First Watt SIT-3 power amplifier Measurements

Sidebar 3: Measurements

I performed a full set of measurements on the First Watt SIT-3, using my Audio Precision SYS2722 system (see the January 2008 "As We See It"). Before doing any testing, I preconditioned the SIT-3 by running it at one-third power into 8 ohms for an hour. At the end of that time the top panel was warm, at 106.3°F (41.3°C), and the heatsinks were hot, at 123.8°F (51°C).

The First Watt's voltage gain into 8 ohms was a very low 11.3dB and the amplifier preserved absolute polarity (ie, was non-inverting). The input impedance measured close to the specified 200k ohms at low and middle frequencies, dropping to a still very high 151k ohms at 20kHz. The output impedance was half that of the First Watt J2, which Herb Reichert reviewed in October 2016, at 0.25 ohm at 20Hz and 1kHz, and 0.265 ohm at 20kHz. As a result, the response with our standard simulated loudspeaker varied by ±0.2dB (fig.1, gray trace). This graph shows that the audioband response is flat up to 10kHz, with a rolloff in the top octave reaching –0.8dB at 20kHz. The SIT-3 reproduced a 10kHz squarewave with short risetimes, and no overshoot or ringing (fig.2).

119FWSIT3fig01.jpg

Fig.1 First Watt SIT-3, frequency response at 2.83V into: simulated loudspeaker load (gray), 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (green) (1dB/vertical div.).

119FWSIT3fig02.jpg

Fig.2 First Watt SIT-3, small-signal, 10kHz squarewave into 8 ohms.

Like the J2, the SIT-3's channel separation was excellent, measuring >100dB in both directions below 1kHz, and still 74dB at 20kHz. The wideband, unweighted signal/noise ratio, ref. 2.83V and measured with the input shorted to ground, was an excellent 87.3dB in the left channel and an even greater 95.4dB in the right. These ratios improved to 94.5 and 98.1dB when the measurement bandwidth was restricted to the audioband, and to 107dB when A-weighted. These high ratios will be related to the low overall gain, of course. Spectral analysis of the SIT-3's noise floor (fig.3) revealed spuriae both at 120Hz and its harmonics, and at 60Hz and its odd-order harmonics; the latter will be due to magnetic interference from the AC power transformer. All of these spuriae are very low in level, however, and will not be audible.

119FWSIT3fig03.jpg

Fig.3 First Watt SIT-3, spectrum of 1kHz sinewave, DC–1kHz, at 1W into 8 ohms (linear frequency scale).

The SIT-3 is specified as being able to deliver 18Wpc into 8 ohms (12.55dBW). With "clipping" defined as when the THD+noise reaches 1%, fig.4 indicates that the SIT-3 clipped at 8.4Wpc into 8 ohms (9.24dBW). Relaxing the definition of clipping to 3% THD+N, the First Watt clipped at the specified 18Wpc. Into 4 ohms (fig.5), the SIT-3 delivered 16.8Wpc (9.24dBW) at 1% THD+N, and 36Wpc (9.55dBW) at 3%. Figs. 4 and 5 both show that while the distortion is low at very low powers, it rises linearly as the power increases. This is typical for an amplifier with a single-ended output stage.

119FWSIT3fig04.jpg

Fig.4 First Watt SIT-3, both channels driven, distortion (%) vs 1kHz continuous output power into 8 ohms.

119FWSIT3fig05.jpg

Fig.5 First Watt SIT-3, both channels driven, distortion (%) vs 1kHz continuous output power into 4 ohms.

The First Watt SIT-3 offered levels of distortion that remained relatively constant with frequency but that increased into lower impedances (fig.6). Fortunately, and as specified by First Watt's Nelson Pass (footnote 1), the SIT-3's distortion was predominantly the sonically benign second harmonic, what appears to be the opposite phase to the fundamental, at least regarding the positive-going half cycles (fig.7). (A single-ended output stage amplifies the signal waveform's negative-going half cycles by a different amount than the positive-going half cycles. This asymmetry primarily produces second-harmonic distortion.) Though the third harmonic was present at a level around 15dB lower than that of the second (fig.8), it didn't increase in level as the power increased or the impedance dropped (fig.9). However, higher-order harmonics became visible when the output reached 5W into 8 ohms and above, as predicted by Nelson Pass (fig.10). Tested with an equal mix of 19 and 20kHz tones at a low power into 8 ohms, the levels of higher-order intermodulation products were relatively low (fig.11), though the second-order difference product lay at –54dB (0.15%). Commendably, the levels of these intermodulation products were similar when I kept the output voltage the same and halved the load impedance (fig.12).

119FWSIT3fig06.jpg

Fig.6 First Watt SIT-3, THD+N (%) vs frequency at 2.83V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (left gray).

119FWSIT3fig07.jpg

Fig.7 First Watt SIT-3, 1kHz waveform at 1Wpc into 8 ohms, 0.244% THD+N (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

119FWSIT3fig08.jpg

Fig.8 First Watt SIT-3, spectrum of 50Hz sinewave, DC–1kHz, at 1Wpc into 8 ohms (linear frequency scale).

119FWSIT3fig09.jpg

Fig.9 First Watt SIT-3, spectrum of 50Hz sinewave, DC–1kHz, at 2Wpc into 4 ohms (linear frequency scale).

119FWSIT3fig10.jpg

Fig.10 First Watt SIT-3, spectrum of 50Hz sinewave, DC–1kHz, at 5Wpc into 8 ohms (linear frequency scale).

119FWSIT3fig11.jpg

Fig.11 First Watt SIT-3, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 1Wpc peak into 8 ohms (linear frequency scale).

119FWSIT3fig12.jpg

Fig.12 First Watt SIT-3, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 2Wpc peak into 4 ohms (linear frequency scale).

Although that relatively high level of second-harmonic distortion, which is a deliberate design choice, is controversial, the First Watt SIT-3 is a well-engineered amplifier.—John Atkinson



Footnote 1: See "Gramophone Dreams" elsewhere in this issue.
COMPANY INFO
Pass Laboratories Inc.
13395 New Airport Road, Suite G
Auburn, CA 95602
(530) 878-5350
ARTICLE CONTENTS

COMMENTS
tonykaz's picture

Off to bad start?, not quite click-bait but compelling.

Well, ok, this seems a thinking man's Amplifier with plenty of personality.

$4,000 while supplies last.

Tony in Michigan

ps. how'd this thing do with Steve G's Klipsch "loud"-speaking" speakers?

John Atkinson's picture
tonykaz wrote:
how'd this thing do with Steve G's Klipsch "loud"-speaking" speakers?

The answer to that question will be found in the April issue of Stereophile.

John Atkinson
Editor, Stereophile

Ortofan's picture

... the level of second harmonic distortion is barely -45dB below the fundamental can be deemed to "preserve the geometry of the original harmonic relationships better than any amp I can remember"?
Wouldn't an amp that adds as little harmonic distortion as possible, such as the Benchmark AHB-2, do a far better job of preserving the geometry of the original harmonic relationships?

mememe2's picture

When Herb writes " I felt I could measure the added reverb. I thought I could feel the air moving in and out of her instrument.". I wondered why he didn't include an actual measurement to back up this "feeling". Next his thought that he could "feel the air moving in and out of her instrument" sounds like pure unadulterated mystical hyperbole. Really? Less hyper-hyperbola please.

johnnythunder's picture

music "feels and sounds" to the individual reviewers, you shouldn't be reading Stereophile or the Absolute Sound, or Hifi News and RR etc. etc. That type of description is WHY we read these publications and websites. Sometimes light is shed on the subjective response by John A's fact checking measurements but not all of it. So we take these written feelings and if they sync with how you like to feel and hear recorded music you add said equipment to your list of pieces to audition. If you want a measurement to back up every comment, you're wasting your time reading audio reviewers. In fact, that's like asking for a chemical analysis of every morsel of food should a food magazine say you're wasting your time with $5 a bottle Extra Virgin Olive oil compared to $20a bottle. I'm sure there are people who feel that the more expensive olive oil is a waste of money" as "taste" is a subective thing.
Ditto with more expensive wines. "Sound" familiar ?

mememe2's picture

According to johnnythunder readers of the mags he quoted only want how a product "feels and sounds With that as the ultimate criteria for reading those mags why do the measurements (sometimes ) appear at the end of the reviews? And let's just do away with the technical descriptions as they don't really matter because they tell us nothing informative about how the gear under review will feel or sound like. Maybe -to shore up the how it "feels and sounds like" we should, at the very least insist on All reviewers to post a recent audiologists measurement (the bane of those that worship at the alter of subjectivity)) of the state of their hearing.

Bertie Bucket's picture

If this was a Chinese amp I'm pretty sure the usual suspects would be all over those measuremets pointing out it must be garbage.

As an aside, relaxing the definition of clipping....why not relax the definition of surface resonance on speakers that have a lot of resonance or any of the other parameters. They are meant to show a standardised approach otherwise, what's the point of them?

This amp bombed producing 8.4W at 1% THD distortion.

The fact amps like these garner a cult following says more about marketing and human fallacies than anything.

X