How can I adjust the gain on the JC 2.
The BOSOZ has a pot that adjusts the intrinsic gain of the circuit...could something similar be done with the JC 2 circuit...would the circuit be "upset" or be taken out of its "sweet spot" if the gain is adjusted???
The BOSOZ has a pot that adjusts the intrinsic gain of the circuit...could something similar be done with the JC 2 circuit...would the circuit be "upset" or be taken out of its "sweet spot" if the gain is adjusted???
myth is still a myth!we can not change what was done to a myth!thank Mr. John Curl was devoted to mankind masterpieces products.Again thank Mr. John Curl and wishing him happy and healthy!sorry if google translated incorrectly
The JC-2 is a very simple discrete transconductance amp, in essence the first stage of many opamps is build more similarly than different to this. There is a difference, but it derails the point. Both of the Analog Devices chipsets will improve upon the discrete JC-2's parameters, including offering lower noise, at least 20dB lower harmonic distortion, and wider bandwidth. A second chipset can be used to cancel phase shift, too. A third cheaper on can be used for DC offset correction, eliminating signal path capacitors.
However, when they are properly implemented, it will be impossible for human ears to discern or identify any tangible between the JC-2 and the chips mentioned. This is because the differences reside below the established thresholds of audibility. One would be surprised how much that is heard is actually the product of expectation bias- when the identity of the device is unknown, they are indistinguishable. The smallest distortion and noise spectra that a human can hear in sighted tests is about -75dB, and it has to be pretty raunchy and cover a wide spectrum to be heard. As for harmonics, 0.5% is around our hearing limits under the best circumstances. No one has ever successfully identified differences lower than this in controlled listening tests.
The JC-2 sits at .003% THD, about -90dB. The chips will do .0002% or better, -120dB with ease. The noise of the chips is also better than -120dB with careful layout, so even after being subjected to 30dB of gain through the power amplifier the system noise floor will be very low in the worst case scenario. These chips all have very high slew rates, so no SID is present.
The only time that chipsets like these can sound different is if very low feedback is used - reducing bandwidth, causing phase shift, and raising output impedance that can't drive capacitive cables - or, if the layout presents noise or interference problems. The JC-2 could benefit from a DC error correction amp on its output. Error correction eliminates capacitive coupling and is one of the more beneficial modifications to a preamp. If you wanted to drive earphones, an additional ouput stage could supply more current.
However, when they are properly implemented, it will be impossible for human ears to discern or identify any tangible between the JC-2 and the chips mentioned. This is because the differences reside below the established thresholds of audibility. One would be surprised how much that is heard is actually the product of expectation bias- when the identity of the device is unknown, they are indistinguishable. The smallest distortion and noise spectra that a human can hear in sighted tests is about -75dB, and it has to be pretty raunchy and cover a wide spectrum to be heard. As for harmonics, 0.5% is around our hearing limits under the best circumstances. No one has ever successfully identified differences lower than this in controlled listening tests.
The JC-2 sits at .003% THD, about -90dB. The chips will do .0002% or better, -120dB with ease. The noise of the chips is also better than -120dB with careful layout, so even after being subjected to 30dB of gain through the power amplifier the system noise floor will be very low in the worst case scenario. These chips all have very high slew rates, so no SID is present.
The only time that chipsets like these can sound different is if very low feedback is used - reducing bandwidth, causing phase shift, and raising output impedance that can't drive capacitive cables - or, if the layout presents noise or interference problems. The JC-2 could benefit from a DC error correction amp on its output. Error correction eliminates capacitive coupling and is one of the more beneficial modifications to a preamp. If you wanted to drive earphones, an additional ouput stage could supply more current.
Last edited:
Don't you think a composite two chip preamp, like that proposed by Walt Jung. can sound better than any of both designs, that is the JC2 or any of those chips?
Here's a headphone amp version that could be used as preamp:
http://www.waltjung.org/PDFs/Showcase_Headphone_Amp_KN_AX_0503.pdf
Here's a headphone amp version that could be used as preamp:
http://www.waltjung.org/PDFs/Showcase_Headphone_Amp_KN_AX_0503.pdf
The JC-2 line stage is not easily replaceable by an IC. In fact, it replaced an IC when it was first implemented.
It should have a slew rate of about 100V/us, has a very simple tru-path and very low order distortion only, with most measuring equipment. It is not perfect, but is a significant departure from any IC available, even today. It is NOT an op amp.
Measurements, below a certain level are an illusion that does not improve the actual listening quality of the gain stage. However, high amounts of negative feedback in order to achieve extremely low distortion seems to be problematic in achieving ultimate high fidelity. At least, I have not found any IC design to favorably compare to a good discrete design, and I design with both.
It should have a slew rate of about 100V/us, has a very simple tru-path and very low order distortion only, with most measuring equipment. It is not perfect, but is a significant departure from any IC available, even today. It is NOT an op amp.
Measurements, below a certain level are an illusion that does not improve the actual listening quality of the gain stage. However, high amounts of negative feedback in order to achieve extremely low distortion seems to be problematic in achieving ultimate high fidelity. At least, I have not found any IC design to favorably compare to a good discrete design, and I design with both.
By the way, it is not necessary to build a JC-2 line amp from scratch. Just get a completed unit E-Bay online from China. I paid $126 + shipping. The case alone is worth that much.
calmart:
it also depends on the job - best to build both and listen.
my short story:
i've never built a jc2 "clone", but i've used the ad744+ad811 between experiments as a default line stage and headphone amp for a LONG time because it does sound good (at least to me).
when i got some new, low impedance headphones (my previous ones were 300ohms; the new ones are <50 ohms i think), i recently built dick marsh's discrete headphone amp (nothing like jc2 schematic btw) and decided i liked it better driving those headphones. I was even using the same power supply for both headphone amps, so it was at least kinda fair test ...
hey, i just realized you should build both anyways - you've got to get more practice in on that Sprint Layout CAD software that you are evaluating!
🙂
mlloyd1
it also depends on the job - best to build both and listen.
my short story:
i've never built a jc2 "clone", but i've used the ad744+ad811 between experiments as a default line stage and headphone amp for a LONG time because it does sound good (at least to me).
when i got some new, low impedance headphones (my previous ones were 300ohms; the new ones are <50 ohms i think), i recently built dick marsh's discrete headphone amp (nothing like jc2 schematic btw) and decided i liked it better driving those headphones. I was even using the same power supply for both headphone amps, so it was at least kinda fair test ...
hey, i just realized you should build both anyways - you've got to get more practice in on that Sprint Layout CAD software that you are evaluating!
🙂
mlloyd1
No, because the distortion of the JC-2 and chipsets, when properly implemented to a board, are sufficiently below the threshold of audibility (1)-(10). All listening tests conducted, to date, conclude that distortion below -50dB is indiscernible (3)(4), some higher (5)-(7), and all attempts to prove otherwise halve always failed. The only time that participants in ABX and blind tests thought there was a difference in sonic quality was when they were under the influence of expectation bias - could see or otherwise knew which design was playing. The actual topology has no bearing on the sound, so long as there is a sufficiently flat response (11)(12), low noise, low distortion, controlled phase shift (15), and both DUTs have been level matched to within 0.1dB tolerance (13). That criteria is quite easy to adhere to today.
Wasn't easily matched 35 years ago, but it's power requirements and parameters as they pertain to audio waveform reproduction are not exclusive today.john curl said:
At the time when the JC-2 was designed, opamp ICs were of low performance standards and they would place notable limitations on use and consistency; low bandwidth, high noise, low gain. However, there was great progress just one decade later. Combined with newer processes in device manufacturing and circuit innovation, later devices surpassed early 1970's chipsets by considerable length.john curl said:
The fact that you accomplished it, is worthy of commendation. It's also worth mentioning that there is an inherent misconception that we often see, one that suggests the rate of change in audio frequencies is somehow exceeds that capability of the devices. Let's take a moment to investigate this.john curl said:
360 Degrees of a 20kHz waveform corresponds to a 50uS time span; each 1/4 of that wavelength is 12.5uS length, the peak rate of change residing at 0, 180, and 360 degrees of each waveform. Recorded audio on CD contains no information above 22kHz and the filter is much lower than this. On other sources, such as SACD, DVD, and DSD the upper limit of the media isn't much greater, even though the medium is capable of greater extension. The bandwidth is simply used to move the post filtering upwards in frequency so that a truly flat response with better square-wave response can be achieved at audio frequencies approaching 20kHz. At full swing when driving an amplifier, this only corresponds to about 1V/uS, placing a very low requirement on a DUT. Vinyl can carry higher frequencies, but the magnitude of those frequencies and its upper rate of change are surprisingly low: a rate of change no greater than 0.55uS at 20VRMS from the output of a 100W amplifier (14). For a preamplifier, it's a fraction of that.
The other data point that is often overlooked is quite telling: the actual magnitude versus frequency of real recordings. Human hearing is not linear, it requires less energy at higher frequencies, a mere fraction, in fact. We never see audio with treble magnitudes as high as bass, and certainly never recorded near 0dB attenuation, which would correspond to a DAC's maximum threshold prior to clipping. Because of this, the treble frequencies never approach the full slew capability of the device. The highest full signal transient in any SACD recording that I knew of occurred under 13kHz, and it was a poor dynamic compressed recording. For virtually all recordings nothing above 8kHz will approach full output magnitude.
According to Douglas Self's publications and others, 20V would only require 5V/us (14) to accurately reproduce all frequencies without slew limiting distortion. A safety margin of 10V/uS is preferable. All credible doctrines cite within this range.
No one has suggested that it was.john curl said:
If I'm reading your context correctly, then I agree that once distortion is below the established thresholds of audibility it makes no difference. I also contend that with a DC error correction amp on its output, the JC-2 is worthy of any audio system. This is subject to the condition that the final power amplifier's gain and the speaker's sensitivity aren't high enough for noise to be audible.john curl said:
The blame that the audiophile community has placed on feedback and opamps seemed to stem from a series of dubious papers submitted to the AES and misconceptions among audio magazines, through the power of suggestion bias. Feedback doesn't create high levels of high-order or odd harmonics, intermodulation distortion, immeasurable distortions, nor does it induce transitory distortions on changing signals. If any change in the audio waveforms existed it would be visible on comparator analysis that references the input against its output. Anytime that there has been a claim that feedback worsens the performance the claimant fails to provide evidence, or fails to participate in properly conducted ABX and blind listening tests. All others that partook in tests were unable to identify which DUT was which, regardless of topology - provided that the levels were matched within 0.1dB so that it was below audibility (13), response was flat from 20-20kHz (11)(12), harmonic and IMD was below -50dB (1)-(10), low phase shift (15), and noise generated by the DUT was below the perceptible threshold in the listening environment.john curl said:
There has been no compelling proof from credible academic sources that feedback causes higher distortion and bad sound. Particularly, none to support the claims that opamps perform inferior to discrete designs. Some have provided evidence to the contrary (16), ref. Walt Jung. The negative claims have been entirely by audiophiles, and only after they read that feedback should've made it sound terrible.
References:
(1) Auditory Perception of Nonlinear Distortion, Audio Engineering Society, Geddes, Earl R.; Lee, Lidia W.
AES E-Library Auditory Perception of Nonlinear Distortion
(2) Audibility Perception of Nonlinear Distortion, Audio Engineering Society, reprint 5891.
(3) Harry F Olson, JDD level of .7% using 40 Hz to 14 kHz bandwidth test system
(4) D.E.L Shorter, "Just perceivable distortion values of 0.8% to 1.3%".
(5) P.A Fryer, 2% - 4% distortion
(6) James Moir, Just detectible distortion (JDD) "level can be no lower than 1%, ie -45dB"
(7)Von Braunm ü hl & Weber, 1% - 2% at frequencies > ~ 500 Hz
(8) M. E. Bryan & H.D.Parbrook also embarked on tests involving the audibility of harmonics.
(9) Just Audible Thresholds for Harmonic Distortion, By Bryan, M.E.; Parbrook, H.D.
Just audible thresholds for harmonic Distortion: ingentaconnect
(10) Audibility of linear distortion in loudspeakers, Sylvain Choise
http://www.almainternational.org/ya...audibility_of_linear_distortion.106172825.pdf
(11) Measurement of Harmonic Distortion Audibility Using a Simplified Psychoacoustic Model. By Temme, Steve; Brunet, Pascal; Qarabaqi, Parastoo
https://secure.aes.org/forum/pubs/conventions/?elib=16446
(12)"A change by 1 dB is about the smallest change a human being can detect."
Sound
(13) Dr. Floyd Toole: "There is some evidence that we can detect slopes of about 0.1 dB/octave, which translates into a 1dB tilt from 20Hz to 20kHz - not much. Such a spectral of error, is likely to be quite benign and subject to adaption."
The Acoustics and Psychoacoustics of Loudspeakers and Rooms.
(14) Douglas Self, Audio Power Amplifier Design Handbook, Page 256, direct references to the results of Peter Baxandall and Nelson Pass.
https://books.google.ca/books?id=TL...a=X&ei=_Bm9VK6jGISZgwSd8YGgBg&ved=0CAsQ6AEwAA
Douglas Self, Small-Signal Audio Design: "Consider also that many amplifiers have RC filters at the input, not so much for EMC immunity, but more as a gesture towards what used to be called 'transient intermodulation distortion' (TID); but this is actually just old-fashioned slew limiting and highly unlikely in practice."
(15) Dr. Leach, The Differential Time-Delay Distortion and Differential Phase-Shift Distortion as Measures of Phase Linearity, Audibility of Phase Shift, Journal of the Audio Engineering Society, vol. 37, no. 9, pp. 709-715.
(16) Walter Jung, The Impossible Void, SID and Slew Induced Distortion.
"The subject of this first installment is slewing induced distortion which I call SID. If that sounds remote, please note that SID includes transient inter-modulation (TIM) and many other distortion buzzwords so much in vogue these days. I believe SID is a broader and more penetrating view of the distortion phenomenon than has been presented to date, and I hope this article begins to vindicate the IC op amp in the minds of those who have heretofore been convinced of its alleged “inferiority.”'
http://www.pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Jung_W/SID_and_TIM_W_Jung_77-79.pdf
There are two JC2 schematics on this eBay ad:
JC 2 Preamplifier Class A Parallel Power Kit WLX 6 | eBay
That is, on the second version there's another output. They do not specify (I asked them) which is the kit they sell.
Besides additional current and more linearity, what other advantages would that stage bring? It looks as if it could drive a headphone too.
JC 2 Preamplifier Class A Parallel Power Kit WLX 6 | eBay
That is, on the second version there's another output. They do not specify (I asked them) which is the kit they sell.
Besides additional current and more linearity, what other advantages would that stage bring? It looks as if it could drive a headphone too.
Some additional questions:
1) Has anyone used the JC-2 with Lightspeed LDR attenuator?
2) Has anyone compared it with the TVC passive attenuator?
3) What pots and values give better results at the input?
1) Has anyone used the JC-2 with Lightspeed LDR attenuator?
2) Has anyone compared it with the TVC passive attenuator?
3) What pots and values give better results at the input?
I think bcmbob might have tried a lightspeed with a JC-2, not sure if he's still posting here.
I did compare a JC-2 i built from the Chinese kit with my DCB-1, the latter is a bit more transparent. The JC-2 does add a bit of warmth and a powerful bass. If you need the gain it gives, it probably wont be beaten on price v performance.
I also noticed that the JC-2 clone i built sounded better with 19 volts rail voltage, gets quite hot though.
I tried a DACT minature stepped attenuator (ebay) and a Tocos pot, both 20K, both worked well but the Tocos is more user friendly. I really dislike the alps blue so have not tried one in years but some folk use them with their JC-2 clones. I found them to be veiled sounding things with poor bass.
Last edited:
The Levinson JC-2 line stage was designed over 40 years ago, and while quite good at the time, and still one heck of a bargain from China today, is not the latest of this series of designs, but it has all the basic changes that we made to get the best performance in order to exceed what we could get from opamps then, and I still think, today.
It is a radical departure from the OP AMP model. It is intrinsically lower distortion, open loop, than the IC op amps normally available. It is similar in philosophy to the Otala Power Amp, written up in the AES Journal in the early 70's. The actual THD measurement is secondary, the slew rate, simple thru-path, and relatively high open loop bandwidth is the difference.
I don't use ABX tests to hear differences. I hope that you out there do not, either. Waste of time, trust your ears.
It is a radical departure from the OP AMP model. It is intrinsically lower distortion, open loop, than the IC op amps normally available. It is similar in philosophy to the Otala Power Amp, written up in the AES Journal in the early 70's. The actual THD measurement is secondary, the slew rate, simple thru-path, and relatively high open loop bandwidth is the difference.
I don't use ABX tests to hear differences. I hope that you out there do not, either. Waste of time, trust your ears.
I purchased the Chinese jc-2 amp based on John Curls tip. It is indeed good value with a nice case, power supply and connectors, although it's a shame (imho) that they finished the exterior in silver instead of brushed aluminium. But at less than 140 bucks landed I'm not complaining. A nice touch is that all the screws are non-magnetic.
PSu section is regulated and with a tl431 voltage reference. Jfets are the Toshibas and look to me to be genuine.
Gain is 13db and it will swing 10v rms before clipping sets in.
What was odd though was the amount of noise - most of it >20khz but it doesn't seem to be mains borne so I don't yet know it's source. It's clearly evident even with a good mains filter and didn't go away even when the measurement was carried out with the mains unplugged (ie. during the 10 secs or so before the caps discharged.) I intend to power this preamp with a known good lab PSU to confirm the issue.
I saved some curves when i had this on the test-bench so I can put some results here in the next few days (am traveling) in case anyone is interested.
Thanks JC for the circuit and the generous tip - now for a bit of work to get rid of the hash. If I can definitively rule out the PSU, this will take some doing.
PSu section is regulated and with a tl431 voltage reference. Jfets are the Toshibas and look to me to be genuine.
Gain is 13db and it will swing 10v rms before clipping sets in.
What was odd though was the amount of noise - most of it >20khz but it doesn't seem to be mains borne so I don't yet know it's source. It's clearly evident even with a good mains filter and didn't go away even when the measurement was carried out with the mains unplugged (ie. during the 10 secs or so before the caps discharged.) I intend to power this preamp with a known good lab PSU to confirm the issue.
I saved some curves when i had this on the test-bench so I can put some results here in the next few days (am traveling) in case anyone is interested.
Thanks JC for the circuit and the generous tip - now for a bit of work to get rid of the hash. If I can definitively rule out the PSU, this will take some doing.
Last edited: