This is a review and detailed measurements of the Audio Precision's APx516 audio analyzer. It is on loan from the company and costs from US $6,000. As tested, the cost is around $9,000.
The APx516B is featherweight compared to the much, much heavier APx555 analyzer I currently have. You can lift and carry it with one hand with no problem. Importantly, while it has a fan, it never came on during testing whereas my APx555 screams loudly requiring wearing a headphone to block it. A noisy analyzer could raise the noise floor of headphone measurements making the APx516 superior to my analyzer in that regard (I have mostly mitigated that however).
What one of the major reasons to get an Audio Precision analyzer is its software. It is easy to learn and given its closed loop nature, highly functional. The latter means the software can instruct the analyzer to do something, measure the results, and then continue with the next step. Such sweep measurements are critical and super useful. Other analyzers tend to run a test signal, capture the results and that is that. Note however that features are unbundled from the analyzer and may require additional licensing to use them.
Needless to say, the company was not about to obsolete its near US $30,000 APx555B series with this unit so performance is degraded. How much is the question. Let's find out.
Audio Precision APx516 Measurements
Let's start with a set of "loopback" measurements where the analyzer's analog signal generator feeds itself. Here, I simulated a balanced analog device with voltage set to 4 volts, measuring the combined performance of the signal generator and analyzer:
Distortion is comfortably low at around -119 dB. Noise however is high enough to sap fair bit of that performance bringing SINAD down to 106 dB. Here is my APx555 analyzer running the same test:
We see vanishingly low distortion courtesy of a dedicated analog generator instead of the DAC used in APx516. In addition, the APx555 uses a pair of ADCs to capture the signal independent of the noise and distortion resulting in much less distortion.
Performance however is voltage dependent so let's sweep that and see what is going on:
We see that the gap is small during start of the sweep but once we get to around 0.22 volt, the APx516 auto-scales the input signal to better match its ADC dynamic range. This is much earlier than APx555 which runs "native" until 2 volts. Still, at lower measurement voltages, the gap is not that large.
Keep in mind that some other capabilities are limited such as what is noted: output voltage on APx516 is limited to around 14 volts. This limits testing of some ultra-low-gain amplifiers which may require as much as 20 volts to drive them to full power although this is rare.
We can bypass the analog generator effects by measuring a DAC, specifically the Topping D70s that I have on my desk. That way, only the analyzer performance is in play. Here is APx516:
Compare that to APx555:
The APx555 has much lower noise floor so visually the distortion spikes look taller but they are not. Not that it is material but the lower noise floor of APx555 reveals higher order distortions that are masked by the APx516.
Surprisingly, dynamic range measurements are the same between the two analyzers:
What this means is that for testing most of the recommended DACs, the APx516's own performance gets in the way. This much, I knew. What I wanted to find out was whether it could be good enough to measure amplifiers as their performance is not as good as DACs. So I powered up the Purifi reference amplifier and tested it with both analyzers:
As noted, up to about 18 watts (8 ohm load) performance is essentially the same. This means that we could compute the SINAD at 5 watts with either analyzer and arrive and close enough results. The rescaling of APx516 at 18 watts, sets it back a bit but that effect diminishes as the amplifier voltage increases, closing the gap between the two analyzers. Even at the limit, we are not talking about big difference.
Conclusions
The APx516 addresses two big shortcomings of the top of the line APx555 analyzer: cost and noise. It is also much more portable and having had to lug mine to events, can be an advantage if you are so situated. Naturally, you give up performance at the top end, limiting you to SINAD of about 105 dB or so. This is good enough to test many amplifiers especially since SINAD at 5 watts is close to my analyzer.
Yes, there are (much) cheaper analyzers out there. But you give up one major thing in the context of ASR: identical test parameters and graphing that I perform with APx555. This allows readers to much more easily comprehend the measurements.
If you are a manufacture and can't afford the APx555, and aim to have your product reviewed by me, I highly recommend getting the APx516. It will give you familiarity with my testing and very nice preview of what I may find.
Overall, I really like the APx516 with its light weight, much more reasonable price and good enough capabilities.
I am going to put the Audio Precision APx516B on my recommended list.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
The APx516B is featherweight compared to the much, much heavier APx555 analyzer I currently have. You can lift and carry it with one hand with no problem. Importantly, while it has a fan, it never came on during testing whereas my APx555 screams loudly requiring wearing a headphone to block it. A noisy analyzer could raise the noise floor of headphone measurements making the APx516 superior to my analyzer in that regard (I have mostly mitigated that however).
What one of the major reasons to get an Audio Precision analyzer is its software. It is easy to learn and given its closed loop nature, highly functional. The latter means the software can instruct the analyzer to do something, measure the results, and then continue with the next step. Such sweep measurements are critical and super useful. Other analyzers tend to run a test signal, capture the results and that is that. Note however that features are unbundled from the analyzer and may require additional licensing to use them.
Needless to say, the company was not about to obsolete its near US $30,000 APx555B series with this unit so performance is degraded. How much is the question. Let's find out.
Audio Precision APx516 Measurements
Let's start with a set of "loopback" measurements where the analyzer's analog signal generator feeds itself. Here, I simulated a balanced analog device with voltage set to 4 volts, measuring the combined performance of the signal generator and analyzer:
Distortion is comfortably low at around -119 dB. Noise however is high enough to sap fair bit of that performance bringing SINAD down to 106 dB. Here is my APx555 analyzer running the same test:
We see vanishingly low distortion courtesy of a dedicated analog generator instead of the DAC used in APx516. In addition, the APx555 uses a pair of ADCs to capture the signal independent of the noise and distortion resulting in much less distortion.
Performance however is voltage dependent so let's sweep that and see what is going on:
We see that the gap is small during start of the sweep but once we get to around 0.22 volt, the APx516 auto-scales the input signal to better match its ADC dynamic range. This is much earlier than APx555 which runs "native" until 2 volts. Still, at lower measurement voltages, the gap is not that large.
Keep in mind that some other capabilities are limited such as what is noted: output voltage on APx516 is limited to around 14 volts. This limits testing of some ultra-low-gain amplifiers which may require as much as 20 volts to drive them to full power although this is rare.
We can bypass the analog generator effects by measuring a DAC, specifically the Topping D70s that I have on my desk. That way, only the analyzer performance is in play. Here is APx516:
Compare that to APx555:
The APx555 has much lower noise floor so visually the distortion spikes look taller but they are not. Not that it is material but the lower noise floor of APx555 reveals higher order distortions that are masked by the APx516.
Surprisingly, dynamic range measurements are the same between the two analyzers:
What this means is that for testing most of the recommended DACs, the APx516's own performance gets in the way. This much, I knew. What I wanted to find out was whether it could be good enough to measure amplifiers as their performance is not as good as DACs. So I powered up the Purifi reference amplifier and tested it with both analyzers:
As noted, up to about 18 watts (8 ohm load) performance is essentially the same. This means that we could compute the SINAD at 5 watts with either analyzer and arrive and close enough results. The rescaling of APx516 at 18 watts, sets it back a bit but that effect diminishes as the amplifier voltage increases, closing the gap between the two analyzers. Even at the limit, we are not talking about big difference.
Conclusions
The APx516 addresses two big shortcomings of the top of the line APx555 analyzer: cost and noise. It is also much more portable and having had to lug mine to events, can be an advantage if you are so situated. Naturally, you give up performance at the top end, limiting you to SINAD of about 105 dB or so. This is good enough to test many amplifiers especially since SINAD at 5 watts is close to my analyzer.
Yes, there are (much) cheaper analyzers out there. But you give up one major thing in the context of ASR: identical test parameters and graphing that I perform with APx555. This allows readers to much more easily comprehend the measurements.
If you are a manufacture and can't afford the APx555, and aim to have your product reviewed by me, I highly recommend getting the APx516. It will give you familiarity with my testing and very nice preview of what I may find.
Overall, I really like the APx516 with its light weight, much more reasonable price and good enough capabilities.
I am going to put the Audio Precision APx516B on my recommended list.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited:
