Parasound Halo A 21+ power amplifier Measurements

Sidebar 3: Measurements

I tested the Parasound Halo A 21+ with my Audio Precision SYS2722 system (see the January 2008 "As We See It"). I preconditioned the amplifier by operating it at one-third the specified power into 8 ohms for an hour. (With a class-AB output stage, one-third power results in the maximum dissipation in the output devices.) At the end of this time, the side-mounted heatsinks were very hot, at 162.8°F (72.7°C). The top panel was also hot, at 116°F (46.7.1°C).

The Parasound's voltage gain into 8 ohms measured 29.2dB from the balanced inputs, 29.4dB from the unbalanced inputs. In bridged-mono mode, the gain was 34.9dB into 8 ohms. The amplifier preserved absolute polarity (ie, was noninverting) with both balanced and unbalanced input signals and with the right channel used as the input in bridged-mono mode, with the outputs taken from the two positive loudspeaker terminals as described in the manual. The balanced input impedance was high at 90k ohms at 20Hz and 1kHz, dropping to a still-high 84k ohms at 20kHz. The unbalanced input impedance was 46k ohms at low and middle frequencies, 37k ohms at the top of the audioband.

The Parasound's output impedance was a very low 0.077 ohm at 20Hz and 1kHz, increasingly slightly to 0.1 ohm at 20kHz. (These figures include the series impedance of 6' of loudspeaker cable.) The modulation of the amplifier's frequency response, due to the Ohm's law interaction between this source impedance and the impedance of our standard simulated loudspeaker, was therefore minuscule, at ±0.1dB (fig.1, gray trace). The response into an 8 ohm resistive load (fig.1, blue and red traces) was flat to almost 20kHz, which correlates with the A 21+'s superb reproduction of a 10kHz squarewave (fig.2). Fig.1 was taken with the balanced inputs; the response with the unbalanced input was identical, though in bridged-mono mode, the output at 20kHz was down by a still-trivial 0.25dB, due to the doubling of the amplifier's output impedance in this mode. A slight 0.35dB channel imbalance in favor of the left channel (blue and cyan traces) can be seen in fig.1, even though the level controls for the two channels had both been set to their maximum positions.


Fig.1 Parasound A21+, 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.).


Fig.2 Parasound A21+, small-signal 10kHz squarewave into 8 ohms.

Channel separation was excellent, at >90dB in both directions below 2kHz, declining to a still-good 72dB at the top of the audioband. Measured with the unbalanced inputs shorted to ground, the amplifier's unweighted, wideband signal/noise ratio was 78.1dB ref. 1W into 8 ohms (average of both channels), this ratio improving to 88.7dB when the measurement was A-weighted. Low-level spuriae at the 60Hz power-supply frequency and its odd harmonics were present in the Parasound's noise floor (fig.3). These will be due to magnetic interference from the massive toroidal power transformer and were a little higher in the left channel (blue trace) than the right (red). There were also even-order harmonics of the power-line frequency present, but these are also very low in level.


Fig.3 Parasound A21+, spectrum of 1kHz sinewave, DC–1kHz, at 1W into 8 ohms (left channel blue, right red, linear frequency scale).

Parasound specifies the A 21+ as delivering 300W into 8 ohms (24.8dBW) with both channels driven and 500W into 4 ohms (24dBW). Using our definition of clipping, which is when the output's percentage of THD+noise reaches 1%, the Parasound easily exceeded its specified powers with both channels driven with a 1kHz signal. It clipped at 400Wpc into 8 ohms (26dBW, fig.4) and at 620Wpc into 4 ohms (24.9dBW, fig.5). The maximum power into 2 ohms with one channel driven was 900W (23.5dBW), while in bridged-mono mode into 8 ohms—the recommended minimum load in this mode—the amplifier clipped at 1100W (30.4dBW).


Fig.4 Parasound A21+, distortion (%) vs 1kHz continuous output power into 8 ohms.


Fig.5 Parasound A21+, distortion (%) vs 1kHz continuous output power into 4 ohms.

The shape of the traces in figs.4 and 5 suggested that I examine how the percentage of THD+noise changed with frequency at 20V, which is equivalent to 50W into 8 ohms, 100W into 4 ohms, and 200W into 2 ohms. The THD+N was very low in the midrange into 8, 4, and 2 ohms (fig.6) but rose at higher frequencies. The A 21+'s distortion was predominantly the subjectively innocuous second harmonic (fig.7), and higher harmonics are all very low in level (fig.8). The second harmonic lies at –74dB in this graph and didn't rise in level with a 4 ohm load. When the amplifier drove an equal mix of 19 and 20kHz tones at 50W into 8 ohms (fig.9), the second-order difference product at 1kHz lay at a very low –86dB (0.005%), and higher-order intermodulation products were even lower in level. Again, the levels of the distortion harmonics didn't rise when I kept the output voltage the same but reduced the load impedance to 4 ohms.


Fig.6 Parasound A21+, THD+N (%) vs frequency at 20V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta) 2 ohms (left gray).


Fig.7 Parasound A21+, 1kHz waveform at 25W into 8 ohms, 0.01% THD+N (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).


Fig.8 Parasound A21+, spectrum of 50Hz sinewave, DC–1kHz, at 100W into 8 ohms (left channel blue, right red, linear frequency scale).


Fig.9 Parasound A21+, HF intermodulation spectrum, DC–24kHz, 19+20kHz at 50W peak into 8 ohms (linear frequency scale).

The Parasound A 21+ offers a lot of low-distortion power and isn't fazed by impedances as low as 2 ohms.—John Atkinson

Parasound Products, Inc.
2250 McKinnon Ave.
San Francisco, CA 94124
(415) 397-7100

Glotz's picture

and thank you for the comparisons of the Halo A21+ with the Benchmark AHB2, as I am in the market for a power amp directly in this price segment. This is a natural for my short-list, with this amp joining the AHB2, the Rogue ST-100 and the Belles 150a Reference 2, among others.

The Benchmark HPA4 preamp, the sister to the LA4, is fantastic on every level and I am so grateful for the review which led to my satisfied purchase!

Oh, one other thing- Regarding the Audio Research MP-1: The IC's that are in the unit can be replaced with more modern ones for much less hiss and self-noise. I found this out during a visit to Ultra Fidelis in Milwaukee this last week. They have performed the modification in the past for customers as well. Food for thought?

Kal Rubinson's picture

Food for thought, indeed. Thanks.

rmkessler's picture

I read with interest your previous review of the AHB2 amp. While overall you appeared to be impressed with the AHB2, you reported that it did not mate well with certain speakers. I see that you have been using two AHB2 amps in mono. Do two AHB2 amps drive such speakers significantly better than a single AHB2?

Kal Rubinson's picture

The only problem that I had was with an AHB2 with the Monitor Audio Silver 8s in my CT system and I have not tried them with the monoblocs. The Benchmarks live in the city where, so far, they have not disappointed.

Bogolu Haranath's picture

Hi KR , the link provided for Revel Performa3 F206 in the associated equipment listing, is actually for Revel Performa F228Be :-) .......

Kal Rubinson's picture

Nope. I no longer have the Revel Performa F228Be but I bought a pair of the Revel Performa3 F206 speakers for surround channels, primarily.

Bogolu Haranath's picture

When I click on the link, that connects me to your review of F228Be ...... You can try clicking on the link :-) ......

Kal Rubinson's picture

Aha! You are correct. I cannot fix that but I will notify someone who can.

Glotz's picture

The sister site has Hi-Fi News used to have other measurements for the AHB2 amp that showed the amps performance under dynamic conditions vs. continuous conditions. (One can still search for the original re-printed review though.) There is no headroom with the amp whatsoever, but it still does 200Wpc into 4 ohms, and the monoblocks close to 500Wpc. Greater overall capacity with the monos, obviously. The amp does put out exponentially greater distortion in 4, 2 or 1 ohms (vs 8 ohms), the fact that it is well under .01% still (vs. 0005%) makes it a non-issue.

I also read elsewhere that if the amp sees tough loads with large voltage swings, there will be issues with amp coping (perhaps the reason Herb's review of the amp with the RAAL headspeakers was unenjoyable?), something the A21+ above avoids completely, due to its far greater power capabilities.

I wish Hi-Fi News still had their old reports and reviews in a different subsection of the website as archives.

rmkessler's picture


Many thanks for your thoughts regarding the ABH2. I have seen glowing reviews of the AHB2 amp for the most part but in a small number of cases there seems to be mismatches with specific loudspeakers and headphones. I suspect that you have put your finger on the reason for these mismatches. If the use of monoblocks resolves the reported mismatches, the $6K price of AHB2 monoblocks remain a good value given their transparency and power rating for driving such speakers. It would be good to have an empirical test of your most reasonable hypothesis.

Bogolu Haranath's picture

For the price of two mono AHB2s, one could also consider the bigger model stereo Parasound JC 5 :-) ......

Glotz's picture

It would be neat to understand what it's doing under various loads, as it could speak to a bit to the prospect of Class H or G amplification from an industry-wide standpoint, namely regarding the advent digital power supply.

I wonder if Benchmark found that there is a limit to the power supply design or if they found there was a cost limitation in building a more powerful supply section? I'd like to see this design implemented over Class D any day...

Ortofan's picture

... (and one-third the price) Parasound NewClassic 2250 v.2, which has a 2-channel power rating of 275W into 8Ω and 400W into either 4Ω or 2Ω. Bridged/mono power rating is 750W into either 8Ω or 4Ω.

Bogolu Haranath's picture

Another amp which could also be considered ......... AudioControl Avalon G4 ($2,200), favorably reviewed by TJN for S&V magazine ....... 2, 3 or 4 channels, 600 WPC 8 Ohms, 2 channels :-) ........

Ortofan's picture

... (for $999 each) and still have money left over?

Bogolu Haranath's picture

Agreed ....... Parasound may be discontinuing some of the NewClassic lines ....... They seem to have the new ZoneMaster lines ........ I suggested the AudioControl, because it was reviewed by TJN and he compared it with the Parasound Halo 5 channel amp :-) ........

Bogolu Haranath's picture

Ortofan, I have a subwoofer for you ....... Ascendo Seismic Subwoofer (see, S&V website) ....... 50" diameter driver ...... 88 pound magnets, 14" diameter voice coil, 6 feet tall, 140 db max SPL ....... You can hear lowest organ frequences, as well as earthquakes, train crashes and MOAB bomb explosions :-) ........

trmoore2's picture

I'm sure this is a wonderful amplifier, and I know there are significant differences in speakers' quality, but I never see A/B tests among amps (like Revel does with their speakers). Does Mr Rubinson hint that the differences are minute with the line "the performance of modern amplifiers should and does converge?" How much does my sound "suffer" by using a $2000 Marantz integrated amp/receiver or an Emotiva amp? If this amp is 10% "better," that number is not worth testing the waters of diminishing returns. Tell me if I'm wrong...

Kal Rubinson's picture

How much does my sound "suffer" by using a $2000 Marantz integrated amp/receiver or an Emotiva amp?

Dunno. I have not had an integrated amp or AVR in house in a very long time. When I had an Emotiva (, it did not satisfy me.

If this amp is 10% "better," that number is not worth testing the waters of diminishing returns.

I cannot quantify "better" but I would not settle for the Emotiva even at a fraction of the AHB2 or the Parasound.

Tell me if I'm wrong...

I cannot.

trmoore2's picture

I tested a Revel set up (2.0) with a MacIntosh amplification at the dealers and it sounded wonderful. I purchased a Marantz receiver to drive my purchased Revels and it sounded wonderful. Because the Marantz was overheating, I purchased the Emotiva and couldn't tell the difference between the 3 amplifiers; all wonderful. I mention this not to take away from the Parasound, but to help people mystified or susceptible to the charms of high priced audio equipment. Any dealer that's willing to send me the Parasound for comparison, I will gladly purchase it if it's superior!

RockMan85's picture

But in the article you linked, this is a direct quote...

" Still, for $899. . . heck, forget that conditional: I am greatly impressed with the Emotiva XPA-5."

Can you please explain this discrepancy?

Kal Rubinson's picture

No discrepancy. Quality product but one can do better.

sjeffers's picture

Just curious what you mean by "the dynamics are demo-quality". Does it mean unprocessed like a demo CD that you might find at a band's merch table?

Kal Rubinson's picture

The statement merely acknowledges that RR recordings are known for their wide dynamic range (R128 dynamic range is 22.2LU on the track discussed) and that they are often chosen to demonstrate that at audio shows. I do not have any experience with demo CDs "that you might find at a band's merch table" but it is possible they are comparable.

Long-time listener's picture

"Unless an amp designer imposes her will ... "

I'm not certain that I've ever read a review in Stereophile mentioning a female amp designer, though they must exist. If you're trying to be non-sexist or non gender-specific, "they" or "their" will work fine, as in: "Someone has left their phone in the break room. They can claim it at the front desk."'s picture

Parasound needs to promote the designer that put the gold trim strip in the faceplate. Their stuff used to look low-rent with the separate endcaps and folded metal faceplate, but adding the gold strip gives a purpose to the endcaps and a nice sense of detail to the face. well done.

Kal Rubinson's picture

If you say so. I hadn't noticed any change.'s picture

Interesting. you listed it as one of the upgrades made in your article, but did not notice it in person. Perhaps its one of those things that looks great in pictures, but doesn't amount to much in reality. I will have to take a gander at my local dealer. maybe even listen. haha

Bogolu Haranath's picture

No meters? :-) .......

Kal Rubinson's picture

Ya got me! I am color-blind. :-)'s picture

got ya! now I am King of the Forums!

I tried to tell my wife I was gold/silver color blind but it didn't work.

Bogolu Haranath's picture

That may not be a problem ....... Traffic lights are not gold or silver :-) .......

tlathbury's picture


Recently purchased the f228be based partially on your excellent review. Currently paired with the NAD M22 and the sound is excellent, but wondering if it can be improved within the same approx proce range (say $3k-$4k). Maybe larger soundstage, fuller/richer mids. Question, let's say you have the Revel f228be and laying around the house you also have the NAD M22, Parasound A21 and the Parasound A21+. Which one to you keep connected to the speakers, and why ? You personally.

Thanks !

Ortofan's picture

...a larger (apparent) soundstage, then try a tube amp.

In his review of the ($3,495) Rogue Audio Stereo 100 power amp, HR stated that the amp in triode mode "delivered bigger, deeper, more CinemaScope soundstages populated with denser, more vivid three-dimensional aural images" and "in my view, the Stereo 100 is everything—sonically, mechanically, aesthetically—that we could hope for in a 21st-century tube amp. Everything about it screams "End-game tube amp, long-term keeper!""

tlathbury's picture

Thank you for the suggestion.

Kal Rubinson's picture

Me, personally? Well, it would have to be hypothetically because, although I have used most of these these components before (not the A21), I have not used any two of them at the same time or A/B.

If I was forced(!) to buy a pair without audition, I would mate the A21+ with the f228Be. I would anticipate(!) that the pair would have "fuller/richer mids" than the F228Be with the M22. I have not heard/used the M22 v2. That's all I can offer (with included qualifications).

tlathbury's picture

OK thank !

Bogolu Haranath's picture

F228Be have two 8" woofers in the bass frequency ....... Rogue 100 amp may not provide enough control ....... A 21+ could provide sufficient control of all the drivers in F228Be ....... The bass could be more impactful with A 21+ ....... Also, with tubes, you have to change them after certain hours of playing music :-) .......

Ortofan's picture

... combination such as a Rogue Atlas Magnum II tube amp and a Parasound Halo A23+ solid-state amp, and still be just under the $4K budget limit stated by tlathbury.

Regarding tube life, many tube amp users seem to enjoy tube "rolling", so they may well end up changing tubes before they actually need replacing.

Bogolu Haranath's picture

If $4k is the budget, one could choose the Rogue Audio DragoN tube/transistor hybrid amp ....... Tube input and Hypex Class-D output ...... 300/500 WPC 8/4 Ohms :-) .......

tlathbury's picture

Thanks, looks interesting...can't find any reviews.

Bogolu Haranath's picture

I know ...... DragoN hybrid amp is a new model from Rogue Audio ....... That is probably the reason why no reviews, yet ....... Rogue Sphinx hybrid integrated amp was favorably reviewed by HR for Stereophile ....... Given the track record of Rogue Audio, my guess is that the DragoN also should sound good :-) .......

Bogolu Haranath's picture

Rogue Audio similar hybrid, less power Hydra was favorably reviewed by TAS :-) .......

tlathbury's picture

Thanks again !

tlathbury's picture

Thanks, may be more boxes than I'd like to deal with.

Bogolu Haranath's picture

NAD M22 is a pretty good, powerful amp ........ Why you want to change it? :-) .......

tlathbury's picture

It's difficult to audition amps. There's no where I can go to listen to all the different amp options with a specific speaker model. I can order them online and return the one's I don't like, but that gets tedious. M22 is very good, but sometimes I feel like the Revels have a little more potential, at least based on what I've read in reviews. So trying to validate with opinions I feel are reliable.

davemill's picture


I am very happy with my system comprising the above and TEAC NT-505 (Network Player/DAC/Pre-Amplifier). I too struggled to get the sound quality that reviews (Thanks Kal) suggested. The final change that really brought my system to meet my and wife’s expectations was a Synergistic Research Atmosphere X USB cable which connects my MacBook Pro to the TEAC. I had also tried several other USB cables such as AudioQuest Carbon and Diamond and Kimber Kable CU. None of these sounded nearly as good as the SR. TEAC connects to the A21+ using a set of AudioQuest Earth XLR cables. Selected the A21+ based on Kal’s review of the F228Be’s. He tried the Classe amplifier which is a class D like the NAD and found the other 2 amps to be better match with the F228Be so I decided to go with the new Parasound. Good luck.

growboxguy's picture

Greetings, excellent review. I am down to two amplifiers either a Parasound Halo 21+ or a Bryston 3b3. The price is nearly identical, about $400 more on the Bryston. I am running on a pair of Focal Aria 926 but plan on upgrading to a set of Paradigm Persona 3f's as soon as finances allow. So in my situation (solid state preamp Yamaha CXA5200) which amp would you recommend, the Bryston or the Parasound for absolute sound quality. Thank you so much for the hand.

davemill's picture

Hi growboxguy,

I don’t know where you are located but AudioAdvisor is a US online dealer for both Bryston and Parasound. I actually purchased my A21+ from them. They currently have the 3b3 on sale for $5215.50 while the A21+ is $2995. Although price isn’t indicative of which would be better, a 3b3 for only $400 more than the A21+ sounds like a better deal. The less powerful Bryston 2b3 is on sale for $3865.50 which still more than $400 than the A21+.

Vintageaudio66's picture

After many years I'm getting back into quality audio. Bryston 3B3 is high on my list, but considerably more expensive. Is it that much better? Stereophile has also reviewed the PS Audio Stellar M700. Similar power and price. How does it compare with the Parasound 21+. Thanks

Max Earshot's picture

Great review. Thank you!
I am considering the following Power Amps to drive my Monitor Audio PL300ii. Which one would you recommend?
- Parasound A21+
- Anthem MCA225 Gen2
- Cambridge Audio Azur 851W

Thank you.

davemill's picture

I owned an A21+ for about 13 months. I used it with my Revel F228Be speakers. Midrange clarity and bass response were lacking. I say that after trading in the amplifier for a Pass Labs XA-30.8. The Pass hasn't been in my house for more than a couple of hours and I can't believe how great the F228Be's really are. OK, I thought they would be but the supporting equipment makes a HUGE difference. An A21+ just isn't up to the task. If Stereophile gives it a Class A rating, the XA-30.8 is a Class A+++! Class A ratings should not be cost dependent and especially not advertising dependent.