ATC SCM7 v.3 loudspeaker Measurements

Sidebar 3: Measurements

John Marks sent me the review pair of this elegant British, sealed-box minimonitor to run through my loudspeaker test regime. I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the ATC's frequency response in the farfield (I measured serial no.1002), and an Earthworks QTC-40 for the nearfield responses. All measurements were taken without the grille.

As expected from its diminutive size, the SCM7 v.3 is not very sensitive; my estimate of its voltage sensitivity came in at 82dB(B)/2.83V/m, which is 2dB below the specified 84dB. The speaker's electrical impedance, specified as 8 ohms, is shown as the solid trace in fig.1. The sole peak in the bass, reaching 48 ohms at 57Hz, indicates that that is the tuning frequency of the woofer in the sealed enclosure, which in turn implies only modest low-frequency extension. The impedance averages 6 ohms in the midrange and low treble, with a minimum value of 5.25 ohms at 1100Hz. Though the electrical phase angle (fig.1, dotted trace) reaches –61° at 73Hz, the impedance magnitude is high at that frequency; the ATC won't present any drive difficulty to the partnering amplifier.


Fig.1 ATC SCM7 v.3, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

There is a discontinuity at 600Hz in the impedance traces that, all things being equal, would suggest the presence of some kind of enclosure resonance at that frequency. However, when I investigated the vibrational behavior of the cabinet walls with a simple plastic-tape accelerometer, I found nothing untoward at 600Hz. I did find a strong mode at 715Hz on the sidewalls (fig.2), as well as three slightly less intense resonances centered on 410Hz. The 410Hz mode was also present on the ATC's top panel, but at a much lower amplitude (not shown). Given the relatively high frequency of this cabinet behavior and the small radiating areas of the affected panels, I doubt they will result in any coloration.


Fig.2 ATC SCM7 v.3, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of side panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

Fig.3 shows the individual responses of the woofer (blue trace) and tweeter (red). Below 350Hz, the blue trace shows the woofer's output measured in the nearfield. There is virtually none of the rise in the upper-bass response that usually results from measuring in the nearfield—the assumption behind a nearfield measurement is that the speaker baffle extends to infinity in all directions, which boosts the measured low-frequency output below a frequency related to the actual baffle dimensions. The SCM7's output is down by 6dB at the cabinet/woofer tuning frequency, as anticipated from the impedance graph. This measured behavior suggests that the little ATC will benefit from some boundary reinforcement; otherwise, the speaker's bass will sound, as JM found, "'respectable' rather than 'convincing.'"


Fig.3 ATC SCM7 v.3, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield response of woofer (blue) plotted below 355Hz.

Fig.3 also indicates that the crossover between the two drivers occurs just above 2kHz, and that each rolls out with an initial 12dB/octave slope, though the tweeter's slope steepens an octave below crossover. Overall, each driver has a fairly flat output within its passband, but the woofer's response is disturbed by a small peak and dip between 500 and 700Hz. This may well be related to the discontinuity in the impedance traces noted earlier, and also appears in the SCM7's farfield response, averaged across a 30° horizontal window centered on the tweeter axis (fig.4). Other than that, the ATC's response is commendably smooth from the upper bass to the top of the audioband. The speaker's horizontal dispersion (fig.5) has a very slight off-axis flare at the bottom of the tweeter's passband but is otherwise wide and even, at least up to 8kHz or so, where the tweeter starts to become directional. As the SCM7's on-axis response lacks the usual top-octave boost, the ATC will sound rather mellow when listened to at a distance in a largish room. This speaker will sound best when the listener sits relatively close to it.


Fig.4 ATC SCM7 v.3, anechoic response on tweeter axis at 50" without grille, averaged across 30° horizontal window and corrected for microphone response, with nearfield woofer response plotted below 300Hz.

The traces in fig.5 are normalized to the tweeter-axis response. The horizontal radiation pattern without this normalization (fig.6) indicates that the uneven behavior between 500 and 700Hz is maintained off axis, as it is in the vertical plane (fig.7). Fig.7 also indicates that the SCM7 is relatively tolerant of vertical listening axis: suckouts centered on the crossover frequency do not develop until 20° above and below the tweeter axis.


Fig.5 ATC SCM7 v.3, lateral response family at 50" without grille, 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 ATC SCM7 v.3, lateral response family at 50" without grille, from back to front: responses 90–5° off axis, reference response, responses 5–90° off axis.


Fig.7 ATC SCM7 v.3, vertical response family at 50" without grille, 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 ATC's step response on its tweeter axis is shown in fig.8. Both drive-units are connected with positive acoustic polarity, and the slight discontinuity between the decay of the tweeter's step and the initial rise of the woofer's step suggests that the optimal axis actually lies just below the tweeter axis. This is what you would expect from a small monitor likely to be positioned atop a mixing console's meter bridge. The cumulative spectral-decay plot (fig.9) is superbly clean, though some delayed energy is apparent in the midrange coincident with the on-axis anomaly. (Ignore the ridge of delayed energy at 16kHz, which is due to interference from the measuring PC's monitor.)


Fig.8 ATC SCM7 v.3, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.9 ATC SCM7 v.3, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

I was impressed by the measured performance of the latest version of ATC's SCM7. This speaker seems optimally suited to its role as a nearfield monitor. As John Marks concluded, "the updated SCM7 seems designed to be a thoroughly modern minimonitor." Amen to that.—John Atkinson

ATC Loudspeaker Technology Ltd.
US distributor: Lone Mountain Audio
7340 Smoke Ranch Road, Suite A
Las Vegas, NV 89128
(702) 307-2727

Supperconductor's picture

Headline typo.

John Atkinson's picture

Thanks - JA

audiolab's picture

I know you are right in that low c on a piano is 27.5hz, I will not argue that low c on a bass guitar is 41hz (I do not know of such things), but who decided low c on an organ is 32hz. Any half decent sized organ in a cathedral or concert hall will have a 32ft stop that yields a low c of 16hz. Some people would point you to two organs in the world that have a 64ft stop that yields a low c of 8hz, some would even point out that technically one of those two could play a resultant tone of a 128ft stop resulting in a almost purely accademic low c 4hz. I am quite happy to settle with it at 16hz and what a wonderful note it is to !

John Atkinson's picture

audiolab wrote:
I will not argue that low c on a bass guitar is 41hz

The low E string (not C) on a bass guitar or double bass is tuned to 41.2Hz, but most of the energy lies an octave higher, at 82.4Hz. See fig.3 at

John Atkinson
Editor, Stereophile

jmsent's picture

In the absence of enclosure resonance as an explanation, that fairly substantial dip in the measurement between 600 and 1 kHz might be the result of a woofer "edge resonance". Surprising, given the pedigree of the company.

John Atkinson's picture

jmsent wrote:
that fairly substantial dip in the measurement between 600 and 1 kHz might be the result of a woofer "edge resonance".

I had wondered if that were the problem, but I thought that cone/surround termination problems occurred a little higher in frequency, between 1kHz and 3kHz.

John Atkinson

Editor, Stereophile

jmsent's picture

I would normally take place a bit higher up in frequency on a driver this size. But certain cone/surround combinations can push it down. But whatever the cause, that sharp dip is fairly severe, and I'm still surprised they'd let it go to market with this flaw.

RoryB's picture

Looking at the magnitude of the peak/dip couplet, which does look like a surround resonance (probably due to the relatively wide surround; a Klippel scanning laser vibrometer could reveal the true cause), it is visible in the graph but at +3/-3dB in magnitude, it is not terribly severe. At the lower frequency where it occurs, you are less likely to notice it. You might notice it more as a coloration if it occurred at 2-3 kHz, and more would need to be done to correct it. A thicker surround or a variable thickness surround could correct this problem, but it would reduce broadband sensitivity of the driver.

Bob Loblaw's picture

In the article the term active and powered to describe speakers with built-in amplification is used interchangably. They are not the same thing. An active speaker has an active crossover and 1 amplifier for each driver, such as with ATC's active speakers.

A powered speaker is a passive speaker with an amplifier built in. The Audioengine A5+ is a powered passive speaker.

Active speakers have several advantages over passive and powered passive speakers and it's a disservice to your readership not to acknowledge the difference when there are those out there who may not know better.

ryder's picture

I am well aware that small monitors cannot go too loud or too deep in the bass. Having said that, can this SCM7 v3 go loud without losing control at moderate to high volume levels? Can it go louder than say the Harbeth P3ESR?