Thanks
No
> or are they individually compromised because they are keyed to
a master?
>
> Thanks
>
>
Yes but mainly against picking..
With master pins added to the locks,
there are more chances to manipulate the pins to their individual
shear lines..
--------
"Keyman"
11113, 11131. 31111, 33111, 31313, 13131, 33111, 33311, (and any
other combination of ones and three you can think of) will work
the lock.
Hold and vary master keying is a bit more secure, but that is the
subject of another post.
The locks are less secure, but it is not a big screaming deal.
They are also easier to lock pick, but that isn't a major deal,
either.
Christopher Young
. wrote in message ...
>I have 7 Schlage deadbolts, all different keys, that are keyed
to the same
>master. I am the only one who has the keys as well as the only
master. Are
>all of these locks just as secure as if all seven were keyed
alike? or are
Actually, the quote is from G.L. (Gerry) Finch.
BTW, they could be six pin locks as well, in which case there could be
64 keys that operate provided the system was generated correctly.
BBE.
On Fri, 15 Sep 2000 06:00:14 -0400, "Billy B. Edwards Jr."
<BBEd...@thelockman.com> wrote:
>b...@upn.com wrote:
>>
>> They are less secure. In a master keyed lock AT LEAST 2 keys fit it
>> and maybe even 32 keys. A master keyed lock will never be as secure
>> as a keyed alike lock. Master keying is "the controlled destruction
>> of security"...you can quote me if you like.
>
I wonder if the question is being answered too tersely. Yes, master
keying decreases security. Master keying also provides convenience.
The *real* question is "how much" for each side of this trade-off.
IMHO properly done single-level masterkeying provides a slight decrement
in security, not a great one. Being able to carry 1 key vs. 7 is *for
me* a major increase in convenience.
--
--henry schaffer
h...@ncsu.edu
In a standard 5 pin lock, there are 5 pins that match
the depths cut into the key. Just one key operates it.
To allow for master keying, additional sheer lines have
to be created to allow those keys to work the lock.
This is achieved by adding extra pins to the lock.
The number of keys that work a given lock depends upon
how many additional pins were added to it.
1 additional pin: 2 keys
2 additional pins: 4 keys
3 additional pins: 8 keys
4 additional pins: 16 keys
5 additional pins: 32 keys
6 additional pins: 64 keys
7 additional pins: 128 keys
8 additional pins: 256 keys
As you see, it can get quickly out of hand.
A large hotel with 250 rooms and 5 floors could
have a system where there are several keys, other
than the room key, that work the lock.
A Grand Master Key: 1 key for ALL the locks,
A Floor Master or Maids key: 1 key for the locks on a floor.
In our simple example above, using 6 pin locks, and a
maximum variation of 7 depths, there are only 206
possible combinations. Even this optimal system means
that each lock is capable of being opened with 8 keys:
Room key, GMK, MK, and 5 "Ghost" keys.
To achieve 250 combinations, an additional pin needs
to be added making each lock capable of being opened
with 16 keys: Room key, GMK, MK, and 13 "Ghost" keys.
(Source: Master Key Systems by Harold S. George)
You can see why electronic or card keys are so popular
with hotels now. It allows for greater access control with
instant changeability. After you check out, that room's lock
code is changed from the desk so your card will no longer
operate the lock, and when you turn in the card, it is
re-programmed to the new code the room lock was set to.
Raymond Schreiner
Midtown Locksmith, Inc.
--
Roger Shoaf
If you are not part of the solution, you are not dissolved in the solvent.
Billy B. Edwards Jr. <BBEd...@thelockman.com> wrote in message
news:39c200d0$0$99051$726...@news.execpc.com...
> b...@upn.com wrote:
> >
> > They are less secure. In a master keyed lock AT LEAST 2 keys fit it
> > and maybe even 32 keys. A master keyed lock will never be as secure
> > as a keyed alike lock. Master keying is "the controlled destruction
> > of security"...you can quote me if you like.
>
16 keys out of (say) 700,000 ie 1 chance in 40,000 (or so) in the above
250 combination case is not that bad, especially when a less common profile
is used.
A 6 pin masterkeyed cylinder using an uncommon profile would be significantly
more pick resistant (and less susceptible to stray keys fitting) than (say) a
5 pin 'C' profile Schlage or a Kwikset.
Peter
I believe Gerry's the first I heard use it, for whatever that's worth.
Of course he could have swiped it in turn. I always liked it precisely
because it _IS_ a strong statement, and that's sometimes what's needed
to help folks understand when it is and isn't appropriate, and why they
need to start with better quality locks if they want the result to be
acceptable.
The problem isn't just the ghost keys. You also lose a considerable
amount of pick resistance, which tends to be a more serious concern.
--
------------------------------------------------------
Joe Kesselman, http://www.lovesong.com/people/keshlam/
Opinions expressed are solely those of the author
Can you direct me to the source of that quote? I have a copy of the
booklet 'Master Keying, Brush Method' and can't seem to find it in
there. Thanks in advance,
BBE.
If you do find a reference I would be interested in knowing about it. I
have written a couple of books on the subject myself and like to be
aware of such things. Gerry was a good friend, he published his first
book on MK'ing the same year my first come out, 1979. Regrettably, I
never got to meet Mr. Brush, but was well acquainted with Harold George,
Hollis Devine and George Deahl.
BBE.
>In article <39c18cf7....@nntp.we.mediaone.net>, <Bob> wrote:
> I wonder if the question is being answered too tersely. Yes, master
>keying decreases security. Master keying also provides convenience.
>The *real* question is "how much" for each side of this trade-off.
>IMHO properly done single-level masterkeying provides a slight decrement
>in security, not a great one. Being able to carry 1 key vs. 7 is *for
>me* a major increase in convenience.
>--
>--henry schaffer
>h...@ncsu.edu
Well I don't know a lot about Master Keying but it seems to me since
inside the lock it's just a matter of stacking chips, that you can
make extra keys work the lock while keeping the number of chips you
put in the lock low. In other words if you need two keys to opperate
a single lock then instead of puting chips in that add up to 32
potential combinations why not just put chips in that add up to say
half that or even a quarter or eighth of that. It's certaintly
mathmatically feasible Instead of puting say 5 chips (2 to the power
of 5 =32 ) why not put in just one or two chips (2 to the power of 2
=4 , to the power of 3 =8) then out of the potential 10,000 keys that
can opperate the lock besides the master and slave key there are only
6 others? And besides 32 out of 10,000 seems awfully high to me. Let
me ask you pros how many times have you ever ran across a system in
which you learned that a potential key (i.e. one not intentionally
designed to work the lock) actually opperated the lock? I'll wager
that on average you may run across something like that happening about
once in every 10 years and maybe once every 3 years if you do buisness
in a dense urban area such as NYC. And what's all this talk about it
being easier to pick? So what a lock is inheritly flawed that way.
If they can't be picked other meathods of entry are available. If you
don't want the lock to be picked use what we use at work, an
electronic key that you hold up to the reader and it releases the
latch on the lock. Talk about master keying this device allows you to
lock users out on holidays, it lets you program new users instantly (
I know because I run the system for my department) it lets you remove
someone when they have been transferred or terminated. I've had this
key for over 3 years and it's never given me any problems. No
lubricating the lock. And now were swithching to badges with our
pictures on them and integrating the locks into our time clock. This
can't be done with master keys. Locksmithing is old hat anyway. What
a waste in the 21st century unless it's for your home which doesn't
need to be master keyd.
______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
With Servers In California, Texas And Virginia - The Worlds Uncensored News Source
As well, your comments regarding electronic controlled vs.contempory
locksmithing is just spurious and irrelevent. I would estimate that a full
half of my masterkey jobs are for small business owners that would like
their top level key (masterkey) for their businesses AND their homes that
will involve just a few change keys.
As you seemed quite intent on letting the world know of the merits of your
card access system, praytell just how much did it cost? Can you send me
literature so that I can tell my "Mom and Pops" business owners that they
TOO can have card access at the $$$$ that they cost
Darren
Wildwood Lock
b...@upn.com wrote:
>
> I have heard Hollis speak on many occasions when I first started in
> the business. It was George Deahl's son Bill that taught my first
> master keying class. I had the unique experience of going to class
> tuesday and thursdays at night and using the information the next day
> at work while we were rekeying 3 high rise buildings in the city of
> Orange. There were thoudands of locks and it took weeks to rekey.
> When we started the job I was the one asking all the questions but by
> the end I was solving some of the problems, thanks to Bill.
>
> On Sun, 17 Sep 2000 00:58:36 -0400, "Billy B. Edwards Jr."
> <BBEd...@thelockman.com> wrote:
>
Like the SARGENT -
LA, LB, LC can be opened with LD, LDH, LDM & LN blanks
LE, LF, LG can be opened with LH, LDH, LDM & LN blanks
LJ, LK, LL can be opened with LM, LDM, LHM & LN blanks.
This works great for multi story offices, or hotels.
These systems are vulnerable to key profiles being mutilated by users.
A 8-pin cylinder with 1 profile is preferable to 6-pin cylindrers with multiple
profiles, assuming such cylinders are manufactured (there are alternatives to 8
pin cylinders that are just as effective).
A conventional 8 pin cylinder would have 70 million or so usable differs. If
it is fully masterkeyed, 256 keys would fit each cylinder ie 1 chance in
quarter million of a stray key fitting. (Chance of a stray key fitting 9 pin
cylinder with 9 spacers would be about 1 in one million and this would diminish
the more pins there are).
However with 8 pin cylinders, 4 or 5 spacers only would normally be required
(using rotating constants), thus decreasing the probability to 1 chance in one
million or better.
Moreover a 8 pin cylinder even with 8 spacers would present a fair picking
challenge. Not to mention that the number of 8 pin keys in circulation would
be far less than Schlage 'C' keys in circulation, hence the probability of any
stray pin tumbler key fitting would be significantly less than winning a major
lottery.
In practice the 5 'auxiliary' pins in ASSA and Primus cylinders make up for
loss of security resulting from masterkeying the 6 main pins.
The Corbin 'master sleeve' cylinders and Yale bicentric (ie duplex) cylinders
were developed to eliminate (or at least minimise the number) of 'ghost' keys,
but the latter is now of historic interest only.
Peter
--
Roger Shoaf
If you are not part of the solution, you are not dissolved in the solvent.
Peter <pmi...@the.net.nz> wrote in message
news:RBxx5.20282$cr3.5...@ozemail.com.au...
> In article <20000918011838...@ng-cb1.aol.com>,
> joc...@aol.comphuckoff pontificates...
> >
> >This is the reason that SARGENT, SCHLAGE, RUSSWIN, CORBIN, YALE & others
use
> >MULTI KEYWAY SYSTEMS allow a master & submaster & tennant blank fit the
> >cylinder.
> >
>
> These systems are vulnerable to key profiles being mutilated by users.
>
> A 8-pin cylinder with 1 profile is preferable to 6-pin cylindrers with
multiple
> profiles, assuming such cylinders are manufactured (there are alternatives
to 8
> pin cylinders that are just as effective).
>
> A conventional 8 pin cylinder would have 70 million or so usable differs.
If
> it is fully masterkeyed, 256 keys would fit each cylinder ie 1 chance in
> quarter million of a stray key fitting. (Chance of a stray key fitting 9
pin
> cylinder with 9 spacers would be about 1 in one million and this would
diminish
> the more pins there are).
>
> However with 8 pin cylinders, 4 or 5 spacers only would normally be
required
> (using rotating constants), thus decreasing the probability to 1 chance in
one
> million or better.
>
> Moreover a 8 pin cylinder even with 8 spacers would present a fair picking
> challenge. Not to mention that the number of 8 pin keys in circulation
would
> be far less than Schlage 'C' keys in circulation, hence the probability of
any
> stray pin tumbler key fitting would be significantly less than winning a
major
> lottery.
>
I don't know about OZ, but in the US the Bicentric cylinder and the
Duplex lock are different things with Duplex locks being made only in
the 1800's and the Bicentric available until just recently, if not still
available.
BBE.
You have kept up! There are others as well, Dorothy Friend, Mike
Kirkpatrick, Gordon Morris, Lloyd Seliber and Dave Killip. Yes, but I
am afraid it is coming to an end. All of those old masters are gone and
there are just three of us left still doing it daily. Some have gone
into other areas and others have just gone. The really sad thing is
that there doesn't seem to be a bunch of new ones waiting in the wings.
BBE.
Of course the increase in the number of differs also increases exponentially
with more pins and completely overwhelms the number of 'ghost' keys.
Try using partial quotes.
Like this response..
That way we can tell exactly what you are responding to..
--
"Keyman"
>I have 7 Schlage deadbolts, all different keys, that are keyed to the same
>master. I am the only one who has the keys as well as the only master. Are
>all of these locks just as secure as if all seven were keyed alike? or are
>they individually compromised because they are keyed to a master?
It depends how they were set up.
The short answer, which most of the other follow-ups to your question
have given, is that if the locks were set up randomly, there will be
lots of extra keys that will work in them, and lots of extra shear
lines for the bad guys to take advantage of if they know how to pick
locks.
It's been pointed out in dozens of other threads that bad guys
generally don't have th patience to pick locks and prefer brute force
methods of entry. If your locks were installed with the strike
reinforcement plates that came with them, you've done far more to
strengthen your security than any amount of master keying would weaken
it.
That said, in a master key system as small as yours, there need not be
more than one master pin per lock, meaning that there would be NO
other keys besides the master and the particular change key for each
lock.
Compare this system to the other examples posted in this thread:
In a Schlage system you can have a maximum of 5 different depths used
per chamber, and you can't generally put the lowest and the highest
depths in an adjacent chamber. (You also shouldn't put several cuts
in a straight line, but this is an example, not an entire master
keying class!)
Master key is 33333 (to modify the simplified example from another
reply.)
Change keys are 33331, 33335, 33337, 33339, 33313, 33353, 33373.
In a lock set up to work with only the first change key and the master
key, there would only be one master pin, a #2, in the last chamber,
which is also the hardest to pick. In the lock set up to work with
the second key, there would only be a #4 in the last chamber.
That's pretty secure.
If you went all out and wanted a lock on the front door that could be
opened by all 7 keys plus the master key, it would have 3 extra pins
in the 4th chamber and 4 extra pins in the last chamber, and could
thus be opened by all possible keys with the cuts in your keys, such
as 33315 or 33359. Besides your 8 keys, there would be an additional
12, which still isn't too shabby compared with some of the other
examples you've been given, and you didn't say you wanted to do that.
You could also do fun things like have a lock on a door to the 2nd
floor stairway and have it set to work with all the keys that start
with 3333, thus locking out all the keys that don't have a 3 in the
4th chamber.
The other disadvantage of a master key system is that if someone has
one of your 7 change keys and knew that the locks were master keyed
and knew how to determine a master key from a lock and had access to
the lock the key went to when you weren't around, then he could take
the lock apart, decode the pins and then figure out what your master
key is and then open the other locks.
By the way, it's also possible to do the same thing without any of
your keys if he could take apart two of your locks and decode all the
keys that worked in them. It would then be a simple matter to look at
the list of possible keys for both locks and find master keys that
would open both of them and possibly more in the same system.
Lee
|Lee Most, CRL, San Francisco, California
|lee...@slip.net
|----------------------------------------
|This is no time for moderation!
>Instead of puting say 5 chips (2 to the power
>of 5 =32 ) why not put in just one or two chips (2 to the power of 2
>=4 , to the power of 3 =8) then out of the potential 10,000 keys that
>can opperate the lock besides the master and slave key there are only
>6 others?
A good point. In fact, as I just pointed out in another post, you can
put in only one chip and not have any unintentional keys work in the
lock.
> And besides 32 out of 10,000 seems awfully high to me. Let
>me ask you pros how many times have you ever ran across a system in
>which you learned that a potential key (i.e. one not intentionally
>designed to work the lock) actually opperated the lock? I'll wager
>that on average you may run across something like that happening about
>once in every 10 years and maybe once every 3 years if you do buisness
>in a dense urban area such as NYC.
Unfortunately, a lot more often than that. Even now that I'm an
institutional locksmith, I still get called on several times a year to
"fix" badly designed master key systems in apartment buildings where
one tenant could get into another's apartment, and until rather
recently, the schools had a big problem with teachers from one school
being able to access locks in other schools.
>And what's all this talk about it
>being easier to pick? So what a lock is inheritly flawed that way.
Some are.
>If they can't be picked other meathods of entry are available.
Other methods of travel are available than cars. Does that mean that
you stop using one? As people intent on reducing all possible methods
of entry, yes, we are concerned about whether something we are doing
to a lock will make it more susceptible than it was to one of the
known methods of entry. This is a discussion about master keying, and
that has been known to increase the ease of picking a lock, so it gets
discussed. In other threads, we discuss how to prevent other methods
of entry. Whether or not a lock is master keyed has nothing to do
with how easy the door is to be kicked open, so it's pointless to
mention it here and makes more sense to discuss in a thread about
doors, or which lock makes a door stronger, or ways to reinforce a
door frame.
>If you
>don't want the lock to be picked use what we use at work, an
>electronic key that you hold up to the reader and it releases the
>latch on the lock. Talk about master keying this device allows you to
>lock users out on holidays, it lets you program new users instantly (
Yes. That's great, isn't it!
>I know because I run the system for my department) it lets you remove
>someone when they have been transferred or terminated. I've had this
>key for over 3 years and it's never given me any problems. No
>lubricating the lock.
Really? What system do you use? We had so many problems with the
last one that we scrapped the whole system and went with a Northern
system that uses proximity cards. Now the readers are pretty stable,
but people are breaking the cards right and left. They are brittle
and snap in two far too easily.
You never lubricate your locks? What are they, those $300
electromagnetic ones? You still have to clean them or crud prevents a
good bond.
I'd love to install prox card systems everywhere, but it's impractical
at an installed cost of about $3500.00 per door (our Security Dept.
demands that all electronically opened doors be hard-wired to their
computers at the central monitoring station across town, so modems and
fixed phone lines are involved.)
And, of course, being a railroad as well as a bus, trolley, and cable
car transportation system, more than half of our locks are padlocks,
which are hard to find electronic versions of, especially hard-wired
to Security. <g>
>And now were swithching to badges with our
>pictures on them and integrating the locks into our time clock. This
>can't be done with master keys.
It also can't be done when the Payroll department won't accept a
report of a badge being shown to a reader as a substitute for a signed
time card cross-signed by a supervisor standing over your shoulder
while you punch it because too many people got others to punch in or
out for them. ("Hey, Joe! Here's my badge. My girlfriend's picking
me up to go skiing after lunch and I won't be back until Monday. I'll
get it from you just before we punch in, okay? This 'll make up for
when I did it for you every day during the World Series.")
>Locksmithing is old hat anyway. What
>a waste in the 21st century unless it's for your home which doesn't
>need to be master keyd.
Well, actually, my home is master keyed, but that's besides the point.
I assume you mean mechanical locksmithing. After all, there must be
someone who maintains that electronic card access system you have
there, and someone else who will replace it in a few years when the
next upgrade comes along.
So you seriously think that there won't be a need for locksmiths to
repair locks on file cabinets, desks, padlocks, closets, tool boxes,
lockers, automobiles, gates on fences, and all the other things that
electro-mechanical locks are really not suited for? You may have a
neat little card to activate the lock on your door, but there's still
something locking it, and that something can break. Some people may
believe that we'll reduce crime by putting lots of people in prisons
because they had a few joints on them or a needle full of some
forbidden substance, but people keep learning from other people around
them, no matter where they are, and what they are learning in prisons
has little to do with being better citizens. Do you really think
crime will end in the 21st century, that locksmiths won't be needed?
Or did you mean only that master keying is old hat? Having seen the
messes that untrained people have made of master key systems in
hotels, schools, and even here before I got here, I certainly hope
that electronic locks come down in price and up in reliability enough
so that everyone can use them, not just large businesses, governments,
and utilities.
Until then, we're stuck with master keying.
Lee
|Lee Most, CRL, San Francisco, California
|lee...@slip.net
|----------------------------------------
|Nothing is so smiple it can't get screwed up