I'm going to start this thread with "Dave taught me better"... several hours later... now I understand.

I had this 227 apart and back together several times before I found the smoking gun. I was so focused on the gear train from the center wheel out, I kept missing the problem.

This watch would run like a top, good beat error, consistent beats per hour in 5 positions, but then it would just up and stop at random intervals. 8 hours, 16 hours, 3 hours... I could nudge the center wheel with a piece of peg wood and she would run for several hours again... then just up and stop.

It turned out to be verdigris, however tiny, barely visible with the eye, but pretty obvious with a loop... it was built up in the valley of the teeth on the barrel. Since the barrel only rotates a few times in a day, the three deposits of verdigris would meet the center wheel pinion at odd and unpredictable times and stop the gear train.

When I did the (novice) COA, I didn't give the barrel much attention. It looked good at a glance. Now that it has received the attention it should have had in the first place... the watch is happy again.

Movement

NAWCO Security 10Kgf Rail Road Model

Interesting story Paul,
There's nothing more satisfying than resolving an issue such as this. And, I'll try to remember it when I'm trying to figure out what's wrong too.

Nice job Paul. You might want to check to see if the balance wheel has mean time adjusting screws. Your regulator is moved toward slow and I'm guessing that is where it is keeping decent time right now. Timing screws are usually located at the ends of the balance arms and sometimes 90 degrees from the balance arms as well. There may be 2 or 4. The screws will have threads longer than the other balance screws and are made to turn in and out and stay there. Frequently an inexperienced person sees the screws turned out and figures they got loose and turns them all the way in. This will make the watch run fast. So first, check to see if the balance wheel does indeed have timing screws. Next see if they are turned all the way in or have some adjustment available to them. If so, you can center the regulator which will make the watch run faster and turn out each timing screw an equal amount which will slow it down. If you have 4 screws, I would go 1/2 turn out on each and check it. If 2 screws I would go 1 turn out and check. Repeat until timing is very close. If everything else is set up properly this should get your watch timed to the point where the micro-regulator then needs only a very small adjustment one way or the other for final timing. NEVER attempt to make timing screw adjustments with the balance in the movement. Remove the balance cock and balance and put it on a tack and work on it there. This will save you from bending or breaking a balance staff.

Thanks Roger, I read up on the timing screws today and tweaked out a half turn on two screws... this made almost no difference in beat rate.

I strong armed Dave in to restaffing this balance wheel with only the pillar plate and balance cock while I worked on the bridges and wheels at home. He tried to tell me that timing would be an issue... It is way fast. 18090 plus BPH with the regulator at center. Down to 18035 dialed all the way to slow and backing the timing screws out a half turn only slowed her down another 5 BPH to 18030...

I don't see a balance spring issue... no out of place coils, no coils married due to oil or gunk... so, I think I may head back to Dave's shop with the complete watch for a tutorial.

Looks like main spring winding tools and tools to remove the hair spring so I can poise balances may be the next in line for my bench.

You're fortunate to be able to go to see Dave...I didn't realize you were that close. He'll definitely steer you in the right direction. You may need to add timing washers to add a little weight to the balance wheel or the regulator pins may be too close. Poising is very important if you want the watch to run well in different positions. It can be very time consuming but very rewarding when you get it poised right with the correct total weight on the balance so the regulator can be centered. Have fun and good luck.

I just realized that I should have added something very important in my first post that just slipped my mind. When adjusting the timing screws, NEVER do it while the balance is in the watch. Remove the balance cock with the balance and place it on a balance tack and work on it there. This will save you from bending or breaking a balance staff. I'm going to go up to that post and edit it so someone doesn't get into trouble. Sorry for my oversight.

Paul thank you for the mention of my work. Please let me add that I hesitated at first to do Paul’s staff at all only because proper operation of the watch really needs a systematic restoral INCLUDING any repairs to the parts. In this case we were lucky and the watch "came in" at only a couple minutes a day which can be very easily tuned up.

As for poising, when you re-staff a Balance wheel you must EXPECT poise(Balance) to go out the window because even the slightest change in the replacement staff pivot centers to center of Balance wheel hub rotation has a BIG effect on balance of a wheel that is probably already a little whacko anyway.

So I always re-poise a balance wheel as part of the re-staffing process because it is necessary for the watch to have ANY position accuracy at all and has to be done before re-assembling the Hair Spring.

Re-poising requires trimming too heavy or replacing too light balance wheel weights to bring the wheel back to poise and will always change the rate of the watch.

On the finished re-staffed watch, I treat errors of 5 or more minutes a day (18,124-17,876 Beats/Hr) with further attention to the Balance weights to bring the watch within range of the mean time screws, or in the absence of M.T.Screws (as with 99% of Elgins), even greater care to bring it within a few Beats per hour that can be adjusted by means of manipulating Hairsprings, Regulator pins or the micro-regulator.

Dave,

Like you, I always poise the balance when replacing a staff, but I wonder if it was common practice for this important step to be bypassed for the sake of $$$. I find so many watches running poorly in positions and it's almost always because of a balance that hasn't been poised with evidence of a previous staff replacement. It makes me wonder if there were a significant number of 'watchmakers' who just didn't take pride in their work or were in a production environment to push the work out the door. Any thoughts?

This is finally a happy watch. I had to come out 1-1/2 turns on all four screws, a half a turn at a time... I ended up at 17998 to 18002 in the various positions on my electronic timer...and the watch has not gained or lost so much as one second in the six hours I've been carrying it...

Thank you to everyone who had a piece of this project. I learned a ton.