background image

Benchmark Media Systems, Inc. 

Page 22 of 39 

UltraLock™

 … What is It? 

Accurate 24-bit audio conversion requires a very low-jitter conversion clock.  Jitter can very easily turn a 24-bit 
converter into a 16-bit converter (or worse).  There is no point in buying a 24-bit converter if clock jitter has not 
been adequately addressed. 
 
Jitter is present on every digital audio interface.  This type of jitter is known as “

interface jitter

” and it is present 

even in the most carefully designed audio systems.  

Interface jitter

 accumulates as digital signals travel down a 

cable and from one digital device to the next.  If we measure 

interface jitter

 in a typical system we will find that it 

is 10 to 10,000 times higher than the level required for accurate 24-bit conversion.  Fortunately, this 

interface 

jitter

 has absolutely no effect on the audio unless it influences the conversion clock in an analog-to-digital 

converter (ADC) or in a digital-to-analog converter (DAC). 
 
Many converters use a single-stage Phase Lock Loop (PLL) circuit to derive their conversion clocks from 
AES/EBU, Wordclock, or Superclock reference signals. Single-stage PLL circuits provide some jitter attenuation 
above 5 kHz but none below 5 kHz.  Unfortunately, digital audio signals often have their strongest jitter 
components at 2 kHz.  Consequently, these converters can achieve their rated performance only when driven 
from very low jitter sources and through very short cables.  It is highly unlikely that any converter with a single-
stage PLL can achieve better than 16 bits of performance in a typical installation.  Specified performance may be 
severely degraded in most installations. 
 
Better converters usually use a two-stage PLL circuit to filter out more of the 

interface jitter

.  In theory, a two-

stage PLL can remove enough of the jitter to achieve accurate 24-bit conversion (and some do).  However, not all 
two-stage PLL circuits are created equal.  Many two-stage PLL’s do not remove enough of the low-frequency 
jitter.  In addition, two-stage PLL circuits often require several seconds to lock to an incoming signal.  Finally, a 
two-stage PLL may fail to lock when jitter is too high, or when the reference sample frequency has drifted. 
 

UltraLock™

 converters exceed the jitter performance of two-stage PLL converters, and are free from the slow-

lock and no-lock problems that can plague two-stage PLL designs. 

UltraLock™

 converters are 100% immune to 

interface jitter under all operating conditions.   No jitter-induced artifacts can be detected using an Audio Precision 
System 2 Cascade test set.  Measurement limits include detection of artifacts as low as –140 dBFS, application of 
jitter amplitudes as high as 12.75 UI, and application of jitter over a frequency range of 2 Hz to 200 kHz.  Any 
AES/EBU signal that can be decoded by the AES/EBU receiver will be reproduced without the addition of any 
measurable jitter artifacts. 
 
The 

DAC1

DAC-104

 and the 

ADC-104

 employ Benchmark’s new 

UltraLock™

 technology to eliminate all jitter-

induced performance problems. 

UltraLock™

 technology isolates the conversion clock from the digital audio 

interface clock.  Jitter on a DAC digital audio input, or an ADC reference input can never have any measurable 
effect on the conversion clock of an 

UltraLock™

 converter.  In an 

UltraLock™

 converter, the conversion clock is 

never phase-locked to a reference clock.   Instead the converter oversampling-ratio is varied with extremely high 
precision to achieve the proper phase relationship to the reference clock.  

Interface jitter

 cannot degrade the 

quality of the audio conversion.  Specified performance is consistent and repeatable in any installation! 
 
 

How does conversion clock jitter degrade converter performance? 

  

Problem #1:

 Jitter phase modulates the audio signal.  This modulation creates sidebands (unwanted tones) 

above and below every tone in the audio signal.  Worse yet, these sidebands are often widely separated from the 
tones in the original signal.   
 
Jitter-induced sidebands are not musical in nature because they are not harmonically related to the original audio. 
Furthermore, these sidebands are poorly masked (easy to hear) because they can be widely separated above 
and below the frequencies of the original audio tones.  In many ways, jitter induced distortion resembles 
intermodulation distortion (IMD).  Like IMD, jitter induced distortion is much more audible than harmonic distortion, 
and more audible than THD measurements would suggest.   

Содержание DAC1 HDR

Страница 1: ...Benchmark Media Systems Inc DAC1 Instruction Manual 2 Channel 24 bit 192 kHz Audio Digital to Analog Converter...

Страница 2: ...iau L DAC1 Manual Rev L Doc 01 31 07 F G Allen Burdick John Siau M DAC1 Manual Rev L Doc 01 10 09 F G Elias Gwinn N DAC1 Manual Rev L Doc 10 26 09 F G Elias Gwinn Manual Revision Notes Revision Note I...

Страница 3: ...4 FEATURE SUMMARY 5 FRONT PANEL 6 REAR PANEL 7 CALIBRATION 7 REMOVING TOP COVER 13 JUMPER LOCATIONS 14 JUMPER SETTINGS 15 COMPLIANCE AND SAFETY INFORMATION 18 CERTIFICATE OF CONFORMITY 19 CONTACT INF...

Страница 4: ...dio Precision System 2 Cascade test set Measurement limits include detection of artifacts as low as 140 dBFS application of jitter amplitudes as high as 12 75 unit intervals UI and application of jitt...

Страница 5: ...switch Preset output levels use multi turn trimmers one 10 turn trimmer per output 20 dB Range 2 dB turn Total jitter immunity with Benchmark phase accurate UltraLock technology Two Error LED s no di...

Страница 6: ...out distortion The HPA2 delivers the full rated performance of the DAC1 into two 60 stereo headphones outputs The HPA2 is fully protected against short circuits and overloads VARIABLE GAIN CONTROL The...

Страница 7: ...d monitors The factory preset levels may be increased by 5 dB or decreased by 15 dB in order to conform to other studio reference levels This range of levels is also well suited for direct connection...

Страница 8: ...outputs are studio grade line level outputs and should not be plugged into microphone inputs without the use of the internal 20 or 30 dB pads or a suitable external pad Unbalanced Analog Outputs The L...

Страница 9: ...l We have chosen to use a BNC connector because of its superior reliability The DAC1 comes with a BNC to RCA adapter to allow easy interface with consumer equipment The BNC input has a 75 internal ter...

Страница 10: ...Output Level Calibration Controls 10 turn trimmers 1 per output Calibration Adjustability 2 dB turn Output Level Range at 0 dBFS In Calibrated Mode 9 dBu to 29 dBu Attenuator off 1 dBu to 19 dBu Atte...

Страница 11: ...itter Induced Sidebands 141 dB 10 kHz 0 dBFS test tone 12 75 UI sinusoidal jitter at 1 kHz Maximum Amplitude of Spurious Tones with 0 dBFS test signal 126 dB Maximum Amplitude of Idle Tones 128 dB Max...

Страница 12: ...M Non compatible digital audio format on active input 1 Red AC POWER REQUIREMENTS 50 60 Hz 8 Watts Idle 8 Watts Typical Program 16 Watts Maximum Fuses 2 required 5 x 20 mm 100 V to 130 VAC operation 0...

Страница 13: ...omponent failures may affect the long term reliability or may degrade the audio performance Use a static control wrist strap when changing jumper settings Caution Disconnect AC power by unplugging the...

Страница 14: ...4 of 39 JUMPER LOCATIONS Position Function N O C X XLR only BNC only Optical only Switchable input Position P5 P8 Attenuation 1 2 3 4 5 6 7 8v 30 dB 20 dB 10 dB 0 dB JP7 75 ohm BNC Termination Left Ri...

Страница 15: ...to 1 can be used to disable the front panel source selection switch This feature is useful when the DAC1 is in a critical audio path and only one of the three digital inputs will be used In such an ap...

Страница 16: ...ed if the user wishes to loop a single coaxial feed through several other pieces of equipment using a BNC T adapter on the DAC1 A 75 termination must be applied at the last device on the loop and ther...

Страница 17: ...ws 0 dB Attenuator disabled Jumper plug between pins 1 and 2 of each header 10 dB Jumper plug between pins 3 and 4 of each header 20 dB Jumper plug between pins 5 and 6 of each header 30 dB Jumper plu...

Страница 18: ...e received including interference that may cause undesired operation Safety Information Do NOT service or repair this product unless properly qualified Only a qualified technician should perform repai...

Страница 19: ...Benchmark Media Systems Inc Page 19 of 39 CERTIFICATE OF CONFORMITY...

Страница 20: ...ystems Inc Page 20 of 39 CONTACT INFORMATION Benchmark Media Systems Inc 203 E Hampton Place Ste 2 Syracuse NY 13206 USA 315 437 6300 Phone 315 437 8119 Fax http www benchmarkmedia com mailto sales be...

Страница 21: ...warranty to a period of five 5 years from the date of delivery For the extended warranty to become effective the original purchaser must register the product at the time of purchase either by way of...

Страница 22: ...ock when jitter is too high or when the reference sample frequency has drifted UltraLock converters exceed the jitter performance of two stage PLL converters and are free from the slow lock and no loc...

Страница 23: ...y remove most of this unwanted noise before it can alias down into lower audio frequencies These crosstalk problems may not become obvious until jitter is present Stop band attenuation can be measured...

Страница 24: ...can be done to remove the damage Jitter induced sidebands are extremely complex and cannot be removed with any existing audio device It is therefor important to attack jitter at both ends of the audio...

Страница 25: ...operating at a 48 kHz sample rate The top graph shows that the differential phase is better than 0 5 at 20 kHz The bottom graph shows the amplitude response on a highly expanded 0 05 dB division scal...

Страница 26: ...e The top graph shows that the differential phase is better than 0 5 at 20 kHz and better than 1 at 43 kHz The bottom graph shows the amplitude response on a highly expanded 0 05 dB division scale Not...

Страница 27: ...Analysis of Idle Channel Noise The above graph demonstrates that the DAC1 is free from idle tones and clock crosstalk The highest spurious tone measures 128 dBFS and is AC line related hum The highes...

Страница 28: ...hase between 10 audio channels using 5 DAC1 converters operating at 96 kHz The DAC1 converters were chosen from stock at random and measurements were made using a random combination of Coaxial XLR and...

Страница 29: ...tion from 20 Hz to 20 kHz is very slight Note that at worst case the distortion is 109 dB less than the 3 dBFS test tone and 112 dB less than the full scale output of the DAC1 This implies that the di...

Страница 30: ...Media Systems Inc Page 30 of 39 THD N vs Level at 1 kHz Balanced Outputs Below 4 dBFS distortion is lower than the noise floor of the converter Above 3 dBFS distortion reaches a maximum value of only...

Страница 31: ...e Outputs This graph shows the output of the HPA2 headphone amp driving a 60 Ohm load at a very high level 14 dBu Even under these conditions the HPA2 delivers the full rated performance of the DAC1 C...

Страница 32: ...a Systems Inc Page 32 of 39 THD N vs Level at 1 kHz Unbalanced Outputs This graph demonstrates the performance of the unbalanced outputs Note that the performance is nearly identical to that of the ba...

Страница 33: ...f 39 THD N vs Sample Frequency The above graph shows that the DAC1 provides consistent performance at all sample rates Distortion is not a function of sample rate The minor variations in the above plo...

Страница 34: ...op curve is on the right hand side of the graph and is calibrated in UI of jitter The bottom green curve shows the THD N of the DAC1 as the jitter amplitude and frequency is varied at the inputs of th...

Страница 35: ...e the sensitivity of the test We set the interface jitter amplitude to its maximum value of 12 75 UI 2075 ns of jitter We then swept the jitter frequency from 2 Hz to 9 kHz and plotted the THD N from...

Страница 36: ...dia Systems Inc Page 36 of 39 DAC1 Cable Jitter Immunity The above FFT plots demonstrate that the performance of the DAC1 is not degraded in any way when long cables are used to transmit digital audio...

Страница 37: ...37 of 39 XLR Digital Input Sensitivity The above graph shows that the performance of the DAC1 is not a function of the signal level at the XLR digital input When the signal is too low to decode 160 mV...

Страница 38: ...f 39 Coaxial Digital Input Sensitivity The above graph shows that the performance of the DAC1 is not a function of the signal level at the coaxial digital input When the signal is too low to decode 12...

Страница 39: ...the AES In addition the above plots show that while the AES minimum eye pattern specifications are barely met at the end of 1000 feet of Category 5 UTP cable the DAC1 receivers have enough sensitivity...

Отзывы: