.:: Keypicking :: Anything And Everything For Locksmiths, Hobbyist Lock Pickers, Locksport, Lock Picking, lockpicks, cutaways and Security. Simply The Best Lockpicking Forum Around!

I have on my workbench a lock that drove me crazy. I gave up and finally decided to open it with an autodialer.
After removing the cover I confirmed the spline key was loose.

Now that I know what the problem is, I'm going to try and manipulate it as it is. Please tell me not to...

Cheers

It is one of last big adventures of mankind. If Stephen King is not horrifying enough a loose spline key is the way to go. I really wonder why the loose spline key has not yet been patented as a method for a manipulation protected lock. If you get it open you can improved it with one unlocked wheel.

What?!
You have an autodialer? I must have missed that thread.
Good luck with your adventure, and looking forward to hearing the story of how you get that lock open.

Martin, the loose spline is simulated with the off centre roller on the nose of a S&G lock, but you know it already. But an actual loose spline is a nightmare. They could make a lock with a similar "defect", if mechanical locks were still interesting for manufacturers.

jarveee, I don't have one, but a couple of colleagues have their own and I can borrow one if I need it. With this safe, it saved me!



The idea is to reach the latest point that I can read. I touch the contact point and read, then touch it again a couple times until it no longer moves. That should give me a consistent reading, since the spline key is at its movement range limit.
I knew what was going on inside the lock, so the first graph wasn't that difficult to make. It took more time than a normal graph because I had to read the contact points multiple times. It was clear there was a gate at L9.
The high low tests were unclear, and I decided to make them again. The gate at L 9 could be on wheel #1 or #2. I needed a break.

A few days later I decided to make the second graph with both wheel #1 and #2 at L9, testing only wheel #3 Around Left.
I obviously had to dial AWL, then made almost two turns Right so I could test wheel #3 Around Left, starting at 10, much more convenient since it's the normal dialing direction.
There's nothing wrong in checking a wheel the other way around, but then you have to convert the gate center and there's always the risk to make a mistake.
This graph was rich in extra contact points, so I traced a line across the most external points to get a better view.
It became clear that wheel #3 had a gate at L26!

Now it's time to decide the next step. I could brute force wheel #2 and if it fails brute force wheel #1, second choice because it takes longer.
Or I could test a few combos to see if I can determine if the gate at L9 is on wheel #1 or #2.

Cheers

I don't have good news. After a few checks, during graph #3, contact points became wider and wider, a lot wider. I stopped graphing and tried to brute force wheel #2. I failed.
I need some time to rest, otherwise I'll remove the back cover and check what happened.

Cheers

I decided to refrain from opening the back cover and went on like the lock was still in a safe. As said, the lock didn't open, neither revealed anything, during graph #3. For the sake of it I decided to brute force wheel #1 because the high-low tests were not so clear. I got nothing, but have no more doubts, 9 is on wheel #1.
So it's time to try 9 AR 26 again, but with a gentler touch, since contact points are moving a lot. If I try how far they go, I can reach an increment more than the previous one. Completely useless.
I started at 0 and was going pretty well, until 80. Then the contact points started wandering again. I forced myself to finish the graph, hoping that even if it was almost impossible to read, or use the info I could register, it was worth a try. And then there was my idea that during the test the lock would suddenly open. But it didn't.



After this last graph I gave up and removed the back cover. The first thing I noticed was a small brass piece acting as a spline key.
The lock was reassembled exactly like it was in the safe, exactly as I asked. I wanted to give another try and possibly learn how to deal with a loose spline key. Well, this is too much.
Looking at the wheels, I confirmed the correct combo was 9 - 42 or 43 - 26. But it doesn't always open with the correct combo: The spline key is so loose that the last wheel has a couple numbers play. I imagine it's the same with the other two wheels.

My conclusion is that if the spline key is a bit loose, the lock can still be manipulated. But if it gets too loose, then it's beyond my limits.

Cheers

Now you have the capacity for suffering to tackle all manipulations!

Was the spindle modified? Did they widen the spline key slot to make it fit to the pin? I have seen drilled spindle and cam with a round pin, spindle slots cut with something like an angle grinder, much to small spline keys. The angle grinder slot didn't create problems in use. Apparently the gap was so large that the key shifted reliably back and forth. It was only found because I checked the safes. The to small spline key did work at first, but apparently became then loose and that is when opening problems started.

I really do!

The slots in spindle and cam are not modified.
The spline key looks just a brass piece. It may be a rectangle cut out of a brass sheet. It's not L shaped like usual. Just a squarish profile.
The lock cover prevented it from falling off, luckily.

Cheers

femurat, Could you please test how far you have to turn the dial to get with the drive cam from one side of W3 to the other side?

Sorry, I can't. After taking the last picture, I installed a brand new spline key. Now I'm waiting for a friend to change the combo so I can play with standard conditions.

IIRC the play was 2 increments on wheel #3.

With the angle grinder spindle it was 4.