Phone scrambler of 1966

Someone's got to reproduce these 1966 "phone scramblers" in all their silky, chunky plasticky glory, leaving all but one tiny corner hollow, filling that tiny corner with a modern crypto device.

This scrambler keeps private phone conversations safe from wiretappers and eavesdroppers. Fitted to an ordinary handset, it needs no electrical connection, has its own power source. To hear, a person needs an unscrambler coded identically. Delcon Division, Hewlett-Packard Co., Palo Alto, Calif., sells it for $275, keeps your name and code locked in its vault

Wiretap-proof telephone (Jan, 1966)

• [Oren Beck]

  See THIS:

  http://zfoneproject.com/

  Few thing move a project from Beta to daily use quicker than a large pool of early adopters.

• [Anonymous]

  “But wait! That’s what you wore yesterday!”

• [igpajo]

  Leila: Is it you? This is Leila. Are you using a scrambler?
  Parnell: I can’t hear you. I’m using a scrambler.

• [jahknow]

  “As you may recall, sir, one of the provisions of Plan ‘R’ provides that once the go-code is received, the normal SSB Radios on the aircraft are switched into a specially coded device which I believe is designated as CRM-114. Now, in order to prevent the enemy from issuing fake or confusing orders, CRM-114 is designed not to receive at all. Unless the message is the correct three-letter recall code prefix.”

• [prodpoke]

  coke dealers need that

• [Anonymous]

  If they kept your “name and code” in their vault, then you weren’t safe from wiretapping by law enforcement – or the company itself. Or even an unscrupulous employee.

• [vamidus]

  It sounds like a great idea – have a little encrypting gadget latch on to the receiver. The moth piece would encrypt the sounds coming out of your mouth and send them as a series of notes, a lot like a modem. The earpiece would decrypt sounds coming from the receiver and play them in your ear.

  There’s a little technical difficulty – the sound of your voice will still reach the receiver’s microphone and the eavesdropping party will be able to hear your unencrypted speech.

  The only way I see to solve this little problem is to have this “gadget” completely replace the receiver and have it jack directly into the receiver port on the phone.

  Do we have a CPU powerful enough to do a on-the-fly encryption of a binary stream (audio) with no or minimal delay? Small enough to fit in your pocket? Efficient enough to run off a cellphone battery?

• [GregLondon]

  Do we have a CPU powerful enough to do a on-the-fly encryption of a binary stream (audio) with no or minimal delay? Small enough to fit in your pocket? Efficient enough to run off a cellphone battery?

  Depends on what kind of encryption you’re doing. Public Key encryption is processor intensive. But you might use that to exchange AES keys or something, which isn’t quite so processor intensive and is something you could implement in hardware.

  An alternative is to use one-time-pads. Here’s a simple schematic for a truly-random noise generator. You could use that to generate a digital one-time-pad.

  The limitation is that you would have to meet with the person you want to talk to before you talk, and exchange one-time-pads. Voice-over-phone is pretty low bandwidth, and with a bit of voice compression, a simple 8 gig secure digital card would hold enough one-time-pad data to encrypt days worth of phone conversation.

  The whole thing could be implemented with a small, embedded processor, a USB interface chip, and that random-noise-generator.

  http://www.maxim-ic.com/appnotes.cfm/appnote_number/3469

• [Teapunk]

  I really want her glasses. Really. They are deliciously über-geeky. Want.

• [neurolux]

  It looks like something that really would cause brain tumors.

• [slappin]

  i cannot wait for the papercraft phone scrambler post!

• [Anonymous]

  reinvented and build some similar thing. using time transposition (so, there’s a slight signal delay…) implemented on a cheap avr controller.
  http://www.kielnet.net/home/julien.thomas/tech/HEKTOR_en.htm

• [AliasUndercover]

  I wonder if it had tubes…

• [bfarn]

  I’m just curious as to why Dana Carvey would be advertising encryption gear…

  http://www.sethbarnes.com/blogphotos/sethbarnes/www/church_lady.jpg

• [Modusoperandi]

  jahknow;

  “So let’s get going, there’s no other choice. God willing, we will prevail, in peace and freedom from fear, and in true health, through the purity and essence of our natural… fluids. God bless you all.”

  …although he might have sent that in the clear.

• [Belinda]

  I like it very much, I think I could use it :)

• [vamidus]

  The whole thing could be implemented with a small, embedded processor, a USB interface chip, and that random-noise-generator.

  ..we could use radio and/or phone line static as seed for the random-noise-generator :)

• [glenn]

  IMHO, any “modern crypto device” more complex than ROT13 would be a waste of time: your unencrypted voice will leak through the plastic into the phone.

• [jtegnell]

  I’d like to get one of these fitted for my Motorola Krave. That would be awesome!

• [madsci]

  I had a pair of these… picked them up at a ham radio swapmeet or something, I think. They were fun for about 20 minutes, but the volume was rather low and they had an annoying buzz.

  I’m pretty sure they were just sideband inversion, which is trivial to descramble today.

  Also, they don’t fit well over most modern phones. My dad was a telephone repairman, though, and we had no shortage of old phones in the house. We had at least 8 active phones in the house and barn (technically one of them was a central office test board console, and the one in the garage was a payphone) and many more on the wall for display.

  Oh, and we had field telephones – I ran one up to the top of the pine tree and had the other in the barn. Never thought to try the scramblers on those.

  I wish I knew what had happened to them. I can picture the metal case they came in, and I know where they were in the barn (in a pile of old pinball machine parts) about 15 years ago, but I can’t recall having seen them in the last two moves.

• [madsci]

  #5 – Look at the picture. There are foam pads on the mating surface, and that handle lets you (in theory) pull them tight against the receiver. In practice it made your hand pretty tired keeping it tight. Notice that she doesn’t have a good seal in the picture.

  Also, processing power is no problem. I can sample audio at 19.2 kHz and encrypt it using XXTEA on a $2.50 8-bit processor. The delay would come more from the fact that you’re encrypting in blocks, and your transmission is in frames with synchronization information and forward error correction. And you need compression – a 33.6k modem (can’t do 56k peer-to-peer) gives you about 3.3 kbytes/second, which is not enough to reproduce decent quality voice uncompressed, even without FEC overhead. Compression is done in blocks as well.

  It’s been a long time since I used a STU-III, but I don’t recall the delay being particularly bad, and that’s not exactly cutting-edge technology. So yes, it can certainly be done, and the delay is more a result of the other processing you’ve got to do for a digital signal than the encryption itself.

• [Oren Beck]

  I defer to an authority on this subject:

  http://www.philzimmermann.com/EN/essays/WhyIWrotePGP.html

• [zeroy]

  Caption: “What are you wearing”?

• [BADO]

  Not bed phone))

• [GregLondon]

  your unencrypted voice will leak through the plastic into the phone.

  Ideally, the new and improved version would be in teh form of a bluetooth headset. It would encrypt the voice data before transmitting it to your cellphone via bluetooth RF.

  Obviously, if you’ve got listening devices in your home, then you’re pooched. But this would at least stop Patriot Act style sweeping and snooping from remote.

• [reginald]

  the lot
  but no pineapple
  and
  extra
  garlic
  oregano
  and …
  chilli