Quite frankly, it's unlikely there's any reason for me to pick up hardware like
the Enigma. I'd probably not recognize one if it bit me in the leg. Unless I
have reason to take a board home, I leave it in the junkyard. I've walked right
by large (DEC, Unisys, CDC, CRAY, DENELCOR, etc, hardware without so much as a
look. Likewise, the SGI, HP, SUN, etc, stuff is safe from me. I also generally
don't even bend over for a MAC. It's the old Multibus-1 and S-100 stuff, which
I hardly ever see around here, that interests me, and I very seldom go looking
for classic hardware, except in my basement. Sometimes I even find what I want.
If I see something YOU might want, I leave it where it lies for you to find it.
Dick
----- Original Message -----
From: "Tony Duell" <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Wednesday, June 13, 2001 5:39 PM
Subject: Re: Ebay horror ...
> >
> > +AD4-tony asked:
> > +AD4-Do you have any clue as to how an Enigma works and what it consists of?
> > +AD4-Simple rotary switches indeed?????
> >
> >
> > Of course, like most of my answers to this list it
> > will likely be corrected later on, here's a quick
> > answer:
>
> My original question was aimed at Richard.... Anyone who even considers
> scrapping an Enigma should be LARTed. And anyone who thinks you can get
> 'simple rotary switches' from one doesn't really understand how the
> machine was built.
>
> > Keyboard (1 letter press +AD0- 1 signal line)
> > goes to...
> > -+AD4- Arbitrary plugboard (''fixed'' rotor)
> > -+AD4- Rotor: N signal points in,
> > N out in scrambled order,
> > angle dependant
> > -+AD4- Rotor, with diff. scrambling AND angle
> > -+AD4- Rotor, again
> > -+AD4- Light array (1 signal line +AD0- 1 Illuminable Letter)
> >
> > The rotor as you may guess increments with each
> > keypress. Later improvements included a fourth
> > rotor and a reflection arrangement which apparently
> > allowed each rotor to pass the signal both to its
> > right AND to its left without ever having crossed
> > wires (neat). It would reach the far wall and then
>
> Close. AFAIK, the reflector came in pretty early on.
>
> I've once used an Enigma machine (at Bletchley Park, of course). I wasn't
> able take said machine apart, of course, but I did manage to deduce some
> bits of how I think they were built.
>
> Firstly let me assume there are 26 keys and light bulbs. That might be
> wrong, but it's a working figure.
>
> Each key is a chageover switch. The normally open contact of all the keys
> goes to one side of the battery. The normally closed contact to one side
> of a lamp. The other side of all the lamps goes to all the batteries.
>
> What this means (of course) is that if the common contacts of 2 of the
> keys are linked, pressing one key lights the bulb associated with the
> other and vice versa. This means the same setting can be used to encryt
> and decrypt a message -- it's self inverse.
>
> These connections between the common contacts of the keys are made by the
> rotors. There are 26 contacts in a circle fixed to the chassis of the
> machine (one contact to each key common contact, of course). The first
> rotor has 26 contacts on one side that make contact with the ones on the
> chasiss. These are wired (essentially at random although all machines
> were, of course, the same) to 26 contacts on the the other side of the
> first rotor. This is repeated for the other 2 rotors (with different
> wiring patterns). Finally, the last side of the third rotor makes contact
> with the reflector. This has 26 contacts on it which are wired together
> in 13 pairs.
>
> When a key is pressed a current (from the battery) flows :
>
> Through the N/O contact of that key to one of the contacts on the
> chassis.
>
> From there through a wire on the 1st rotor to a contact on the other side
> of that 1st rotor
>
> Ditto for the 2nd and 3rd rotors
>
> Then though one contact on the reflector, along one of the 13 wires in
> the reflector to another contact on the reflector.
>
> Then back along a different wire in the 3rd rotor
>
> Ditto along different wires in 2nd and 1st rotors
>
> To a contact on the chassis (which must be different from the first
> contact I mentioned)
>
> Then to the common terminal of another key (which is not pressed)
>
> To the N/C contact of the that key to a lamp
>
> And back to the battery.
>
> As you said, there's a mechanical linkage to step one of the rotors on
> one position after each key press. And a 'carry' linkage that moves the
> next rotor on one step after the first rotor has gone round once. And
> when the seocnd rotor had gone round once, the third rotor was advanced
> one position.
>
> Ah yes, the plugboard.
>
> It has 26 2 pin sockets, each polarised (different size pins IIRC). One
> size of pin goes to the N/C contact of the corresponding keyboard switch,
> the other plugboard pin goes to the lamp. There's a contact inside each
> socket that shorts the socket pins together if there's no patch lead
> inserted, The patch leads had a 2 pin plug on each end, with the
> connections crossed over (large pin at one end to small pin at the other).
> Inserting a patch lead between 2 of the sockets effectively crossed over
> the external connections associated with those 2 sockets
>
> There are 2 places that the patchboard might have been connected (I don't
> know which is correct). One is between the common contacts of the
> keyswitches and the chassis contacts for the rotors. The other is between
> the N/C contacts of the keys and the corresponding lamps.
>
> The latter position would allow the Enigma to encrypt a letter as itself
> (that's impossible without the plugboard since as soon as a key is
> pressed, the corresponding lamp is out of circuit). With the plugboard
> between the keys and the lamps it would be possible for the encrypted
> form of a character to go via the plugboard to the lamp for that
> (unencrypted) character.
>
> Since I believe that the fact that a letter was never encrypted as itself
> was one of the weaknesses that was used to crack Enigma, maybe the
> plugboard was in the the first position I mentioned.
>
>
> > be sent back eventually to the rotor origin. A wire
> > cross anywhere would cause non unique mappings, IE 2
> > letters out at once, which was designed out.
>
> Actually, wired as I described, no matter how the rotors are wired, and
> how the reflector is wired, you must end up with a 1-1 self-inverse
> mapping between keys and lamps for a given rotor position.
>
> So, are my guesses anything close to correct?
>
> -tony
>
>
>
Received on Wed Jun 13 2001 - 23:18:13 BST