Update 1/21/18:
I wanted to let everyone know how impressed I have been with this unit by just using the stock 5532 op-amps with this Douglas Self based circuit design. I am using a rectifier supplied by Audio Sector (Peter Daniel) The parts are explained further below. It uses MOSPEC U08A60 ultra fast recovery diodes with a SILMIC II for each rectifier to smooth out the DC output. With his design I can actually run twin 9V secondaries into the rectifier and have stable +15V and -15V at the output (this is how mine is running now). Using the XLR outputs of the DAC vs. the RCA outputs gives us many benefits, especially in two areas.
The first area is the floor noise, it doesn’t exist!! This setup is as silent as it could possibly be when nothing is playing. To me, this is as important as having quality sound. The second area is the nuances in the tracks you hear more clearly, usually a secondary instrument or a background vocal. I am currently listening through foobar2000 and Jack Johnson’s “All The Light Above it Too” at 24-bit 96 kHz. Everything just seems more alive, more musical and pleasing to listen to without fatiguing your ears. I will have to do more comparisons before I can be sure. I also threw my Kenwood Basic C-2 Pre-Amp into the mix to allow easy switching for comparison. Stay tuned for the op-amp review Part 2 and more on this project including throwing discrete op-amps in to see if that changes much!
Update 1/22/18:
I also wanted to thank Tom Hoffman for allowing me to test the XLR setup with some Belden 1800F “Brilliance” XLR cables using Neutrik plugs. He gave me some other goodies most of which I haven’t been able to utilize yet. I am utilizing a Cryo treated 5x20mm HiFi-Tuning.com Fuse from Tom for my ES9038 DAC vs. the stock metal and glass ones. This will definitely require more supposed HiFi fuse samples (I have 3 on the way) before I can say if this makes any noticeable difference on my setup. I can already tell you from other HiFi Fuse Reviews, even the people who say it makes a difference would rate it around a 1-2 level change out of 10 (so don’t blow big bucks on fuses go for below $20).
I have also utilized the “Kimber Kable KWIK-12 E120909 M 14/4 (UL) CL2 F2027013″ speaker cable which is bi-amping my Polk 70’s from one run of cable! Again this is from Tom! So a big thanks to my new friend Tom Hoffman!! Without Tom I would have had to wait another 1-2 weeks before testing the XLR to RCA converter. This speaker cable is quad conductors, but one of the special things about Kimber cabling is that they use different gauges of wire to make up one conductor x4 (with double Teflon jackets).
From Kimber’s Website on KWIK-12 cables:
Premium in wall speaker cable for premium sound. Dual gauge strand construction in PE dielectric. Four conductor design to be used in hybrid braid style two conductor termination. 12 AWG / 3.31mm2. (UL) CL2 rated. One conductor of each color contains finer gauge copper strands while the other conductor of each color contains heavier gauge copper strands. When the two gauges are combined they serve to help minimize resonance within the cable stranding, thereby optimizing performance. Far superior, both electrically and in terms of sound quality, to the commonly used PVC dielectric. PE provides a smoother, cleaner and more grain free sound. The off-white PVC outer jacket is durable and pulls smoothly through studs without sticking or binding. The neutral color of the jacket is less objectionable, with regard to decor, should the cable become visible. In addition, convenient foot labeling makes it easier to track cable usage.
Kimber cables are actually pretty reasonable with regard to price for what they can offer to the right amp and speaker combo. You can get 20 feet for $50 which is actually cheaper than some of the mid to high end cables found on Amazon! I will say one thing about the Kimber cable setup I have right now on my front left Polk 70, it’s harder to drive than the Cables Matter speaker wire setup I have on the right speaker. BUT, the Kimber side sounds fuller and has more texture than the right speaker. Tom didn’t think this should happen (being lower in volume), so I am going to try running my bi-amp runs with two separate Kimber KWIK-12 runs, one for each amplifier (bi-amp = 2 amps per speaker).
HERE is the spec sheet for the MOSPEC U08A60 MCDN Ultra Fast Recovery Diodes. I am using quad balanced microphone wire for both the XLR and RCA jacks, as shown below. This is cable sold by Monster cables, which is why I have it cut it into pieces!! Just kidding, this cable actually uses a helix design with double Teflon jackets. The PCB is the part that is based on a Douglas Self circuit, the rest is my design.
The D. Self based design layout and parts selection was done by someone else and I came in and added Nichicon FW instead of the cheaper stock ones. The black capacitors are Rubycon. Then we have MKT type caps along with older style metalized film. The board’s four op-amps require +18V 0V -18V, so this tells us we do not run AC power into this board. The only way to get +18 and -18 volts is by using a rectifier of some sort. It can be done many ways, see here. Most of the time my rectifiers end up with too much voltage which is handled temporarily with a variac.
I will be using this in my op-amp review which is why I wanted to go ahead and get this built! This design uses four Dual-DIP op-amps! Below is an almost finished picture, I just need to decide on how I want to mount the rectifier board along with the transformer. This is 90% complete by my guess.
When soldering the ground (shield) of the RCA don’t go for the areas sticking out to keep wires apart. This is the thickest piece of metal on the RCA jack and thus the hardest to use for solder. It took me about 10 minutes of fiddling around with the shield inside the enclosure before I realized that and got solid test tones and all below 1 ohm readings for continuity.
Almost every voltmeter (VMM) has a continuity test, but even if it doesn’t just switch the meter to read resistance and see what the reading is from the PCB to a reading taken on jacks outside of the enclosure. I use these little RCA to hot and ground plugs that make it super easy.
“What are the big differences between going to balanced (XLR) vs unbalanced (RCA)?”
Interference (from Electromagnetic Induction) Reduction:
A typical balanced cable contains two identical wires, which are twisted together and then wrapped with a third conductor (foil or braid) that acts as a shield. The two wires form a circuit carrying the audio signal; one wire is in phase with respect to the source signal, the other wire is reversed in polarity. The in-phase wire is called non-inverting, positive or “hot” while the out-of-phase wire is called inverting, phase-inverted, anti-phase, negative or “cold”. The hot and cold connections are often shown as In+ and In− (“in plus” and “in minus”) on circuit diagrams.[2]
The term “balanced” comes from the method of connecting each wire to identical impedances at source and load. This means that much of the electromagnetic interference will induce an equal noise voltage in each wire. Since the amplifier at the far end measures the difference in voltage between the two signal lines, noise that is identical on both wires is rejected. The noise received in the second, inverted line is applied against the first, upright signal, and cancels it out when the two signals are subtracted.
This differential signal recombination can be implemented with a differential amplifier. A balun may also be used instead of an active differential amplifier device.
The wires are also twisted together, to reduce interference from electromagnetic induction. A twisted pair makes the loop area between the conductors as small as possible, and ensures that a magnetic field that passes equally through adjacent loops will induce equal levels of noise on both lines, which is canceled out by the differential amplifier. If the noise source is extremely close to the cable, then it is possible it will be induced on one of the lines more than the other, and it won’t be canceled as well, but canceling will still occur to the extent of the amount of noise that is equal on both lines.
The separate shield of a balanced audio connection also yields a noise rejection advantage over an unbalanced two-conductor arrangement (such as used in typical home stereos) where the shield must also act as the signal return wire. Any noise currents induced into a balanced audio shield will not therefore be directly modulated onto the signal, whereas in a two-conductor system they will be. This also prevents ground loop problems, by separating the shield/chassis from signal ground.
Differential signaling:
Signals are often transmitted over balanced connections using the differential mode, meaning the wires carry signals of opposite polarity to each other (for instance, in an XLR connector, pin 2 carries the signal with normal polarity, and pin 3 carries an inverted version of the same signal). Despite popular belief, this arrangement is not necessary for noise rejection. As long as the impedances are balanced, noise will couple equally into the two wires (and be rejected by a differential amplifier), regardless of the signal that is present on them.[3][4]
However, there are some minor benefits to driving the line with a fully differential output:
The electromagnetic field around a differential line is ideally zero, which reduces crosstalkinto adjacent cables, useful for telephone pairs.
Though the signal level would not be changed due to nominal level standardization, the maximum output from the differential drivers is twice as much, giving 6 dB extra headroom.[3]
Increasing cable capacitance over long cable runs decreases the signal level at which high frequencies are attenuated. If each wire carries half the signal voltage swing as in fully differential outputs then longer cable runs can be used without the loss of high frequencies.
Noise that is correlated between the two amps (from imperfect power supply rejection, for instance), would be cancelled out.
At higher frequencies, the output impedance of the output amplifier can change, resulting in a small imbalance. When driven in differential mode by two identical amplifiers, this impedance change will be the same for both lines, and thus cancelled out.[3]
Differential drivers are also more forgiving of incorrectly wired adapters or equipment that unbalances the signal by shorting pin 2.[3]
Source: https://en.wikipedia.org/wiki/Balanced_audio
Internally Balanced Audio Design:
Most audio products (recording, public address, etc.) provide differential balanced inputs and outputs, typically via XLR or TRS phono connectors. However, in most cases, a differential balanced input signal is internally converted to a single-ended signal via transformer or electronic amplifier. After internal processing, the single-ended signal is converted back to a differential balanced signal and fed to an output.
A small number of audio products have been designed with an entirely differential balanced signal path from input to output; the audio signal never unbalances. This design is achieved by providing identical (mirrored) internal signal paths for both the “non-inverting” and “inverting” audio signals. In critical applications, a 100% differential balanced circuit design can offer better signal integrity by avoiding the extra amplifier stages or transformers required for front-end unbalancing and back-end rebalancing. Fully balanced internal circuitry has been promoted as yielding 3 dB better dynamic range, as explained above,[5] though at increased cost over single-ended designs.
Source: https://en.wikipedia.org/wiki/Balanced_audio
Keep an eye out for the testing of this module. Can this simple circuit take the ES9038’s balanced outputs as inputs and then provide us an unbalanced output with higher quality sound than just using the unbalanced jacks found on the DAC? Is this like that saying, “Going around your ass to get to your elbow”? Stick around and find out along with the op-amp head to head showdown that is already in progress. Although I am still waiting on a full size 4 pin 22.5792 MHz oscillator to use the Amanero again.. It’s almost here, I hope.