We had an interesting debate a while back on the subject of automatic winding movements, and in particular over just how good (or poor) the early “bumper” movements were. I contributed to this in bold defence of the good old cal 3xx series of half-rotor movements, but I have to say that a lot of your opinions were much more negative. Lots of users, and the watch companies themselves, seem to imply that you must fully wind even a modern automatic watch from the crown before using it (otherwise it will run down overnight, presumably). My idea that the watch might wind itself up (after being fully discharged) on wrist movement alone was greeted with some scepticism. And as far as the old fashioned ‘bumper‘ movements were concerned, the received wisdom often is that they were really very inefficient and could only “top-up” a regular hand-winding. So I decided to run an experiment in more controlled conditions.

The subject of the test was my 1953 Seamaster with a cal 354 providing the power -

The movement was completely run down and stopped. When I got up the next morning, I rocked the watch side-to-side five times to get the movement started, then I pulled the crown and set the time. No winding turns were made from the crown at all. The watch was then strapped on for a day’s wear, and the start time was 8.0 am.

Now of course comes the great unknown, and the factor we cannot control, which is the wrist movement of the individual wearer. My amount and nature of activity is bound to be different from yours. Nonetheless, I had what I consider to be an average day -

* The morning was mostly spent at my desk, working on my PC. I did have a break to unpack a delivery of some groceries.

* After lunch, I did a couple of hours in the garden, which included some hedge cutting and general tidying, sweeping leaves etc.

* Later I went for a nice walk around the lanes, about three miles.

* After dinner, I chilled out with a beer, a few jazz albums, and a good book.

The watch was taken off at 10.0 pm precisely and placed face-up on my desk – remember, it had had no winding other than from my wrist movement during the day, 14 hours in total.

Next morning (day 2) it was still going strong, and was spot-on for timekeeping. That evening at 10.0 pm, 24 hours later, it was still running. The next morning (day 3) it was still running fine, about 3 to 5 secs slow. It finally stopped a couple of hours later, at 10.0 am precisely (!), after 36 hours on my desk. Which means that the movement was quite close to being fully charged after one day’s wearing on my wrist and no winding input from the crown at all.

All of this confirmed more scientifically what I already knew, that I can wear this watch for weeks on end and never wind it up. I can even put it down and not use it for a day, and it will carry on the next day and never miss a beat. Now as far as I know, there is nothing special about this 354. I purchased it from Tim Mackrain, who serviced it before sale. So it does look as if Tim knows a thing or two about setting these old bumpers up . . .

It would be interesting if a few more forum members ran a similar test on their own favourite auto movements

John

 

Sorry, folks, my image attachments have got completely scrambled and I can't seem to edit them

There are only supposed to be two pictures . . .

 

Certainly bumper wind watches work, but they are quite inefficient when compared to most full rotor designs - I think this is the point that people were making. This is why no one uses the bumper design anymore.

"All of this confirmed more scientifically what I already knew, that I can wear this watch for weeks on end and never wind it up."

Not sure how scientific this test was. As you point out, one big variable is how much the watch is moved during the day, and how it's moved. This is why when we test watches after service, we use a test winder under very controlled conditions. Also the condition of the movement can have a huge impact on how long a watch runs - if it is dirty it won't run as long as when it's clean and properly lubricated.

Also, you state that it was near full wind when you took it off. How are you determining that? If you have not wound it fully by the crown and let it run down to see how long it runs (to confirm how long it runs when truly fully wound), then I'm not sure you can say if it was close to full wind or not when you took it off.

Omegas stated reserve for this watch is 38 hours, but often watches run far longer than the reserve indicates. For example a modern 1120 is rated for 48 hours, but they often run 50 or 51 hours. In addition, it's possible a mainspring of a different length was installed in the watch (in this case not completely unreasonable since Omega no longer supplies mainsprings for this caliber) so if someone put a spring in that was slightly longer than the original, it could give you skewed results - one reason why checking it against the reserve when wound by the crown will give you a better picture of how fully wound your specific watch was.

Now there are certainly full rotor designs that are not very efficient either - I have an old Gruen on my bench now that has quite a large dead angle - this is the number of degrees when the rotor is turning but the gears that actually wind the watch are not engaged yet, so no winding is taking place. Consequently it is not a very efficient winder compared to many other watches out there.

Now watch companies state that you should wind the watch using the crown before you put it on because they have no idea how active you are. So rather than this being a strict requirement in a technical sense, what it is for is to head off people who have no clue how an automatic watch works coming back to the dealer saying "my battery must be dead" because my watch isn't working...don't laugh, it happens more often than you would think.

Also, these watches are meant to be fully wound for best timekeeping results - one of the big advantages of an automatic watch is more consistent torque coming from the mainspring compared to a manual winding watch. If you start out with the watch fully wound, and wearing the watch all day keeps it fully wound, you won't have to worry as much about this:



Anyway, automatic winding can be quite a subject on the technical side of things, so just a few thoughts from me on this topic.

Cheers, Al

 

John, was the amount of caffeine figured in to the equation?

Clever. Always liked those games.

 

well, I suppose I could have been a bit hyper that day - we'd better add in five cups of tea and two brews of coffee, just for the record.

As usual, very perceptive comments from Al. I knew that my 36 hours from this test was close to a fully wound result, because I have indeed checked the same watch from 40 winds of the crown, and it comes out to just over 40 hours, so that's easy to remember. This is not my best bumper watch in this respect, my 354 Connie will beat that by a couple of hours.

Just to set the record straight - I am not recommending that you should not give a watch a manual wind in ordinary service. As Al says, you will get the best timekeeping accuracy from a fully wound movement. This was a rather special experiment to isolate just the automatic rotor and see how good it really was. And of course the full-rotor movements like the 500 series are even more efficient, as they wind both ways, for a start. But I wanted to challenge this idea that the bumper movements were really poor performers - they obviously were pretty good, in fact. And when you think about it, they must have been good in their day, otherwise how could Omega have competed against Rolex?

 

Hey John - not being too critical of your test - hey I didn't even mention it was a sample size of one! Okay I just did...sorry.

Anyway, just to give some feedback on the 36/40 hours (thanks for letting me know that you did that test BTW) if I had serviced a full rotor watch (modern or vintage) and found a 4 hour shortfall in the reserve after it being on the test winder, I would consider that a pretty big failure, and would be tearing it apart again to see what the problem is. On a bumper, not so much.

From a practical standpoint you can certainly not wind it up - it's your watch after all - but the bumpers are not as efficient as a full rotor watch.

And you have brought up another related topic in your last comments. There is certainly no firm conclusion in the industry that a watch that winds in both directions is more efficient than one that winds in only one direction. That pesky dead angle comes into play again, as well as the fact it must overcome the mainspring tension in both directions in a watch that winds in 2 directions, affecting inertia. There is evidence that for less active people, one direction winding actually may be more efficient, like in an ETA 7750.

Cheers, Al

 

Well, I did encourage others on the Forum to have a go themselves, so we could get a bigger sample with a bit of luck . . .

That is really interesting, and something I did not know. And of course if you only wind one way, you don't need the complication of reversing gears.

 

3 years or so ago I would have joined in. Now, I'm finding myself less and less interested in statistics unless they relate to rarity. Only once in the last year have I bothered to check a watch's accuracy, and that was only because I wanted to see for myself if the caliber 8500 lived up to the hype. (It did and then some.)

 

Several years ago I was asked to test the beta version of the Orbita Sparta (?) winder with a bumper movement watch. This winder is battery powered and works by swinging the watch once an hour. The watch swings for about 30-40 seconds each time. Basically the test was to let the watch wind down until it stops and then place it on the winder. All I can say is that the watch started working after the 1-2 wind and kept on going for a couple of weeks, until I took it off the winder. Not how significant this is, but what it told me was that this movement (cal 354) did not need to be hand wound before putting it on the wrist/winder. My .02 cents

 

This 354 serves me quite well, as long as it stays on my wrist, it needs no additional winding

 

I have 16 watches, mostly omega, all full rotor autos - 505, 552, 562, 564, 1128, 3313, and a few more. Every single one winds quite happily on the wrist from a dead stop, with the exception of my black dialled 561 which also has a worn crown which makes the initial wind a nightmare. Im not so precious about timekeeping that i mind a few seconds while it winds....

 

I've got a 351. Not a great pr but the last guy that worked on it was a hack. It generally lasts through the night but thats about it. I do find however that my bumper cv very sensitive to the auto wind and I barely have to motion it and it will start right up. My autos seem a second or two sluggish.

 

Both my 354s seem to wind just fine and keep their power reserve overnight after about a day's wear, similar to your experience I suppose. I find it takes more prolonged wear to get a full 24 hours or more out of them after taking them off. This usually isn't a problem as I love wearing them and feeling the "bump-bump" sensation as I walk down the stairs

 

When completely run down, I simply give my 354 bumper about 20 winds, and it then runs without any issues and keeps very good time. Once idle in the watch case, it will keep running for about 30 hours. Not bad for a nearly 70 year old piece!

 

I have noticed that carrying the Omega watch box (versus the not-Omega watch box) around will get my cal 355 going. It is freshly serviced and my watchmaker bushed the holes that elongate on this series of movements. I have not tested reserve, because I also wear these watches to bed.

But I do wonder how much the condition of the plates matters with respect to these bumpers?