If you have ever used a timegrapher to measure the timing of a mechanical watch, you will have come across the term ‘lift angle’.

In a Swiss lever escapement, the lift angle defines the arc that the balance impulse jewel sweeps between its initial contact with the notch in the pallet fork and the point that it breaks contact on the other side.

The timegrapher measures the rate at which the watch is running by recording the timing of the sounds that the escapement makes as the balance wheel, pallet fork and escape wheel perform their synchronised dance.   The machine has a sensitive microphone that can detect the beat noise associated with each tick and tock of the movement.  Each of these percussive sounds derives from three dominant sounds associated with 1) the impact of the impulse jewel as it makes its initial contact with one side of the pallet fork, 2) the sound of the exit pallet stone as it releases from a tooth on the escape wheel and 3) the almost simultaneous meeting of one of the teeth on the escape wheel with the locking plane on the entry pallet stone and the contact of the pallet fork arm with one of the two banking pins.

The timegrapher uses the time between the first and third pulse in each tick or tock in its calculation of the amplitude of the watch balance.  That time is the time it takes for the balance to sweep out an angle equal to the lift angle.

The machine will then use that time to calculate the total amplitude using a formula that takes account of the beat rate of the movement in beats per hour (for a beat rate in beats per second, the factor of 3600 in the numerator becomes redundant).   

For example, for a movement running at 18,000 beats per hour with a lift angle of 52°, and a measured interval between the first and third pulses of each tick and tock of 12 ms, the calculated amplitude will be

The amplitude of a watch movement is defined as the maximum displacement from the mean position of the balance wheel and so the actual total angle of rotation of the balance between each turning point is twice the amplitude.  In the example described above, the balance wheel will be swinging through an angle of 552°.

The animation above shows a watch balance swinging through 390 degrees total angular displacement which corresponds to a displacement from the mean position of 195 degrees (the amplitude).

The most common lift angle in mechanical watch movements is 52 degrees and this is the default value used in many modern electronic timegraphers.  However, only about 36% of the mechanical watch movements for which records exist have a lift angle of 52 degrees which means that the comfortable majority do not have a lift angle of 52 degrees in spite of the fact that this is the most common value. 

Lift angle values can vary from as low as 34 degrees (Citizen 66) to as high as 70 degrees (Piaget 25P).  Seiko movements span a somewhat less wide-ranging 47 to 58.4 degrees with about 35% of Seiko movements using a lift angle of 52 degrees.  The highest value in the Seiko list belongs to the Bell-matic 4006 movement but it is worth noting that there is some doubt about whether this value is correct.

When measuring the timing of any mechanical watch movement therefore, you can take one of two initial approaches:  you can assume that the default lift angle value is likely to be close enough to the actual value for your movement, in which case the measured amplitude is likely but not certain to be ‘close enough’ or you can look up the documented lift angle for your movement from any one of a number of databases that are available on the internet.  See for example: https://watchguy.co.uk/cgi-bin/lift_angles.  However, these lists are by no means complete and as we shall see, are not guaranteed to be correct. 

In surveying the list of Seiko movements for example, I would note that many of the early 1960’s hand wind movements are missing, such as the Marvel, Crown, 57-series Grand Seiko, the early King Seikos and later 44 series, as are the high beat 45-series of hand-wind movements as well as 51-, 62-, 76-, and 83-series of automatic movements.  The high beat 6145/6 are absent too although both 6156 and 6159 are present, but with different lift angles, in spite of their close relationship to each other.  Finally, I would note again the suggestion that the value of 58.4 degrees for the 4006 may not be correct.

With all of that in mind and prompted by a recent dalliance with a Seikomatic 8306, I thought it might be instructive to set about measuring the lift angle for a representative of the 83 series as well as a Bell-matic 4006A.

The approach that I propose taking is one that I employed previously in measuring the amplitude of a Seiko EL-370 electronic watch (see here).  The plan is to film the running movement in slow motion at 100 frames per second and with a fast enough shutter speed to capture the two turning points at which the balance is briefly stationary.  There are two methods you use to determine the lift angle when employing this approach:

• the first is to gradually wind power into the initially exhausted movement until the watch is running at 180 degrees amplitude (i.e. the balance is swinging through 360 degrees in total).  At this point, you take a measurement on the timegrapher and adjust the lift angle setting on the machine until the indicated amplitude reads 180 degrees.

• The second method does not require you to achieve a particular state of wind but instead, just to wind the movement to an arbitrary degree of reasonably high power – this could be a full wind – and to measure the total swept angle from the position of the balance wheel at its two turning points, as captured in the slow motion footage.

I will be taking both approaches with the first of our two guinea pigs, a 8346A automatic movement fitted to Seiko Business-A from 1967 (see here).  I removed the auto-winding weight and released all the power from the mainspring.  I then marked the balance wheel with a Sharpie permanent marker pen and slowly wound in power until the watch was running as close as I could manage to 180 degrees.  Here is the 100 fps footage captured as I approached 180 degrees. 

With the movement running at just slightly higher than 180 degrees amplitude, I captured two still images at each of the two turning points.

I then measured the difference between the two positions using a protractor app on my phone.

The measured angle here of 2 degrees 31 minutes is the angle beyond the total swept displacement of 360 degrees resulting in a total displacement of 362.5 degrees.  The amplitude is obtained by dividing this value by 2, resulting in an amplitude of 181.25 degrees.  I would suggest the error here is at least ±1 degrees so let’s call it 181 degrees.

The timegrapher was reading 174 degrees at the default lift angle value of 52 degrees and so I increased the lift angle setting to 53 degrees and took another reading:

I reckon that’s close enough to suggest that 53 degrees is about right for the 83 series.  Let’s wind the watch to full power, let it settle and repeat the process.

The much greater amplitude has meant that the Sharpie mark on the balance rim is obscured by the balance cock but I can still capture the two turning points as the basis of a measurement of the total angular displacement.

Once again, the protractor app on my phone does the trick:

This time, the measured angle of 22 degrees 48 minutes is the angle beyond the total swept displacement of 540 degrees, the balance having rotated more than a full one and a half turns between each turning point.  This measurement yields a total displacement of just shy of 563 degrees.  Dividing this value by to gives an amplitude of 281.5 degrees.

At the point at which I was recording the video described above, the timegrapher was reading 279 degrees at the adjusted lift angle of 53 degrees but had been wandering between about 274 and 283 degrees.  Given the obvious scope for small errors, this would seem to validate the earlier conclusion that the lift angle for the 8346 is 53 degrees.

The second movement to come under scrutiny is a 27 jewel 4006A fitted to a Seiko Bell-matic 4006-7011, dating from June 1970.  I’ve had this watch a very long time and it’s not been serviced since about 2010 but it still runs very strongly with decent amplitude and very low beat error.  Once again, I marked the balance wheel and then started gradually winding in power to the mainspring until I had achieved around 180 degrees.

It is really quite difficult to eyeball exactly 180 degrees but the static images of the two turning points suggest we are close enough for the purposes of this exercise.

The measured angle beyond 360 degrees total angular displacement is 5 degrees 26 minutes.

 This translates to a measured amplitude of just under 183 degrees.  The timegrapher measurement at its default lift angle of 52 degrees showed the following:

Pretty much bang on the nose.  It would seem that contrary to the received wisdom, the Bell-matic lift angle is not in fact an exotic 58.4 degrees but a rather more humdrum 52 degrees.

This method is a bit hit and miss but if conducted with care can probably be used to deduce lift angles to an accuracy of ±0.5 degrees.  My method could do with some refinement – a finer Sharpie pen to allow more accurate measurement of the turning points; a faster shutter speed slow motion capture to better freeze both turning points; but ultimately I don’t think that matters a great deal.  Differences in lift angle of a degree won’t make a big enough difference to timegrapher-measured amplitudes to matter much but differences of 2 or more degrees in the lift angle probably do matter and this is where it is important to know your movement’s lift angle AND to have faith that the source of that information is reliable.

With all of that having been said, it is somewhat ironic that for both of the candidates chosen for study here, the lift angles are either the same as or slightly higher than the default.  However, in one case, that figure is a full 6.8 degrees lower than the documented figure and in the other, there was no documented figure.  So all in all, a worthwhile exercise.

More content will come after this long hiatus.  I’ve not lost my appetite, just too many of life’s distractions elbowing their way into my available disposal down-time.