BeoloverRIAA Preamp: First Installation of Finalized Design in Beogram 4002

I received the redesigned PCBs for my Beogram 4002 internal RIAA pre-amplifier a couple days ago. The fact that this pre-amplifier runs from the single rail Beogram-internal power supply posed some interesting design challenges that made it necessary to go through a couple of design interactions to achieve a satisfying signal-to-noise ratio. See my previous post for more details on the design process.

But now I think this design is ready for primetime. I populated one of the boards and implanted it into  a first Beogram 4002. Here are a few impressions. This shows the redesigned board with its 3D printed mounting support:

The board grew a bit in size to accommodate my improved power supply design.

This shows the mounting support on the pins of the board:

The pins fit directly into the solder points of the original output relay of the Beogram 4002. In fact, the RIAA preamp board maintains the full functionality of the non-amplified standard output of the Beogram. That is the reason that there are two relays on the board. One for the amplified output and one performing the function of the original output relay

This means the user can easily switch the deck between RIAA and standard output by simply plugging the connectors either into the amplifier board or into the original connectors on the output board (see below).

This shows the board installed on PCB 8. I had to move the grounding switch that I had initially installed to a different location underneath the original input connector:

This shows PCB 8 installed back in the Beogram 4002:

The input and output connectors are shown inserted into the amplifier board, i.e. this is the configuration for amplified output. The original non-amplified output can be configured by inserting the connectors into the original board-to-wire connectors on PCB 8:

Here an impression from the front of the deck:

After reassembling the Beogram I connected the output cable to my QA400 audio analyzer and measured signal-to-noise curves:

Nothing changed from the prototype as expected. The red curve shows the noise floor with the Beogram started and the arm down next to the platter (i.e. not touching the record). The noise floor at 1kHz is still about -100dB.  That gives the baseline for all further noise considerations. The large noise dB numbers stated for many available RIAA preamps are measured with the inputs shortened to ground, which is pretty much meaningless since the main noise source other than from the vinyl itself is the cartridge. Due to its own impedance (and the necessary 47k input resistance on the amp) the generated Johnson noise vastly exceeds the input noise of modern low-noise opamps such as the LM833 used here. 

The blue curve illustrates that the noise generated by the vinyl surface is yet another factor 3 or more higher. It was measured while playing the 'rumble track' on my Analogue Test LP. I used a good condition MMC20EN for these measurements. It is obvious that the RIAA amp performance is such that the noise while playing a record is not increased significantly. 

This beolovely performance inspired me to celebrate my latest addition to the 'Beolover parts catalog' with one of my favorite 45s of my youth (don't judge me by this!...;-), 'Dance Hall Days' by Wang Chung:

Allright, looks like this Beogram is ready to be shipped back to its owner!