Hey, thanks for the nice feedback. Note that there is/was only one version of the Sapphire 4. There have been a couple of minor changes to the board layout for different size caps and so forth, but the circuit is unchanged.

Kits and boards are available for $150/$25 US, respectively, including shipping. (link) Updates, feedback, questions, and discussion at diyaudio (thread) Current board revision is 4.2b, though the circuit has remained unchanged since the Sapphire 4 was first introduced a few years back. It's...

It's more likely you'll find someone to answer that over at the development thread on diyaudio. Current revision is 4.1t, photos are of rev 4.1m, cosmetic changes only.

The open loop and closed loop configurations sound slightly different, open loop having more 2nd harmonic and therefore more 'bloom'. From someone else who has built one (and with whom I am in 100% agreement): "natural, clean, airy, and revealing recording space like nothing else I've heard."...

I like to think of it as a diamond buffer input, current mirror "pumps" for voltage gain, and a Sziklai pair output buffer.

A new version has been developed, the Sapphire 4. It's an all-discrete transistor current feedback amplifier. The board features switchable gains, and can be configured with the buffer either open loop or closed loop. Please see the web page for more details. http://phonoclone.com/diy-sap4.html

They are film type, the orientation doesn't matter. You just replace the four original ones with new ones in the same position. I don't claim it made any difference, but it doesn't hurt. The more time I spend with the stx the more I like it. Both the output and inputs are very low noise and...

Sorry for the long delay getting back to this.   I think you nailed the sound perfectly. Dirty but enjoyable, weighted to the lower midrange / midbass.   Yesterday I switched the output op amp briefly back to the LM4562 (DAC I/V still 5532) and confirmed that the balance does indeed shift up...

The output DC offset voltage is application dependent. It depends on the gain, the input offset voltage, input offset current, input bias current, and the impedances seen by the IN+ and IN- inputs.     (And shouldn't you be the one telling me this, and not the other way around? This whole...

  @Burson Audio   I was referring to the input bias currents and the resulting, possibly high, output voltage offset. This has nothing to do with the power consumption or the current draw of the package.   Please measure the output offset voltages of the op amps when used in the STX. They...

@Burson Audio With input offset bias currents for the SS V5 module in the hundreds of microamps (compared to nanopamps for the NE5532) I'm a little concerned about the output offset voltages. From your published datasheet and the PCM1792A example circuit I estimate the offset might reach as...

Yes, those should do.   By the way, about the NE5532/4, it's worth noting that these are recommended by default in the datasheet for the IV stage as being guaranteed to reproduce the datasheet performance figures for the DAC. Don't let the low price put you off, they are very good for general...

Sure, let me give it a shot.   Here's the stock board with the cover off,     The three socketed 8 pin op amps for the I-V stage and output stage are visible top left. Zooming in,     The I-V stage is to the right, two JRC2114D shown. The output differential-single converter / low...

Greetings. Today I modified my STX by upgrading the four 2700 pF caps in the LP filter to WIMA FPK and, while I was in there, changing out the LM4562 and 2114D for 3x NE5532A.     To my ears the TI NE5532A sounds as a good as any op amp for these kinds of low impedance, low gain...

The next release version, Sapphire 3.0f, has current sources and paralleled output transistors, as well an optional hi/lo gain switch. The boards are drop-in replacements for the previous version.      

Well, some water flowed under the bridge and we got this circuit,   which gives the following board layout (one channel),

It comes with the card, but is not for sale separately. Onkyo might have spares but this is unlikely.   It's a standard DB15 (VGA-type) connector, so you could perhaps wire something up from a modified VGA cable if you had the pinouts.

The circuit below started off life as a pretty standard diamond buffer, and I'm planning next on building this modified version:   It sims out nicely in LTSpice: low distortion, good PSRR, and (as far as I can tell) reasonable bias stability towards voltage and temperature changes.   I'm...

Pro Tip: open the X-Fi console and untick all the DSP which is enabled by default. Untick the equalizer, EAX, 3D, and the "crystallizer". (I can't believe they left this all on by default .. groan.)   Also, select 32/192 in the Windows audio device properties/setup. Finally, just to be on the...

I've been doing some research reading the Japanese reviews of the card.   Large majority of users seem very happy with the sound, but a small number of people with experience with the previous version consistently note the change from the 200PCI in the same generally unfavorable light. Namely...

http://phonoclone.com/diy-sapp.html   I've finished updating the web page to the current ver. 2.0 of the Sapphire circuit, and the new boards are due in momentarily.     (two-layer board, bottom trace shown)       The main change is to the diamond buffer, with new bias resistors R9...