mainspring 'working' out?

[taterbug]

The mainspring on this wants to work it's way away from the plate while cycling the hammer.  It doesn't come completely off the plate (or hasn't yet), but does it after only a couple of cycles after I've pressed it back into place.  it is working away from the bolster first, it appears.  Not sure if the surfaces there (end of spring and bottom of bolster) are shaped in the wrong directions and making the tip of the spring want to walk away from the plate.

I think there may be a small twist in the spring, somewhere, but haven't been able to verify.  Could it be an inconsistent bend?  I need more lenses on my visor! 
For me, I see lots of potential failure modes in trying to locate and fix any twist in the spring...

Short of locating and correcting the twist, I'm almost thinking of putting a small notch on the bottom of the bolster close to the plate.  And then creating a small lip on top end of mainspring to catch in the notch in the bolster.  That would be about a 20 minute task, and I think I can do it. 

The gun has been functioning and shooting well for over 45 years now.  Just the first time Ive had the lock completely apart in about 10 years, but all this talk about lock tuning made me look at it more closely this time. 

[rich pierce]

I’m not a mainspring guru but often the peg that holds the mainspring in place is not 90 degrees to the plate when this occurs on a new gun, or the whole side that fits against the plate is not flat. In a well used gun the tumbler toe or mainspring toe could be worn and not square to the lock plate anymore. Just a guess.

[Bob Roller]

That lock may have developed a wear pattern on the OUTER side next to the bridle and the contact area of the mainspring also has this wear and will cause a matching angle.In the match.When I made the old style,slip and slide tumblers I would angle the contact area of the tumbler toward the lock plate and made the mainspring to match.You may be able to do this to the lock shown.
Bob Roller

[rich pierce]

That lock may have developed a wear pattern on the OUTER side next to the bridle and the contact area of the mainspring also has this wear and will cause a matching angle.In the match.When I made the old style,slip and slide tumblers I would angle the contact area of the tumbler toward the lock plate and made the mainspring to match.You may be able to do this to the lock shown.
Bob Roller

Great info there. That tip is a keeper for sure.

[Bob Roller]

That lock may have developed a wear pattern on the OUTER side next to the bridle and the contact area of the mainspring also has this wear and will cause a matching angle.In the match.When I made the old style,slip and slide tumblers I would angle the contact area of the tumbler toward the lock plate and made the mainspring to match.You may be able to do this to the lock shown.
Bob Roller

Great info there. That tip is a keeper for sure.

Rich,
It has been decades since I thought of this.I think I started making linked tumblers in 1970.Seeing this problem revived the memory in an
increasingly feeble mind.I polished BOTH contact surfaces and put a "speck"of STP on the tumbler to enhance the performance of the lock.
 also found that the linked mainspring and tumbler is easier to make.at least it was for me.
Thank you for the rapid response/comment.
Bob Roller

[taterbug]

thanks for the tips guys!  I was able to get my visor lenses stacked up so I could get a REALLY close look at things.  Looks Like I have almost all the things mentioned above. 

The end against the bottom of the bolster has a slight angle (wrong direction of course!), making the top of spring want to work away from the plate.

And the tip of the spring against the tumbler has the (incorrect again) angle, making the spring want to work out at the top.  I'm not able to verify if the tumbler toe is square to the plate, but the foot of the spring does not try to work off the tumbler while cycling the lock.  I see no wear or witness marks on the plate from the long arm of the spring, so I think I'll leave the tumbler alone for now.

I guess after 45 years or so, those things have worn the retainer hole in the plate a bit, causing the top of the spring to walk.  And the nub on the spring has worn a bit too.

I can fix the first two problems with a bit of judicious stoning.  Not sure I can fix the last one, or if I need too.  I'll see if fixing the first two makes the last one less of a factor.  My thinking is the hole and nub should not have to provide lateral retention of the spring, only up/down, fore/aft location.  Let me know if I'm off in left field with that thinking! 

if the spring still wants to walk after stoning a bit, I may go ahead with the notch on bottom of bolster with a slight lip on the top end of the spring.  It doesn't get shot too much these days, so not too much of an issue considering how it got to this point. 

 I'll let you folks know if my 'work' makes things better, or worse, or no change.  Not saying the advice isn't valid, only that my handiwork may not have the ability to fix it these days. 

Thanks again folks!!

[D. Taylor Sapergia]

The tip of the upper arm of the spring should be filed so that there is a vertical lip that fits into a corresponding notch cut into the bottom of the bolster at the plate. This cut can be made with a graver or die sinker's chisel, or with a rotary tool and an abrasive disc. The spring's tension against the cut will keep the top arm of the spring from working away from the plate.  That is usually enough to cure all of the issues you seem to be having.

[taterbug]

Taylor, that's pretty much what I did.  Well, only had to do half the job, as there was already a bit of an undercut ledge at the bottom of the bolster.  Just had to make if a bit bigger.  The rest was mostly clean-up of some 45 year old 'youthful enthusiasm' from when I originally built it.

Spring stays in place now.  it probably would have stayed in place the way it was, but 'whilst I was in there'...

Maybe it'll participate in the April postal match.  see how my wifes schedule fits me...