This project started out with my building a version of the VSPS (Very Simple Phono Stage). It is a very simple single opamp project with RIAA equalization in the opamp's feedback loop. The design is fine, but I found the sound to be a bit less than I had been hoping for. However, I don't listen to enough vinyl to spend an inordinate amount of money on a phono stage, so I needed something that would not cost a lot more.

The solution I decided upon was to simply try to get as much out of the VSPS circuit as I could. I had already built the circuit out of pretty high quality parts (OPA627 opamps, Riken Ohm and Draloric resistors, silver mica caps, Solen ouput caps), so it seemed that circuit design, and not parts, was what the problem was. There are a number of tweaks that can help any single opamp circuit, I more or less simply copied what people have been doing to Chu Moy's single opamp headphone amplifier.

The first tweak I wanted to try was to add a buffer inside the opamp's feedback loop. For this, I used a BUF634 buffer as I happened to have them around. This serves several purposes. Primarilly it relieves stress from the opamp allowing it to work a little better. Second, it provides current and thus helps drive the interconects. Both of these are important, but for this circuit, the thing it really does is allow the opamps to be biased into class A. In order to do this, one simply adds a constand current drain to the opamp from the output to the negative rail. However, this trick only works of the opamp sees a sufficienty high impedence load, which is what the buffer presents.

There are several methods to add a small constant current drain including JFets and voltage regulators, but I went with the easy and small CRD as, again, I had some on hand.

The last circuit tweaks were to bypass the output coupling caps with small Vitamin Q caps and too a add small high quality film and foil decoupling caps close to the power supply pins of the opamps and buffers. There is a lot of disagreement about whether bypassing output caps is a good thing or not. I am generally agnostic in the debate as I have found it to work well in some places and not so well in others. It just so happens that it works well here. Whether this is becaus the Vitamin Q is much higher in quality than the Solen or something else, I don't know. I would guess that using a very high quality single cap would be the best option here, but since that can get very expensive, I went with the second best.

Thus, with all of those changes to the circuit, I ended up with the following (the component values not on the schematic are available on the VSPS web site):

The next step was to work on the power supply. There were two primary considerations. The first was to make the supply (at least marginally) dual mono. I have found that this is one of the most important considerations in the quality of a circuit. The second was that, due to the class A nature of the circuit, I needed more capacitance on the rails as otherwise the phono stage was a bit bass shy. To do this, I used a pair of LM317 based power supplies to provide +/-15V. I then used a second level of regulation based on a pair of LM7812's and a pair of LM7912's for the dual +/-12V supplies.

For caps, I really just kept adding more and more until it sounded how I like. Thus, this is worth experimenting with.

There are still a few things worth doing. Jim Hagerman (of HagTech) has suggested removing the buffer from the feedback loop and instead placing it after. I have not gotten around to this yet, but will try it one of these days. As it stands, the phono stage sounds good enough that it does not seem like the limiting factor for my in vinyl reproduction. Last, for what it's worth, the case is made from some leftover sheets of aluminum I had around, and a couple pieces of walnut for the sides.