Your browser (Internet Explorer 6) is out of date. It has known security flaws and may not display all features of this and other websites. Learn how to update your browser.
X
Post

Safe Cracking Robot

This isn’t one of those telephone spamming machines that call you during dinner time or while you’re sleeping. This is robotic safe cracking. Two curious MIT studentswith a mysterious safe and a bit of free time built a laptop-controlled robotic fixture that opened a “manipulation proof”, high security safe in just a few hours.

My epic battle with a high security safe began about a year ago when a friend of mine, Grant Jordan, managed to get his hands on an old safe – with an unknown or long forgotten combination. It could have been filled with cool stuff like gold coins, ancient relics, or even mummified body parts. Of course, we had to get it open or we would have died of curiosity.

Fortunately, Grant had been tinkering with locks for several years and had become quite knowledgeable about the subject, but had never tried to open something this difficult. We did a bit of research and discovered that, according to the books, we were pretty much out of luck. Grant’s safe was fitted with a Sargent and Greenleaf 8400 lock. This lock is a “manipulation proof” group 1 lock.

The S&G 8400 is one of the most advanced mechanical locks ever built. It was used by the government to lock up classified documents for nearly 30 years. It cannot be manipulated by any traditional attacks used on group 2 locks, such as the techniques described in the paper “Safecracking for the computer scientist”. This paper is a great read if you have some time, but I’ll be showing you the real way computer scientists crack safes…

It is worth noting that the standard lock for classified documents has since been upgraded to an even more advanced electronic lock, so our machine is not a national security threat. I’m going to be describing our process under the assumption that the lock really is “manipulation-proof” and that the only way to open the safe is to try every possible combination.

Combination space optimization is the key. By exploiting of the mechanical tolerances of the lock and certain combination “forbidden zones”, we reduced the number of possible combinations by about an order of magnitude. Again, read the paper mentioned above for details. Grant implemented our algorithm in Java and was able to test it far before we started constructing the dialer.

We used a custom stepper motor to rotate the dialer head. The dialer head transmits torque to the dial via a piece of heavy duty surgical tubing. The stepper motor we chose has more than enough resolution to implement our algorithm, but it’s not quite as fast as it could be. Stepper motors have an extremely high “holding torque”, which is ideal in this situation since the dial must be held in place while the butterfly knob is being turned.

The head also contains an RC servo motor with a machined knob to mesh with the butterfly knob. This setup enables independent rotation of both the dial and butterfly knob. The stepper motor shaft is also connected to a high resolution optical encoder for position feedback. The encoder is mainly used to detect when safe is successfully opened. The torque required to open the safe when the correct combination is entered is much higher than the maximum torque of the stepper motor, so the encoder is programmed to report when the position error exceeds a certain threshold. Basically, the stepper motor stalls and the encoder flips out if the safe actually opens.

Instead of buying off the shelf motion controllers and hacking these together to build a complete system, I opted for an all-in-one controller. I built an Atmel microcontroller based control board to connect Grant’s laptop to the stepper motor, RC servo, optical encoder, limit switches, and an optional LCD screen. The control board connects to a laptop via USB and talks to a computer just like a serial port. I wrote the microprocessor firmware in C and used an in-circuit programmer to download code to the chip. There are about two thousand lines of code in the firmware, and that does not include any of the actual dialing algorithms. After two rounds of PCB’s and about a dozen firmware revisions, we had a fully functional dialer.

The Autodailer successfully detected the correct combination after running for about 21,000 cycles. I’m sorry to report that there were no gold coins, ancient relics, or mummified body parts inside the safe.

UPDATE: We built a new autodialer that works on Sergent and Greenleaf 8500 locks. Check it out here. I’ve also included a video of it running below:

  • [...] Safe Cracker [...]

  • [...] you spent the big bucks and got that fancy safe but if these guys can build a robot to brute-force the combination you can be there are thieves out there who can pull it off too. [Kyle Vogt] mentioned that we [...]

  • [...] want to subscribe to the RSS feed for updates on this topic.Some curious students have built a robot that mechanically cracks a electronic safes. It is worth noting that the standard lock for classified documents has since been upgraded to an [...]

  • [...] Safe Cracking Robot – “This isn’t one of those telephone spamming machines that call you during dinner time or while you’re sleeping. This is robotic safe cracking. Two curious MIT studentswith a mysterious safe and a bit of free time built a laptop-controlled robotic fixture that opened a “manipulation proof”, high security safe in just a few hours….” [...]

  • I love smart people!, rock on guys!

    Axel

    January 19, 2011

  • Hi, This is a very nice build, I am impressed!
    I have been wanting to do this exact same thing for a while using a Mark 3 Manifoil combination lock which is used extensively here in the UK and had the similar thought pattern as you guys (using an encoder to detect a win and utilising the fact that the lock has a number of forbidden zones) though unfortuantly university and the lack of a combination lock got in the way :-)

    One thing I did always wonder though was how fast you could accellerate / decellerate without causing the internals to spin / mis-align – you video shows it operating at a very high speed?

    Regards,
    - Alex

    Alex Grant

    January 20, 2011

  • i love your brain, kyle.

    lizziejtv

    January 20, 2011

  • [...] robot “brute-forces” high end lock combinations – [Link] Tags: brute-forces, lock, Robot Filed in Robots | 1 views No Comments [...]

  • This is not a new thing, it is pretty old. The big advantage is that the lock will run out in fast speed :-) )

    Ian

    January 20, 2011

  • This is a very interesting project. Of course, use brute force can in theory can crack any security measures that rely purely on numerical permutations. The main challenge for this type of mechanical hacking is to develop mechanisms that are fast enough to perform such attack.

    While software based deciphering can rely on multiple machines working concurrently, mechanical hacks can not be easily paralleled.

    kwong

    January 20, 2011

  • [...] informacijos galite rasite vaikinukų iš Masiučiuseto technologijų universiteto svetainėje. Nors nieko tokio jų puslapyje nesužinosite ko ne būtų galima įsivaizduoti pagalvojus apie [...]

  • I’d be interested to know how long it to actually open the safe once the device was in place, versus the theoretical time needed to exhaust all possible combinations.

    ldillon

    January 21, 2011

  • How much time is “21,000 cycles”?

    Uthor

    January 21, 2011

  • [...] jest dziełem dwóch studentów MIT. Pokonany sejf ma zamek Sargent and Greenleaf [...]

  • amazing..how many combinations did it took to crack this safe ?

    rocky

    January 22, 2011

  • @Uthor: First paragraph: “…in just a few hours”

    Anders Tornblad

    January 24, 2011

  • While it’s great that your GSA security container with the S&G 8400 series lock was opened, machines such as this have been around for several decades. Commercial versions which can open common types of safe locks called “Group 2 Locks” are available to the safe & vault industry and law enforcement agencies. This type of lock called a “Group 1 or 1R Lock” depending on the materials of the combination lock wheels has had specialized autodialers manufactured before. That is exactly what the machine is, an autodialer that performs an additional function which in this case is to turn the butterfly every time each combination is dialed.

    It takes advantage of known mechanical tolerences and manufacturer’s specifications and then dials every combination bringing the dial back to zero, then turning the butterfly right and finally turning the dial further right to see if the lock is unlocked.

    Not a simple concept or machine to make but it has been done before which is the main reason the newer more advanced electronic lock you mentioned above has been in service since 1991. This version of locks called the X-07, X-08 and X-09 locks used on GSA security containers and vault doors is an electromechanical lock which makes it’s own power by a small generator powering up the lock for a limited time for the end user. Are you up to another challenge??? Design and manufacture a machine for the newer electromechanical locks.

    Dave LaBarge, CPS-CML-CMS

    January 24, 2011

  • Thanks for the article – certainly very interesting to see how you went about cracking the safe lock!

    Gary

    January 24, 2011

  • Would y’all mind disclosing what the combination was?

    I’d like to see if it fits into my idea of the limited keyspace.

    Marc

    January 25, 2011

  • tank you

    lord

    January 25, 2011

  • I saw one of these years ago at MITRE, and watching it in operation is impressive.

    Have you considered front-loading a dictionary based attack? I used to see 6-letter words turned into a combination via the standard telephone keypad mapping, and have always wondered just how big a keyspace that would encompass.

    David Schuetz

    January 26, 2011

  • dremel will do the same thing… =D

    Fernie

    January 26, 2011

  • [...] Blowing safes open with dynamite is so twentieth century. When Kyle Vogt wanted to know if an old safe contained “cool stuff like gold coins, ancient relics or even mummified body parts” the MIT student roped in a friend and built a safe cracking robot. [...]

  • @David Schuetz That’s where I was heading when I asked for the combo.

    there are ~15,553 6-letter words, minus all the ones that make forbidden combinations – i.e. for the 8400 lock: no numbers between 35 & 55 for the last number, no numbers ending in 0 or 5, and no rising or falling sequences.

    Marc

    January 28, 2011

  • @David Schuetz That’s a really clever idea.

    It doesn’t seem intuitive to me to do the phone keypad mapping, but that would make a great first pass of combinations.

    The reason we spent so much time thinking about combination space optimization is that we thought the dial would seize up after something between 20,000 and 100,000 combinations. We had to turn the safe on it’s side and let some lubricant drip down into the lock to loosen it up after the first 5,000 or so combinations were tried.

    kyle

    January 28, 2011

  • So, the combination was…

    Marc

    January 28, 2011

  • [...] it might be the first one using Legos. Massachusetts Institute of Technology student Kyle Vogt has posted a blog that describes an autodialing device that he and a colleague configured and programmed to open a [...]

  • Like Mr LaBarge’s comments above, the dialers for the mechanical locks, including those used by the government, have been around for decades. The Mas X-07/08/09 was designed to detect mechanical motion. I know people who know the combinations to them and can’t open them because of their dialing methods, and after . They have three modes of operation, one allows access
    by dialing a six-digit combination (xx-xx-xx), one allows access only when two separate codes are entered within 40 seconds of one
    another, and one allows access by a subordinate’s code only after a supervisor’s code has been entered. Additionally, they have a full complement of auditing features, including a non-resettable record of openings and an unsuccessful attempts record (audits after 3 unsuccessful attempts) that resets once the proper access code is entered. 10-14 errors results in a 3 minute time out, 15 errors or
    greater results in a 4 minute time out. The error count and penalty time resets with valid combination. All of this is built into the lock, no external reader is required. Want a real challenge, build an automatic dialer that can open one within 20 man-hours.

    Doug Kyle

    February 15, 2011

  • [...] Build a robotic safe cracker, of course! It’s either that or die of curiosity – the Magic Safe could contain anything! Recommend on Facebook Share on Linkedin Tweet about it Subscribe to the comments on this post [...]

  • [...] Build a robotic safe cracker, of course! It’s either that or die of curiosity – the Magic Safe could contain anything! [...]

Leave a comment  

name*

email*

website

Submit comment