ichiban said:
Yest for 100% only the Benchmark use a dual opamp in the last stage, and the second half of it have the function of a unitygain buffer. Between the output and the input of the second half of the dual opamp you find the nasty elco's in the signalflow. I don't get why no other mod company bypass al those opamps after this circuit, because the decrease the sound. My advice take the output from the second opamp after the I-V conversion. The dac1 original use 4 dual opamps in the signal flow before it finaly reach the outputstage!!
All values for resistors and feedback cap's are equal from the datasheet. If you still want to use the headphone amp, replace the unity gain buffer in the secondhalf of the opamp with a wire. The signal will be buffert again be the next opamp in the Buf634 circuit.
It is raining opamps in the dac1...
Best regards,
Bas
Dear,
To tune out DC on the output can be diffficult. Analoge Devices (and so Benchmark) already did it by feed a voltage of 2.7 volt into the negative inputs of the I-V cconvertors to compensate the DC offset of the dac chip. However you still get a remaining 200mV DC. For DIY-ers that is ok because most of us know how to deal with it and take care of it in our pre-amplifier. But for comemrcial use the dac should have near zero DC offset on the output. Compensate it by change to voltage applied to the negatieve input of the I-V covnerter wont work 100% because the DC drift is dependend on temperatuur etc. We designed a dedicated DC servo circuit, that does the job verry wel, and much better then every CAP or Elco in the signalflow IMHO.
Best regards,
Bas
To tune out DC on the output can be diffficult. Analoge Devices (and so Benchmark) already did it by feed a voltage of 2.7 volt into the negative inputs of the I-V cconvertors to compensate the DC offset of the dac chip. However you still get a remaining 200mV DC. For DIY-ers that is ok because most of us know how to deal with it and take care of it in our pre-amplifier. But for comemrcial use the dac should have near zero DC offset on the output. Compensate it by change to voltage applied to the negatieve input of the I-V covnerter wont work 100% because the DC drift is dependend on temperatuur etc. We designed a dedicated DC servo circuit, that does the job verry wel, and much better then every CAP or Elco in the signalflow IMHO.
Best regards,
Bas
marinsimina said:
Dear Marin,
Wel, the most important thing is to replace at first those SMD caps in the signal flow.
A: I don't like SMD caps (ment for power supplys) in the signal flow.
B: I don't like the capacitance of 1000mF!!! and
C: I don't like al those bypass caps.
Replace bij a simply 47mF blackgate hiq NX makes miracles (but I prefer a good designed DC servo though).
This give the dac more freedom and less grain. The grain some people complain about is gone already for 60%
The next step is to remove the NE5532. Doesn't matter what Benchmark says about how great those perform... I don't like the sound
. Especialy not as an I-V converter with the NE5532 slewrate limitations. Even though the OPA2134 measure less good I like the sound so much more. Replacment of opamps give the DAC1 more depth and insight in the recording room. With the NE5532 I feel like the DAC1 sound dark and a little bit compressed (in an unconsious way) and nasal. The AD1896 upsampler ic (what benchmark called "ultralock") works nice with the FIFO buffer. We can discuss about those upsampling and FIFO buffer principes but let's not do it in this topic
. For me it works fine with knowing the compromises and the limitations. However the performance is now for a big part dependent on the internal asynchroon clock. Sadly the DAC1 have the XO and the AD1896 fed bij a noisy 7805 voltage regulator wich decrease the performance. The Tentlabs XO-2 (www.tentlabs.com) clock with an super clean onboard voltageregulator provide a better and more stable clock. Here the DAC1 gains great in performance. Simply tighter lows more definition and details I never heard before. Now the dac1 sounds like reference class. The power supply mod. like we do (12000uF of capacitance instead of 2000) gives the dac1 simply more rest and bass slamm. And we like slamm and tigh bass.
(I'm I bass player you can't blame me )The standard dac1 straight out of the box perform great for the money. It give some detail and resolution unheard of for this price point. But it is also not very involving and musical sounding. It is detailed but I miss a texture and natrual flowing of the sound. After all the mods
-I-V conversion for AD797
-Buffer opamp for OPA627
-Bypass of all the other opamps signal straight after the second opamp to the output like the analoge devices datasheet
-Power wuply upgrade with 12000uF capactitance
-Local powersupply capacitance of 220uF near the opamp supply pins
-new masterklok
-New decoule cap in the digital input
Now the dac1 sounds big with a lot of depth. No grain but a big inside in the recording room with plenty of detail Tight and fast bass with a real chest feeling slamm.
I still think the AD1853 is a little big dark sounding, but I really like the performance of the upgraded dac1. Especialy the resolution and the details and the musicality.
Offcourse there is always more to modify, but you have to put a limit. If you want to do it really better you can better design a whole new dac with the same principe, but then good from the ground up. Not that I want to say the dac1 is not good, because it is good. But I mean substantionaly better. Offcourse that have a pricepoint, and I respect Benchmark Media very much for what they reach for this price. The did a great job!
Best regards,
Bas
marinsimina said:
Dear Marin,
You are correct. We remove the 1000uF and bypass caps after the 1/2 opamp. We inject the servo output into the negative input of the 1/2 opamp. With the Op27 and a 50Khom trimpot between pin 1 and 8 you can tune dc offset as low as 0,01mV. proper care have to be taken of the filter caps of the servo and local powersupply. One problem is, is that the servo acts slow when first turn on. It needs 8 seconds to tot make the 200mV dissapear
So wait 8 seconds with turning on your preamp . Our servo will have a relais that short the outputs for 12 seconds. Iexperiemnt with many DC servo ideas, but for now it turn out the most simple way sounded the best. But we are always searching for beter and stille experiment with servo's with a higher order of filtering with cascade connection.
If I had to design a dac from the groun up I would not inject the servo output to the inverted input of the buffer opamp, but I would try to let the servo ajust the offset compensation pins of the opamp tot ajust the bias generators.
Best regards,
Bas
Sebastiaan said:
Well, I dont bypass all of the op-amps because:
1) it eliminates some of the features, such as: volume control and balanced outputs
2) the unity gain buffers are there to buffer the transmission-lines
Too many op-amps is certainly not good, but long traces in the circuit board are equally as bad. This is like plugging a really bad interconnect in the middle of your system. I have been planning to bypass these traces in an updated mod.
Also, the volume control circuit could sound good with a different pot etc.., but it has some problems, including strange breaks in the ground plane. It would be nice to fix this as well.
I think bypassing the "calbrated" pot op-amps makes a lot of sense, or at least put fixed resistors in place of the pots and remove the B/W limiting caps in the feedbacks.
It is interesting to see other approaches to improving the DAC-1.
Best regards,
Steve N.
Empirical Audio
audioengr said:
Dear Steve,
You are right in that part. You will miss some functions. But you can still leave the headphone amp working, just like the balanced outputs. Our redesign is is goaled on people that use it as an standalone DAC with the unbalanced outputs.
I don't like the balanced outputs anyway. It don't make sense.
First they make the diffferential signaal a-symetric. And end of the road the use an NE5532 to invert fase again for the pin3. (a ugly way to make a signal balanced, because pin 2 and 3 will enver be 100% equal inverted) For custommers that use the Benchmark Balanced we designed a new balanced outputstage that is differential straight from the DAC ic to the output.
Best regards,
Bas
Sound Evolution PDF in English part1-2
Redesign thank Evolution for downloading ours informatie-PDF concerning our redesign programme for the benchmark DAC1 because the benchmark DAC1 is a multi-purpose apparatus and on several use manners is, have we our redesign programme opgespitst in two categories.
Catogorie1:
Use benchmark DAC1 as D/A converter without volume monitoring of the ongebalanceerde RCA terminations. For this situation we can offer the most fraaiste redesign possibility because we can skip now a large part of the original circuit in our adapted design. Less is more! The other terminations remain be in functioning but not ge-upgrade.
Catogorie2:
Use benchmark DAC1 the terminations balanced with and/or the volume control preamplifier function. In this case we leave the circuit intact, and improve all versterkertrappen in the current design. RCA, XLR and head tel. terminations. On the next page you find firstly an overall, step by step description of our redesignpunten. To be able offer upgrade for a broad public, we have in the programme several upgrade possibilities, from evolution 1 to evolution 4. evolution 4 of category 1 is a "statement non compromise" upgrade which brings DAC1 on solitary altitude. The price for this evolution 4 upgrade has been stipulated by the very precious BurrBrown OPA627 opamps which we apply (8 stuks!). These are considered as the best opamps in the audio industry and are normally only applied in statement audio components as for example a mark of Levinson and Wadia. With regard to benchmark redesign programme we are worldwide concurrentieloos. Nobody of whom we the existence know do so serious design interventions in this product. What you, however, her and of the will encounter much on the www to be, as we call that "voodoo" acquired characteristics where a lot of money for in demand becomes. In contrast to last mentioned we have deeply dove effective in the design and have the design on technical points have effectively improved. No border or challenge goes our too far, and it can a wonder be called that we in benchmark DAC1 sound Evolution redesign programme Rev. C Redesign options and prices Sebastiaan freezes, 1 October 2006 Opamps because of the vele possibilities which DAC1 must offer, is the signaalweg longer than necessary is for use as D/A converter with direct ongebalanceerde RCA terminations. Although the D/A converterchip differentiële have terminations, this balanced indicator direct ongebalanceerd is made in the further design. For extra opamps in the signaalweg to occur which must make the indicator later balanced, we for audiofiel thuisgebruik with short interlinks the RCA recommend using terminations. If DAC1 the as such use becomes we let us can the amplifier circuit simplify drastically and instead of with 13 the present opamps, we let us can now from the feet with only 4 (less is more!). Moreover replaced we by benchmark used NE5532 opamps by quality burrBrown copies. Opamp are in fact line amplifier module in a chip (IC) and therefore very klankbepalend. Used NE5532s lowcost opamp devise those tribes from the years 80 now in more modern times are for that many are more fraaiere (, however, more precious) alternatives, such as we that apply. This the first large step is to a wide consonance improvement. In category 2 we leave the design intact and replace we all 13 opamps by quality copies. Lowpassfilter each DAC or CD player needs lowpassfilter in the analogous uitgangstrappen hoogfrequente indicators from the D/A convertersignaal filter. Because these filter components in feedback course of the versterkertrappen sit, have this influence on eventual consonance quality. In DAC1 (as from evolutie2) we build the lowpassfilter again with high-quality components. Elco's signaalweg D/A converterchips and termination amplifier always produce a bit gelijkspanningscomponent (DC) what comes in the audio indicator. Amplifiers and two-delicate are possible not well along with gelijkspanning and for this reason must each audio component no send out gelijkspanning. Elco's have to let through as a property wisselspanning (REPORTING OFFICE, what a music indicator is also) and gelijkspanning to block. For that reason elco in serial with the indicator are in many cases put to make gelijkspanningsvrij the termination. The music indicator however passings away entirely by this component. Elco's forms in combination with the capacitance and induction of the circuit a filter. Elco's joints distortion to the music indicator and are moreover generally polarisen, what makes them unsuitable for high resolution hifiweergave. Benchmark applies SMD voedingselco. This is terrible sin and such a component in the signaalweg of so high influence that these as well as possible Redesign points have been described Redesign points described (continuation) must be. We remove the triple condenser circuit of the benchmark and apply a non polar type of Blackgate. These components stand confessed if the bests what there to get is for this application and is moreover non polar and has a negligible distortion factor. The second step to more resolution, rests and commonplace in consonance has been put. In our redesign we use by canal opamp for IV conversion (flow tension conversion) and opamp for lowpass filtering and buffer. As from we leave there the rest of the circuit for what it is and shuts we the RCA terminations new by means of tailored made screened wiring directly to. Feeding an audio component is as weak as its feeding. After much experience with several feeding concepts are appropriate we to this end wherever possible in our redesigns very fast voedingsschakelingen. These have a large buffer feeding this way possible keep clean. We use plural changed gear very fast elco's for a fast response time on fast pulses. This way there is always sufficient flow for each conceivable situation in muziekpassage. Also we have a heavy oppression of asperities of the lighting system. In the benchmark DAC1 we can create entirely new voedingsschakeling which the buffer increased of 2000uF to 12000uF! Each opamp give we an extra local feeding buffer directly on the voedingspinnen.
Redesign thank Evolution for downloading ours informatie-PDF concerning our redesign programme for the benchmark DAC1 because the benchmark DAC1 is a multi-purpose apparatus and on several use manners is, have we our redesign programme opgespitst in two categories.
Catogorie1:
Use benchmark DAC1 as D/A converter without volume monitoring of the ongebalanceerde RCA terminations. For this situation we can offer the most fraaiste redesign possibility because we can skip now a large part of the original circuit in our adapted design. Less is more! The other terminations remain be in functioning but not ge-upgrade.
Catogorie2:
Use benchmark DAC1 the terminations balanced with and/or the volume control preamplifier function. In this case we leave the circuit intact, and improve all versterkertrappen in the current design. RCA, XLR and head tel. terminations. On the next page you find firstly an overall, step by step description of our redesignpunten. To be able offer upgrade for a broad public, we have in the programme several upgrade possibilities, from evolution 1 to evolution 4. evolution 4 of category 1 is a "statement non compromise" upgrade which brings DAC1 on solitary altitude. The price for this evolution 4 upgrade has been stipulated by the very precious BurrBrown OPA627 opamps which we apply (8 stuks!). These are considered as the best opamps in the audio industry and are normally only applied in statement audio components as for example a mark of Levinson and Wadia. With regard to benchmark redesign programme we are worldwide concurrentieloos. Nobody of whom we the existence know do so serious design interventions in this product. What you, however, her and of the will encounter much on the www to be, as we call that "voodoo" acquired characteristics where a lot of money for in demand becomes. In contrast to last mentioned we have deeply dove effective in the design and have the design on technical points have effectively improved. No border or challenge goes our too far, and it can a wonder be called that we in benchmark DAC1 sound Evolution redesign programme Rev. C Redesign options and prices Sebastiaan freezes, 1 October 2006 Opamps because of the vele possibilities which DAC1 must offer, is the signaalweg longer than necessary is for use as D/A converter with direct ongebalanceerde RCA terminations. Although the D/A converterchip differentiële have terminations, this balanced indicator direct ongebalanceerd is made in the further design. For extra opamps in the signaalweg to occur which must make the indicator later balanced, we for audiofiel thuisgebruik with short interlinks the RCA recommend using terminations. If DAC1 the as such use becomes we let us can the amplifier circuit simplify drastically and instead of with 13 the present opamps, we let us can now from the feet with only 4 (less is more!). Moreover replaced we by benchmark used NE5532 opamps by quality burrBrown copies. Opamp are in fact line amplifier module in a chip (IC) and therefore very klankbepalend. Used NE5532s lowcost opamp devise those tribes from the years 80 now in more modern times are for that many are more fraaiere (, however, more precious) alternatives, such as we that apply. This the first large step is to a wide consonance improvement. In category 2 we leave the design intact and replace we all 13 opamps by quality copies. Lowpassfilter each DAC or CD player needs lowpassfilter in the analogous uitgangstrappen hoogfrequente indicators from the D/A convertersignaal filter. Because these filter components in feedback course of the versterkertrappen sit, have this influence on eventual consonance quality. In DAC1 (as from evolutie2) we build the lowpassfilter again with high-quality components. Elco's signaalweg D/A converterchips and termination amplifier always produce a bit gelijkspanningscomponent (DC) what comes in the audio indicator. Amplifiers and two-delicate are possible not well along with gelijkspanning and for this reason must each audio component no send out gelijkspanning. Elco's have to let through as a property wisselspanning (REPORTING OFFICE, what a music indicator is also) and gelijkspanning to block. For that reason elco in serial with the indicator are in many cases put to make gelijkspanningsvrij the termination. The music indicator however passings away entirely by this component. Elco's forms in combination with the capacitance and induction of the circuit a filter. Elco's joints distortion to the music indicator and are moreover generally polarisen, what makes them unsuitable for high resolution hifiweergave. Benchmark applies SMD voedingselco. This is terrible sin and such a component in the signaalweg of so high influence that these as well as possible Redesign points have been described Redesign points described (continuation) must be. We remove the triple condenser circuit of the benchmark and apply a non polar type of Blackgate. These components stand confessed if the bests what there to get is for this application and is moreover non polar and has a negligible distortion factor. The second step to more resolution, rests and commonplace in consonance has been put. In our redesign we use by canal opamp for IV conversion (flow tension conversion) and opamp for lowpass filtering and buffer. As from we leave there the rest of the circuit for what it is and shuts we the RCA terminations new by means of tailored made screened wiring directly to. Feeding an audio component is as weak as its feeding. After much experience with several feeding concepts are appropriate we to this end wherever possible in our redesigns very fast voedingsschakelingen. These have a large buffer feeding this way possible keep clean. We use plural changed gear very fast elco's for a fast response time on fast pulses. This way there is always sufficient flow for each conceivable situation in muziekpassage. Also we have a heavy oppression of asperities of the lighting system. In the benchmark DAC1 we can create entirely new voedingsschakeling which the buffer increased of 2000uF to 12000uF! Each opamp give we an extra local feeding buffer directly on the voedingspinnen.
Sound Evolution PDF in English part2-2
Jamming is from outside oppressed this way still further, and the opamp have always direct stroomreserve to its arrangement. Bell the benchmark DAC1 is because of its FIFO buffer virtually onbeïnvloed by digital jitter from outside. The eventual quantity internal jitter is at the benchmark DAC1 however total dependent on the used internal bell. Unfortunately benchmark uses XO bell which is fed by a standard 7805 regulator. The feeding noise of this regulator reduces the performance and exact timing of the bell. Now we the beautiful Ultralock have concept, we let us can exploit this technology still much further by placing a very good reference bell in the benchmark. XO-2 of Tentlabs which we use for that its own discreet regulator to border and a deviation have lower than 50 picoseconde. The original benchmark up/down sampled all entering indicators on sample rate of 110Khz. we adapt this with the new bell and the new internal sample rate become real 192Khzth remaining redesign elements at the redesign of the benchmark DAC1 packages we except feeding, bell and termination training period also small points. - The digital entrance gets new DC disconnects pulsecondensator (Blackgate NX hiq); - the RCA chassis parts are replaced quality copies; - the DAC has chip himself fraaie, relative quiet regulator got of benchmark. However AD1896 the upsampler chip and formerly the klokschakeling had do it with 7805 a regulator which would have in fact prohibited in digital circuits. This regulator replaces we by a much quieter alternative of Linear Tech or Tentlabs shunt regulator. Redesign points described (continuation) category the 1 benchmark DAC1 use as D/A converter on the RCA ongebalanceerde terminations without gebruikmaking of the preamplifier volume controlfunctie. Redesign and prices. Evolution 1 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - Replacement RCA chassis parts; - Local feeding buffer by opamp. Price including wages and VAT 300 euro. Evolution 2 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit. Price including wages and VAT 600 euro. Evolution 3 - Bypass whole orginele circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; Category the 1 (continuation) benchmark DAC1 uses as D/A converter on the RCA ongebalanceerde terminations without gebruikmaking of the preamplifier volume controlfunctie. Redesign and prices. - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. Price including wages and VAT 800 euro Evolution 4 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion opamps for the bests AD797 opamps for this circuit made by means of tailored SMD printjes which of two single opamps dual make for the benchmark DAC1; - Replacement lowpass filtering and buffer opamps tailored SMD made by the best BurrBrown OPA627 copies by means of printjes which make dual of two single opamps for the benchmark DAC1; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. Price including wages and VAT 1100 euro Evolution 1 - Replacement of all 13 opamps by audiofiele Burrbrown copies; - Replacement RCA chassis parts; - Local feeding buffer by opamp. - Price 500 euro including wages and VAT. Evolution 2 - Replacement of all 13 opamps by audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit. Price including wages and VAT 750 euro. Evolution 3 - Replacement of all 13 opamps by audiofiele copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement entrance elco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. With the balanced terminations and/or the volume control preamplifier function including wages and VAT 900 euro benchmark DAC1 prices category 2 use. In this case we the circuit intact and improve all versterkertrappen in the current design, RCA, XLR and head tel. leave. General conditions if you have interest in redesign of your benchmark DAC1 read on you then the following conditions just as. You must mail our preferably for an appointment. You use for this of the liaison form on our Internet site. In times of upheaval there a chance exists that you do not get help direct at our. We plan your product in in our agenda in consultation between u. you must take into account redesign time of maximum two weeks in which you your product is lost. In times of upheaval that can lead. We communicate this vantevoren. All by our ge-upgrade products go 24 hours in the duration test. We pass on guarantee of redesign sound all along the line done Evolution to a year. The Tentlabs XO-2 bell has a guarantee period of 5 years. Guarantee does not pass to fall in case of foutief use of the product or technical acquired characteristics and/or adaptations sound carried out Evolution or the original manufacturer of this product. At present sales we (still) no benchmark DAC1. You must the product therefore elsewhere buy if you do not possess a benchmark DAC1 prices and technical information is with reservation. Prices have been indicated valid as from 1 October 2006 to other one on our Internet site. Sound Evolution preserves itself the right for prices, descriptions and if specifications to modify. Sound Evolution preserves itself the right redesign requests to refuse. You can turn visibility at any time to our for further questions. Specific custom made options are always open to discussion. For additional information and photograph material concerning these and other audio upgrades we refer you to our Internet site, www. soundevolution.nl for further information on the Tentlabs XO-2 bell which we apply in Evolution 3 and 4 we you refer to www.tentlabs.com.
Jamming is from outside oppressed this way still further, and the opamp have always direct stroomreserve to its arrangement. Bell the benchmark DAC1 is because of its FIFO buffer virtually onbeïnvloed by digital jitter from outside. The eventual quantity internal jitter is at the benchmark DAC1 however total dependent on the used internal bell. Unfortunately benchmark uses XO bell which is fed by a standard 7805 regulator. The feeding noise of this regulator reduces the performance and exact timing of the bell. Now we the beautiful Ultralock have concept, we let us can exploit this technology still much further by placing a very good reference bell in the benchmark. XO-2 of Tentlabs which we use for that its own discreet regulator to border and a deviation have lower than 50 picoseconde. The original benchmark up/down sampled all entering indicators on sample rate of 110Khz. we adapt this with the new bell and the new internal sample rate become real 192Khzth remaining redesign elements at the redesign of the benchmark DAC1 packages we except feeding, bell and termination training period also small points. - The digital entrance gets new DC disconnects pulsecondensator (Blackgate NX hiq); - the RCA chassis parts are replaced quality copies; - the DAC has chip himself fraaie, relative quiet regulator got of benchmark. However AD1896 the upsampler chip and formerly the klokschakeling had do it with 7805 a regulator which would have in fact prohibited in digital circuits. This regulator replaces we by a much quieter alternative of Linear Tech or Tentlabs shunt regulator. Redesign points described (continuation) category the 1 benchmark DAC1 use as D/A converter on the RCA ongebalanceerde terminations without gebruikmaking of the preamplifier volume controlfunctie. Redesign and prices. Evolution 1 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - Replacement RCA chassis parts; - Local feeding buffer by opamp. Price including wages and VAT 300 euro. Evolution 2 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit. Price including wages and VAT 600 euro. Evolution 3 - Bypass whole orginele circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion and lowpass filtering opamps for audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; Category the 1 (continuation) benchmark DAC1 uses as D/A converter on the RCA ongebalanceerde terminations without gebruikmaking of the preamplifier volume controlfunctie. Redesign and prices. - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. Price including wages and VAT 800 euro Evolution 4 - Bypass whole original circuit after the first 2 versterkertrappen for as a termination training period pure as possible; - Replacement of IV conversion opamps for the bests AD797 opamps for this circuit made by means of tailored SMD printjes which of two single opamps dual make for the benchmark DAC1; - Replacement lowpass filtering and buffer opamps tailored SMD made by the best BurrBrown OPA627 copies by means of printjes which make dual of two single opamps for the benchmark DAC1; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. Price including wages and VAT 1100 euro Evolution 1 - Replacement of all 13 opamps by audiofiele Burrbrown copies; - Replacement RCA chassis parts; - Local feeding buffer by opamp. - Price 500 euro including wages and VAT. Evolution 2 - Replacement of all 13 opamps by audiofiele Burrbrown copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement ingangselco digital circuit. Price including wages and VAT 750 euro. Evolution 3 - Replacement of all 13 opamps by audiofiele copies; - construction Lowpassfilter with MKP condensers; - Replacement RCA chassis parts; - Local feeding buffer by opamp; - New analogue feeding circuit with 8 very fast elco's. A total feeding capacity of 12000uF has been created; - Replacement 7805 feeding regulator for upsampler chip; - Replacement entrance elco digital circuit; - Plaatsing Tentlabs XO2 bell circuit for upsample/reclock circuit. With the balanced terminations and/or the volume control preamplifier function including wages and VAT 900 euro benchmark DAC1 prices category 2 use. In this case we the circuit intact and improve all versterkertrappen in the current design, RCA, XLR and head tel. leave. General conditions if you have interest in redesign of your benchmark DAC1 read on you then the following conditions just as. You must mail our preferably for an appointment. You use for this of the liaison form on our Internet site. In times of upheaval there a chance exists that you do not get help direct at our. We plan your product in in our agenda in consultation between u. you must take into account redesign time of maximum two weeks in which you your product is lost. In times of upheaval that can lead. We communicate this vantevoren. All by our ge-upgrade products go 24 hours in the duration test. We pass on guarantee of redesign sound all along the line done Evolution to a year. The Tentlabs XO-2 bell has a guarantee period of 5 years. Guarantee does not pass to fall in case of foutief use of the product or technical acquired characteristics and/or adaptations sound carried out Evolution or the original manufacturer of this product. At present sales we (still) no benchmark DAC1. You must the product therefore elsewhere buy if you do not possess a benchmark DAC1 prices and technical information is with reservation. Prices have been indicated valid as from 1 October 2006 to other one on our Internet site. Sound Evolution preserves itself the right for prices, descriptions and if specifications to modify. Sound Evolution preserves itself the right redesign requests to refuse. You can turn visibility at any time to our for further questions. Specific custom made options are always open to discussion. For additional information and photograph material concerning these and other audio upgrades we refer you to our Internet site, www. soundevolution.nl for further information on the Tentlabs XO-2 bell which we apply in Evolution 3 and 4 we you refer to www.tentlabs.com.
PartsConnection Mods
Attached is a list of mods suggested by Chris Johnson, the president of Parts Connexion, for DAC1. According to Chris:
"Essentially, we upgrade the analog output stage...from the DAC chips to the output...including critical signal path caps, resistors, op amps (these are the most expensive and most critical items in the mod...Burr-Brown OPA627 op amps on dual vertical adaptor boards), wire and output connectors. Plus, we upgrade the sonically critical PS diodes and resonance damp the chassis. The op amps we use in this mods are several levels above the NE5532's...the OPA627's are the very best audio op amps available....better dynamics, more authouritative bass, silkier highs, more 3 dimensional mid-range. The caps and resistors simply have better tone and ambience retrieval. The wire and connectors we use allow for better detail and imaging."
Attached is a list of mods suggested by Chris Johnson, the president of Parts Connexion, for DAC1. According to Chris:
"Essentially, we upgrade the analog output stage...from the DAC chips to the output...including critical signal path caps, resistors, op amps (these are the most expensive and most critical items in the mod...Burr-Brown OPA627 op amps on dual vertical adaptor boards), wire and output connectors. Plus, we upgrade the sonically critical PS diodes and resonance damp the chassis. The op amps we use in this mods are several levels above the NE5532's...the OPA627's are the very best audio op amps available....better dynamics, more authouritative bass, silkier highs, more 3 dimensional mid-range. The caps and resistors simply have better tone and ambience retrieval. The wire and connectors we use allow for better detail and imaging."
It seems to me the most critical mods are the replacement of the op-amps and the tent clock plus perhaps the rectifiers.
I cannot comment on changing the regulators to LT because I have not heard them compared to the standard ones.In fact although they are in fashion with some tweakers I have not seen any concrete feedback on any sound differences.
A lot of pro modding is bad value for money IMHO when you start using boutique resistors and caps.
Anyway for the high price some are asking one would think someone would come up with a nice discrete jfet I/V stage ala Borbely or a mini tube output stage or even some discrete jung type superregulators.
OK if there are people willing to pay double the price of the DAC1 for a handful of parts that cost say 200 usd maximum let them enjoy it.But I think the audiophile mod kitchen industry is not giving good value for money.Probably most of these mods will be available soon as standard on a chinese ``copy`` of the Benchmark for 400 usd the way it is going.
I cannot comment on changing the regulators to LT because I have not heard them compared to the standard ones.In fact although they are in fashion with some tweakers I have not seen any concrete feedback on any sound differences.
A lot of pro modding is bad value for money IMHO when you start using boutique resistors and caps.
Anyway for the high price some are asking one would think someone would come up with a nice discrete jfet I/V stage ala Borbely or a mini tube output stage or even some discrete jung type superregulators.
OK if there are people willing to pay double the price of the DAC1 for a handful of parts that cost say 200 usd maximum let them enjoy it.But I think the audiophile mod kitchen industry is not giving good value for money.Probably most of these mods will be available soon as standard on a chinese ``copy`` of the Benchmark for 400 usd the way it is going.
Dear Sebastian,
I feel the idea “less is more” quite appealing, but I don’t understand why you changed the upsampling frequency to 192 KHz. Benchmarch specifically chose 110 KHz because they measured that the dac or the filter (I don’t remember) was doing much better at that frequency.
Did you try the original 110 KHz upsampling rate with a Tent clock?
On your web site you show mods of other CD-player, how do they rank soundwise compared to the Benchmark?
Best regards.
Philippe.
I feel the idea “less is more” quite appealing, but I don’t understand why you changed the upsampling frequency to 192 KHz. Benchmarch specifically chose 110 KHz because they measured that the dac or the filter (I don’t remember) was doing much better at that frequency.
Did you try the original 110 KHz upsampling rate with a Tent clock?
On your web site you show mods of other CD-player, how do they rank soundwise compared to the Benchmark?
Best regards.
Philippe.
plep said:
Dear Philippe,
Honestly we only measured jitter performance accoarding to the "Dunn methode" (http://www.nanophon.com/audio/diagnose.pdf) Mebey not the most accurate methode but it gives us a picture of what really happend on the analogue outputs. In first instance I really like the idea of what Benchmark calls "ultralock" but I don't like the noisy clock enviroment with a bad 7805 that feeds both the clock XO as well the AD1896. Anyone with a DAC1 should messure the supply pin of the XO with a scope or a simple spectrumanalyzer, and see what a noise...
I don't even dare to doubt about bechmarks decision to use 110Khz. sample rate. However Tentlabs have not an XO clock of that frequency in program right now, so we decided to try out what happend if we use upsampling to 192KHz. We really liked the more clear soundstage and the improved depth. It is more then obvious and a great improvment. I think that says more about the quality of the Tentlabs clock then the move to 192Khz. I can not comment about a Tentlabs clock with 110Khz. samplerate. It is waiting till Tentlabs will have the right frequency in stock. The messured jitter is lower with the Tentlabs clock.
The DAC1 compared to other players. We are more then lucky that we receive a lot of great stuff on the office because we are a fulltime modification company. If you refer to my site. I like the Marantz SA1/CD7 more then the DAC1. I like the 4 TDA1547's the way the are configured in both the SA1 and CD7. Offcourse we get rid of all those HDAM stages becos I really hate them and they suck resolution. But that is another story. I have much experience with the AD1953 DA chip. And even though I like it very much, there is an certain "dark" character I dislike a little bit. I also appears in my own dac design prototypes made with this chip. I also like dacs with the way cheaper but in my opinion better sounding new AKM chips. My Apogee PSX-100 dac have one of them as well.
However the detail, and resolution and insight of the recording with a (modded) dac1 is still unheard of especialy for the price.
Best regards,
Bas
protos said:
Dear Protos,
Wel you have to put some limit somewere. We designed a dedicated DC servo circuit for the dac1. Replace almost the whole power supply board. We only go a little step further to ad a seperate power supply in a seperate enclosure. Only custom made for some custommers we made complete new outputstage boards with onboard regulation. But this is only for people that use the dac1 true balanced.
Any way. Like I said in the start of this post you have to put a limit somewhere. I had 1000 of ideas for the dac1, but then you really reach the point you can better come with a own design from the ground up. The dac1 is a very good product the way it is and unique the way it is. We want to keep it that way in respect to the designers but only push more out the little box. It can competete with the competition 3 times as expensive.
About the prices. Well... our bills have to be paid to. LikeI said we are a fulltime modification company. And trust me I don't get rich from it.. far from it.
Best regards,
Bas
I attached a picture below, which show C13 and C14 coupling caps (these are the numbers on the latest version of DAC1; see page 14 in the pdf version of the manual), which are the only 1000uf/6V caps on the board (near the headphone jacks). I will let Bas to give you more details.
Marin
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.