Need info on turn-of-the-century S&G C-Series lock

[flyingbovine]

A local non-profit has given me access to an old safe which I'm trying to manipulate. From what the director tells me, it's combination was lost sometime around the twilight years of the Nixon administration. I first asked questions about it in the thread "Spindle lube for ancient safe". Since then, I've found that just about everything I was doing was all wrong... At any rate, I now have a much more lucid view of what I'm working with...

The safe is in the rage of 100 years old. From what the director tells me, it was installed several years after the 1906 San Francsico Earthquake. It's a single door unit with a strange formation of concrete poured around it. The lock mechanism is (I think at least) a Sargent & Greenleaf C-Series Lock. It didn't take long to figure out that this unit was in pretty rough shape. There was a considerable amount of play in the dial, and the bearing surfaces of the spindle shaft squealed loudly whenever I turned it. I've since dropped a bit of Tri-Flow behind the dial, and it apparently found it's way to the spindle. That took care of the squealing. I also found that the spindle dial was not only loose...it was bent. I brought a dial indicator on a recent visit. The dial (when you "shake" it), at the largest point on the place where you turn it with your fingers, has .010 in movement from side to side, and .025 in movement from top-to-bottom. When rotated with a dial on the same spot, it shows a eye-popping wobble of .055" TIR.



When I first started working on this unit, I was relatively sure it was a 4-number combo. It took me a day to figure out that you start a 4 number lock by dialing the first number to the right....not to the left. After considerable tinkering, I'm questioning whether or not this is a 4 number lock. When I gather up all the wheels to zero and then dial in the opposite direction, I can feel the driver pick up the first wheel at around 2 (going left). It seems to pick up the next two wheels at the same point, prompting me to conclude that his lock has a fixed drive pin mechanism on the first wheel, but moveable flys on the rest of them. I can't figure out why they would do this... At any rate, there's a barely perceptible change in resistance when the drive picks up the first wheel. There's noticeable resistance when you pick up the second, and you can definitely feel the third. All three wheels seem to pick up at 2 on the dial. If I keep turning to the left, it starts picking up what *I think* is the 4th wheel. At 8 gradations before the expected pick-up point, the dial feels like it's going into quicksand. Soon thereafter dialing resistance goes through the roof to the point where I prefer to grip the dial with thumb-and-forefinger wrapped around it rather than with the tips of my fingers. My guess? Someone else tried dribbling lubricant down the dial of this thing. Over the years, that lubricant has coagulated.

At the moment, my plan is to manipulate the lock. I don't have the equipment (or the know how) to drill it, and the non-profit will let the safe sit for another 100 years before they shell out money for a professional vault technician. On what I've learned via attempting to manipulate the lock: The left contact point seems to be at 99, while the right contact is somewhere in the "forbidden zone", a place on the dial between 0 and 20 where there are no numbers or gradations. When I gather up the wheels and go check the contact points every 2 gradations on the dial, there's a 3 number range where the left contact goes to 99.5. It does this whether the wheel pack is moving right or left. This doesn't make sense, as I'd expect it to shift with the drive pin variation I'd mentioned. When gather up the wheels to the left, then reverse direction and stop at the middle of this range on each turn, the left contact stays at 99 until I pass the target number for the fourth time. At this point, it rises to 99.5 for all subsequent turns. I interpret this to mean that it's either the the first number in a three number lock, or the 2nd number in a four number lock. I don't get any more significant variation if I park this number and scan the wheels in front of it. This suggests it's a four number lock, and the next lowest point is on the first wheel (meaning the one furthest from the drive cam). Bend and play in the spindle seem to be complicating any further manipulation. There are other observations, but they're long winded and beyond the scope of this post at the moment.

Questions:

Does this lock sound like it's three numbers (a C-66 or a C-68) or four (C-62 or C-64)? I found a pic of what *I believe* to be the inside of this thing.



I got a serious "AH HA!" moment when I saw this...it answered *a lot* of questions that I had while dialing this thing. Note the machined wheel pack. I don't see those varying much in size or eccentricity, suggesting that the vast bulk of my manipulation clues will come from fence lean (which would suggest 3 numbers). On the other hand, parking the 3rd wheel and scanning the two in front of it yields nil (note, possibly because of spindle wear-damage), suggesting there might be one more wheel in back, and that I need to map it out two gradations at a time. Time constraints (classes and activities that the non-profit sponsors) have so far prevented me from checking some of these things.

Secondly, I'd like to straighten the spindle. Does anybody have a suggestion for straightening the spindle on a locked safe? First off, does anyone have any details on the construction of the spindle? (i.e., material, solid or hollow, etc) and how difficult it will be to bend? I can see two approaches to this problem: 1.) Place an dial indicator on the dial, point the high point on top and impact the dial downward or 2.) Indicatge the dial so that the high point is downward, and push the dial up from the floor with some kind of a jack-screw arrangement. In either method, start with light forces, and gradually increase until the indicated run-out is minimized. I'm leaning towards the jack-screw method, as I can keep the indicator on the dial while pressure is being applied. On the other hand, is this something that I shouldn't tangle with? Bear in mind that the spindle-lock bearing on this unit is already shot, and has LOTS of play....and that if I give up on this unit, it will probably sit locked for another 100 years.

Thanks! FB

[flyingbovine]

A week later, no responses. (Heavy Sigh).

I racked my brains over this one, and showed up at the site today armed with all sorts of diagnostic goodies....a bolt-on vernier (it was too heavy to read contact points...I'm using it to indicate on), several dial indicators, a caliper, and a small piece of paper to put numbers on the "dead zone" of the dial. I also brought along a make-shift, brute-force implement with which I could attempt to straighten the dial. After doing a bunch of things with dial indicators, I found that the dial had .050 of radial run-out from the lock's axis, but almost no wobble (meaning angular variation from the lock's axis). The two axes were .025 (.050 divided by two) apart, but they were almost parallel. How this happened through accidental impact is beyond my comprehension. In light of this new-found information, I opted to scratch the plan to straighten the spindle. Today's plan was to treat the lock as a 4-number lock, ignore "parking the 2nd wheel and scanning the 3rd and 4th", but instead dial 5x R the unknown number, then 4xL the known number and seeing if i got any gate variation. The first number was in the first third of the dial....I opted to start at the middle of the dial and work up. A ways into this, I noticed the contact points narrowed drastically after dialing the 2nd number within a three-number range. The Contact zone dropped from about 8 wide to 1.5. It appeared that with only two numbers dialed, the cam gate and the nose had meshed. After going in right and stopping in this range, I could reverse direction and get a hard stop at around 19 (in the "forbidden zone"). After that, the door handle would rotate 1/8th of a turn, and the door itself could be pulled out about 1/8" (when locked, it moves about .020").

http://www.flyingbovine.com/images/pub/almostopen.jpg

I figured the cam nose was in the gate, but for some reason, not all the way. I tried parking wheels #1 and #2, while scanning #3 and #4 for another spot where the cam and gate would line up. Nothing worked. If I turned the cam anywhere where it touched a fly, the gate and nose would stop meshing. After about an hour of this, a horde of screaming kids hit the area for a scheduled activity...I had to clear out. I did some research and found out (not the first time) that I'd made some bad assumptions, and had been doing everything wrong.

First off, I'd read in the S&G manual that all all four-number S&G locks both started dialing the first number and stopping when dialing right. I also assumed that all S&G three number locks started left, and stopped on right. The last two days of my efforts had been assuming this. Then I took a closer look at the article that accompanied the picture in the previous thread:

http://en.allexperts.com/q/Antique-Safes-3774/2012/2/old-s-g-safe.htm

The guy pointed out that most safes have the lock on the left, and the hinges on the right. This particular safe had the lock on the right, and the hinges on the left, which meant it was a "Left Hand" lock. He also mentions that it's opening instructions started and stopped opposite of what he expected. I was working on a similar safe....lock on the right, hinges on the left. I'd been dialing it backwards. From a mechanical sense, I'm still not sure why it's not opening....but I'm pretty sure I've corrected all my incorrect assumptions, and can have the unit opened on my next visit. I also looked at the picture of the lock again, and visualized it's rotation as to where it would be inside the safe...with the view I'm seeing facing away from me rather than toward me. Yes indeed, this unit was set up to stop when turning left. But why did the lock mesh after two numbers? Picking through an S&G manual for a four-number lock, I found mention of setting a lock to a two number combination. Then it hit me....this was how the lock was set. I just have to do what I did today, but backwards.

The plan for my next visit is to do what I did today...only starting left and stopping left. Within 30 minutes I'll have either an open safe or another batch of incorrect assumptions.

Wish me luck...and I hope this post helps out some hapless soul like myself sometime in the future.

[femurat]

Why don't you park the wheels at 50 instead of 0 to count the wheels? This way the contact points aren't in the way and don't distract you.

Straightening the spindle may be risky. But you may try and align the dial ring bu carefully hitting it with a rubber hammer.
This is a delicate operation and may be better to avoid it. But it's better than straightening the spindle of a mounted lock because if done carefully there are less chances of screw up.
EDIT: I went on reading your second post and found out you decided not to straighten the dial. Great.

About the direction you dial the combo, it doesn't matter if you start left or right, as long as you have the correct numbers, that may be slightly different depending on the direction you dial.
The important thing is to reverse direction on the last turn, after you reach zero and the nose drops in the cam, otherwise the bolt won't be retracted.

Good luck

[flyingbovine]

I went to the site today, and I got it open. Technically, I had it unlocked last night when the handle turned 45 degrees. When I showed up today, figuring out the reverse numbers took about 15 minutes. After dialing them, I got the same thing....Handle turning 45 degrees, door coming open maybe 1/8". Then I noticed the door had been painted...by hand, with a brush. There was paint in the upper area between the door and the jamb, effectively gluing the top of the door shut. I gave it a good, majority-of-body-weight pull, and the paint broke. Whee! It's unlocked! But it's *not open*. Why? About five years ago, the organization that owns the safe did some re-modeling. they put in a new floor, and the new floor is about 1/2 inch higher than the bottom of the safe door. Because of this, I could only open the door about an inch and a half. I could see the door bolts....but noting inside the safe itself. The director talked to their maintenance guy, who said that "the situation with the floor was not going to get fixed any time soon". The entire staff came down and marveled that this monstrosity that they'd wondered about for years was finally open. Their office is located adjacent to one of the local food banks. The food bank gives them some of their surplus. One of the ladies took me into the kitchen area and gave me a large bag of baby carrots, six packages of "El Monte" frozen tacos, and *huge* bag of frozen blueberries. They have my contact info, and will be in touch when they can do something about the excess flooring material. All in all, a fun time! It may be a while, but I'll try to post pics of the inside when I can...