廠家研發dac1時的思路 (擇自head-fi 論壇)︰
John Siau - Director of Engineering, Benchmark
We designed the DAC1 for maximum transparency. If you want to add warmth, you can't add it with a DAC1. Personally, I do not like what warm sounding equipment does to the sound of a piano. Warmth is wonderful on vocals, guitars and certain instruments, but it beats against the streched overtones of a piano. The overtones in a piano occur at slightly higher than harmonic ratios, and these create beat notes with the exact integer ratios produced by electronic equipment (and speakers). Too much harmonic distortion will make a piano sound out of tune.
Put on a pair of HD600 headpones, plug them in to the DAC1 headphone amp, and listen to a solo piano recording. This combination will deliver far lower distortion than any speaker/amplifier combination. The 0-Ohm headphone drive on the DAC1 is extremely clean under load.
The audio path is well isolated from the power supply rails, and the power supply rails are bypassed to their own ground system (not to the audio ground reference). The DAC1 will run just fine with lots of ripple and noise on the power supply rails. The regulators (as generic as they are) are orders of magnitude better than they need to be. The DAC1 can actually run with several volts of ripple on the +/- 18 volt rails (we have run this test in the lab) without any measureable change in performance. The +5V and 3.3V regulators are the most critical. Excessive noise on the 5 and 3.3 volt rails can cause jitter. The filter caps on these rails are oversized, and there is no measurable jitter due to noise on these rails. Again, by injecting noise onto these rails, we know what the margins are, and we know that we will not gain anything by changing the regulators.
This is also why the DAC1 will not benifit from high tech AC line cords and/or AC line filters.
Here is one way to domonstrate this: The DAC1 is designed to operate with AC line voltages as low as 90 VAC. Below 90 VAC the +/- 18 volt regulators will drop out, and AC ripple will increase rapidly as the input voltage is decreased further. At about 80 VAC the DAC1 will gracefully shut itself down by applying each of three digital mute circuits. Just above 80 VAC the DAC1 will opearate normally with no detectable hum or distortion in spite of the fact that it is operating with the +/- 18 V regulators in a drop-out condition. At 80 VAC there is about 3 Vpp AC ripple on the +/- 18 V rails. The only negative impact on performance is that the ripple reduces the headroom of the analog outputs. Normally the DAC1 can deliver +29 dBu to the XLR outputs. When the power supply is out of regulation, the maximum output level is reduced to about +27 dBu.
I highly recommend gold-pin Neutrik connectors, and Canarie L-4E6S cable. These have excellent electrical characteristics, and are exceptionally durable. All benchmark products, including the DAC1 are tested using these cables. The cables were made to order for us by H.A.V.E. 1800-999-4283, and by Markertek. Cost is about $10 to $15 per cable. These cables have been subjected to harsh use at our test stations for 6 to 8 years without a single failure. These cables are connected and disconnected from product many times a day. Markertek sells XLR to RCA adapter cables.
Inportant: When ordering XLR female to RCA male cables, please specify pin-3 floating (do not connect pin 3 of the XLR to ground).
We injected 20 Hz to 200 Hz noise onto the power supply rails to test for PSRR. We injected RF frequencies into all interconnect cables and into the AC line cord. We subjected the DAC1 to electromagnetic fields at frequencies up to 1 GHz at a magnitude of 3 V/M. We conducted electrostatic discharge tests, surge and transient tests, and more. See the "Certificate of Conformity" in the back of the DAC1 manual for a complete list.
The DAC1 continued to operate normally throughout all of these tests, and in most cases no change in performance was measurable. Under very severe test conditions at a few specific frequencies, we measured some very minor crosstalk, noise, or increase in distortion. In the very worst cases, the interfering signal or distortion still measured lower than -90 dBFS, while in most cases it was well below -107 dBFS THD+N spec of the DAC1. These are severe tests, and were designed to insure that the DAC1 will perform normally in a variety of adverse conditions. We sell equipment to Radio and TV broadcast facilities were the equipment can be subjected to strong RF fields. It is important that our products perform as advertised when conditions are less than ideal.
As for the tests in the Stereophile magazine, these reflect the performance of the A/D converter card in the Miller test set. In all cases the test results were limited by the test set converter card and not by the performance of the DAC1. The Miller test set cannot be used to measure the performance of the DAC1. The noise, distortion, and jitter of the Miller test set are much higher (in many cases an order of magnitude higher) than that produced by the DAC1 . We duplicated all of these tests our Audio Precision System2 Cascade. The AP System2 shows that the true performance of the DAC1 is much better than what the Miller test set indicates. Some of these tests are published in the manual, and some are available on our web site. The jitter immunity of the DAC1 is much better than the Sterophile tests indicate. We have been encouraging them to upgrade their test equipment.
Don't underestimate the NE5532. We have just completed a very comprehensive set of tests on the NE5532 and on various substitutes. We have not found anything that will equal the tranparency provided by the NE5532.
Some have suggested susbstituting OPA2134 op amps in place of the 5532s. This is a bad idea! Our tests show an increase in both 2nd and 3rd harmonic distortion, and the addition of higher order (4th, 5th, 6th, and 7th) harmonics that are virtually absent from a stock DAC1. In addition, IMD will increase, and SNR will degrade.
The NE5532 is power hungry, it has high input bias currents, and high offset voltage, but it can drive high-level low impedance circuits with ease. The 5532 should not be used with low signal levels, and it should only be used in low gain circuits. Also, the offset voltage must be managed with appropriate design techniques. I believe the NE5532 has aquired a bad reputation because it has often been missapplied. The DAC1 is carefully designed specifically for the NE5532 op amp. It may surprise you that the NE5532 was selected for transparency and not on the basis of cost.
