Empirical Audio Overdrive SE USB DAC/Pre

Device Type: Digital to Analog Converter
Digital Inputs: Asynchronous USB Audio Class 2.0: Up to 24bit, 44.1, 48, 88.2, 96, 176.4, 192 kHz; S/PDIF Coax Canare 75 ohm BNC – up to 24/192 kHz, I2S (Empirical standard RJ-45) up to 24/192 kHz
Output: 1 Pair RCA (Unbalanced) and 1 Pair XLR Balanced
Dimensions (H x W x D): 2.5 inches X 6.75 inches X 6.75 inches
Weight: 3.11 pounds

Substation AC Power Supply: (IEC AC Connector)
Dimensions (H x W x D): 2.5 inches X 6.75 inches X 6.75 inches
Weight: 2.9 pounds

Price: $5,999 (Overdrive SE DAC/Pre and Substation AC Power Supply)
Availability: Direct Sales
Website: www.empiricalaudio.com

Final Drive: Passive Transformer Device Inserted between DAC and Amps, or between preamp and amps.
Device Type: 2 Passive Transformers Buffers/ Selectors
Inputs: RCA, XLR 1, XLR 2
Outputs: RCA, XLR
Dimensions (H x W x D): 2.63 inches X 4.27 inches X 5.25 inches
Weight: 2.18 pounds
Price: $2,999

Short Block
Device Type: USB Filter
Price: $199

The Overdrive USB DAC/Pre is the creation of engineer-designer Steve Nugent of Empirical Audio. Empirical Audio was formed in 1994 and initially concentrated on the sales of patented cable products designed by Steve Nugent. Steve’s background included a degree in Electrical Engineering with 25 years of digital design experience. His resume included work for the Intel Corp. where he served as a design-team lead in the development of the Pentium II processor. Steve has also been a passionate audiophile for over 39 years and this interest lead him into the modification of other manufacturer’s audio products in 2002. His modifications not only included DACs, but preamps, power amps and CD players / transports. The modification of numerous designs gave Steve a wonderful opportunity to see what sounded best in audio and DAC design. During this time, Steve developed the Off Ramp Turbo USB Converter; a device that offered a USB interface to digital products. In 2009, Empirical Audio ceased equipment modification and concentrated on original computer audio product design.

The Design
The Overdrive SE DAC/Pre begins with a separate DC power supply with 3 high-current switching supplies followed by 3 linear analog Hynes voltage regulators. Empirical Audio believes that the design of the Substation Power Supply results in low noise while providing a fast reaction time for regulation. The power supply has over 60,000uFd of energy storage.

The DAC utilizes a USB asynchronous interface with the addition of S/PDIF Coax and an i2S input that can be driven from the Empirical Audio Off-Ramp or Pace-Car USB reclockers. Short signal paths are a key design feature of the Overdrive SE. The analog circuits incorporate Class-A design with a fully balanced analog path. High quality parts include a large number of Black-Gate electrolytic capacitors with point-to-point silver wiring. High quality output coupling capacitors are also used (V-Caps and Sonicap Platinum).

The analog path consists of an I/V stage op-amp that drives a single output transistor through a coupling cap. All stages are run Class-A. There are no resistors in the analog signal paths save for a very small damping resistor on the output.

The Overdrive SE has 2 clocks in the USB module section that provide for the 2 sample rate ranges on USB: 44.1/88.2/175.4 and 48/96/192. A dual Turboclock is utilized as an extremely low jitter dual clock with 2 Hynes voltage regulators.

The volume control is a combination of digital and analog design for direct drive of amplifiers from the Overdrive SE.

The Overdrive SE also features a small chassis size for short digital and analog signal paths to reduce noise from ground-plane coupling and crosstalk. An AC-coupled output is also included because DC-coupling would not allow a pure Class-A operation.

I asked Steve to describe the features of the Overdrive SE DAC/Pre design:

“The Overdrive design philosophy was one of minimalism. Because the #2 sound quality deficiency with modern DACs is poor Digital Filtering (#1 is jitter), the strategy was to create a DAC that sounds like a NOS DAC, but renders higher detail than NOS chips and supports up to 24/192. To accomplish this we bring-out three digital filter selections and recommend using the 192kHz filter for all sample-rates. Since the digital filtering is minimized for 44.1kHz, we add just enough low-order analog filtering to eliminate the visible "stair-steps" in the output signal. We determined with listening tests that this is optimum. This makes our measured distortion at ultrasonic frequencies a bit higher than the typical DAC; but these ultrasonic frequencies are attenuated by the preamp, amp and speakers anyway. The trade-off is worth it for the sound quality gains. Another benefit of this is measured impulse response at all sample-rates. The Impulse Response has very little pre and post ringing and achieves the full amplitude of the impulse, unlike most DACs that truncate the impulse due to slow-reacting power subsystems.”

“Another step we take to minimize the design is in the analog signal path. We utilize a single stage to do I/V conversion, analog filtering, gain and output drive. There are literally no resistors in the signal path except small snubbers in the output to minimize transmission-line effects in the cabling. Resistors add thermal noise to the system, so we avoid them. Multiple active stages compound the compression, noise and distortion so we use one stage. The stage runs pure class-A, so there is no crossover distortion. The output is capacitively coupled as a result. We have taken a lot of care in choosing the optimum output capacitor combinations that deliver the most natural and extended response. The 4 analog paths are identical and provide an active balanced output. We realized that active balanced outputs can never actually be TRULY BALANCED, so we offer the Final Drive transformer buffer/selector to create a truly balanced output. This tends to make amplifier balanced inputs very happy and has the added benefit of galvanic isolation, eliminating the ground-loop between DAC and amps. It also allows one to bring-in other analog sources such as home theater and vinyl after the DAC.”

“A third step we take is in the volume control technology. No one would disagree that the best performance will be achieved when the line-out of a DAC is directly connected to an amplifier. This of course is hazardous with the typical DAC and requires the use of DSP software volume control to reduce the volume. Speaker damage is obviously a concern with this technique, as well as bit decimation. The Overdrive uses a volume control technology that connects the line-out directly to the amps, but uses the D/A conversion reference voltage to reduce the volume rather than DSP. The reference voltage is the voltage that is used by the D/A to generate the height of the stair-steps that make up the analog output waveform. There are no other devices or gain stages added or being changed with volume adjustment in the Overdrive. One gets the benefit of direct line-out to amp connection without having the hazards or the decimation that occurs with DSP software volume control. This volume technology is unlike all other techniques in that it actually reduces distortion levels as the volume is decreased. This is exactly what one wants from an ideal volume control. Direct amplifier connection is what we recommend.”

“The fourth step we take is to minimize jitter at the system level. Our independently powered Async 192 USB module delivers very low jitter using I2S direct connection over a 6GHz stacking connector from module to baseboard. The I2S clocks are delivered to the D/A through 50-ohm embedded coax connectors. This allows for very fast risetimes, minimizing system jitter and radiated noise. The clock used in the Overdrive is the Turboclock, which is a separately powered module using custom oscillators. The Turboclock connects to the USB module using 6GHz-rated coax cables.”

“The Overdrive circuitry is not minimized in all areas however, only the important ones. In areas of power supply we take it to the limit, providing 16 separate regulators and three separate power supplies/voltages from the Substation to the Overdrive. This is combined with optimized capacitive power decoupling to achieve unprecedented separation, imaging, dynamics and bass extension/control. I would be remiss to not mention that these voltage regulators are all Paul Hynes designs implemented by Empirical Audio. We found that Hynes regulator circuits use a unique topology that enables them to be both extremely low-noise as well as fast responding, which is ideal for maintaining low-jitter in digital.”

User Features
The Overdrive SE’s cases are attractive aluminum with a very open vented design due to the Class-A operation and the resultant warm operation. The short power umbilical from the Substation Power Supply allows the DAC to be placed alongside the power supply for optimum heat dissipation. There is a power switch that provides DC to the Overdrive DAC unit. 3 LED indicators display the 3 DC voltages supplied to the DAC. When the power switch is off, only the Substation is powered with no DC output.

The Overdrive SE DAC is quite simple and attractive in design. A toggle switch selects from the 3 inputs while another toggle switch selects from Low, Medium, and High frequency digital filter roll-off. The High is the down position and is the recommended filter. A De-emphasis Enable toggle switch when enabled applies an equalization at the high-frequencies that eliminates hardness due to high-frequency pre-emphasis in some recordings. The down or Normal position is recommended.

There are 2 recessed switches; one allows High/Low gain. The other selects the volume control In or Out. Two Volume Range jumpers are also provided and are located just inside the front panel to optimize the operation of the volume control. 3 LEDs indicate USB operation, Data flow, and DC power.

The Volume Control is designed to provide best performance when at near maximum position.

Set-up
The Overdrive SE utilizes M2Tech OSX and Windows drivers. I used the “Snow Leopard” drivers for OSX Mavericks as recommended by Empirical Audio. No matter what I tried, I could not get the OSX driver to work in Mavericks. Steve’s preferred OSX playback device is a 2009 Mac Mini with Snow Leopard driven by a Hynes linear power supply. He also likes to use the Amarra software player.

The Windows 64 bit driver had no issues installing into my Boot Camp Windows 8.1 Pro 64 operating system. JRiver Media Center Player 19 was successfully used in my listening evaluation of the Overdrive SE. I also listened to JPlay 5.2 paired with JRiver, but had infrequent lock-ups with this combination and the Overdrive SE driver. I did most of my listening using JRiver with the Kernal Streaming output selection.

I first listened to the Overdrive SE played into my Ayre KX-R preamp with single end and balanced outputs from the DAC. I found that I preferred the sound of the balanced outputs from the Overdrive. I will also be discussing my experiences with direct drive of my amps from the Overdrive SE and the use of the Final Drive units.

Initial Sonic Impressions
The Overdrive SE’s resolving capability is one of the most impressive that I have yet heard from any DAC. The Overdrive SE is able to delineate simultaneous playing multiple voices and instruments in a natural and musical fashion without over accentuation or an unnatural etched sound. This DAC is very detailed from top to bottom with superb bass control and slam. The Overdrive SE is never hard sounding or fatiguing to listen to for extended periods. The DAC is very quiet, and no doubt, this quality contributes to its revealing quality. The soundstage is wide and deep running Windows 8.1 Pro 64 even without the soundstage enhancement found when running JPlay 5.2 with JRiver.

Extended Listening
Reference Recordings HRx recordings played back with finest sound I have yet heard from these titles. I have always found a slight dryness to the Reference Recording sound of their digital HRx recordings. The Overdrive SE played these titles without this quality resulting in a totally engaging musical experience. Eije Oue / Minnesota Orchestra Rachmaninoff Symphonic Dances (24/176.4) never sounded as good as it did with the Overdrive SE. The Overdrive did a superb job reproducing the hall acoustics with a wide and deep soundstage. The air and bloom around the instruments was reproduced with rich tonal colors.

But it was not only orchestral music that benefited from the Overdrive SE, but well recorded vocals such as Reinaldo Brahn’s Brasileiro Soul. This 24/176.4 Reference Recordings HRx also sounded better with the Overdrive SE than any other DAC I have heard it reproduced with. The voice and instruments were reproduced without image smearing and excellent image solidity. The Overdrive SE had a slight forward quality to the sound that I really liked. Again, the studio soundstage was the largest and most open that I have heard with this excellent recording of Brazilian samaba-jazz style music.

The Overdrive is very good at reproducing dynamic contrasts with good pace and rhythm. The Dunwells Blind Sighted Faith 24/96 (HDtracks) delivered solid bass with good impact. The resolution of transient detail was also very good with this DAC endowing instruments and voices with pleasing clarity.

While I have always enjoyed Young The Giant’s Mind Over Matter 24/96 (HDtracks), I have felt that the voices were not well recorded and sounded lost in the instrumentation. Not so with the Overdrive SE. The Overdrive SE reproduced the vocals with far more clarity and focus than I have previously experienced.

The Final Drive
I began my evaluation of the Final Drive by listening to the Overdrive SE connected with balanced cables directly to my Ayre MX-R amps. While the sound was good, I preferred the result obtained when listening to the Overdrive SE through my preamp. I did feel that using the Overdrive’s volume control to directly control my amps sounded better than any other DAC I can remember that had an internal volume control be it analog or digital.

Connecting the Final Drive between the Overdrive and my amps was a different story. The Final Drive is sold as a pair of passive transformer buffers/selectors that are inserted between the DAC and amps or between a preamp and amp. The Final drive allows one to have multiple inputs; 1 set of RCA and 2 sets of balanced inputs that can be selected with a front panel rotary input switch. Gain can also be selected between 0 and +6db boost for each input. There is a choice of RCA or XLR outputs.

Theory of Operation (From the Manual)
The Final Drive is a passive device requiring no power. One is required for each analog channel. It consists of a signal transformer, switches and connectors with connecting wiring. The main functions are:

  • Isolation of ground
  • Amplitude/level shifting
  • True balanced output
  • SE to balanced conversion
  • Balanced to SE conversion
Using the Final Drive between the Overdrive SE and The Ayre MX-R amps improved the overall sonic results considerably. I felt that there was significant improvement in terms of a lower noise floor with enhancement of microdynamic nuances and rhythmic drive. The sound from the Overdrive SE combined with the Final Drive was more relaxed with greater purity and liquidity. Did I find the Overdrive SE/Final Drive combination to be superior to the Ayre KX-R preamp? I actually liked both combination, but did find them to sound different from each other. I had an overall preference for the Ayre KX-R, but I guess it’s a matter of taste. All of this demonstrated the positive effect of the Final Drive and just how good the Overdrive SE performed when combined with this product.

The Short Block USB Filter
Most of my listening was done with this USB filter that is placed at the input of the DAC. The Short Block reduces jitter caused by common-mode noise on the USB cable. This passive device severs the +5V wire in the USB cable. The Short Block cannot be used with USB DACs that rely on the +5v connection between the computer and DAC. I did notice an improvement in low end definition and high end detail. The soundstage seemed to be a little larger using the Short Block. Background silence was also enhanced with the Short Block. I was unable to try the Short Block with other DACs on hand since these DACs required the +5v USB line to function.

Comparisons with Other DACs
The MSB Technology Analog DAC at $6995 welcomes comparison with the $5999 Overdrive SE. These two DACs sound quite different from each other. The Analog DAC is bigger sounding with a less forward presence to its sound. The Overdrive SE is faster sounding, particularly in the low end. The Analog DAC has a more “relaxed sound” overall compared to the Overdrive SE. If one adds the Short Block to the Overdrive SE, there is an increase in overall definition and further reduction of noise with an even blacker background than that heard with the Analog DAC and its Desktop Power Supply.

The Overdrive SE sounded closer to the tonal balance and general sound of the Bricasti Design M1 DAC. The $8995 Bricasti was even faster and more resolving in the midrange and highs than the Overdrive SE and while possessing much of the general ease of the Analog DAC. Adding the Final Drive, Short Block, and direct balanced connection brought the Overdrive SE’s performance closer to that of the Bricasti Design M1.

Summary
I enjoyed the time spent with the Empirical Audio Overdrive SE USB DAC/Pre. I found this DAC to be an excellent performer with speed and definition that rivals some of the most expensive designs I have heard. When combined with the Final Drive, the Overdrive SE is more than capable of replacing an expensive preamp using its built-in volume control. Many will find the Overdrive SE/ Final Drive combination to be superior to their expensive analog preamps. I feel that the Short Block makes a very positive improvement to the Overdrive SE’s sound, and should be purchased with this DAC.

Hopefully, Empirical Audio can find a solution to the OSX Mavericks driver issue. The DAC does work with Snow Leopard, Lion, and Mountain Lion using the Snow Leopard driver posted at the Empirical Audio site. While Snow Leopard supports native integer playback, Lion and Mountain Lion do not. I personally feel that Lion and Mountain Lion, even without integer support, sound better overall than the Snow Leopard operating system with integer. Apple's latest operating system, Mavericks, supports native integer playback; a feature that is a real sonic plus for DACs that can utilize this format.

That issue aside, I feel that the Empirical Audio Overdrive SE DAC/Pre offers substantial performance with considerable attention to design detail that should appeal to those interested in a quality high-end DAC.

COMMENTS
tubefan9's picture

I like Steven's reviews because I feel like they come from the heart.

Steven Plaskin's picture
tubefan9, The compliment is appreciated. Thanks, Steve
ketcham's picture

Steven,

Did you ever place the final drive between your Ayre preamp and amp monoblocks? What was your impression? I found when I bypassed my $12,000 preamp the effect was like removing thick wool blankets that were wrapped around my speakers three times. This was not subtle. A passive preamp at a fraction of the cost has significant benefit.

Steven Plaskin's picture
I did try the Final Drive between my preamp and amp with no noticeable improvement in sound. As I stated in the review, I really liked the Final Drive. Just how all of this plays out will depend on your preamp-amp and setup. But going to the Overdrive-Final Drive-and my amps did not result in a substantial improvement compared to the Overdrive-Ayre Preamp and Ayre amps.
scully280's picture

Steven,
After adding a Empirical Overdrive to my system several years ago I was initially quite happy with every aspect, until I tried to use my FM tuner. Wideband noise from a new, unknown source had made most FM listening useless. Disconnecting the Overdrive power, however, cleared up the interference. I took the Overdrive to work, where we have an Electromagnetic Compatibility test chamber and ran the standard FCC test for consumer products. The Overdrive failed the radiated emissions test miserably. Fortunately the application of a clip-on ferrite bead to the USB cable allows the Overdrive to pass the test, and it does significantly reduce the FM band interference in my system, but doesn't eliminate it completely completely. I was disappointed by this oversight and possible violation of regulations, and I was wondering if you have noticed any EMI compatibility issues with the equipment you have reviewed?

Steven Plaskin's picture
I'm sorry to hear this about the Overdrive SE. My setup and amplification does not appear to be sensitive to RFI / EMI; at least not in my limited reviewing experiences. My Ayre KX-R preamp, I hope to update it to the Twenty Model eventually, is pretty resistant to most extraneous RFI / EMI. Empirical Audio should respond to this issue.
scully280's picture

I've made a an error with the name of the device i have, it's an earlier Off-Ramp 4 not an OverDrive. I have never tested or heard an Overdrive and I can't comment on any aspect of it's performance. My experience with the older Off Ramp made me wonder how a current Empirical product behaved. I regret that my error with the name of the two products may have caused some confusion.

To put the technical issue in perspective, the frequencies involved are all over 30MHz, and the levels are around 40 micro-volts, so only wide-open components will allow the signals to get into their circuits, and I suspect most non-radio audio gear isn't wide open and will reject the interference. So it's quite possible most audiophiles will not experience any direct interference except with FM tuners. As to the regulatory issues this is something Empirical Audio may want to contemplate.

Regardless of all this, I continue to use my Off Ramp with the added ferrite bead as my primary DAC driver!

Empirical's picture

The EMI compatibility problem you experienced was likely due to the cable and computer you were using and the size of the ground-loop created with attached power cords. The computer is actually the primary emitter here and the Off-Ramp mainly the receiver. The emissions were due to the ground-loop with the computer, so putting a ferrite on the cable will solve it, however this will also effect the rise-time and therefore increase the jitter possibly impacting audio quality. A better solution is to add the Short-Block USB cable filter, which primarily minimizes HF in that ground-loop (common-mode noise) without impacting the differential signal lines, actually reducing jitter. We have found that different computers USB interfaces create different levels of common-mode noise on the USB cable, some worse than others.

When you performed the emissions testing, did you also have a coax cable and DAC attached to the system?

Were you using the OR4 wall-wart or a third-party power supply for the Off-Ramp? We offer a new wall-wart used on the OR5 and the Dynamo linear supply to improve on the OR4 wall wart.

All of the Off-Ramp products also have very fast risetimes on the S/PDIF coax outputs in order to deliver low jitter signal to the DAC. It is critically important that you use a well shielded coax cable that is very close to 75 ohms characteristic impedance in order to avoid emissions from that source. We offer an excellent cable for this called the BNC-BNC coax with RCA adapters for $250. There is also a new mod available for the Off-Ramp for $150 that changes the S/PDIF output transformer to a better one, improving the signal integrity of the S/PDIF output and reducing jitter even more. All of these will improve SQ and reduce emissions.

Thanks,
Steve N.
Empirical Audio

scully280's picture

Indeed the ferrite is on the USB cable. I don't recall the exact details of the test because it was done informally over a year ago, and the test technician who did most of the work has left the company. But we did the test with a different computer than I have at home. One is an apple mac mini the other was a Dell laptop, i believe. But I would be happy to personally repeat the tests that your own test lab did when you had your FCC and CE certifications completed. I suspect that the setup details are in the report that your test lab sent you.

Empirical's picture

I'm sorry that you had this problem. I wish you would have contacted my directly when it happened. The problem with retesting is that my products have so many options that its really impractical and very expensive to test every single configuration. Different power supplies, DACs and computers add a lot of additional configurations to the mix also. I must test only one configuration and hope that the other configurations are okay. In addition, I often make improvements to my products mid-stream, and I typically do not retest for these. I try to keep the costs down so that I am not forced to charge even more for my products.

It would be helpful for me to know if your OR4 has the USB Hynes reg or the Turboclock and whether you are using the standard wall-wart that came with it.

I'll tell you what, since you had this problem, I can offer you a Short-Block for free. This is what I would rather you use anyway. Just shoot me an email. I'm building another batch in the next 2 weeks.

Regards,
Steve N.

scully280's picture

I was going to let this drop, but I’m a bit insulted by the last reply. Here’s where we are; I found what I think might be an FCC EMI violation in an earlier product. This is high stakes; large fines and large seizures of equipment have resulted from single violations. Gambling that a version of a product passes is a big risk, but Mr. Nugent has told us that’s a risk he takes. This is something the company I work for, and many other companies can’t tolerate because they have found that even the best engineers can’t always be right on the first pass and the potential legal liability is too great. We test in our own lab until we know that our equipment will pass the FFC as well as the much more inclusive CE regulations, then we go to an outside test lab with all the accessories that will be sold with the product and test multiple setups.

So here’s the part that really bothers me. What is the offer for a free upgrade all about? Is the freebie to make this potential big liability go away, in which case I’m really cheesed off? Because, Steve, you gotta know I’m not your usual audiophile customer. I’m an electrical engineer who knows his way around this stuff, is more interested in results than trade jargon, and I have a lab at my disposal to verify results. Even if this is just honest customer relations do you think I would really be happy with an upgrade? At this point that happens only if you can show me data that tells me the upgrade really solves the problem, and now I think I want that data from an independent lab. Anyhow I’m happy with my solution, and when I get around to it I may add another ferrite.

So here’s my bottom line for this review. How does the reader know when you bring the DAC under discussion home that it will play nicely with your AM or FM radio, with your cellphone, your Bluetooth keyboard, or your television? Or how about with your laptop, or your wireless network? By trusting Steve? If that works for you, fine, but I would have to see data quoted from an independent test report.

And that’s all I’ve got to say.