CE PS-1 & PS-2 theory

[K6JEK]

I'm trying to understand the "differential phase shift networks" in the Central Electronics equipment.   They are the PS-1 in the early gear, the PS-2 in the 100V.   They are resistor/capacitor networks that take in 0˚ and 180˚ degree audio signals and put out 0˚ and 90˚.  Except in the sideband slicer where the PS-1 is used differently. 

I've been content to to treat them as black boxes but I want to really understand the slicer so I think I better figure out the PS-1. 

Can any of you point me to a reference that discusses beasts like this?

Thanks,

Jon, K6JEK

[Steve - WB3HUZ]

I'm pretty sure it's covered in some of the older handbooks and maybe even some of the more recent ones too. I think they show the schematic of the venerable B&W 2Q4. By differential they mean each leg of the network shifts the signal by 45 degrees, for a total of 90.

[Pete, WA2CWA]

You might try:
QST Jan. 1950, page 42, Audio Phase-Shift Networks
Ham Radio, June 1970, page 34, Design Criteria for SSB Phase-Shift Networks

[K6JEK]

Thanks guys.   These are great leads.   I have figured out how the slicer uses the PS-1 but really understanding the phase shift network is worth doing.

Jon

[Bacon, WA3WDR]

A good reference is R. B. Dome, "Wideband phase shift networks," Electronics, vol. 19, pp. 112-115; December, 1946.  I saw a copy of this once, long ago, in W3PHL Fred's library.

I don't recall if it covers the passive PSNs widely used in amateur SSB gear at one time, but it describes some theory.

A basic technique was to create an all-pass network with phase shift that changes with frequency.  Then create another such network, but start its phase shift either higher or lower in frequency.  The difference between the outputs of these two filters can have all-pass flat response, and 90 degree relative phase difference over a wide frequency range.  Dome was talking about something like 80 to 5500 Hz with the analog tube circuitry of the day, if I recall correctly.  With more phase shift sections in the two filters, the range can be wider.

[Steve - WB3HUZ]

All-pass filter designs using 1% components can yield the desired phase shift tolerance over a range of at least 10 kHz, maybe more. Pete SOV and Chuck FJJ have built some around op-amps for sync detectors.

[Pete, WA2CWA]

I was going through some Central Electronics paperwork and came across some of the tri-fold brochures that dealers handed out to customers. Here's front and back for the Sideband Slicer.

[Steve - WB3HUZ]

Does that thing slice cheese. I got my crackers ready!

[w3jn]

That slicer, by the way, really has outstanding opposite sideband rejection.  I've messed with a bunch of outside SSB adapters and the CE is by far the best.  Ashtabula Bill cannot be wrong!