Subjective Blind ABX Test of EnABLed FF85WK - Round 6

[xrk971]

This is a subjective blind test, 6th in series of tests of full range drivers, but with this one concentrating on whether or not eNable treatments to a FF85WK driver are audible. In a related thread, I showed that the differences were measurable as small 1dB variations in the 1kHz to 10kHz range. I then developed a passive FAST speaker to allow the testing with a state of the art class AB MOSFET amp (VHEX+) with an external DAC (UCA202) driving the amp, a passive crossover design, and a transient perfect FAST speaker utilizing an RS225-8 woofer in a sealed cabinet. The driver under test (DUT) is mounted in a Dagger style short aperiodic (vented) TL with a trapezoidal baffle for a clear, open sound signature with minimal box coloration or diffraction effects from a rectangular baffle.

The usual warnings: please do not post your selection plainly visible for others to see as it may sway votes. If you wish to have your vote recorded for posterity, create a txt file and put your vote in there and upload as an attachment. This way, people have to purposely click on attachment if they want to be swayed by popularity.

Related threads that led to this thread:

http://www.diyaudio.com/forums/full-range/292130-ff85wk-rs225-8-passive-fast.html

http://www.diyaudio.com/forums/full-range/291855-planet-10-measured-improvements-ff85wk.html

Here is a photo of the test setup:

Here is a photo of the VHEX+ amp used:

Here is the measured frequency response and XO response of the FAST system:

Here is the measured phase:

Here is the measured distortion:

Here is the measured impulse and step response:

The crossover frequency is at 700Hz (relatively high for a FAST but was necessitated by practical limitations of a passive crossover components being of reasonable size/cost). Here is the schematic diagram for the XO used (caps C2 & C3 are of 2.2uF 250V MKP film, C1 is a 50uF 450V film in oil industrial motor run cap, and resistors R1 & R2 are 10W cement filled and R3 is a 2W thin metal film for the Zobel, inductors were all ferrite core and cobbled together from 3 separate inductors (2.50mH+1.02mH+0.56mH) and R4 is the inherent DCR of the 3 inductors in series:

Measurements made at 0.5m on axis with DUT and using UMIK-1 microphone calibrated by third party (Cross Spectrum labs) and speaker excitation drive voltage was 2.83v (as measured with sine wave at several fixed frequencies).

Here is the detailed measurement of the differences between the drivers when viewed at 1dB/div vertical and with 6-cycle frequency dependent window (FDW) gating applied to remove external room contributions:

A more conventional view of the differences without gating looks like this (1/12th octave smoothing):

Even farther back, it looks like this (high pass only on XO):

For reference, here is the Fostex factory frequency response for FF85WK:

An externally hosted image should be here but it was not working when we last tested it.

Attached below are the level-matched reference sound clips used for the test recordings which were all made at the same time as the above measurements in one session with the mic and speaker untouched and only drivers swapped out.

The test will be to choose the set of sound clips (A or B) which you think sounds most like the unknown driver under test (X).

You will not know the identity of X a priori, the test is not which sounds better, but which one sounds the same. Although since you have the reference clips, you are free to make up your own mind which you think sounds closer to the source.

As usual, to circumvent the max file size limit, 320kbps MP3's are named with .asc extension. Change to .MP3 to listen.

Thanks for playing and have fun.

I will ask a moderator to make this into a poll with choices A or B and leave poll open for about 2 weeks - let's say by 12:00 GMT on June 19, 2016.

Finally, I would like to acknowledge Byrtt's generosity with help in buying some of the drivers under test and for designing the passive XO for this speaker. Although many have assisted in providing drivers in earlier tests, without Byrtt, none of these Subjective Blind Test threads would have been possible. Thank you, Byrtt!

Edit: Instructions on how to rename files on a Mac

From Peter Brorsson:

On my Mac, one-finger tap on touchpad, wait two seconds tap again. Erase asc and type mp3

Left corner click touchpad and same procedure as above works also. But it depends how track pad is configured....

Maybe there is a quicker way, but this is how I solved it.

Peter

From Vzaichenko:

Also, tapping the track pad with 2 fingers with the pointer on particular file brings up a context menu.
Then "Rename", etc.

Results of Poll - the big reveal here:

http://www.diyaudio.com/forums/full...t-enabled-ff85wk-round-6-a-8.html#post4752205

 

[xrk971]

Unknown X sound clips

Here are the four sound clips for the unknown driver under test.

Change extension to .mp3 to listen.

 

[xrk971]

Unknown driver A sound clips

Here are the sound clips for unknown driver A.

Change extension to .mp3 to listen.

 

[xrk971]

Unknown driver B sound clips

Here are the sound clips for unknown driver B.

Change extension to .mp3 to listen.

 

[Cal Weldon]

Adding a poll, please be patient.

 

[xrk971]

Thanks, Cal. I hope you decide to play his time.

I finally listened to the clips and am quite surprised how good they sound - close to the source material. The measurements confirm that a flat response will sound like how the source material was meant to to be heard. Use good headphones.

 

[Cal Weldon]

Poll added. I couldn't find the option of setting the end date as you requested so I set it at 14 days, hence the odd end time.

 

[freddi]

one of the unknowns may sound closer to the reference tracks but I prefer the other and think perhaps a tiny bit of ambiance is added from the room despite close micing. Anyhow, the results are quite good but I'm on a set of bass heavy cans as my current amp won't power my Fostex. Somewhere are cheap AT which are more lively than the T50/20rp. May I vote for the preferred or do you wish the duller ?

 

[Bangla H]

FF85K

Hallo X.

I see you are crazy quick to come up with new arrangements in way of Fast
systems.
Very nice to see your engagement that way and also your big experience.

I have a problem to open the asc- Files, my computer do not open any conversation to me to dismiss that problem - not a good collaboration - should i pull out my hammer?- angry!
Who can it go work?- i am real to dumb, i think.

Because i have not the much possibilities in costs i must go simpler way, but
always interested what other do and find.

I like the little FF85K. If you look to technical details, you will like it more.

With lower listening level on Desk in TP140 Horn without XO correction i feel
the most of its magical is alive there.
Of course i only trust my ears- no budget to by much equipment.

Regards Bangla.

 

[BYRTT]

xrk971,
Thanks your work and sharing data will add my vote later : )


Bangla H,
Regarding asc-files you need to rename dot asc extension to dot mp3 as in below visual, that is example from Win7 OS where folder options into control panel have settings that allow one to see extension for file types.

 

[Jay]

this one concentrating on whether or not eNable treatments to a FF85WK driver are audible.

How far have you made sure that the difference is due to eNable (and not something else)??

 

[xrk971]

The only difference is the driver. It is treated per Planet10 as an eNabled driver - although that includes some putty in the basket to smooth the airflow as well.

Otherwise all is identical in test setup.

 

[Jay]

The only difference is the driver. It is treated per Planet10 as an eNabled driver - although that includes some putty in the basket to smooth the airflow as well.

From your measurement I think I saw the eNabled ones have higher sensitivity??? Where do you think that came from?

 

[xrk971]

From your measurement I think I saw the eNabled ones have higher sensitivity??? Where do you think that came from?

I don't know but speculate a few reasons.

1. The gloss coat adds a bit of stiffness to the cone making it behave more pistonic so less energy is lost on bend modes. Note that the resonant 10kHz peak is shifted slightly lower in frequency on both enabled units.

2. Planet10 tests the TS parameters from a large batch to match the parameters for the customer to within 3%. There is sensitivity variability in drivers as I have seen in my two pristine units (1.5dB). It may be luck of the draw that I was given two matched "top shelf" FF85WK's from the set of 20 or so that were tested. Lucky me.

3. The putty smoothing treatment on the basket provides better airflow on the back side and this improves sensitivity a bit.

4. The dots actually do something with sensitivity? Probably not.

 

[Jay]

The mass you add to the cone will lower sensitivity.

 

[Godzilla]

Planet 10 said the drivers he sent you were less efficient after treatment. He measured them. I'm not sure the treatments raise efficiency. If the pristine drivers were treated then measured we'd have accurate information to examine.

 

[xrk971]

So maybe these were really "hot" units to start with and even after the treatment they are 1dB above the stock ones I bought? Certainly 1dB difference is audible but very minor difference.

I think the question is whether or not we can detect a difference between a set of Planet10 eNabled drivers and a box stock pair I buy from Madisound. The results of this blind test will reveal that I think.

Folks should make note of which group they prefer the sound of (privately in txt file).

 

[jimbro]

X - I think you should have given some evaluation requirements such as this quote from you from the 'measured improvements' thread. "You should use good headphones and the best signal chain you can afford or have." It should also be made clear that even with the best equipment this does not meet the standards for true blind test evaluation and is basically for fun and interest and therefore no definite conclusion should be made on the merits of enabl. That said, it's guys like you who keep this forum going. Thanks

 

[xrk971]

Yes, I forgot to say: please use headphones and a good DAC and good headphone amp. It makes a difference!

For two very good sounding yet cost effective solutions - here is what I use:

UCA202 DAC ($20) - a well-reviewed DAC by Nwavguy (creator of O2 head amp).

JLH 1969 class A head phone amp ($20):

 

[xrk971]

I would be curious to know which clip people think sounds better, A or B? Please post as an attached text file.

Thanks,
X