Xpre preamplifier with buffer and UGS-6 module as a DIY project according to Nelson Pass

In front of my beloved stacked quads, a DIY chain works according to the concept of the Pass’s amplifier. And now a buffer?
Nelson Pass and Wayne Colburne are American contemporaries who have made history with ingeniously sounding hi-fi components and are enjoying a growing fan base.
A circuit concept based on the XP20 (american patent!) Is released for the replica of the “do it yourself” (private use only), and such a preamplifier with the Muses74230 component has been working to my full satisfaction for several years.

Volumecontroller with Muses74230


And then someone comes from the fan base and offers the opportunity to “drill out” this really good sounding preamp ….
How??
Behind the UGS6 module – an ingenious supersymmetric circuit – a buffer is used that enables pure Class A operation, i.e. a high-current concept.

PLEASE DO NOT INQUIRE ABOUT WIRING DIAGRAMS AND / OR BOARDS.
THE CONCEPTS ARE PROTECTED BY COPYRIGHT AND ARE SUBJECT TO THE COPYRIGHT OF THE INVENTORS.
ONLY THE REPRODUCTION FOR PRIVATE PURPOSES IS ACCEPTED.
THE COMMERCIAL SUCCESS OF THE XP30 MUST NOT BE HAZARDED BY UNAUTHORIZED DISTRIBUTION.

The conversion on the previous modular motherboard was not possible 1: 1.
The mainboard had to be “knitted” again because the SE amplifiers had to make room for the buffer.
The UGS6 module, the Muses-based volume controller and in principle the controller could continue to be used.

The controller module has also been changed and replaced by a faster and more readable display (new heaven display).

The housing construction has to wait a little longer, because a larger CNC milling machine is on the way.

So I was excited about the rebuild, which currently still functions as a “board construction” ……..

XP30 Clone

Power Supply DC Filter and snubber-Netzwerk (left)

Controllerboard, SMD Technik, Controller Chip is a fast STM32L431CC …..

…Controller – a lot of technic and the “brain”

Display is a NewHaven LCD-OLED Technik

The mainboard as a component grave;
ULN2803 as Darling tone driver, PCF as I2C bus controller in front of the relay circuit, BDW93c as Darlington replacement behind the cap multiplier, TL431 as constant current source, polypropylene capacitors, clarity-cap coupling capacitor between volume controller and UGS module, higher-quality film capacitors as bypass to electrolytic capacitors, …. ..

….the Mainboard

All sound-relevant assemblies are manufactured as single boards and can be removed using a jumper. This opens the way for further adjustments, versions, optimizations, ……

UGS6 Modul:

UGS6 Buffer
….audiophile heart


The MOSFet’s are matched (ZVN2110 and ZVP2110). I wrote down the Vgs voltages by hand.
But: I made a big mistake well-meaning. And that should cost a lot of time and measurements ……..

…a bad ZVN Type, measurment after unsoldering

The 2SJ74 and 2SK170 as input transistors. Both belong to the species dying out !! Good if you still have some of it in your drawer.

The buffer module:
Coming from the XP12, XP22, the signal is transmitted potential-free via a special optocoupler and “passed through” to the power MOSFET. Here there is 630mV voltage over 27 Ohm resistance. The power loss requires generously dimensioned heat sinks.

The disaster:
I am happy to describe a “failure” when trying to match the power MOSFET in the UGS6 module …..
In the idea that a component tester (Atlas)of the better kind could do this, I measured my inventory of around 30 ZVP or ZVN and equipped the UGS modules with it.
Unfortunately, the adjustment of the offset did not work at all.
I spent a whole afternoon trying to adjust, and finally the amplifier rustled audibly during commissioning – which is not possible !!
“For fun” I used the UGS6 module from the old Preline (praise for the modular structure): and great silence, no hum, perfect sound ……
But what was the reason?

…..SMD MOSFET

I made a quick decision about the actually recommended method of “matching”.
And there was a considerable spread in the entire collection and another interesting phenomenon.
The ZVN types scattered from 7mV – 51mV voltage measurement at 5mA current. So I opened and measured a set of ZVN types from Mouser. All 10 (!!!) individual types were exactly at 18mV.
I do not name the source of the “scatter types” here.

……scattered ZVN Type

The ZVP also collected around 18mV (another Mouser order), so that I was finally sure to install types that were very close to each other.

…non scattered ZVP’s

After careful desoldering of the SMD types, the equipment was replaced and the trimmers, which had meanwhile been taken away, were replaced.
Again, I will only use trimmers from Bourns in the future.
The bulk goods from R. and C. (german distributors) are simply too widely dispersed for these demanding applications and extremely mechanically sensitive!
In the end, the UGS6 module could be adjusted to exact offset values in warm operation within 5 minutes.

And now a few words about the sound:
it is not for nothing that Pass uses the buffer not only in the Preline’s of the modern series.
The sound is increased again.
Without being too flowery to describe, the space, depth of the stage and richness of detail become audible.
And the whole music scene sounds “more precise”.

Dortmund, 10.12.2019

Quellen:
diyaudio.com
rstaudio.de (thanks Ralph)
D.M. Supporting UGS 6 and famose programming
Fotos passlabs.com
German Pass Distributor Audio Components
mouser.de