Received: by 10.216.194.136 with SMTP id m8mr804774wen.2.1350352603644;
        Mon, 15 Oct 2012 18:56:43 -0700 (PDT)
MIME-Version: 1.0
Path: q11ni134324916wiw.1!nntp.google.com!feeder2.cambriumusenet.nl!94.232.116.12.MISMATCH!feed.xsnews.nl!border-2.ams.xsnews.nl!feeder3.cambriumusenet.nl!feed.tweaknews.nl!216.196.110.142.MISMATCH!border3.nntp.ams.giganews.com!border1.nntp.ams.giganews.com!border4.nntp.ams.giganews.com!border2.nntp.ams.giganews.com!nntp.giganews.com!news.panservice.it!feeder.erje.net!newsfeed.datemas.de!eternal-september.org!feeder.eternal-september.org!mx04.eternal-september.org!.POSTED!not-for-mail
From: Ecnerwal <MyNameForw...@ReplaceWithMyVices.Com.invalid>
Newsgroups: rec.crafts.metalworking
Subject: Re: Need to open a DIEBOLD SAFE
Date: Tue, 09 Oct 2012 10:19:44 -0400
Organization: disorganized
Lines: 45
Message-ID: <MyNameForward-D49EFB.10194409102012@news.eternal-september.org>
References: <ln9378dloir0o9tdli0r6e4ig79cneqkrs@4ax.com> <seGdnWQttPMmuO7NnZ2dnUVZ_rOdnZ2d@giganews.com> <slrnk76u99.15u.BPdnicholsBP@Katana.d-and-d.com> <deidnWXtea9mAu7NnZ2dnUVZ_tWdnZ2d@giganews.com>
Injection-Info: mx04.eternal-september.org; posting-host="422da6038cbfbafb4c31b60cd7fdc051";
	logging-data="14766"; mail-complaints-to="ab...@eternal-september.org";	posting-account="U2FsdGVkX1/owLuPjyzLKNMSrt5xcMEfkYk5QA8LFIo="
User-Agent: MT-NewsWatcher/3.5.3b3 (Intel Mac OS X)
Cancel-Lock: sha1:PhPwyrFW06tnZxFIca1TpMldcnw=
Bytes: 3290


Read Feynman's books - they are worth the time.

He noted the last number from inspecting open safes (actually "secure 
file cabinets" IIRC). It was a flaw in the design.

He found (IIRC without re-reading the book right now) that the numbers 
were not overly precise - so 100 numbers might really be 25-33 if 
running trials. This is a pretty common flaw of 100 number safe dials, 
at least on the lower end. If the spec says 100 numbers, they put on 100 
numbers, but if the numbers are all off by one or two, the safe still 
opens...

He pointed both of these flaws out to security, which rather than fixing 
them, responded in typical idiotic fashion - at least until they needed 
him.

He also used the same tricks that any password cracker uses now - common 
numbers people might choose - anniversaries, children's birthdays, 
numbers scribbled on the desk drawers, etc.

And if he cracked one in 20 minutes, he made sure to read something for 
another couple of hours before opening the office door, so as to not 
make it look too easy.

For a safe you actually own, the correct way to computerize the process 
would be a robot dial-spinner - doesn't need to sleep and won't fudge up 
the numbers if it's built right. But you do need to know the correct 
directions to spin for that model of safe (no matter what you are 
using.) Just the thing for a spare servo (or stepper) motor and 
controller, plus a linear actuator or something for the handle part. Let 
it grind away until it pops, and figure some way to note when it pops so 
you save the right numbers. The you could re-run trials to see how wide 
the band of numbers that work is to figure what the center numbers (ie, 
the real combination) should be.

If making new combination locks now I suppose you might include some 
means of noting too much dial twiddling, but I guess the serious safes 
all went to time locks to deal with that problem anyway. Any safe you 
can pick up (with a forklift, if needed) is ultimately not all that 
serious, is it?

-- 
Cats, coffee, chocolate...vices to live by
Please don't feed the trolls. Killfile and ignore them so they will go away.