3
\$\begingroup\$

From https://www.cryptomuseum.com/crypto/rs/pu104/index.htm :-

PU-104 was a key generator for perforated paper tape, made around 1986 by Rohde & Schwarz (R&S) in München (Munich, Germany). The device was able to create long sequences of truly random numbers and send them to external paper tape puncher via its built-in V.24 interface.

I'm trying to understand how random noise was sampled by the CPU. I don't think that they used an dedicated analogue to digital converter, but rather went directly from analogue to digital logic. I can't follow what happens after the noise enters the purple square in the following schematic extract:-

noise

The extract is from the complete user manual at https://www.cryptomuseum.com/crypto/rs/pu104/files/pu104s2_manual.pdf. See pages 24-27 of the PDF file. The 'bottom' board photograph clearly shows two PM156 op-amps which I cannot locate on the schematics. They're not of a style I'm accustomed to.

It don't know why there are two schematics.

\$\endgroup\$
  • 2
    \$\begingroup\$ "directly from analogue to digital logic" <-- that is an ADC, just a 1-bit ADC. (don't laugh, these exist, even as separate components.) \$\endgroup\$ – Marcus Müller Mar 10 '19 at 15:31
  • \$\begingroup\$ page 29 defines D21–D24 to be SN75150AP line drivers, if that helps. An RS-232 line driver, to be exact. \$\endgroup\$ – Marcus Müller Mar 10 '19 at 15:36
  • \$\begingroup\$ V for transistors and diodes, that's interesting. Never seen that before. Probably short for a German word, since it's a German company. \$\endgroup\$ – Hearth Mar 10 '19 at 16:19
  • \$\begingroup\$ @Hearth - the Vx naming convention is a holdover from the days of vacuum tubes (valves) which were used in equivalent transistor and diode applications. \$\endgroup\$ – Peter Smith Mar 10 '19 at 16:22
  • \$\begingroup\$ @PeterSmith Ooh, that makes sense. I didn't realize the designator got carried over to semiconductor devices. \$\endgroup\$ – Hearth Mar 10 '19 at 16:23
5
\$\begingroup\$

This doesn't look like a noise generating circuit.

According to the BOM on page 29, D21 to D24 are SN75150AP, so RS-232 line drivers, which need a dual +-12V supply. That's exactly what V1 and V2 are doing: being rough, regulated power supply. I'm not sure, but V3 and V4 being BZX79-C13 have a ~ 13V Zener voltage – these might simply be protection diodes.

Generally, your PDF is the manual and service manual to the V.24 serial interface card, not to the board that generates the randomness.

\$\endgroup\$
  • \$\begingroup\$ I see what you mean. That transistor/diode arrangement is the same as for producing Zener avalanche noise, which I've used and see all over the Interweb. I didn't expect R23/24 to be so low though. 39K perhaps, but not 390 Ohms so perhaps you're right. The implication is that the entire RNG circuit is missing from the PDF. \$\endgroup\$ – Paul Uszak Mar 10 '19 at 16:51
  • 2
    \$\begingroup\$ as said, you're barking up the wrong PDF! You've got the service manual for the V.24 board, not for the RNG board. \$\endgroup\$ – Marcus Müller Mar 10 '19 at 16:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.