Interesting locks

[VashTSPD]

not something I could pick with stuff I would carry on me. And a thief probably wouldn't know how to pick one if it was his first time seeing one.

Security through obsecurity.

[mh]

I really wonder how they make them. It's either a high-tech process where the way the keys have to be bent is calculated with complex formulas - or they just bend the key at random (kind of), and then select the appropriate discs - or better still, insert a key, then cut the notches...

mh

[greyman]

mh, that is a very good question! My guess would be that the twisting of the key is calculated to match the discs. Any other method would result in a small number of differs and/or unreliable tolerances. But I don't know either...

[mpj]

I really wonder how they make them. It's either a high-tech process where the way the keys have to be bent is calculated with complex formulas - or they just bend the key at random (kind of), and then select the appropriate discs - or better still, insert a key, then cut the notches...

Atrix's keys have the number '916' -- could this be the key's code? Looking at the keys, it seems that there aren't really that many twists (or rather changes in slope) in them. Perhaps there are only three places (front, middle, back) where the numbers are 'controlled', ie twisted by the key shaping machine, and in between you have intermediate values from those. Naturally, these discs would also be necessary for the smooth operation of the key.

All this makes me wonder how they can claim 100 million = 10^8 combinations. One possibility: you could say that adjacent discs can be at most one "step" apart, so they can be either in the same position, or offset by one step either clockwise or counterclockwise. If I counted correctly, there are 18 discs; the number of pairs of adjacent discs is therefore 17, and 3^17=129 million. At least that would be in the right order of magnitude.

Yes, I'm just guessing...

[greyman]

mpj - I think you are on the right track. The MACS of 1 sounds right for a lock with a twisted key as it models that maximum slope (twist in this case) on the key pretty well. I'm not sure if you can estimate the number of combinations as 3^17 though. The problem is that there are constraints on the way you can twist a key that might mean that the key twists are dependent across more than just the adjacent positions on the key blade. I think I'll have to go away and have a think about this some more.

[mpj]

I'm not sure if you can estimate the number of combinations as 3^17 though. The problem is that there are constraints on the way you can twist a key that might mean that the key twists are dependent across more than just the adjacent positions on the key blade.

I agree. In fact, 3^17 probably represents the upper bound on the possible number of different keys. In practice this is likely to be much lower and (if I was right in my previous post) they are not even trying to reach this number. Perhaps the 100 million they claim is just marketspeak.

[mh]

For sure to get all these key differs, you would have to use milling/cutting, not bending. The keys on the picture look bent, though.

However, marketing people could assume that the key manufacturing process could be upgraded once the first part of the keyspace (accessible by bending) is used up.
There's a similar option with the EVVA MCS, where the relative magnet orientation right now apparently uses only 45 degree steps, but could potentially have smaller steps.

Cheers,
mh

[TOWCH]

I would think they would heat the blanks, then insert them and twist the discs to the appropriate code.

[greyman]

Good point. Perhaps there's a metal alloy that's soft at low temperature, that can be twisted into the right shape and then hardened by another process. Where is the resident metallurgist?

[Shrub]

Ive not thought about this too much but i am guessing that they could simply make the keys and then put a disc pac on them then mill the sidebar grove in the pack of discs,

Only another idea,

[atrix]

hi all
new photo:


Disks have no false notches




With this tool It is possible to open for 10 min.


No other forums allowed to be linked to
Friends excuse, if I sometimes do not answer questions. I never learned English, and to me happens difficultly to translate. But I shall try to give the maximal information on questions interesting you

[atrix]

Hi all
Link about Yale Scorpion
http://www.locks.su/test/yale_ml14/test.shtm#def

[mh]

atrix - thanks for these interesting pictures. I love the first one with the twisted key. I haven't seen anything quite like that before, although it looks a bit like a DOM Diamant.

Hi,

This thread is already a bit old, but it's one of the few that actually mentions the DOM Diamant around here...

I wonder if someone's here who had a close look at a cutaway or similar -
when are the discs aligned? (a) While the key inserted, or (b) afterwards, when the key is turned?

Thanks,
mh

[Schuyler]

oh no atrix's photos have disappeared I loved all of his posts.

someone really has to do a harvest of this website.

[Shrub]

I think his pics are in the advanced section but looking at the titles it appears they shouldnt have been in the open sections?

Ill wait until they are back up then leave or remove i guess,