I don't have any schematics, but I did look inside one once (Rob, any chance they'd let you have a look inside one of the ones you saw recently???) I remember green St.Ives transformers and board components and layout similar to the Neve version. That was before I really began to learn about Neve stuff, though, so I don't know that much about the PAST design.
Without truly knowing otherwise, my guess is that Steve probably used the 31105 design as the basis for this unit. At the time PAST started (1986, I think) the Neve 1081 and 1073 craze hadn't yet fully begun, though some engineers were clearly beginning to appreciate the older Neve designs. The 31105 is the most 'refined' of the 4-band channel amplifiers, so it would make sense that this, coupled with the fact that it is also the most feature-packed discrete Neve module, would make it a good candidate for cloning at the time. Also, the knobs on the PAST unit are the same black and blue plastic types used on the 31105.
If this is true, then it wouldn't have the LO2567-type output transformer of the 1081 (basically an ungapped LO1166 wired for +10dB step-up), but instead the TO129 tertiary feedback output transformer (+8dB step-up). Modules with the TO129 is a bit more persnickity to set up properly. Here's a great post from Geoff Tanner that describes it:
"The next 32 series line amplifier(3405)used a 340 amplifier to drive a
LO1173 that had 8dB step up and internally mounted damping resistors.
I have heard very naughty claims that the 3405 will sound like the
1081 when wired as a mic pre.... not so, the circuitry is quite
different even if the amplifiers are the same. The really obvious
difference being that the 1081 drives a chunky LO2567 transformer with
10dB step up compared to the weeny transformer in the 3405.
With the introduction of the 3415 line amplifier the new TO129
transformer was introduced. This was identical to the LO1173, but had
a 1:1 (relative to the input) tertiary feedback winding added to it to
include the distortion in the transformer path with the negative
feedback from the driver amplifier.
Now here I am going to stick my neck out and suggest that Neve got the
circuit diagram wrong for this transformer application... and I'll
explain why....
Any Neve module with an output transformer has an issue with the
capacitance of the cable tied to the output, especially in the cable
is going to a distant remote patchbay or tape machine. The capacitance
of the cable resonates with the leakage inductance of the transformer
and causes a high frequency hump... You know, that famous one Rupert
speaks of at 45KHz that Geoff Emerick could hear! You can check this
either by sweeping the consone or by simply sticking a square wave
though the beast and watching by how much it rings... it shouldn't be
more than a tiny blip... around 5% max overshoot.
You fix this by loading the path with a resistor to dampen the effect,
indeed you can tune the circuit to flat at 20KHz and very gentle roll
off beyond this point by selecting the best value for this resistor...
say around 3Kohm max and 600 ohm minimum.
Having explained the nastyness of having capacitance hanging on a
tranformer coupled output... look what Neve did with a TO129... they
hung around 2 nanofarad directly across secondary, rather than the R/C
network on every other Neve transformer (the resistor on these dampens
the effect of the C).
It would have been a far smarter thing to have changed the value of
that capacitor.... after all, it's total value and effect depends on
the capacitance of the load applied to the module output, and to fit
an internal damping resistor... as done on the 3405.
On the topic of TO129, all those transformers were made by Marinair
Radar and I know for a fact that "latter day copiers" have major grief
reproducing it because of the difficulty in squeezing that extra
winding onto an already crowded small bobbin. I have heard horror
stories on this matter and personally don't know why anyone should be
bothering to recreate it. I'd leave the feedback off and wire the
module like a 3405... noting that the gain setting resistors need
changing to compensate for the lack of extra feedback.
This only applies to copiers... if you have a regular Neve console
with modules fitted with a TO129, just sweep the output and adjust for
best response. " -Geoff Tanner
Rob, perhaps you can use this info to sneak a peek inside these, you know, to make sure they have them set-up properly :wink: .
The only other consequencial differences between the 31105 design and the 1081 that I am aware of is the use of different inductors in the mid bands, same type, just different values, I believe. Oh yeah, and BA4xx modules throughout instead of BA3xx blocks.
Rob, if you can have a look inside one I'd be interested to know if my assumptions are accurate or not.
Peace,
JC