cares what order is the contributor.
  • Compression is the best I have measured to date as well. Another incredible performance with only about 0.4dB of compression at 12vRMS input.
  • Multitone distortion gets above about 1% distortion thresholds between 2-4kHz. One-percent. At 101dB @ 1 meter equivalent. That’s low.
  • When supplied an 80Hz to 5kHz signal the multitone distortion is even lower up to 111dB @ 1 meter.


This speaker has the best distortion values I’ve seen with even higher SPL output capability than larger drivers. The 6.5-inch nominal speaker size combined with its excellent distortion and notable frequency response linearity make this a great option for those who want a smaller, compact 2-way bookshelf speaker. And it just so happens this particular drive unit came from a 2-way bookshelf DIY speaker that I am reviewing soon.


Contribute
If you like what you see here and want to help me keep it going, you can donate via PayPal here.

You can also join my Facebook and YouTube pages if you’d like to follow along with updates.
 
Last edited:

Doodski

Major Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
5,532
Likes
4,511
Location
Canada
#2
Another veryyy extensive review. Well done. :D They must be phenomenal drivers for ~$700.00 per pair.
 

briskly

Active Member
Joined
Jul 10, 2019
Messages
105
Likes
138
#3
Those tables don't transfer very easily.

Curiously, the manufacturer's Kms chart suggests parabolic stiffness rather than progressive. It was expected to be the first thing to give up linearity based on the provided data. But this is not a serious limitation for a driver.
Edit: I was remarking how badly it transferred; I was already reading the link.
 
Last edited:
OP
hardisj

hardisj

Major Contributor
Joined
Jul 18, 2019
Messages
2,299
Likes
10,171
Location
North Alabama
Thread Starter #4

yourmando

Active Member
Joined
Aug 14, 2020
Messages
132
Likes
158
#5
Here's the link to my original review because some of the tables and such won't easily translate here:
https://www.erinsaudiocorner.com/driveunits/purifi_ptt65w04/



Purifi Audio PTT6.5W04-01A 6.5 Inch Woofer Review
  • Sunday, Aug 16, 2020





Information
Per the manufacturer:
The PTT6.5W04 is a 6.5” driver that truly cracks the long-stroke code. PURIFI’s research has identified the parameters that have so far prevented long stroke drivers from breaking through in truly high performance audio. The distortion of the acoustic output of a driver is a combination of several separate distortion mechanisms in the motor, cone and suspension. When testing a complete driver with sine waves, a situation often arises where a small tweak seemingly improves harmonic distortion (HD) by letting two mechanisms counteract each other. This always leads to a clear worsening of intermodulation distortion (IMD) that becomes obvious when testing the unit with a more complex signal. This is why traditional HD tests fail to predict subjective sound quality. PURIFI takes care to optimize the various distortion mechanisms separately, thus guaranteeing that any measured improvement truly reflects a real audible improvement no matter what the signal is. Accurate mathematical models are developed for the motor, suspension and the vibroacoustics of the dome and surround. These models clearly explain several distortion mechanisms in each domain, which provides fundamental insights into better ways of constructing the motor, cone and suspension. Finally, the same models are used to numerically fine-tune the geometry. This way of working reliably reduces multiple distortions mechanisms which secures low IMD for complex signals.​
Retail price is about $740/pair from Madisound.com.

Product Spec Sheet can be found here.

Test Results
Foreword: Subjective Analysis vs Objective Data of Drive Units (click here)

Small Signal Testing (Thiele-Small Results)
Using Klippel’s Distortion Analyzer 2, Linear Lumped Parameter Measurement Module, Pro Driver Stand and provided Panasonic ANR12821 Laser along with Klippel’s Training 1 - Linear Lumped Parameter Measurement tutorial, I measured this drive unit’s impedance and small-signal parameters. Below are the results.
Note: The impedance graph can be found in the Frequency Response Linearity graphic later in this review.
Electrical Parameters
Re3.69Ohmelectrical voice coil resistance at DCLe0.218mHfrequency independent part of voice coil inductanceL20.385mHpara-inductance of voice coilR22.06Ohmelectrical resistance due to eddy current lossesCmes440.29µFelectrical capacitance representing moving massLces62.52mHelectrical inductance representing driver complianceRes84.51Ohmresistance due to mechanical lossesfs30.3Hzdriver resonance frequencyMechanical Parameters(using laser)Mms26.119gmechanical mass of driver diaphragm assembly including air load and voice coilMmd (Sd)24.391gmechanical mass of voice coil and diaphragm without air loadRms0.702kg/smechanical resistance of total-driver lossesCms1.054mm/Nmechanical compliance of driver suspensionKms0.95N/mmmechanical stiffness of driver suspensionBl7.702N/Aforce factor (Bl product)Lambda s0.007suspension creep factorLoss factorsQtp0.297total Q-factor considering all lossesQms7.092mechanical Q-factor of driver in free air considering Rms onlyQes0.309electrical Q-factor of driver in free air considering Re onlyQts0.296total Q-factor considering Re and Rms onlyOther ParametersVas26.2642lequivalent air volume of suspensionn00.228%reference efficiency (2 pi-radiation using Re)Lm85.78dBcharacteristic sound pressure level (SPL at 1m for 1W @ Re)Lnom86.13dBnominal sensitivity (SPL at 1m for 1W @ Zn)Sd132.70cm²diaphragm area


Large Signal Modeling (Linear Xmax Results)
Using Klippel’s Distortion Analyzer 2, Large Signal Identification Module, Pro Driver Stand and provided Panasonic ANR12821 Laser along with Klippel’s Training 3 - Loudspeaker Nonlinearities tutorial, I measured the linear, nonlinear and thermal parameters of this drive unit.
Nonlinearities
Traditionally, Xmax has been defined in one of the following ways:
  • the physical overhang of the voice coil (height of the voice coil relative to height of the gap)
  • 115% times the physical overhang above
  • the point where displacement limit(s) is/are exceeded
The third option is where the Klippel LSI module comes in to play. It permits a more “apples to apples” approach of defining the displacement (Xmax) limits based on the XBL, XC, XL and Xd. The displacement limits XBL, XC, XL and Xd describe the limiting effect for the force factor Bl(x), compliance Cms(x), inductance Le(x) and Doppler effect, respectively, according to the threshold values Blmin, Cmin, Zmax and d2 used by the operator.
There are one of two sets of thresholds which can be used to define linear excursion:
  1. Non-Subwoofer Drivers: The thresholds Blmin= 82 %, Cmin=75 %, Zmax=10 % and d2=10% generate for a two-tone-signal (f1=fs, f2=8.5fs) 10 % total harmonic distortion and 10 % intermodulation distortion.
  2. Subwoofer Drivers: The thresholds Blmin= 70 %, Cmin=50 %, Zmax=17 % create 20 % total harmonic distortion which is becoming the standard for acceptable subwoofer distortion thresholds.
These parameters are defined in more detail in the (Klippel) papers:
  • “AN04 – Measurement of Peak Displacement Xmax”
  • “AN05 - Displacement Limits due to Driver Nonlinearities”
  • “AN17 - Credibility of Nonlinear Parameters”
  • “Prediction of Speaker Performance at High Amplitudes”
  • “Assessment of Voice Coil Peak Displacement Xmax”
  • “Assessing Large Signal Performance of Loudspeakers”


Below are the displacement limits’ results for this drive unit obtained from Klippel’s LSI module:
Displacement Limits
Code:
X Bl @ Bl min=82%    9.1    mm    Displacement limit due to force factor variation
X C @ C min=75%    6.9    mm    Displacement limit due to compliance variation
X L @ Z max=10 %    >11.2    mm    Displacement limit due to inductance variation
X d @ d2=10%    42.1    mm    Displacement limit due to IM distortion (Doppler)


Per the above table, this drive unit's linear excursion is limited to 6.9mm due to exceeding the Cms displacement limit of 75% for the distortion limit of 10%.

We can break the above information down further. The below text is written by Patrick Turnmire of Red Rock Acoustics and used with his permission, substituting data from this drive unit’s test results where applicable.
Large Signal Modeling
At higher amplitudes, loudspeakers produce substantial distortion in the output signal, generated by nonlinear ties inherent in the transducer. The dominant nonlinear distortions are predictable and are closely related with the general principle, particular design, material properties and assembling techniques of the loudspeaker. The Klippel Distortion Analyzer combines nonlinear measurement techniques with computer simulation to explain the generation of the nonlinear distortions, to identify their physical causes and to give suggestion for constructional improvements. Better insight into the nonlinear mechanisms makes it possible to further optimize the transducer in respect with sound quality, weight, size and cost.
Nonlinear Characteristics
The dominant nonlinearities are modelled by variable parameters such as
Bl(x)instantaneous electro-dynamic coupling factor (force factor of the motor) defined by the integral of the magnetic flux density B over voice coil length l as a function of displacementKMS(x)mechanical stiffness of driver suspension a function of displacementLE(i)voice coil inductance as a function of input current (describes nonlinear permeability of the iron path)LE(x)voice coil inductance as a function of displacement

Nonlinear Parameters
Bl change with excursion



The electrodynamic coupling factor, also called Bl-product or force factor Bl(x), is defined by the integral of the magnetic flux density B over voice coil length l and translates current into force. In traditional modeling this parameter is assumed to be constant. The force factor Bl(0) at the rest position corresponds with the Bl-product used in linear modeling. The red curve displays Bl over the entire displacement range covered during the measurement. You see the typical decay of Bl when the voice coil moves out of the gap. At the end of the measurement, the black curve shows the confidential range (interval where the voice coil displacement in this range occurred 99% of the measurement time). During the measurement, the black curve shows the current working range. The dashed curve displays Bl(x) mirrored at the rest position of the voice coil – this way, asymmetries can be quickly identified. Since a laser was connected during the measurement, a coil in / coil out marker is displayed on the bottom left / bottom right.

Suspension Stiffness change with excursion


The stiffness KMS(x) describes the mechanical properties of the suspension. Its inverse is the compliance CMS(x).

Inductance change with excursion





The inductance components Le (x) and Bl(i) of most drivers have a strong asymmetric characteristic. If the voice coil moves towards the back plate the inductance usually increases since the magnetic field generated by the current in the voice coil has a lower magnetic resistance due to the shorter air path. The nonlinear inductance Le(x) has two nonlinear effects. First the variation of the electrical impedance with voice coil displacement x affects the input current of the driver. Here the nonlinear source of distortion is the multiplication of displacement and current. The second effect is the generation of a reluctance force which may be interpreted as an electromagnetic motor force proportional to the squared input current.
The flux modulation Bl(i) has two effects too. On the electrical side the back EMF Bl(i)*v produces nonlinear distortion due to the multiplication of current i and velocity v. On the mechanical side the driving force F = Bl(i)*i comprises a nonlinear term due to the squared current i. This force produces similar effects as the variable term Le(x).

Fs shift with excursion


Qts change with excursion



Asymmetrical Nonlinearities
Asymmetrical nonlinearities produce not only second- and higher-order distortions but also a dc-part in the displacement by rectifying low frequency components. For an asymmetric stiffness characteristic the dc-components moves the voice coil for any excitation signal in the direction of the stiffness minimum. For an asymmetric force factor characteristic the dc-component depends on the frequency of the excitation signal. A sinusoidal tone below resonance (f<fS) would generate or force moving the voice coil always in the force factor maximum. This effect is most welcome for stabilizing voice coil position. However, above the resonance frequency (f>fS) would generate a dc-component moving the voice coil in the force factor minimum and may cause severe stability problems. For an asymmetric inductance characteristic the dc-component moves the voice coil for any excitation signal in the direction of the inductance maximum. Please note that the dynamically generated DC-components cause interactions between the driver nonlinearities. An optimal rest position of the coil in the gap may be destroyed by an asymmetric compliance or inductance characteristic at higher amplitudes. The module Large Signal Simulation (SIM) allows systematic investigation of the complicated behavior.
Bl symmetry xb(x)
This curve shows the symmetry point in the nonlinear Bl-curve where a negative and positive displacement x=xpeak will produce the same force factor Bl(xb(x) + x) = Bl(xb(x) – x).
If the shift xb(x) is independent on the displacement amplitude x then the force factor asymmetry is caused by an offset of the voice coil position and can be simply compensated.
If the optimal shift xb(x) varies with the displacement amplitude x then the force factor asymmetry is caused by an asymmetrical geometry of the magnetic field and cannot completely be compensated by coil shifting.



Kms Symmetry xc(x)
This curve shows the symmetry point in the nonlinear compliance curve where a negative and positive displacement x=xpeak will produce the same compliance value kms(xc(x) + x) = kms(xc(x) – x).
A high value of the symmetry point xc(x) at small displacement amplitudes x » 0 indicates that the rest position does not agree with the minimum of the stiffness characteristic. This may be caused by an asymmetry in the geometry of the spider (cup form) or surround (half wave). A high value of the symmetry point xc(x) at maximal displacement x» xmax may be caused by asymmetric limiting of the surround.


Parameters at the Rest Position
For accurate system modelling “Large + Cold” parameters are preferable to “Small Signal” parameters because they more closely reflect the parameters in their typical operating range.
Symbol Large + Warm Large + Cold Small Signal Unit Comment Note:for accurate small signal parameters, use LPM module———————–———————————————————————————————————————————————–Delta Tv [referenced]79.70Kincrease of voice coil temperature during the measurement referenced to imported Re(Delta Tv=0K)Delta Tv = Tv-Ta52.700Kincrease of voice coil temperature during the measurementXprot12.712.71.9mmmaximal voice coil excursion (limited by protection system)Re (Tv)4.804.004.00Ohmvoice coil resistance considering increase of voice coil temperature TvLe (X=0)0.220.220.20mHvoice coil inductance at the rest position of the voice coilL2 (X=0)1.161.160.47mHpara-inductance at the rest position due to the effect of eddy currentR2 (X=0)0.550.550.77Ohmresistance at the rest position due to eddy currentsCmes (X=0)544544490µFelectrical capacitance representing moving massLces (X=0)100.37100.3780.02mHelectrical inductance at the rest position representing driver complianceRes (X=0)95.3695.3637.27Ohmresistance at the rest position due to mechanical lossesQms (X=0, Tv)7.027.022.92mechanical Q-factor considering the mechanical system onlyQes (Tv)0.290.250.31electrical Q-factor considering Re (Tv) onlyQts (X=0, Tv)0.280.240.28total Q-factor considering Re (Tv) and Rms onlyfs21.521.525.4Hzdriver resonance frequencyMms26.87526.875g(calculated from imported Bl) mechanical mass of driver diaphragm assembly including voice-coil and air loadRms (X=0)0.5180.5181.591kg/smechanical resistance of total-driver lossesCms (X=0)2.032.031.35mm/Nmechanical compliance of driver suspension at the rest positionKms (X=0)0.490.490.74N/mmmechanical stiffness of driver suspension at the rest positionBl (X=0)7.707.707.70N/A(imported) force factor at the rest position (Bl product)Vas50.426750.426733.4917lequivalent air volume of suspensionN00.1640.1970.168%reference efficiency (2Pi-sr radiation using Re)Lm84.385.184.4dBcharacteristic sound pressure levelSd132.70132.70132.70cm²diaphragm area

Frequency Response
Frequency Response data is generated using Klippel’s Transfer Function Mueasurement module in conjunction with Klippel’s In-Situ Compensation module. This pairing enables me to generate anechoic-quality measurements in the farfield with a resolution of response down to 20Hz increments; far better than splicing nearfield with gated farfield data (the latter of which is often limited to 200Hz or 300Hz resolution and will not show high-Q resonance in the midrange and bass frequencies). Data is represented at 2.83v/1m.

On-Axis Linearity and Impedance




On and Off-Axis Frequency Response






Total Harmonic Distortion (THD) and Compression:
Klippel’s 3D-DISTORTION MEASUREMENT (DIS) module is used to calculate the Total Harmonic Distortion and Compression for this drive unit.
Distortion and Compression measurements were completed in the nearfield (approximately 0.3 meters). However, SPL provided is relative to 1 meter distance.
Harmonic Distortion and Compression: What does this data mean? (click me for info)











Maximum Long Term SPL (Multitone Distortion Testing)
Klippel’s Multi-Tone Measurement (MTON) module is used to calculate the maximum SPL for this drive unit.
The below data provides the metrics for how Maximum Long Term SPL is determined. This measurement follows the IEC 60268-21 Long Term SPL protocol, per Klippel’s template, as such:
  • Rated maximum sound pressure according IEC 60268-21 §18.4
  • Using broadband multi-tone stimulus according §8.4
  • Stimulus time = 60 s Excitation time + Preloops according §18.4.1
Each voltage test is 1 minute long (hence, the “Long Term” nomenclature).
The thresholds to determine the maximum SPL are:
  • -20dB Distortion relative to the fundamental
  • -3dB compression relative to the reference (1V) measurement
When the speaker has reached either or both above thresholds, the test is terminated and the SPL of the last test is the maximum SPL. In the below results I provide the summarized table as well as the data showing how/why this SPL was deemed to be the maximum.
This measurement is conducted twice:
  • First with a 20Hz to 20kHz multitone signal
  • Second with a limited bandwidth multitone signal
The reason for the two measurements is because it is unfair to expect a small bookshelf speaker to extend low in frequency. Applying both will provide a good idea of the limitations if you were to want to run a speaker full range vs using one with a typical 80Hz HPF. And you will have a way to compare various speakers’ SPL limitations with each other. However, note: the 80Hz signal is a “brick wall” and does not emulate a typical 80Hz HPF slope of 24dB/octave. But… it’s close enough.
You can watch a demonstration of this testing via my YouTube channel:
Test 1: 20Hz to 20kHz
Multitone compression testing. The red line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Multitone distortion testing. The dashed blue line represents the -20dB (10% distortion) threshold for failure. The dashed red line is for reference and shows the 1% distortion mark (but has no bearing on pass/fail). The green line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Test 2: 80Hz to 5kHz
Multitone compression testing. The red line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


Multitone distortion testing. The dashed blue line represents the -20dB (10% distortion) threshold for failure. The dashed red line is for reference and shows the 1% distortion mark (but has no bearing on pass/fail). The green line shows the final measurement where either distortion and/or compression failed. The voltage just before this is used to help determine the maximum SPL.


The above data can be summed up by looking at the tables above but is provided here again:
  • Max SPL for 20Hz to 20kHz is approximately 97dB @ 1 meter. The compression threshold was exceeded above this SPL.
  • Max SPL for 80Hz to 5kHz is approximately 105dB @ 1 meter. The compression threshold was exceeded above this SPL.


Bottom Line
T/S Parameters and Linear Excursion:
  • My measured linear excursion is limited to about 6.9mm one-way. This is based on 10% distortion limits. This jives with the spec provided if one were to take the graphics and apply the displacement limits (75% for Cms). The product spec sheet states a linear xmax of 10mm one-way linear likely defined by the physical overhang of the voice coil [VC length (23.7mm) - gap height (4mm)].
  • This is good linear excursion for a speaker this size. However, the Scan Illuminator 18wu I tested in October 2011 still takes the cake in the 6-7 inch midwoofer linearity category with a Klippel verified 9.1mm one-way excursion (also limited by compliance).
  • My specs fall well within line of the manufacturer’s spec. Not much else to here.
Frequency Response:
  • The average sensitivity is measured at about 87dB from 300Hz to 1000Hz.
  • On-axis response linearity is ±1.5dB within 100-1kHz with some mild bumps here and there.
  • On-axis response linearity is ±3.0dB within 60-2.5kHz.
  • Breakup is kept to a very respectable +5dB above the mean SPL at 3kHz.
  • Off-axis response shows nice linearity until about 4kHz (evidenced by the 60° measurement showing a +4dB resonance at ~3.9kHz).
Distortion and Compression:
  • This is the best speaker I have tested (I don’t have the same distortion data for the Scan 18wu I mentioned above). An incredibly low 2% THD at 80Hz with 12vRMS (making for about 101dB @ 1 meter) and well below 1% THD above 100Hz.
  • The dominant mode of distortion is 2nd-order. But, really, it is so low… who cares what order is the contributor.
  • Compression is the best I have measured to date as well. Another incredible performance with only about 0.4dB of compression at 12vRMS input.
  • Multitone distortion gets above about 1% distortion thresholds between 2-4kHz. One-percent. At 101dB @ 1 meter equivalent. That’s low.
  • When supplied an 80Hz to 5kHz signal the multitone distortion is even lower up to 111dB @ 1 meter.


This speaker has the best distortion values I’ve seen with even higher SPL output capability than larger drivers. The 6.5-inch nominal speaker size combined with its excellent distortion and notable frequency response linearity make this a great option for those who want a smaller, compact 2-way bookshelf speaker. And it just so happens this particular drive unit came from a 2-way bookshelf DIY speaker that I am reviewing soon.


Contribute
If you like what you see here and want to help me keep it going, you can donate via PayPal here.

You can also join my Facebook and YouTube pages if you’d like to follow along with updates.
Excellent review! That was a lot of work!

May I ask which DIY bookshelf speakers using the Purifi’s that you will be reviewing? It‘s nice that’s some DIY kits are coming out.
 
OP
hardisj

hardisj

Major Contributor
Joined
Jul 18, 2019
Messages
2,299
Likes
10,171
Location
North Alabama
Thread Starter #6

yourmando

Active Member
Joined
Aug 14, 2020
Messages
132
Likes
158
#9
Your reviews and videos are awesome!

Do you have the full Klippel with near field robot whirligig like Amir?

I’ve noticed you can produce full Spinorama and 360 degree polar response plot—is that possible w/o the ~$100k in options?

May I ask what it would roughly cost to get a Klippel test kit + modules for 2034 spin loudspeaker testing?
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
1,760
Likes
6,364
Location
NYC
#10
Your reviews and videos are awesome!

Do you have the full Klippel with near field robot whirligig like Amir?

I’ve noticed you can produce full Spinorama and 360 degree polar response plot—is that possible w/o the ~$100k in options?

May I ask what it would roughly cost to get a Klippel test kit + modules for 2034 spin loudspeaker testing?
I'm not Erin but he talks about figuring out his measurement process in another thread. He does not have the Klippel NFS Amir has but does have some Klippel gear which I believe is primarily oriented at testing drivers-- I'll let him clarify.

From what I recall he is using gated measurements for the high frequencies and ground plane for the low frequencies, but he's manually doing each of these measurements so it's very labor intensive and definitely the most thorough manual speaker measurements I've seen. He also has a custom Matlab script for presenting the polar and spinorama data.

Lazy bum that I am, I settle for using gated measurements and a nearfield splice of the on-axis bass. VituixCad can 'guess' the rest of the angles, which is normally fine but becomes tricky for speakers with unusual bass directivity.
 

yourmando

Active Member
Joined
Aug 14, 2020
Messages
132
Likes
158
#11
I'm not Erin but he talks about figuring out his measurement process in another thread. He does not have the Klippel NFS Amir has but does have some Klippel gear which I believe is primarily oriented at testing drivers-- I'll let him clarify.

From what I recall he is using gated measurements for the high frequencies and ground plane for the low frequencies, but he's manually doing each of these measurements so it's very labor intensive and definitely the most thorough manual speaker measurements I've seen. He also has a custom Matlab script for presenting the polar and spinorama data.

Lazy bum that I am, I settle for using gated measurements and a nearfield splice of the on-axis bass. VituixCad can 'guess' the rest of the angles, which is normally fine but becomes tricky for speakers with unusual bass directivity.
Excellent. Love the ‘selfie’ trick to make sure the mic is centered on the tweeter. I’ll read that whole thread later.

@napilopez, did you happen to write up how you do your testing, or have you followed a particular process? I‘ve run across a couple of your measurement posts with interest as well. Thanks!
 
Top Bottom