This review page is supported in part by the sponsors whose ad banners are displayed below
Malcolm: "I understand that you are still a bit suspicious regarding the shunt, but keep an open mind as you might be surprised. Remember there are two ways to reduce voltage here: One is to increase the series value in the signal path turning, as you said, much of the signal to heat (this works but the music ends up lifeless and boring.) The other approach is much more interesting and in keeping with the philosophy for the shunt. This is to lower the impedance of the shunt so that more unwanted voltage is shunted to ground. This allows you to use a lower-value series resistor in the signal path, preserving all the life in the music. Of course you have to present a load to your amplifier that it is comfortable to drive. So it is a bit of a balancing act. Each amplifier has its own requirements so it is impossible for a passive pre to be all things to all men. We cannot predict what it will be married with once it leaves our lab. We can only go through this balancing act with our own Bosangwhas to get something that works in combination ... in fact we are still fine-tuning the pot and series resistor values as we speak*."

This e-mail suggested that Musical Laboratory* had contacted me at the very beginning of their launch, with certain design decisions still to be finalized. I'll not dwell on possible implications beyond this mention, however. As regards the passive Paeonia, readers should also know about Michael LeFavre's Magnequest Ingot designed in collaboration with Steve Eddy. Their iteration of a minimalist preamp circuit uses a small input transformer to inductively stabilize impedance seen by the source while a 10K pot across the secondary handles passive attenuation. Another Ingot version adds a FET buffer.

* Reviews can readily sustain preexisting myths even while quoting manufacturers - in this instance, the belief about what, in a shunt circuit, constitutes the signal path. With the Paeonia, the following amplifier sees the voltage dropped across the shunt resistor's reaction to the current flow through it. It's overly simplistic to view the signal path as being merely the series resistor between input and output.

Now that we have covered the design philosophy, related precursors and alternate options, it's time to take a closer look at Musical Laboratory*'s actual goods with closeups and candid shots. And to know what we're looking at, Malcolm provided these descriptive additions: "The Paeonia Revision 2 changes include some mechanical and construction improvements; low-impedance potentiometers; Sfernice P11s configured in a shunt/Fake Law arrangement*; hand-made next-generation low noise audio resistors in the signal path. The Paeonia Rev 2 changes allow for better tracking at low listening levels and improved channel balance for more detail."


Not understanding this particular description, I requested an explanation which Malcolm duly furnished: "What we have done now is to combine the shunt with linear rather than logarithmic pots. With a linear style it is possible to get better tracking and channel balance -- and therefore stereo imaging -- but of course the ear responds best to a logarithmic volume curve. So we solve this dilemma by converting a linear response to a logarithmic curve by adding a Fake Law resistor to ground at between 12-15% of the impedance of the pot. This way you get better tracking and balance with a response curve which is pleasing to the ear yet allied to the clarity and detail possible from keeping the pots away from the direct signal path. The Sfernice P11 is a cermet pot which has a small but avid following in DIY circles."

"The Bosanghwa Revision 2.2 changes include: Mechanical/construction improvements; new high-speed PCB substrate selected from a number of exotic substrate materials we tried; improved specification of copper tracing on the board; improved/simplified circuit layout further reducing resistance; larger ground plane; single ground connection achieved; use of hand-made next-generation low noise audio resistor in the signal path; DC offset is limited to within 2mV, with DC unity at around 10Hz -3dB (small DC offsets not addressed can be amplified x 30 by the gain stage into the 0.75V - 2V range. This causes the circuit to heat up, reduces output power and places a constant load on the speakers whether or not music is playing. Looking at the photos from your review, it is not clear if this was actually addressed in the Gaincard circuit.) Use of silver gold conductor foils for both input and output; Nylon 6.6 speaker connections with PTFE washers."

The music box appellation truly fits these unique components. I've not encountered anything even remotely similar in all my hifi years. And what of those mirrors inside the monos? Their power supply is more pedestrian in appearance by far. Worse, the short trip from Luxemburg to Switzerland had severed the rivets holding down the internal assembly which now rattled and clanged. Fortunately, electrical connections hadn't been - um, impacted. Malcolm took the news in stride: "Yes, we used to use M4 bolts on the bottom assembly but switched to rivets in Rev.2 because of the lower profile they offer. Okay, we'll revert that part of the design." Because the extrusion caps are riveted as well, I cannot present shots of the power supply innards.

Need I add that my wife hinted rather insistently how great this kit would look in her upstairs artist studio with her metallic purple Zu Audio Desktop Druids? Really. And there I thought the family purse had been cleanly divided into audio (his) and jewelry (hers) a long time ago. Me thinks our household is suffering a sudden attack of gender reversal...