T+A DAC200 Measurements

Content and analysis made possible thanks to support from https://headphones.com, and https://patreon.com/goldensound supporters.

This unit was loaned to me by a friend to test

Design:

The T+A DAC200 is a digital to analogue converter and analog preamplifier coming in at just over $7000. It’s garnered quite a positive reputation among many enthusiasts, and coming from T+A, who have quite extensive experience designing products in many areas of the high end audio industry from DACs to speakers and headphones, it’s no surprise that there’s a fair amount of impressive engineering behind this one too, so let’s have a closer look!

The DAC200 is relatively busy internally, which is to be expected given the extensive featureset. To begin, there are a number of custom digital filter options the user can select from, as well as an analog 60khz low-pass filter that the user can enable to reduce unwanted ultrasonic noise, particularly useful with many DSD source material which can often contain high levels of ultrasonic content. The user may not be able to hear this directly, but the presence of high ultrasonic content levels can degrade the performance of many amplifiers and other devices later in the chain, which then have audible impacts.

The VU meters can be set to display internal stats such as temperature as well as the signal level, volume can be controlled using a fully analog preamp circuit, and this can also be used alongside the analog input to use the DAC200 as a fully analog preamp, meaning you could connect the output of your vinyl phonostage directly to this.

These features are all controlled by an ARM microcontroller by STM.

UPDATE: T+A has informed me that actually, the features above are not controlled by this component, but instead by an MSP3430 microcontroller on the front panel PCB. This ARM controller is actually serving as the USB receiver and generation of control signals for the 1-bit converter circuit.

But most importantly of course is the DAC section itself. The DAC200 actually uses two different DAC circuits depending on the source material. For PCM, four Burr-Brown PCM1795 chips are employed, two per channel. These can be seen about a third of the way down the image below.

DSD sources however are not converted by the Burr-Brown chips, but instead by T+A’s custom 1-bit conversion circuit. This allows the DAC to take the DSD data, and convert/output in its native form, rather than having to process the data or convert to PCM internally first as many other DACs have to do. This effectively means you have two DACs in one box, and also allows the DAC200 to take extensive advantage of tools such as HQPlayer, which can provide much more powerful and effective DSD modulators than could possibly be run on any sort of internal hardware.

And also worth noting, the DAC section is galvanically isolated from the digital section, and we can see a SI8660BA isolator chip on the board just to the right of the DAC section.

Interestingly, it seems that the clock signals themselves also run through this chip, rather than having the clock source on the DAC side of the isolation barrier. I unfortunately can’t say for certain given as I cannot look at the underside of the board, but there doesn’t seem to be a clock source close to the DACs nor any sort of controller that would facilitate a PLL, so it seems that the PLL and clock itself are on the other side of the isolation barrier.

UPDATE: T+A has confirmed that the clock signal is generated on the digital side of the isolation barrier. Here is what they said:

“The DAC clock is generated on the “digital” side of the isolation barrier. This is done to keep the noise generated by the digital clocking section out of the DAC and analog section of the DAC200.

To avoid possible jitter, that might be caused by distance and by the SI8660 isolatior chip, the clocking signals are separately transferred to the DAC section via impedance controlled PCB traces and all relevant signals are re-synchronized directly at the DACs. To keep the DAC clock jitter free, a special high frequency pulse transformer (instead of integrated isolator chips) is used for its galvanic isolation.

We found that this technique keeps digital noise effectively out of the analog section and it preserves extremely low DAC clock jitter.“

Externally, the design is busier than many devices from other manufacturers, but that’s somewhat necessary given the myriad of features this offers! The unit offers a crisp, clean display, and a pair of VU meters that not only can be set to display signal level, but specs such as internal temperature as well….oh….and they have RGB lighting to help match them to the rest of your setup.
I’m quite a fan of the aesthetics myself.

Measurements:

Test Setup:

– Audio Precision APx555 B-Series Analyzer with 200kOhm input impedance set unless otherwise specified
– USB Source: Intel PC via intona 7055-C isolator
– All measurements shown are with the DAC connected via USB unless otherwise specified
– Measurement setup and device under test are running on regulated 230V power from a Furman SPR-16-Ei
– Audioquest Mackenzie XLR and RCA interconnects
– Intona Reference Impedance Characterized USB Cable
– Audioquest Carbon SPDIF and AES cables (1.5m)
– Exact analyzer/filter configurations for each measurement are detailed in the full reports
– Measurements shown are with the DAC set to no attenuation unless otherwise stated
– FIR 1 filter used unless otherwise specified

Full Measurement Reports:

Reports available here:

Dynamic Range (AES17): 115.2dB

SNR: 116.1dB

IMD SMPTE: -82.2dB

DC Offset: 0.6mV active, 0.6mV idle

1khz Sine (PCM), 0dBfs, XLR Out:

Performance at full output could be better, not quite reaching 16 bit accuracy.

1khz Sine (PCM), -15dBfs, XLR Out:

Best performance is achieved at about -15dB. The distortion increases quite a bit as you go above this level.

1khz Sine (DSD), -1dBfs (HQPlayer vol), XLR Out:

The DSD section on the DAC200 handily outperforms the PCM section, providing a nearly 10dB+ drop in level of harmonics and up to a 13dB improvement in THD+N depending on how you’re comparing the two, and also avoids the rising distortion above -15dB seen on the PCM section. If you own this DAC, you owe it to yourself to try HQPlayer with DSD as the results are quite a bit better than feeding it straight PCM.

1khz Sine 2V Analog in 2V XLR Out (Preamp):

The preamplifier in the DAC200 is very good. Nice to see such performance for an ‘extra’ feature!

Jitter:

Jitter performance is fine, though still a number of deterministic jitter components quite visible. It would have been ideal if this were better. There’s a number of possible reasons for this, the clock signal having a seemingly long signal path to the DAC as well as going through an isolation chip instead of being generated closer to the DAC itself, or possibly just the quality of the clocks used. It seems that these (shown below) are the clocks being utilized which are somewhat out of place in a $7000 product, or when compared to the higher end clocks from manufacturers like Crystek or Accusilicon, which can be found in many products including at lower prices.

It is entirely possible that I am completely mistaken there though and there are some clocks on the underside of the board perhaps. I must however commend the PLL, which provided a roughly 30dB higher reduction in jitter than that of the Schiit Yggdrasil for example when feeding it a very jittery signal (10ns 1khz jitter applied to the SPDIF signal).

Upsampling Filters:

The DAC200 offers a number of digital filters, two traditional FIR filters, one longer/steeper than the other, two Bezier filters which are not a ‘technically correct’ way of doing PCM reconstruction but can provide some specific tradeoffs some will enjoy, so they’re nice to have. And two ‘NOS’ options, which are actually the same, just NOS1 applied the 60khz analog low pass filter whereas NOS 2 does not.

The ‘NOS’ options are not actually NOS however. As the PCM1795 chips are delta-sigma. And so they are still doing further processing, including running a high speed modulator. The NOS options on this DAC essentially just disable the first stage PCM oversampling, but it is not possible to bypass the second stage oversampling/modulator. In fact if you did, this DAC would only have a maximum of 6-bit accuracy which would not be great.

The PCM1795 and similar chips have been used in other producs such as the Teac UD501 and iFi iDSD Signature/Black Label and those products also offer the same functionality of defeating the initial filter. Though the DAC200 performs significantly better than the UD501 generally, and also avoids many of the issues that unit exhibited such as higher levels of high frequency noise in the audio band when in NOS mode. If you’d like to read about exactly what is happening in the ‘NOS’ mode with this chip have a look at the last section in my post on the UD501.

Personally I would recommend using the FIR1 filter as this comes closest to an ideal reconstruction. Though using HQPlayer & the DSD section will still provide the best results.

Analog Filter:

For this I played 768khz white noise through the DAC and measured the output with and without the analog filter enabled. As you can see, with the analog filter enabled (purple) it provides a roughly 14dB reduction for everything above 60khz. Impact on distortion and other areas of performance was insignificant. I’d recommend leaving this on, and personally I wish more DACs had LPFs on the output! Lots of downstream components can see degraded performance in the presence of high frequency signal content and so helping to keep that down is ideal especially if there’s no impact on other performance aspects.

Low level signal output:

Low level signal output is mostly fine, just slightly higher noise than perhaps desired. However, I did notice that when using the DSD converter, low level signals produced quite significant distortion compared to higher level ones:

Note that the scope is set to a 20khz bandwidth, hence the reduced noise compared to the charts above, but the important thing is all that activity around 2-3khz. Now it’s still ridiculously low overall, but only shows up in the presence of low level signals. It also increases in level as the frequency gets lower:

Distortion with a 25hz signal is in total almost the same energy as the fundamental, with a THD+N of 0.5dB.
I don’t exactly think this is a big concern, all of this remains low no matter what, it was just an oddity I found and wanted to share. The DSD converter is still definitely better than the PCM section overall.

Idle Noise FFT:

Some high frequency noise/switching components seen, but nothing major.

THD+N vs Frequency:

Flat as a pancake! Good stuff.

IMD:

The IMD sees a sudden jump above -15dB. Mirroring the behaviour with normal THD. This does not occur using DSD.

Linearity:

Here we see that decrease in performance once you get to -15dB. This could potentially in combination with the same behaviour for IMD lead to an audible effect on macrodynamics, as high level content will see higher distortion.

Crosstalk:

Multitone:

Intersample Overs Test:

When DACs oversample, they can sometimes encounter a situation where the reconstructed/interpolated waveform goes above 0dBfs (the maximum possible digital value).
This is particularly common with poorly mastered music that has been ‘brickwalled’ (To clarify: I mean brick-walled as in the loudness-wars term, not brick-wall nyquist reconstruction filters https://en.wikipedia.org/wiki/Loudness_war ).

DACs ideally should have a few dB of digital headroom to accommodate this and reconstruct properly, and many such as Chord, RME, Benchmark etc do, and will output signals with intersample overs without any issue/distortion.
But many do not do this, either out of lack of awareness of the issue or because doing so usually means sacrificing a few dB of dynamic range and/or THD+N performance. And as a result will not be able to properly reconstruct these signals.

We can test this by creating a signal that induces intersample overs.
See in the image below. All the 44.1khz samples (squares) are below the 0dBfs limit, but the actual waveform itself, is above this limit.

The DAC200 does not have any additional internal headroom and therefore will distort/clip in the presence of intersample overs. Recommended that you set your player volume to -3dB when using this DAC. Given the proprietary processing present in this DAC, seeing it have the proper headroom would have been ideal.

Summary:

The T+A DAC200 offers a very impressive feature set and level of flexibility, including not just actual analog volume control for the DAC but a preamp circuit that accepts an RCA input as well. The overall performance from an objective standpoint is ‘good’ though falls behind competitors in a few areas.

It would have been good to see the DAC200 include headroom to mitigate intersample clipping, as well as a fix for the increasing distortion at higher output levels.

The DSD section outshines the PCM section in the DAC200, and this is a product that if you are planning on getting I would STRONGLY recommend using in combination with HQPlayer. Using it standalone fed straight PCM would be something of a shame.

The video review with my full subjective opinions will be up shortly!