JBL 660GTi Midwoofer Testing

This testing is for only the midwoofer from JBL’s 660GTi Component Set.  This is said to be JBL’s flagship component speaker and the way it’s presented to the consumer, complete with steel briefcase and intricate packaging, certainly make it seem that way.  My friend owns this set and let me borrow it all to test the mid to see just how well this particular driver in the set performs on its own.  At some point, I’d like to test the tweeter in the waveguide, but for now I’ll just test the midwoofer.

As a side note, this driver will be used in a co-hobbyist’s upcoming Subjective Evaluation Test and we both wanted to be able to see if there is a way to correlate subjective opinions vs objective measurements.

Please see this link for information on how the test was performed.  This saves me some much needed time.  😉


Thiele-Small Parameters and Impedance

Electrical Parameters
Re 3.04 Ohm electrical voice coil resistance at DC
Le 0.135 mH frequency independent part of voice coil inductance
L2 0.68 mH para-inductance of voice coil
R2 3.42 Ohm electrical resistance due to eddy current losses
Cmes 379 µF electrical capacitance representing moving mass
Lces 16.97 mH electrical inductance representing driver compliance
Res 72.14 Ohm resistance due to mechanical losses
fs 62.7 Hz driver resonance frequency
f ct 70 Hz driver resonance frequency in enclosure
Mechanical Parameters
(using test encl.)
Mms 10.36 g mechanical mass of driver diaphragm assembly including air load and voice coil
Mmd (Sd) 8.86 g mechanical mass of voice coil and diaphragm without air load
Rms 0.379 kg/s mechanical resistance of  total-driver losses
Cms 0.621 mm/N mechanical compliance of driver suspension
Kms 1.61 N/mm mechanical stiffness of driver suspension
Bl 5.227 N/A force factor (Bl product)
Loss factors
Qtp 0.439 total Q-factor considering all losses
Qms 10.784 mechanical Q-factor of driver in free air considering Rms only
Qes 0.455 electrical Q-factor of driver in free air considering Re only
Qts 0.437 total Q-factor considering Re and Rms only
Other Parameters
Vas 12.8174 l equivalent air volume of suspension
n0 0.669 % reference efficiency (2 pi-radiation using Re)
Lm 90.45 dB characteristic sound pressure level (SPL at 1m for 1W @ Re)
Lnom Zn missing dB nominal sensitivity (SPL at 1m for 1W @ Zn)
Sd 120.76 cm² diaphragm area




Large Signal Analysis

This analysis module from Klippel is the cream of the crop.  It provides significant insight in to the inner workings of the driver under test in a way that some manufacturers can not attain.  What you’ll see below is this driver has a very long linear stroke providing a measured mechanical Xmax of approximately 11.3mm (one-way) with a corresponding linear Xmax (xmax value defined by the 10% distortion limit) at 7.4mm (one-way).  From my previous testing, this is 3rd only to the Exodus Anarchy (with a measured 7.7mm) and the Scan Speak 18wu (9.1mm).  Note the mounting depth of the JBL driver is roughly 20mm less than the Exodus and 24mm less than the Scan.  Nearly 3/4″ and 1″, respectively, less.

Table Summary

Displacement Limits thresholds can be changed in Processing property page
X Bl @ Bl min=82% 7.4 mm Displacement limit due to force factor variation
X C @ C min=75% 8.9 mm Displacement limit due to compliance variation
X L @ Z max=10 % >10.3 mm Displacement limit due to inductance variation
X d @ d2=10% 18.7 mm Displacement limit due to IM distortion (Doppler)
Asymmetry (IEC 62458)
Ak -10.42 % Stiffness asymmetry Ak(Xpeak)
Xsym 0.19 mm Symmetry point of Bl(x) at maximal excursion


As seen above, the suspension has coil out offset at rest of about 0.4mm and a coil in offset at mechanical xmax of 0.52mm.

The force factor shows an outward offset at rest of nearly 2mm with no assymetry at full excursion at 11.2mm.


The above graphs show inductance is a very small contributing factor to the overall distortion contribution where force factor (Bl) is the highest factor, and suspension is 2nd highest.

What this all means is that this driver’s motor and suspension are well implemented designs and keep the driver’s stroke very linear in regards to distortion characteristics which gives it a very high Xmax value, especially given it’s smaller size compared to the two highest rated Xmax drivers I mentioned above.


Frequency Response and Harmonic Distortion at 2.83v/1m

Good usable range in all axes up to about 3khz where the rolloff at 60 degrees begins to become more severe.  At 30 degrees off axis, the response is good to about 4khz.  On-axis response is good to about this same spot where there is a bump in response, which likely helps the off-axis response at 30 degrees.

Overall, HD isn’t bad.  Comprised mainly of 2nd order distortion.

High Output Harmonic Distortion

At higher output levels you can see that 3rd order distortion becomes practically a non-issue.  The separation in odd order harmonics from even (and namely the 2nd order component) is superb with nearly a constant 20dB delta between 3rd and 2nd order above 200hz and 10dB difference at 80hz.

The areas to potentially be concerned with would be the areas with the higher peaks in distortion.  Namely 1khz and 4khz.  However, my personal threshold being 3% THD, these are still below that.  Below 60hz, the driver exceeds that 3% THD mark and 3rd order begins to increase to a point where it more closely matches 2nd order.

From the fundamental response, you can see the driver plateaus before rolling off on the low end at about 72hz.

As mentioned above, this driver non-linear distortion profile is nearly all 2nd order distortion.  Given the tradeoff, this is a good one and will be less offensive (if audible at all) to the listener than 3rd order and subsequent odd order distortion.

Intermodulated Distortion (Voice Sweep)

As with HD testing above, these two results show a higher 2nd order distortion profile than 3rd order distortion, which contributes almost exclusively to the THD results.

You can see the difference in distortion levels as the input voltage is increased as well.

IMD results indicates a rising distortion profile above 2khz in 2nd order, while 3rd order distortion remains nearly constant (keep in mind the scale).

Compression is simply a means to display how much output is lost at higher volumes compared to the initial measurement.  In this case, at about 1khz the largest amount of output is lost.  However, this is only 0.5dB loss from 1v input to 6v input.

Intermodulated Distortion (Bass Sweep)

This is where you can see 3rd order distortion becoming an issue.  However, it’s maximum is 1.8% at about 33hz.  This is well beyond where anyone would likely be running this driver.

Note the larger delta in the stepped voltage measurements between 4.33v and 6.0v compared to the other intervals in this measurement.

Second order distortion is still the top dog here.

Using 50hz as a reference point, there is less than 0.20dB loss due to compression.


The bottom line is this driver is extremely impressive.  The 3rd largest linear Xmax I’ve measured with excellent mechanical throw overall.  Excellent distortion curves with an extremely low 3rd order distortion component.  When used from 80hz to 2khz, this is probably one of the top tier drivers I’ve measured to date.  It can be used beyond that passband but individual user needs will dictate experience.  It definitely has the gusto to take some loud listening levels and contain its composure.

5 thoughts on “JBL 660GTi Midwoofer Testing

  1. Thanks for testing this driver, Erin. Definitely a top-notch driver in its size category. Note that JBL is no longer producing these and they are not available from Harmon/JBL either as a complete component set or as replacement parts. Go figure, another absolute top-tier driver that’s been discontinued! What gives? Maybe Andy Wehmeyer should buy the licensing & tooling from his former buddies at Harmon/JBL and reproduce these under his Audiofrog line! Please Andy? :p

  2. UPDATE: These exact same drivers (and associated tweeters) are now available again in the new JBL 670GTi Component Set (announced at CES 2015) with a MSRP of $999.95. They are available at http://www.jbl.com/new-speakers/670GTI.html and at Sonic Electronix. New to the 670GTi set are “smaller” Crossovers, and now it Does Not include the Leather Speaker Grills for the Mids, nor the large Waveguides for the tweeters. The real Carbon-Fiber Grills are still included for the mids along with multiple mounting options for the tweeters.

Leave a Reply

Your email address will not be published. Required fields are marked *