Fluid Audio FX80

Combining high output, extended low end and remarkable affordability, Fluid's newest point–source speakers offer a compelling option for engineers who need a bit of 'oomph'.

A couple of years ago I reviewed Fluid Audio's FPX7 nearfield monitor. Once a few early sample teething problems had been overcome, that monitor, the dual-concentric ribbon–tweeter–equipped FPX7, turned out to be a capable and engaging one. The subject of this review is somewhat more conventional than the FPX7 in sporting a dome tweeter, however it retains Fluid Audio's implementation of the dual-concentric driver concept. I'll write a bit about the history of dual-concentric drivers a few paragraphs down, but first a bit of description.

The FX80 is the larger of two new entry level-priced, US–designed and Far East–manufactured active monitors. The other model in the range, the FX50, is slightly smaller and slightly cheaper. As perhaps hinted at by the model names, the FX80 incorporates a nominally eight–inch (200mm) bass/mid driver, and the FX50 a five–inch (130mm) unit. The bass/mid driver diaphragms in both drivers are a relatively unusual composite of paper with a thin thermoplastic skin on the front surface. The idea behind the composite is that the paper foundation provides rigidity while the thermoplastic layer provides damping, and perhaps a more high-tech, engineered appearance (paper is a fabulous driver cone material, but it doesn't look particularly 21st century). In other respects, the bass/mid drivers appear to be of entirely conventional manufacture. The FX80 (and FX50) tweeter is mounted slightly forward of the bass/mid driver apex, effectively on an extension of the pole piece. It's a 30mm silk–dome device covered by a metal protection grille and surrounded by a gently profiled waveguide component of around 75mm diameter. The waveguide will help modify the tweeter directivity and aid its acoustic integration with the bass/mid driver, however with the crossover frequency set at 2.4kHz, where the wavelength is around 140mm, the effect of the waveguide will only begin to come significantly into play acoustically from, say, 6kHz upwards. So, below that frequency the tweeter will radiate significant energy backwards towards the bass/mid driver cone where it will reflect and unavoidably cause a degree of disruption to the system response in both time and frequency domains. I'll cover that with a few FuzzMeasure curves later in the review, but first, here's the promised bit on dual-concentric drivers.

Return To Centre

The first and probably best-known exponent of dual-concentric drivers is of course Tannoy. However, the Tannoy dual-concentric principle, where a compression–style tweeter is located behind the bass/mid driver magnet and radiates through its pole-piece, is these days the exception rather than the rule. Contemporary dual-concentric designs, where the tweeter is mounted in front of the bass/mid driver, were to a significant extent made possible by the development of rare-earth magnet materials such as neodymium-iron-boron (NdFeB). Before NdFeB, it simply wasn't possible to position a viable tweeter in front of the bass/mid driver without significant compromise to the bass/mid driver performance — magnet technology powerful enough to enable a small tweeter to reach the required sensitivity simply wasn't available.

The first group of electro-acoustic engineers to realise that NdFeB enabled a 25mm dome tweeter to be made compact enough to be located on the end of a bass/mid driver pole-piece were Laurie Fincham and his team at UK speaker company KEF in the 1980s. Actually, there's some argument that others had similar ideas at roughly the same time (or even before), but KEF won the race to the patent officer's desk. Interestingly, rather than focus on the time–domain benefits of two drivers radiating from the same point in space, KEF christened their technology 'Uni-Q', to reflect that the primary value they saw in the idea was that, through the bass/mid driver cone effectively becoming a waveguide for the tweeter, it equalised the two drivers' directivity (in formal acoustics, 'Q' is used to denote a directivity index). KEF were also keen to differentiate Uni-Q technology from Tannoy-style dual-concentric drivers, by describing it as dual-coincident. This was, and is, more than just a marketing distinction and describes the significant point that along with being aligned concentrically, the acoustic centres of the two drivers are also aligned transversely, so that their acoustic source positions are the same (coincident).

KEF's 1980s dash to the patent office resulted in two decades of effective commercial protection of the dual-coincident principle, protection which was then extended by later patent applications as the technology was further developed (there are still patents in place that concern how the tweeter integrates with the bass/mid cone), and the patent protection is primarily why KEF's principle has not been much more widely used. In recent years, however, especially in the pro-nearfield monitor sector, as the core patents have expired, the KEF–style (tweeter–in-front) dual-concentric technique is gaining in popularity. The FX80 is a perfect example of the phenomenon except that, not being dual-coincident, it would not actually have infringed even the basic KEF patent.

In The Box

Returning back on message to the FX80, I didn't really finish the basic description so probably ought to get that box ticked. In terms of size the FX80 is just about ideal, I think, in terms of being compact enough to suit smaller studio and work spaces, but large enough to contain a reasonable volume of air to load the bass/mid driver and enable both adequately extended low–frequency bandwidth and good maximum level capability. Bearing in mind its entry–level price, it's a lot of monitor for the money. The dual-concentric arrangement enables a compact-looking front panel that's visually dominated by the compound driver, with an ovoid letterbox–shaped reflex port mouth beneath (well, I say beneath, but there's no reason not to install the FX80 in landscape mode). The port mouth incorporates a reasonably, but not generously, flared exit that will help suppress wind noise and delay the onset of compression that results from turbulent air flow. The FX80 front panel is formed entirely from an injection–moulded and painted ABS component but, sadly, for my tastes anyway, it is finished as black as black can be. The edges of the panel incorporate some generous softening that ought to help suppress cabinet edge diffraction, however the compound driver is mounted 15mm or so back in the panel, which results in a significant edge feature all the way around its periphery. Said edge is softened, but it will still likely introduce an acoustic signature that will contribute to the monitor's overall character.

I wrote a paragraph or two ago that the FX80 is a reasonably large monitor, however the very nature of large speaker enclosures built from flat MDF panels, such as the FX80's, is that they tend to be not entirely inert in terms of panel resonance. Back in 1977, Dudley Harwood (who founded the Harbeth loudspeaker company) co-wrote a technical paper at the BBC in which the acoustic outputs from speaker cabinets of different materials and constructions were analysed. It makes for interesting reading and you can actually download it here — http://downloads.bbc.co.uk/rd/pubs/reports/1977-03.pdf — but if you're too busy for all that malarkey, the upshot of the paper was that at specific resonant frequencies the contribution from the cabinet panels could easily reach audible levels. Now, you're perhaps wondering why I've apparently wandered somewhat off-piste for a paragraph, and it's because the user manual for the FX80 describes the cabinet as "reinforced by internal bracing". Now, bearing in mind the size of the FX80 cabinet panels, the speaker designer in me would definitely have tried to find the necessary pennies within the manufacturing budget to include cabinet bracing, and its reassuring that Fluid appear to have done so on the FX80.

Amping Up

The various EQ and placement options are accessed via a set of DIP switches on the back of the monitor.As is ever the case with active nearfield monitors, around the back of the FX80 there's the obligatory amplifier heatsink and connection panel. In this case, along with an IEC mains input socket, on/off switch and input sensitivity knob (that I left at maximum), the FX80 offers just unbalanced phono, balanced XLR and balanced TRS jack inputs. The XLR option is set at +4dBu sensitivity and the jack and phono at -10dBV. The FX80 amplification comprises 60 Watts and 50 Watts of Class–D power for, respectively, the bass/mid driver and tweeter.

Alongside the input sockets is an eight–position DIP switch module that enables a variety of EQ options to be dialled in. Even though, these days, DIP switches definitely require me to don my reading glasses, I quite like them as a means of selecting EQ options. There's something very deliberate about a DIP switch. There are four EQ parameters available and the combination of DIP switch settings provides four settings for each one (a pair of switches are assigned to each parameter providing four combinations of switch position). The EQ parameters and settings are a 7kHz high–frequency shelf with 0dB, +1dB, -1dB and -2dB options; a 1.5kHz mid–range boost/cut with the same level options as the HF shelf; an 'acoustic space' EQ with 0dB, -1dB, -2db and -4dB options; and an LF cutoff. The frequency at which the acoustic space EQ operates isn't specified in the FX80 documentation, but FuzzMeasure will hopefully reveal all.

Measure Centre

Diagram 1a: The FX80's measured frequency response (300Hz-20kHz), both on axis (blue) and off axis (orange).Diagram 1b: The FX80's step response, again measured on and off axis (purple and green traces, respectively).I began my analysis of the FX80 by measuring its basic frequency response linearity. Diagram 1a shows the FX80's on-axis response from 300Hz upwards (blue) along with its response 30 degrees off axis (orange). The main point of interest is that the off-axis response curve is actually significantly smoother than the on-axis curve. In particular the suck-out features at 6.5kHz and 13kHz disappear off axis. The likely reason for this is that those suck-outs are the result of rearwards radiation from the tweeter reflecting from the bass/mid driver cone (or perhaps from the edge feature around the bass/mid driver) and interfering with the direct forward radiation. The suck-outs disappear off axis because they are features of symmetry. Moving the measuring microphone to an asymmetric position where all the path lengths between said mic and speaker are unequal will blur all the reflections and consequently flatten the apparent frequency response. Diagram 1b provides a different and perhaps more revealing illustration of the tweeter reflections. The curves show the FX80's axial (purple) and off-axis (green) step responses (the peak level of the off-axis response is lower because the tweeter has less overall output off axis). The first signal arrivals from the tweeter and bass driver are pretty obvious, but the axial step response is clearly messier, with multiple later arrivals from the tweeter. These are the reflections.

Diagram 2: The FX80's default response is shown in blue; the effect of the EQ options (+1dB MF, -2dB HF) is shown in green.Along with examining the tweeter reflections I checked the behaviour of the FX80 EQ options, and they behaved pretty much as expected. Diagram 2 illustrates the default axial response (blue) compared to the mid–range EQ at +1dB and the high-frequency EQ at -1dB (green). The only thing perhaps to mention is that the high–frequency EQ operates from a significantly lower frequency than the specified 7kHz. It looks to me as if the EQ switch simply adjusts the gain of the tweeter amplifier.

Diagram 3: The -4dB acoustic space option (yellow) versus the FX80's normal LF response.The last FuzzMeasure curves in Diagram 3 show the FX80's low–frequency characteristics measured using a close mic equidistant from the driver and the port. This measuring technique can't provide the last word in accuracy, but it can be revealing nonetheless. In this case I've primarily used it to illustrate the operation of the FX80 acoustic space EQ option. The purple curve shows the FX80's default low–frequency response, and the yellow curve shows the -4dB acoustic space option. The acoustic space EQ operates from around 400Hz downwards and ought to be useful in some installations.

The FX80 produces a usable, wide–bandwidth performance that, once I'd learned its quirks, I'd have no problems using in a mix context.

Listening In

Finally, on listening to the FX80, a couple of characteristics were apparent immediately. Firstly, the FX80's amplification is noisier than I would hope, as hiss was audible from both monitors. The hiss didn't alter with gain control position or the presence of an input connector so it appears to be inherent to the FX80's amplification. It's not a deal-breaker and it becomes effectively inaudible once music is playing, but it's a little disappointing. The second immediately apparent character of the FX80 was that it has a usefully extended low–frequency bandwidth that renders mix elements that earn their keep right down in the bass foundations satisfyingly audible. The presence of low bass makes the FX80 sound particularly grown-up, as if it's more of a midfield monitor than a nearfield monitor. It's not entirely great news though because, along with the bandwidth extension, the FX80 to my ears displays the slight softness and sluggishness typical of ported monitors. FuzzMeasure can provide a reasonable measurement of low–frequency latency and at 40Hz (approximately bottom E on a bass guitar) the FX80 figure is around 17ms. For comparison's sake, 5ms would be typical for a closed–box monitor — although a closed–box monitor that can genuinely play 40Hz is likely to be significantly more expensive than the FX80. While it's a matter of great debate among electro-acoustic geeks, many believe that the low–frequency latency typical of port-loaded monitors is audible. However, as with the amplifier hiss, it's by no means a deal-breaker, and is absolutely comparable with countless other similar proportioned and priced port–loaded monitors. I also had to keep reminding myself just how competitively priced the FX80 is, and adjusting my expectations accordingly.

Moving further up the band the FX80 displays a perfectly usable, if slightly bright, tonal balance, good clarity of mix elements and reasonably low levels of coloration. I found the balance more comfortable listening slightly off axis, by aligning the monitors to fire almost straight into the room rather than angled towards the listening position. Doing this effectively attenuates the tweeter slightly and to my ears also helped focus the stereo image. Perhaps those on-axis tweeter reflections are audible? Having written, however, that the FX80 displays little coloration, it does to my ears have a mild narrow-band character in the low mid–range that adds a bloom to male voices and low–range strings (cellos). I've used this analogy before but if I were recording, say, a cello and heard that character, I'd move the microphone, or maybe swap it for a different one. In my experience the coloration is one that's typical of resonant cabinet panels, so perhaps the FX80's cabinet bracing is not entirely effective. (I had a sneaky look inside an FX80 cabinet after I'd finished listening and not only is the cabinet bracing minimal, there's very little interior acoustic wadding — it's a relatively 'live' cabinet interior.)

Having described a couple of mild flaws in the FX80, the irony is that, especially at their frankly remarkable price, I genuinely enjoyed my time with them. The FX80 produces a usable, wide–bandwidth performance that, once I'd learned its quirks, I'd have no problems using in a mix context. And while I appreciate that monitors are supposed not to impart any of their own character, I couldn't help but hear a persuasive, engaging enthusiasm in the way the FX80 performs. If my audio work mostly involved acoustic instruments and voices, I'm not sure the FX80 would be the right monitoring choice, but if I was on a really tight budget and worked, say, in EDM or indie guitar music, the FX80 would be a pretty smart monitoring choice.