THE CUSTOMER PREMISE

Seems simple enough, right? A nice computer setup running ROON and a pair of Focal Sopra's. The customer previously had Audience 2+2 Main speakers, but he's on the upgrade test path... so... Let's tighten it up.

THEN, then he wanted to add a sub because, well, none of these bookshelf type speakers really have enough low end for some people. Even if the sub integration is gentle and subtle, it always helps with the clarity, and to use a sort-of incorrect but hackneyed phrase, the "extension". So he got the JL Audio E112 sub which HAS a real LR 24dB/octave matched crossover in it; as you adjust the lowpass filter knob it correspondingly sends out everything ABOVE that freq out the RCA's that then go to your power amp.

And here we have it.

But the customer ALSO wanted to experiment (and A/B) the crossover IN the sub with the JL Audio CR1 crossover, which I designed to be THE most comprehensive and clean crossover device on the planet. You can use it Balanced and/or UNbalanced; all the ins and outs are separately buffered, so you can insert it anywhere, and the ins are Gold Neutrik combo jacks (they should be called Jills, right?) so they also accept 1/4" TRS plugs. Plus the absolute BEST killer feature is there's a bypass switch, so you can A/B your mains running full range only, -or- the mains running with the correctly integrated sub. This feature alone is the mind-blowing thing.

Most people have NO IDEA how much better things can be when correctly set up, because first, so few set it up correctly, and second, since humans have no real acoustic memory, plugging cables in and listening and then unplugging and replugging cables in takes WAY too long for an intelligent audio memory comparison.

He also has a DSPeaker Anti-Mode Dual Core after the computer and before the rest of the electronics. Initially he had pretty much everyhing set up wrong, so we had a whole series of emails and phone calls for some 8 months before I went to his location to finalize. The DSPeaker user manual is HERE  

So far so good.

THE FOCALS - WE BOTH "HEARD" IT

But the customer has very good and very interesting hearing. He was discerning the same tweeter "thing" which I have noticed in [other] Focals, there is a sort-of 'anomaly' between about 3.5k and 5k that is discernable, and it's one of those things once you know it's there you can't UNhear it. Distortion? Cone Breakup? Annnoying metal tweeters that shouldn't exist in the first place? Issue with the passive crossover components? Whatever it is, it is there. So yes, it could be removed with very delicate and subtle EQ. If someone had the time and patience they COULD either complain or re-engineer the internal passive crossover... I am also willing to bet — based on years of experience with speaker mabufacturers — that they DON'T use nor understand the CORRECT use of ferrofluid in tweeters, which is a project I have painstakingly researched and experimented with. Even though the DSPeaker device HAS a certain amount of EQ control in it, it would be difficult to put a (let's say) 1 dB dip from 3-5k in and play with it.

SO A BIT ABOUT FERROFLUID

OK, this literally has nothing to do with THIS PROJECT, but it its something audio peeps should know. Ferrofluid comes in a wide range of viscosities, like from water to heavy cold maple syrup. It also is available in different carrier chemicals, from alcohol to mineral oil to silicon based fluids, etc.

But what IS ferrofluid? Imagine you had a piece of recording tape and you scraped the oxide off with a razor blade, then put a bunch of this goo in a blender with some silicon fluid, and blended it until it was smoothly homogenized. Now you have a fluid that is NOT magnetic ITSELF (since the particles are both so small AND randomized), but it is SENSITIVE to a magnetic field, and it will "hang" in a magnetic field, like what's in the tiny gap where the voice coil moves in ANY regular dynamic loudspeaker driver. It's also cute in videos on yoo toob...

So when you put it in a tweeter, (inject it, actually, with like a #32 diabetic syringe) you accomplish a couple of things:
1)  you provide a cooling path for the heat from the voice coil to be conducted away, -and-
2)  you physically DAMPEN the cone movement.

Since EVERY loudspeaker is a physically moving device, most of these things mechanically overshoot. So as a test, you hook up your tweeter to an amp, feed it say, a 2KHz [clean, fast risetime] SQUARE WAVE, you put your laboratory grade test mic (B&K, Earthworks, etc) into your oscilloscope, and you look at the waveform an inch away from the front of the driver. Holy crap! It hardly looks like a square wave! It is typically ringing, which means the voice coil is overshooting, or it MAY look more like a sine wave, which means there is a high freq rolloff, and the voice coil is overdamped. Since there is no crossover connected at this point, the damping is not electrical; it's mechanical.

So you mix up a carefully controlled and measured calibrated mixture of ferrofluid, and you inject the gap area. Now see if the signal looks any better on the scope. No? Then experiment some more. Yes? Then OK, now you have put a square wave in and you get a square wave out. HiFi !

But the previous situations were that manufacturers didn't buy the correct type(s) of ferrofluid, therefore it was drying out mainly because of the heat cycling (especially in subwoofers) and therefore was not only becoming ineffective, but when it dried out the coil might now be rubbing, so the situation is worse. This is why many companies have shied away from using it. It is a critical engineering exercise, and it has to be optimized for each same series of drivers, and very carefully. IDEALLY the manufacturer would then test EVERY DRIVER on a test jig before it goes in the cabinet, but it's safe to say that won't typically happen. But it's ideal.

I now return you to your regularly scheduled programming...

THE ROOM

Carpeted but no substantial treatment; yes IF there were even modest tube traps in the corners it would be tighter. But above bass freqs, in the glorious male and female vocal ranges, this is where the customers' main interest was. He did NOT play this loud at all - in fact I'd have to say that 85 dBA (slow weighted C of course) was considered LOUD to him in this room, and 70-75 dB was a usual area and maximum. So there's not a huge excitement of the room modes, and the speaker feeling is what I'd call "medium intimate".

ABOUT SPL METERS

I often suggest to people that in order to determine the most "realness" in their setup, that they discern how loud a singer really is singing. You get a lovely SPL app for your iPhone, like this: SPLnFFT , or search for it on the APP Store. Note that this app in particular is both affordable and it calibrates the mic in each iPhone so it is quite accurate when compared to a $3000 B&K analyzer for example. If you're on an android device you have my condolences as you have over 250 similar apps and YOU have to somehow discern what really works, and MANY of them are simply named "Sound Meter". In any event, use SLOW WEIGHTED C for home measurements. "A Weighting" is used for legal measurements and reads LOWER; (which also means when there is a bar fight the measurement is in favor of the bar, not the complainant...) but C Weighting more closely approximates the human function.

ALIGNING

I did my usual placement, and the out-of-phase white noise aiming, and that immediately really tightened up the imaging to everyone's surprise. Like other customers, this fellow invited one of his local audio buddies over and so I very much enjoy the opportunity to show them all the details - the teacher in me comes out. When we did the Ethan Winer walk-across-the-room test there was one huge hole at 70 Hz - like 40 dB down (!) but it was not where a listener would be, so it essentially didn't matter. The response AT the listening position was pleasantly even, with some dips down low, 30 Hz or so, that for this customer really were of little concern.

There's the one sub and so after its best position was determined, I phase aligned it with the left main speaker, which it was closest to. While I don't necessarily NEED to experiment with crossover freqs (after 40 or so years of doing this...) often customers don't understand it at all, and so I sort of have to show them. I chose 90 Hz which is also the spot on the CR1 knobs where they are pointing straight up. I designed it that way. Now after a few careful iterations of the tests we were getting tighter and tighter. The various Sinatra tests were wonderfully revealing. I could brag that I could hear the Capitol Recording Studio room, but ok, that's a stretch. Still, it was clean and glorious, and had a beautiful emotional presence. But now both the customer and I were ready for yet another test...

THE KEFS

We both wanted to experiment with A/B'ing a pair of LS50's against the Focals. The Focals are, what, $14k and the KEFs are $1300, so about 10 TIMES as much. Does that mean... well not exactly. Because of this tiny yet noticable thing with the Focal tweeter (and it was identical in each cabinet) I really wanted to A/B and see what "flavor" the KEFs were. I should mention that since this first test I have done this repeatedly and in just about every instance, the customer gets the KEFs.

Don't laugh. It was a wonderful test.

VOICING AND "FLAVORS"

All these speaker companies "voice" their speakers. First, so-called "flat" sounds, well, flat. Second, it is VERY difficult to get both the FREQUENCY response AND the PHASE response flat, and these days, manufacturers are slowly learning that they are both important. For years and years, everyone from manufacturers to reviewers to audiophiles have obsessed about "frequency response". One reason is it's VERY difficult to measure OTHER things. As far as electronics goes, for years people measured THD (Total Harmonic Distortion) and it's an essentially useless measurement because it doesn't tell whether the distortion is even-ordered or odd-ordered, and they sound completely different. Even ordered is "woody", "woodsy", and "warm"; while odd ordered is "cold", "harsh", "metallic", and so on. WHY was this done? Because before there were spectrum analysis type measuring equipment there were famous HP (Hewlett Packard) test meters that ONLY showed THD. So not the "whole story".

So, as I said, speaker companies "voice" their speakers. As you listen to and learn different brands in your audiophile travels you will likely find that "B&W's sound the same", "Tannoys sound the same", "KEFs sound the same", "Focals sound the same", and so on. But they are different from each other! You get the idea. But SOME companies, like Dynaudio, especially years ago, very much surprised me -- nothing sounded "the same". A rep once brought a series of so-called "studio monitor models" into my store and I said, "How can this possibly be? NONE of these sound anything alike! Where's the 'reference'?"

So the customers' local dealer had a pair of the KEF LS50's we could try. So, yes, they are voiced...differently. IMO they are intended for a younger, millenial crowd than perhaps the more affluent crowd that might be buying Focals. Someone MIGHT say, "...well the Focals are more carefully manufactured, and the KEFs are just thrown together so you get what you pay for..." — I don't personally think so. But the KEFs have a slightly more pronounced, forward midrange, more intimate or closer you might say.

Now here was the complaint of mine about B&W's for 40 years: Imagine you are a kid in the US and you always wanted to go to a UK concert, say the Stones, or Led Zep, or even the Beatles. So you save up, take a cheap airline over to London, and you and your backpack stay in a youth hostel. You go to the abovementioned concert except you are sitting 50+ rows WAY back and the performers look like miniatures on the stage. This is my opinion of B&W's. They are SO British reserved, laid back, formal, that the music puts you to sleep it is so un-engaging. Now Tannoys are quite the opposite — in-your-face, beautifully 'present', alive, happy, forward, yet reserved in a MATURE kind of way. See? You CAN describe these things like wine. However like wine and sex, (and loudspeakers) you can talk about it endlessly but nothing is accomplished until you actually DO it.

OTHER SPEAKERS ?

OK, so we can talk about speakers and wine like this until we are nuts. But we have to listen and taste. WIth the KEFs, which by the way were opposite polarity than the Focals which immediately surprised me, we both quite simply liked them better. In fact I personally LOVE them. But in this case I'm not the customer, and this was a test to make him happy. I was also hoping the dealer was going to come by but they probably chickened out...this happens often. So many of them just wanna sell you shit, they don't REALLY want to know about real audio... but it was enlightening enough for us. Now at least there was an interesting "choice".

I am relatively sure that this customer is going to want to play with additional speakers as an experiment. Because the group delay in MAINS is so very small, other than the POLARITY issue, if he simply swaps speakers on the same physical stands, the sub's phase adjustment will be plenty close; the thing that might be different is the relative sensitivity of the mains, therefore the mains LEVEL should be nulled out in the DSPeaker settings. And further, to accomplish the "room curve", the mains should be turned DOWN 1, 2, 3,dB so there becomes the desired TILT in frequency response at the listening position. Yes IF you leave the null levels where they are, the response will be "flatter", but, again, most humans don't really like nor appreciate this.

Onward!