Threshold PCX Crossover Info wanted..

[Daveis]

Recently, I became aware that Threshold made a single-ended 18db/octave crossover back in the late 1980's.

I'm just guessing that it sounds as good as my recent digital crossover/Benchmark DAC1 experiments.

Does anyone know the circuit for this and how it sounds?
I've heard its similar to Threshold preamps made around the same time.

Can this unit be used in a 3-way system? (Is the low-pass and high-pass frequency individually controlled)

Is it hard (info available) for changing the resistor packs?

Can it be modified to 24db/octave?

What modifications could be done to improve its performance today?

[Nelson Pass]

I'm sorry to say that I don't have the schematics for that product,
but I can recall the bulk of it from memory. The 18 dB/octave
crossovers were adjustable by replacing 16 pin dip packages
of resistors. They did not support other slopes, q values, and
were the same for both high and low pass.

[acaudio]

http://www.passlabs.com/threshold/pcx/

a review from 1989...

[Daveis]

The review liked the PCX.

Has anyone compared the XVR1 with the PCX?

Besides numerous features, eg. different slopes and unbalanced operation... does the XVR1 sound significantly better than the PCX operated with the same 18db slope and similar Q ?

[PierreG]

Schematic

http://www.diyaudio.com/forums/showt...7425#post37425

[Nelson Pass]

Cheers for that.

[Daveis]

Discrete opamps using transistors 2N5566 PN4250 A42 A92 ?

Do any of those have better modern replacements?

How does the offset adjust sequence in the schematic work?

[GRollins]

All the original parts are still available I believe, although the 2N5566 is horrendously expensive. If you want to spice the stew, you might give the ZTX450/550 or 453/553 a look for the bipolar positions. The JFET front end is more problematical. The obvious choice would be the 2SK389...except that it's been discontinued. The next best thing would be to buy a handful of 2SK170s and hand-match them. Failing that, try for the 2SK246.
The DC balance isn't all that difficult to follow. The left-most JFET in the differential sees a load of 1.5k, paralleled by a 5k pot. The second stage (some people call it a VAS, meaning Voltage Amplification Stage...but tube folks just call it a second stage--don't need no fancy terms 'round here) sees not only the AC signal across that load, but the DC offset. In fact, that DC offset is how the second stage biases...so the front end load is actually serving two purposes at once. You can take it as Graven In Stone that a bipolar transistor will have something along the lines of .65V between its base and its emitter when it turns on, so at a bare minimum, you're going to need to see that .65Vdc across the front end load.
But...
If you give it a little more bias, a bipolar will respond by delivering more current via its collector. Now you've created a see-saw between the second stage bipolar (aka Q102 to Q502) and the current source down at the bottom (aka D102 to D502). If I'm reading the scrawl correctly, that current source is delivering around 4mA, so to get the DC balance to set properly, you want to goose that PNP up top so that it, too, delivers 4mA. When those two are delivering the same current, the output will balance at the midpoint between the rails and you're good to go.

Grey

[PierreG]

Originally posted by GRollins
All the original parts are still available I believe, although the 2N5566 is horrendously expensive. Grey

http://www.factorydirect.co.uk/fdcom...riceindex.html

3.90 £ (English pounds sterling) for Qty=10
Is this a reliable source?

[Nelson Pass]

So use some 2SK170's or similar.