Development thread starts here.

While I consider this my best work to date, the documentation trail has become hopelessly scattered so let me try to regroup in this blog post and make some sort of gateway/portal to the project. [Update: web page write up now finished.]

The followup to the Sapphire 3 started out as a generic musing on stacked diamond buffers and current mirror amplification, coalesced for a bit as the ill-fated Project Unity before becoming part of the Sapphire line initially as a temporary measure as something I could drop in my existing chassis for testing. The first iteration kept the open loop buffer of previous Sapphire iterations, but now I'm running the 4.1m boards with the buffer inside the feedback loop I realize I've ended up at just the classic, dictionary definition current feedback amplifier (CFA). Sigh. So much for originality. It sounds great though, so I'm not as upset about it as I might have been.

*****...

In hindsight the original CrystalFET suffered from overreach. To obtain the 55 dB midband gain required for an MC phono stage lots of other things I might have wanted to build into the design like degenerative feedback were discarded.

Degenerative feedback as it applies to JFET amplifiers is the practice of leaving the source resistor partially or completely unbypassed. It reduces the voltage gain, but linearizes the dynamic operation of the amplifier - coercing the gain to the ratio of the source and drain resistors rather than leaving it defined by the JFET transconductance. The circuit gain can then be fixed by the designer without worrying so much about matching devices.

The CrystalFET circuit is a traditional two stage passive photo preamplifier design, the only novelty here is the addition of a second JFET configured as a source follower to each amplifier stage. To obtain above 50 dB gain no degenerative feedback could be used, and the JFETs had to be not...

The Phonoclone 4 applies a major revision to the board layout, making it more logical with grouped, standardized I/O connections. The X-Reg has been replaced with the S-Reg shunt regulator circuit. Overall the BOM has been considerably simplified and the parts cost reduced.

Rev. 4.4s boards are back-compatible with earlier Phonoclone revisions, with the anachronistic 2.5"x3.5" mounting holes, so I can drop it into my existing chassis. 4.5b moves to 70x90mm mounting holes to be compatible with the Sapphire, VSPS200, and CrystalFET boards.

*****

Build Notes

4.0b had some low level residual ripple on the output as a result of capacitive coupling or ground loop error. The boards were respun as 4.4s (special test run, limited, unsexy green color) which work as designed. The next wide release is 4.5b.

The BOM and board connectivity checks, no issues.

The Bourns trimmer, R21, comes from the factory...

This is an unfinished design idea.

*****

Another request.

The Sapphire4 and Unity (H) boards were based on the 47 Labs Treasure 0347 amplifier circuit, and the original was not so far away that I couldn't "de-modify" the Unity (H) board back to what is essentially a 0347 (but with neither the original transistors nor the original resistor values).

I changed just about all the resistor values slightly for one reason or another, and changed all the transistor types. I've run my own simulations to check my version basically works as it should. Output BD135(39),136(40) are shown on the schematic as being leftovers from my headphone design, they will work to play music but to meet high current output specifications something bigger in TO-220 or similar thermal package will be needed.

As I hope you can see from the circuit schematic, the design - for a power amplifier - is unconventional. There is no feedback...

Two boards, each board configurable for line and headphone variants. Powered by nominal +12 V -12 V split regulated supply, though 9-15 V rails can be used without modification.

Following on from this post and this post, we arrive in time for the holidays with Project Unity. Merry Christmas everyone.

Four circuits: Unity, Unity H, Unity B, Unity BH. line preamp, Headphone amp, line Buffer, Headphone Buffer respectively, all derived from a common base circuit called Unity Root. Unity Root is conceptual, it exists only as a reference so you can see more clearly how the four working variants relate to each other.

It's the "all for one, and one for all" approach to diyaudio, a single research and development line applied to a range of applications, feedback from any of the applications brought back to apply to the line in general.

This simplifies not just the circuit development, similar efficiency is also brought to the documentation, board layout design, and BOM... about which I'll have more to say in a bit.

I've come this far so I might as well complete the trifecta.

This is the X-Reg circuit, with a new layout and component numbering to match the new S-Reg and Z-Reg boards. Each is a drop in replacement for the other.

Like its siblings, the X-Reg is a low current voltage regulator for line level audio. The output is about 9 V (adjustable up to 12 V) and the maximum load current is 50 mA without heatsinks, 100~150 mA with small heatsinks on Q1,2.

It is not a true regulator as there is no fixed reference and instead the output voltage is defined relative to the input voltage. The high open loop gain of the op amp is harnessed for very low noise and very high ripple rejection. It is necessarily a high feedback approach.

So, there you are: three mix-and-match power supply options for all your low voltage, low current audio needs.

Companion regulator to the S-Reg. Same board dimensions and connection layout. For line level audio.

Same idea as before rectified 2x12 VAC input, +/- 11 V output. 50 mA max output current unless the transistors are heatsinked.

There is a small amount of over-current protection afforded by R3,4 but do not short the output for all but the shortest of transients.

Eagle/Gerber files attached, so you can go get this made yourself, optinally with whatever modifications you need.

I've use this regulator circuit in my Sapphire headphone amplifier. It's a simple and modest circuit but I feel it works really well in practice as long as the audio circuitry it powers has reasonable PSRR. No feedback means no out-of-band noise or instability, even as the output impedance remains low.

S-reg voltage shunt-source regulator for line-level audio circuits.

This is designed to accept rectified 12x2 VAC input, producing +/-12 V output split rails.

Features soft-start and over-current protection, standard configurations for load currents up to 50 mA. High performance: 80 dB ripple rejection up to 100 kHz.

update: BOM uploaded
update: web page (stub)

This is either an ingenious interfacing of the output buffer and current feedback amplifier by adding second arm to the central current mirrors ... or it's another really bad idea(tm).

It simulates nicely though.

[What's happening here is current output of collector Q9 is no longer being asked to bias and drive the buffer Q16. Instead Q2 and Q17 do that job. The change lowers distortion, improves bandwidth, and even raises the PSRR a little.]