Q&A: Wind Indicators and Mainsprings

Q: A grade 454 Elgin with a wind indicate that reads a range of 0 to 40 hours only runs for about 36 hours on a full wind. Is this a mainspring issue?

A: With few exceptions, regardless of whether a watch has a wind indicator, the run times are designed-in the same, with the same requirements.

A watch stops when the power of the mainspring can no longer push through the friction of the mechanism, which is never zero. Exactly when this happens depends on many factors of the condition of the watch and the mainspring. The designers of the watch intended it to support fully winding and setting once a day, generally at the same time each day. And they intended that it still be running at that time. So "normal" is 24 hours plus some slack.

In my experience, the vast majority of usual and typical condition vintage watches run in the ballpark of 28-36 hours on a full wind.

As far as improving that...

The strength of the mainspring isn't the first thing to go after. If you think about it, all that's happening is the spring got turned X turns, it will "un-turn" X turns to move the hands. Power is slowly released by the escapement, but really it's just geometry; X turns in, X turns out. There
are no wasted turns. Even if a watch runs for 3 hours, no turns just disappear. It will just take that much fewer turns to wind it fully again.

In a perfect world. wasted power through friction would be close to zero. In this case the mainspring would need almost no strength to turn the mechanism. It's just about the number of turns. In practice, it really comes close to this. A fair condition watch, properly serviced, can be powered for well over 24 hours even by a nearly dead mainspring.

You can't gain more turns (and longer run time) unless you use a longer, thinner, and weaker spring, so that more turns can fit in the barrel.

A stronger mainspring, that turns the same about, may actually be problematic for an antique, that already has some wear on the mechanism. It will force the bearings to power through friction, more rubbing and grinding, on slightly damaged surfaces, causing more damage, where the watch would have otherwise stopped earlier.

A really unfortunate hack sometimes used in the old days by "botchmakers" was to get a watch to run by replacing a mainspring with a stronger one than intended - thus forcing it to tick. The replacement spring would be thicker, so fewer winds would fit in the barrel, so this was at the expense of run-time. Worse though, it was forcing the mechanism to run under more force than intended, potentially causing damage. Also, some original, actual, problem was left uncorrected. I've actually seen this in watches a few times.

But the short answer is that just because a watch has a wind indicator doesn't mean it's supposed to run that long. In fact, the wind indicator is one more little set of gears the thing has to power - more friction.

I usually set wind indicators so their hand is at zero when the watch is fully run down. In that way it likely does not fully wind up to its maximum reading. That way, the indicator truly shows run-time remaining as it was intended.

Waltham Bond St 1888 Model

Here are the components of Waltham's motor barrel. This same design is found in many Waltham watches, or various sizes.

The two halves do not fix together, they move separately. The arbor turns freely in the outer half (right) and sits in the square hole in the lower, inner half (left).

This watch, as found, was missing a spring in the winding setting mechanism. It was also missing the screw that holds the spring. It took awhile to find these parts.

 Here are the missing parts.

And here are the missing parts in place.

These old Walthams have a really flaky and difficult design for the winding/setting parts, or keyless works. Unlike other watches of its era (which are "negative setting"), the "snapping" function is in the movement, not the neck of the case (so "positive setting", as in modern watches). It's really difficult to get the stem in, and parts are extremely prone to wearing out and breaking - which is probably why this watch was missing its setting spring. Also, these have to go with a specific watch case.

Anyway, there's a problem here. In this photo, the clutch is outward in its winding position. The stem is snapped out.

The way this winding/setting mechanism works, there is a clutch that slides inward and outward. Outward, it engages a pinion at that end that winds the watch. Inward it engages a gear that turns the minute wheel to set the hands.

A lever is moved when the stem snaps in and out. The lever pushes a spring back, to push the clutch in, when the stem is snapped out. When the stem is pushed in, the lever release the spring and the spring pushes the clutch out into winding mode.

Or it is supposed to.

In this picture the stem is in. But the spring does not push the clutch out far enough. There is a visible gap between the clutch and the pinion. The watch will not wind.

So what do we do?

One thought is to use a smaller stop screw - the screw that the inner part of the lever bumps into to stop the snap out, and push the spring in. I checked and this would work. I could have used the lathe to make the screw head a hair smaller in diameter and the the spring would the do its work.

There's three problems with these solution though, 1) never alter existing parts to make a replacement part (the spring) work. And 2) never do anything that is not reversible.  Strictly speaking this is reversible since the screw could be replaced, but still...

Perhaps the worst problem though is that then the stem would actually snap out just a hair further. Although that's a small change, it's a can of worms I want to avoid. Who knows what happens next, the stem has to be shortened so the crown doesn't stick out too far?

So I dug through my Waltham parts and found a few pieces set aside from this grade. It is the same bunch of stuff that didn't include the missing spring that delayed this project (aside: when you have a pile of watch parts from an estate or something, the most needed parts are less likely to be there because those are the parts most often needed, so those have already been used. If you want to know what parts in a watch are reliable and durable, check the spare parts that you have to most of). My parts for this grade did include a clutch. Although this clutch looks just like the one in this watch, I know from past experience that Q/A in the 19th century was not what it is today. Old parts are very often slightly different from each other.

I installed the alternate clutch, and as you can see here, for what ever reason, it fully engages the winding pinion, just as I'd hoped. I suppose it is just a hair longer, or maybe the grove is in a little different place. It was made on a different day, and maybe the machine got set up just a little differently, and it works in this watch.

Ticking away!

Elgin Grade 75, With Before Images

Here are a few "before" photos I took while disassembling this next watch. This isn't the worst I've seen, but I would say it is well in need of cleaning.

Here we see the Swiss style, perpendicular, pallet. This escapement design is style by far the most common in mechanical watches today.

The upper plate has a very wide opening for the barrel. Installing it is no trouble at all.

The secondary serial number stamps are prefixed with an 'E' like symbol.

This watch case had a problem that makes for a chance to show how a typical stem-setting case works.

Most American watches use a "negative setting" system. The default position is for setting mode, but a sleeve-like spring in the neck of the case allows the stem to "snap" inward and be held in winding mode. The snapping is entirely a function of the case.

On this one, the sleeve was too weak and worn and it was possible, without taking care, to pull the stem all the way out of the neck. None of the parts are broken, just too worn. The sleeve and stem can be tricky to replace. One advantage of the negative setting system is that cases are more or less interchangeable. These interface parts are often custom made or altered to fit a given case/movement combination. Luckily, I found a sleeve that is the same part, but much less worn. This will work a lot better. It's lucky because this sleeve is unusually large. It's an odd type.

By the way, "positive setting" mechanisms are usually found on Swiss, and modern, watches. On these the snapping mechanism, and the stem, are part of the watch movement and the case just has a plain tube out the the crown. The default mode on a positive setting keyless works is winding.

Winding or Setting Slips in a Pocketwatch

In most American pocketwatches that are stem set there is a clutch mechanism that moves toward the outer end (second photo) for winding, and toward the inside (first photo) for setting. A "setting lever" moves the clutch when the crown is pulled out or pushed in. There are springs and another lever or two, to press the clutch forward or back to snugly engage at either side.

If all this is working well, fine, if not then the watch may not securely engage at one side or the other and may "skip" or "pop" as it winds and sets as the gear teeth slip.  This is bad because the gear teeth will be being exposed to unusual wear and stress, and will fail before long.

Often the parts that slide the clutch themselves wear at the points they push on other parts. This is the case with this watch, and when that happens the mechanism ends up not pushing some part far enough, and the gears end up not fully engaged.

Sometimes, it is possible to "stretch" the metal at the tip of such a worn part by using the staking set to place a little divot near the edge. It's not really a "fix" because it leaves the part thinner and less precisely shaped, but using a new replacement is not always possible. And in this case, that fix had already been done though, long before I got there.

These are very hard parts to replace with new...  But the long springs that push on these parts are usually secured with a screw, and when that is loose, it is possible to turn the spring just a little so that it will push more, or less. By doing this, over and over, and adjusting the depth of the stem in the watch case, it is sometimes possible to split the difference and get the watch both winding and setting.

The engagement is not 100%, but a very slight push inward on the crown, witch is natural when winding anyway, seems to hold it enough to engage the beveled pinion and start winding. Once there is a turn's worth of power, the mainspring tension is enough to hold the clutch and it feels fine.

The watch is a Hamilton 947 16 size pocketwatch. A winding/setting mechanism like this is very common in American watches.

Stem-Setting and sleeves and Detents and Pocketwatch Cases

Earlier American pendent-set pocketwatches have the mechanism that snaps in and out as part of the case.  In the neck of such a case there is a "sleeve" with spring fingers that grip the winding arbor.  A shoulder on the arbor snaps to one side or the other of the fingers when you pull it out or push it in.

Later watches have an improved design where the snap is part of the movement instead.  The winding arbor is free in the neck of the case, but held in the movement by something called a "detent" which acts as a lever, changing from winding to setting.  This is how the vast majority of mechanical watches work to this day.

This pocket watch has a detent.  Interestingly, I found the broken remains of a sleeve threaded into the neck of the case.  This case may be older than the movement, or perhaps it was designed to work either way.  But it does seem that at some point an older style movement was in this pocketwatch case.

I included in this photo a normal (not broken) sleeve for an example.

Improving a Loose Crown

This watch case has an interesting issue.  There is a lot of play up and down in the crown.  It's a lever-set movement, so it does not impact the watch functionally, and I don't do much case repair, but I'm going to try to improve this one.

This case is an older style, having a screw on cap over the neck of the case.  The sleeve and stem just sit loose in there, held in by the cap.  In this instance, the cap does not screw down all the way on to the top face of the neck.  When the cap is on all the way, it leaves some space.  You can see this in the photo by comparing the height of the cap with the length of the threaded portion of the neck on the case.  The sleeve and stem assembly can move up and down inside the space left inside the cap.

Everything else works and fits.  If a new, shorter cap could be found, then the stem would have to be shortened because the crown would stick out too much.  As a general rule, we never alter existing parts to match a replacement.  So what I think will help to to add a spacer inside the cap to fill the excess space.

It's hard to say why this case has this situation.  I suppose it's most likely that the cap part is not original to the case and perhaps a shorter one was. However, this cap is definitely an old part and is typical of cases like this one.

The first order of business here is to drill out a hole in a brass rod.  The rod is the same diameter as the sleeve.

The hole drilled, this brass rod is now a tube.  It gets cut off, a bit too long so it can be fit carefully.

Here's how the new part will go.  It's a quick job to return it to the lathe to cut is down a bit and round over the top edge so as to conform neatly inside the cap.

Here is the finished and cleaned up spacer and the cap, separate and in place.








The case finally assembled, the cap snug against the top of the neck as before, but now with no free space inside and no up/down play.

Elgin Grade 100, A Convertible Movement with Animation

Here's the bottom plate, dial side, of an Elgin grade 100 pocketwatch, convertible. This unique movement can be assembled in either hunter or open face configurations.  The difference between the two is that the stem is 180 degrees from the seconds dial (the forth wheel) in an open face watch, but 90 degrees from the seconds in a hunter case movement.

This plate allows the winding arbor to be in either of two positions, 90 degrees different from each other. 

In this image we can see a linkage of two similar parts connecting both stem positions to the "snap" mechanism for the crown to snap in and out.  Among the many unique features of this movement, the "snap" is here in the movement, rather than provided by a sleeve spring in the neck of the case.
The "snap" is provided by the business in the upper left.  The hunter position for the winding arbor is directly pointing left.  The open face position is facing down in this image.  The jewel for the lower forth wheel, which is the seconds hand, is at the upper most in the image.

The Elgin convertible models have the ratchet in between, inside, a two part barrel bridge.  The trick to converting from hunter to open face is that a main wheel, which turns the ratchet wheel, which winds the mainspring, can be positioned at one side of the other inside the bridge - a 90 degree difference.
This watch is being assembled in open face mode.  The stem is facing down in these images.  The unused void for the hunter configuration is to the right. 

Adding to its uniqueness, the winding mechanism is on the top, or back side, of the movement, opposite the dial.  

When in setting mode, a pin pushes up from the other side and lefts an arm, causing a main wheel to move up, off of the ratchet wheel, and engage a wheel in the center.

The cannon pinion on the other side rides directly on the arbor of this exposed center setting wheel, thus turning the hands.

Find more examples of Elgin's convertible movements here...

Elgin Grade 303, The Winding/Setting Clutch Mechanism

Here are a few shots showing the winding/setting clutch in a typical American pocket watch, an Elgin 12 size in this instance, a grade 303.

The first images show the normal position, at rest.  The mechanism has a spring that has pushed it into winding mode.  Node that the clutch is pushed upward, engaged with the minute wheel just below the center of the movement.  In this mode, turning the stem turns the minute wheel, and thus the hands.  The hands slip on the center shaft when this happens.

The watch case has a square arbor that engages the mechanism. It snaps in and out on a spring inside the neck of the case. The spring holds the arbor pushed in, pushing the arbor into winding mode.

In the last image that shows the movement in the case, look closely at the clutch. The arbor, now pushed in, actuates a lever that pushed the clutch down, off the minute wheel, and engaged with the winding pinion at the lower edge of the movement.

The clutch moves in the opposite direction from that of the arbor in the case.  That is, when the crown is push inward, the clutch lever presses the clutch down against the winding pinion.  When it does this it is working against the spring in the watch that wants to hold the clutch against the minute wheel

Problems develop if the spring in the case neck, called a sleeve, is broken or not strong enough to hold the arbor inward against the spring in the movement, or if the arbor in the case does not go into the movement far enough, or too far, for the clutch to fully engage in both modes.