Falcon Acoustics LS3/5a loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Falcon LS3/5a's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield response. The Falcon's voltage sensitivity was 0.7dB higher than that of my 1978 pair of Rogers LS3/5a's, at 83.2dB/2.83V/m. Its plot of impedance magnitude and phase (fig.1) was very similar to that of the original (footnote 1) with the magnitude remaining above 8 ohms from 10Hz to 7kHz. However, it was, on average, about 2 ohms lower in the region covered by the tweeter.


Fig.1 Falcon LS3/5a, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

I investigated the vibrational behavior of the cabinet with a plastic-tape accelerometer and the speaker supported on upturned cones, which allows resonances to fully develop. I found a single high-level resonant mode on the side panels at 355Hz (fig.2)—identical to what I found on some of the earlier versions of the LS3/5a.


Fig.2 Falcon LS3/5a, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of sidewall (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

The impedance peak of 80 ohms centered on 76Hz in fig.1 suggests that 76Hz is the tuning frequency of the sealed enclosure, which is identical to the Harbeth and the 1978 and 1989 samples of the Rogers LS3/5a's that I've tested. With the woofer's output measured in the nearfield (fig.3, black trace below 300Hz), this frequency corresponds with where the driver's output is down by 6dB, but it can be seen in fig.3 that the woofer's response peaks by about the same 6dB in the upper bass. This is the classic "LS3/5a bump," which gives the impression that this little speaker produces more bass than it actually does. However, this bump will make the speaker sensitive to excessive boundary reinforcement when not used in free space.


Fig.3 Falcon LS3/5a, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with nearfield woofer response plotted below 300Hz.

Higher in frequency in fig.3, the black trace shows the Falcon's farfield response averaged across a 30° horizontal window centered on the tweeter axis. As with all versions of the LS3/5a (footnote 2), a narrow peak in the upper midrange disturbs an otherwise relatively uniform response. Falcon's version of the original's KEF T27 tweeter produces several narrow peaks in the octave above 15kHz, and the top two octaves are shelved up a little. Fig.4, taken with pink noise on the tweeter axis, compares the in-room response of the Falcon LS3/5a (red trace) with one of my 1978 Rogers examples (blue). The low-frequency outputs are identical, but the Falcon's upper-midrange peak is more pronounced and, overall, the 2015 speaker has 2–3dB more output in the treble. This explains the Falcon's greater measured sensitivity.


Fig.4 Falcon LS3/5a (red) and 1978 Rogers LS3/5a (blue), in-room responses on tweeter axis, 100Hz–20kHz, at 50" with pink noise (5dB/vertical div.)

The Falcon LS3/5a's plot of lateral dispersion (fig.5) looks complex, but this is mainly due to the small dips in the on-axis response tending to fill in to the speaker's sides. However, the broad depression in the presence region off axis might make the speaker sound a little polite. In the vertical plane (fig.6), a large suckout develops in the crossover region immediately below the tweeter axis. In addition, a little more energy is present between 1 and 4kHz between 5° and 10° above the tweeter axis. Both of these factors suggest that the Falcon LS3/5a will work better on a low than on a high stand.


Fig.5 Falcon LS3/5a, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis, reference response, differences in response 5–90° off axis.


Fig.6 Falcon LS3/5a, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45–5° above axis, reference response, differences in response 5–45° below axis.

Turning to the time domain, the Falcon's step response on the tweeter axis (fig.7) is basically identical to those of the earlier versions of the LS3/5a, with the tweeter connected in inverted acoustic polarity, the woofer in positive polarity. However, a prominent ripple in the decay of the woofer's step correlates with the pronounced ridge of resonant energy centered just above 1kHz in the Falcon's cumulative spectral-decay plot (fig.8). The LS3/5a has always sounded a touch nasal, but, all things being equal, from this I would have expected the Falcon's version to sound more nasal than the original version. (HR did comment on the response rise in the upper midrange.)


Fig.7 Falcon LS3/5a, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.8 Falcon LS3/5a, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

Overall, its measured behavior indicates that Falcon Acoustics' re-creation of this classic British minimonitor has been very successful. The company has resisted the temptation to "improve" the sound quality, but has also managed to re-create what made the speaker great in the first place—unlike, for example, the unauthorized Gini Systems version that John Marks reviewed in December 2008.—John Atkinson

Footnote 1: See fig.1 here.

Footnote 2: See fig.3 here.

Falcon Acoustics
US distributor: Big Ear Consulting
Ormond Beach, FL
(800) 752-4018

dalethorn's picture

I bought the ca. 1975 version after reading Holt's review. I particularly remember the phrase "minimum recommended power 25 watts". He noted that maximum power was also specified as 25 watts. I eventually traded them off, fearing I would blow the little drivers at some point. Not an irrational fear given that the pair the seller loaned me (until a new set came in) had a blown driver.

Edit: If I had the original pair today, I'd put them closer to the wall (maybe 12-15 inches) and EQ down the bass hump, which would lessen my anxiety about over-driving them. Once you hit that sweet spot with these little speakers, they'd be almost like the theoretical point-source, having a terrific soundstage.

LS35A's picture

than any other speaker, by far. Even today it's still a favorite.

If you'd told me in 1975 that in 40 years they'd still be being made I'd have said you were crazy.

But in a small room they are still a really nice speaker.

Venere 2's picture

The staples are a nice touch… Really ugly speaker! As good as they supposedly sound, they look like a 99$ DIY project.

RBrooks's picture

No boxes,No horns,No ribbons,No electrostatics,No distortion.
The author of this article thought rather highly of Siegfried Linkwitz Speakers.
As do Steve Guttenberg and Nelson Pass.

audiocaptain's picture

Just wanted to say a big thank you for doing this review. It is a classic and important piece of Audio History that every music lovers deserves to know about. Great writing - Thanks Again! BEC

dce22's picture

Quote : "They played enjoyably well with all—with one exception. Class-D made these hyper-responsive, 15-ohm speakers sound dry, slightly hard, and more generalized than I like. Class-D amps awoke that napping bass bump and turned the slightly rising treble into a distraction. The LS3/5a has always been an exceptionally amp-friendly speaker, not because it makes mediocre amps sound good, but for quite the opposite reason: The LS3/5a excels at letting you hear exactly what your amp really sounds like."

Translation : "I did not try class d amp on this speaker but this is what i think it will sound"

No class d amps were noted in the associated equipment section, message to Herb you need to tell people what kind of class d amp you were using if you were using one in the first place so people need to stay clear of it.

LS35A's picture

Yes, I also was taken with the implication of 'all class d amplifiers are the same' here. Obviously not true.

supamark's picture

Literally the first component listed under the integrated amps, the Hegel H160, is a class D amp. [flame deleted by John Atkinson]

John Atkinson's picture
supamark wrote:
the first component listed under the integrated amps, the Hegel H160, is a class D amp.

The Hegel doesn't use a class-D output stage. But Herb has been recently using the Rogue Sphinx, which does use Hypex class-D modules.

John Atkinson
Editor, Stereophile

jporter's picture

The Magnepan 1.7 is $200 less than this speaker. The Paradigm 15B is $600 less. The PSB Imagine B is $1200 less. Monitor audio has their silver series. All of these sound great and just look so much nicer than the Falcon. I have not heard the Falcon speaker, but those staples are just awful.

audiocaptain's picture

This speaker is made to be played with the grills on. The staples are there to hold the velcro firmly in place and work a great deal better than little plastic nipples or other securing methods. You never see these, ever, in normal use. The image is just to show the original T27 and the B110 drivers.
Although there are other lesser priced speakers they are not authentic classic LS3/5a's.

Venere 2's picture

Authentic, a word that is so elastic that it can be stretched to encapsulate everything that is good and worthy of owning. Even when it is not true.

I did not need to be told the speakers were meant to be played with their grills on, and that the grills would hide the staples. I believe most people picked up on that as well.

And yes, there are lesser priced imitators that claim the lineage of the LS3, and are not truly "classic" and oh so authentic LS3s. But remember, there are also numerous other designs from a myriad number of speaker builders in this price range. Maybe they can't claim the authenticity of nostalgia and the good ole days, but they offer much better build quality, aesthetics, and most importantly great sound.

supamark's picture

[flame deleted by John Atkinson]

This speaker is aimed at a specific, small, market segment (not you obviously, but nothing in Stereophile is). As to build quality, from the photos at least, it looks both exceptional and faithful to the original speaker.

WM's picture

Falcon should be commended for their effort; its one of those rare designs which isn't expensive but has huge 'cred' appeal.. it could easily be in the pages of Wallpaper magazine, if it isn't already

jimsusky's picture

This review arouses no small bit of nostalgia in me. Shimeks Audio (in Anchorage) claimed to have sold more Rogers LS3/5a loudspeakers than anyone else in the world for several years running (so advised, they said, by Rogers). This was undoubtedly a function of the Trans-Alaska-Pipeline construction boom that had started about 1973 (startup was ca. summer 1977). But it was also a function of first-class setup and salesmanship. Terry Shimek could sell the legendary “freezers to Eskimos” (which was not a slur in those much thicker-skinned days).

Shimeks extracted $650 from me as a college student in the Summer of 1978,
which I later discovered was $50-$100 more than stateside retail. Invoking CPI-U we get a 2015/1978 ratio of 3.7 and a current value of $2400 I also dropped $200 (actual retail this time) for the kit version of the Hafler DH100 preamp.

(the SSA informs me I made $1900 that summer – thanks be to Mom who amply subsidized my education, room, board, beer, and audio that year).

By 1978 Shimeks was actively pushing moving coil phono replay with direct-drive turntables, both with stock tonearms and retrofitted with Grace tonearms along with decent solid state electronics (Audio Research and Mark Levinson came a bit later) – their demos with the Rogers monitor were ridiculously compelling – with 3D soundstages that extended well beyond the loudspeakers.

They had a trick which became de rigueur for me – stands with back-tilt. The LS3/5a always imaged well sitting straight up-and-down – but with a 25-degree tilt a good soundstage became cavernous with unparalleled instrumental focus. I was so intoxicated by that effect that I suspect I ignored the tonal-balance tradeoff.

(still I suggest that Herb Reichert give this a try and report back in a Follow Up)

That midrange! Lady singers never sounded so sweet in your living room. That definition! Acoustic guitars were transparently rendered – one could readily hear fingers, picks, nylon, and wound steel. We readily learned microphone techniques – and soon had disdain for mediocre and indifferent production.

Several years later I was lent a fellow student’s uncle’s Bryston 3B and Audio Research SP-3 for only a few hours of alchemical magic with my BBC’s. Rickie Lee Jones was playing when my roommate walked in. He was stunned, stopped dead in his tracks and listened. With astonishment he said “it sounds like she’s in the room!” (thus independent, unsolicited confirmation of electronic wizardry).

Reichert is right – big speakers mostly get in their own way – few of them make that kind of holographic magic.

jimsusky's picture

I always liked the look of the speakers undressed - and dogmatically believed that the grills could not improve the sound.

MGM's picture

I built a pair of these Falcon LS3/5as and would broadly agree with the findings.
They are definitely brighter than my own, but one quickly adjusts to this
The friend I built them for for is extremely happy with them.

Casmar's picture

I have had mine for the past twenty eight years. With all their idiosyncrasies, they were the only constant units in my Hifi for all these years. The most fascinating aspect is how they sing in one room and pale by comparison in another.

Doctor Fine's picture

I guess there are too many riches available nowadays.
I bought the Spendors which can now be found for around $600 mint on Ebay.
And THESE were Herb's favorite 3/5 a few years ago.
At $600 the value is off the charts...
I will keep mine forever and I could easily afford to "upgrade."