On a whim, I picked up a Master Lock 1500iD a few weeks ago. Mostly, this was for physical security research because I was bored at the time. Well, and I wanted a new lock for keeping my bike locked up at the job I used to have.
Posted by Ax0n
I had been using a derivative of the somewhat vulnerable Master 175 Padlock. I have always been a proponent of security in depth, so a somewhat chintzy lock combined with a very sturdy bike rack, a length of towing chain, and a parking garage with patrolling officers and cameras everywhere provided adequate layered protection. Also, in fair weather, several other lesser-secure bikes would be parked with mine, adding a layer of Darwinian Bicycle Security.
Advantages that made me choose this lock were many. First, the shrouded hasp meant it was likely to be resistant to shimming. Also, the "combination" could be entered in low-light conditions and while wearing gloves. This is important, because the parking facility I was using at the time was not heated (so it was cold!) and they'd switched to fluorescent lights that never really warmed up or achieved full brightness whenever it was below freezing. On REALLY cold days, some of the lights would refuse to turn on. All of these factors made this lock look like a solid winner for the situation.
Note: This lock is meant to keep your mobile phone and sunglasses safe in the locker room at the gym. It's meant to keep middle-school kids from stealing your homework. Alone, it's not the best tool for locking up a bicycle or anything valuable.
One of the first things I wanted to know was how it worked inside. I also wanted to know how difficult a task it was to get it open without completely destroying it. To the first end, I stumbled on Michael Huebler's 1500iD visualization flash simulator, and subsequently the PDF breaking down most of the facts on this lock.
In fact, Michael had covered most of the angles I was hoping to discover on my own, and did a better job than I could've done here. Therefore, it's worth the read if you're interested in locks, locksport or mechanical things.
By the way, with a good set of drill bits meant for cutting steel, it took me about 7 minutes to get into the lock on my workbench without completely destroying anything. In practice, an attacker would use a large set of bolt cutters since the hasp isn't completely shrouded. This should make short work of a lock like this one in just a few seconds.
I noticed a few collisions, another point that mh's article brought to attention. The lock opens when the four wheels are in the correct state, and every movement of the joystick changes the state of three out of the four wheels. It is for this reason that there is more than one way to get to almost any given state. Using the state in the screenshot above, Right-Left-Down-Left is the combination shown. The same state can be accomplished with Up-Right-Down-Left.
In short: The number of combinations is unlimited, but the number of mechanically-possible states is markedly finite: 7,501 to be exact. mh likens this to the mechanical version of a hash function. I can't think of a more concise allegory for it.
Mechanically, I think Master did a lot of stuff right. First off, the hasp acts as the wheel reset mechanism. This allows the hasp to be locked with a gate that doesn't rely on a spring. Even without the hasp shroud, there is no way to shim this lock. The best you could hope for is to wiggle a very thin wire in through the reset slot on the back to probe for the various gate positions.
If nothing else, the inner workings are innovative. It's simultaneously bizarre but fitting that Master would test new technology in a "toy" lock like this one. Perhaps there's a way to make it scale, either via more positions per wheel, or more wheels to gain more state space.