Hello all and merry Christmas!
I had an interesting session the other day with a couple of friends (one is a piano player and excellent loudspeaker engineer, the other is the most famous Greek audio press scribe). We were evaluating CD playes and dacs using 2 very high resolution speakers and a wee full range bipole. One is a 2 box 3 way with PHLs and Fountek, other is a 2 way standmount with PHL and Audax, third is Salas TB3 revisited (you can search this forum about TB3 rev.). We had Unison Research moded Unico tubed CDP, Benchmark DAC 1, Deltec vintage hot rodded, Copland 288, Theta Gen V8, Creek Destiny.
Well, we came to the conclusion that CD replay is still a flawed case compared to a turntable (but we cant bear the fact that only 10% of an LP collection can prove this, until it will eventually worn out, so its practically an elite afair) and its replay (CD) did not come such a long way in 20 yrs. Tarted up yes but not fundamentally evolved.
Apart from the above blanket conclusion it was interesting to realise that each manufacturer was trying to present a credible rendering of music, doing his tone hocus pocus. The Benchmark had top resolution but was thin. Theta had the resolution and the guts (but at $$$$), Unison emphasised midrange musicality, Creek was treading a fine line between excess and restrain (overall the safest bet, also the best transport bar the expensive Theta). Copland was resolute enough but grandiose. Funny how the DPA little bit of 15 years old tech was still giving a valid point of view!
I ended up with the notion that I could buy the Benchmark if -and only if- I could flesh out its tone. To get a Theta is out of the question, I am not well off.
I thought of what is making it sound thin...this tone I have heard before...and then it dawned on me! Sounds like 5532! Ok I opened it up...11 NE5532's! silicon bridge rectifiers, smd coupling caps...
Gentlemen! your attention please! Suggestions are in order. Remember, there must be a double opamp thats unity stable and not a rare species with touchy behaviour for throwing out the 5532 among other hot rod suggestions.
P.S. The DPA little bit has 2604 smd, can a resistor from out to V+ or V- help it ? Had good results with 627 doing this once. If yes where is better with this chip? V+ or V-? ...Carlos FM?
See a picture of the benchmark with lid off :
I had an interesting session the other day with a couple of friends (one is a piano player and excellent loudspeaker engineer, the other is the most famous Greek audio press scribe). We were evaluating CD playes and dacs using 2 very high resolution speakers and a wee full range bipole. One is a 2 box 3 way with PHLs and Fountek, other is a 2 way standmount with PHL and Audax, third is Salas TB3 revisited (you can search this forum about TB3 rev.). We had Unison Research moded Unico tubed CDP, Benchmark DAC 1, Deltec vintage hot rodded, Copland 288, Theta Gen V8, Creek Destiny.
Well, we came to the conclusion that CD replay is still a flawed case compared to a turntable (but we cant bear the fact that only 10% of an LP collection can prove this, until it will eventually worn out, so its practically an elite afair) and its replay (CD) did not come such a long way in 20 yrs. Tarted up yes but not fundamentally evolved.
Apart from the above blanket conclusion it was interesting to realise that each manufacturer was trying to present a credible rendering of music, doing his tone hocus pocus. The Benchmark had top resolution but was thin. Theta had the resolution and the guts (but at $$$$), Unison emphasised midrange musicality, Creek was treading a fine line between excess and restrain (overall the safest bet, also the best transport bar the expensive Theta). Copland was resolute enough but grandiose. Funny how the DPA little bit of 15 years old tech was still giving a valid point of view!
I ended up with the notion that I could buy the Benchmark if -and only if- I could flesh out its tone. To get a Theta is out of the question, I am not well off.
I thought of what is making it sound thin...this tone I have heard before...and then it dawned on me! Sounds like 5532! Ok I opened it up...11 NE5532's! silicon bridge rectifiers, smd coupling caps...
Gentlemen! your attention please! Suggestions are in order. Remember, there must be a double opamp thats unity stable and not a rare species with touchy behaviour for throwing out the 5532 among other hot rod suggestions.
P.S. The DPA little bit has 2604 smd, can a resistor from out to V+ or V- help it ? Had good results with 627 doing this once. If yes where is better with this chip? V+ or V-? ...Carlos FM?
See a picture of the benchmark with lid off :
Attachments
salas said:
AD827.
salas said:
You can bias the OPA(2)604, but you won't hear much of a difference, if any. The same for the OPA(2)132.
I have letf using the OPA2604 some years ago, because it sounds too distorted in the midband/treble for my ears, there are much better options.
Try the AD827.
Merry Christmas
Hi Salas. One approach:
http://www.6moons.com/audioreviews/benchmark/dac1_3.html
Can't vouch for it, know nothing more than what's described in the article. I also have a DAC-1, an early model, and have considered making a few mods (lowering the unbalanced Zout for one) but got put off by all the SMT. The bridge rectifiers will eventiually go.
http://www.6moons.com/audioreviews/benchmark/dac1_3.html
Can't vouch for it, know nothing more than what's described in the article. I also have a DAC-1, an early model, and have considered making a few mods (lowering the unbalanced Zout for one) but got put off by all the SMT. The bridge rectifiers will eventiually go.
salas said:
You can say that.
The board is not practical to work on, at all.
Standard rectifier bridges, SMD electrolythic caps, 78/9xx regs... :bs:
I can't see the bottom side of the board, but those regs should be bypassed (input and output to ground) with some small caps (100nF is fine) right on their legs.
No electrolythic psu caps on the analog stage... limits the opamp choice to... NE5532, and some others.
How practical is it to make a decent psu bypass on those op-amps?
Is this an expensive dac?
salas, how old was the Benchmark DAC-1 when you performed your listening test? thin tonality is usually a problem before a component is broken in (i have a class D amp in-house now suffering from that as well). these IC-packed digital designs in particular need several hundred hours of playing time to loosen up.
i would be careful writing off the component choices before the unit is fully broken in and the design and implementation details are understood. audio design is a holistic process, and often substituting "audiophile" parts is more an exercise in futzing around with different colorations than truly improving fidelity. not saying i wouldn't tweak the Benchmark myself (i would), but only with great care, particularly in regard to the SMD components.
i would be careful writing off the component choices before the unit is fully broken in and the design and implementation details are understood. audio design is a holistic process, and often substituting "audiophile" parts is more an exercise in futzing around with different colorations than truly improving fidelity. not saying i wouldn't tweak the Benchmark myself (i would), but only with great care, particularly in regard to the SMD components.
Hello
That unit I tested was a well run in demo. When I got mine 15 days after, I saw that the newest production had far better balance and I found confirmation of differences in that department in a far east forum where the members were advising against a second hand purchase due to better performance of newer units.
My new DAC1 took 1 week to settle and I can see no edge in its sound. Also when I moved the jumper that fixes the digital input (coaxial in my case, position 'C') bypassing the input selection circuit there was a subtle but reasuring improvement for less artificial midhighs.
From what I have gathered, the most documented step by step upgrade is being implemented professionaly in Australia. There they were very reluctant too since they found the Benchmark exceptional right out of the box. But when they covered all bases carefully they declared it their best digital platform.
I am building a woofer extension now for my PHL-Audax 2 ways and when I will be happy with combining it I will listen carefully for long and see if I will proceed to 'dramatise' the DAC1 further. Step by step and prudently. I need to get some THS4032s for evaluation first since I am eying this TI opamp 'with bad intent'.
First step will be the analogue supply with CLC and no regulator substitution with fall back ability care nontheless.
Benchmark engineers told me that they opted for 5532 bcs it does not exibit THD rise in the audio band, thus gives the less IMD. True, its the best 5532 implementation with the less dramatic 5532 signature I have seen. The boys were careful.
Jitter busted or not, the DAC1 shows differences in transports easily I might add.
All in all the DAC1 payed justice to my DIY tubes and speakers and I can unmistakably wallow in a sane correctness eachtime I listen. The details are all there and it sounds like the WEISS in my friend's studio. Warts and all but not artificially.
That unit I tested was a well run in demo. When I got mine 15 days after, I saw that the newest production had far better balance and I found confirmation of differences in that department in a far east forum where the members were advising against a second hand purchase due to better performance of newer units.
My new DAC1 took 1 week to settle and I can see no edge in its sound. Also when I moved the jumper that fixes the digital input (coaxial in my case, position 'C') bypassing the input selection circuit there was a subtle but reasuring improvement for less artificial midhighs.
From what I have gathered, the most documented step by step upgrade is being implemented professionaly in Australia. There they were very reluctant too since they found the Benchmark exceptional right out of the box. But when they covered all bases carefully they declared it their best digital platform.
I am building a woofer extension now for my PHL-Audax 2 ways and when I will be happy with combining it I will listen carefully for long and see if I will proceed to 'dramatise' the DAC1 further. Step by step and prudently. I need to get some THS4032s for evaluation first since I am eying this TI opamp 'with bad intent'.
First step will be the analogue supply with CLC and no regulator substitution with fall back ability care nontheless.
Benchmark engineers told me that they opted for 5532 bcs it does not exibit THD rise in the audio band, thus gives the less IMD. True, its the best 5532 implementation with the less dramatic 5532 signature I have seen. The boys were careful.
Jitter busted or not, the DAC1 shows differences in transports easily I might add.
All in all the DAC1 payed justice to my DIY tubes and speakers and I can unmistakably wallow in a sane correctness eachtime I listen. The details are all there and it sounds like the WEISS in my friend's studio. Warts and all but not artificially.
salas said:
that sounds consistent with my magazine editor's observations. he reviewed both the Weiss and Benchmark recently and if i remember correctly he felt the Benchmark's sound quality was in the same league. given the price difference (more than 10x?), his recommendation for the Benchmark was obviously stronger.
Salas, you might want to check out Steve Nugent's comments on the Benchmark (very favourable). Look under "Component Mods."
Empirical Audio
Empirical Audio
You need some guts to charge $1050 for below mods, on the DAC that can be bought for $700 used:
Two bridge rectifiers replaced with HEXFRED's
Replace regulated voltage filter caps
Improve power delivery to DAC and upsampler chips
Improve power delivery to 9 op-amps - Black Gates and HF caps
Upgrade DAC filtering
Replace all voltage regulators.
Eliminate AC-coupling and tune-out DC offset
Improve digital input coupling
I had the Benchmark for a week and I let it go.
Two bridge rectifiers replaced with HEXFRED's
Replace regulated voltage filter caps
Improve power delivery to DAC and upsampler chips
Improve power delivery to 9 op-amps - Black Gates and HF caps
Upgrade DAC filtering
Replace all voltage regulators.
Eliminate AC-coupling and tune-out DC offset
Improve digital input coupling
I had the Benchmark for a week and I let it go.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.