How much does lift angle really matter?

[14060 or 16610?]

Recently folks have been posting analyzer screenshots, opening up every parameter to scrutiny. One of these parameters is lift angle. Does it really matter if the lift angle is off by a bit? What are the effects? Can a dealer or watchsmith skew the rate and beat error significantly by raising or lowering the lift angle?

I have used 52 degrees as a lift angle on ETA 7750s for years, but recently it was brought to my attention that 49 degrees may be more in line with ETA specifications. So I conducted a little experiment using a recent recording of my now-sold 7750-based B13 Breitling movement (I save and archive recordings of nearly every movement I service). Documented below is the analyzer results using various lift angles:

First, at 52 degrees, which I have used for years and had been well documented as industry standard:

Next, I lowered lift angle to 49 degrees:

Finally, I tried raising to 55 degrees:

What I have concluded is that raising or lowering the lift angle by 3 degrees up or down has minimal effect on rate and beat error. At the most, about half a second and about 0.1 milliseconds respectively.

This info will likely cause you to parse even further the analyzer results that you receive.

 

[honeybear]

Thanks for the experimental data. I was under the impression that lift angle only really affected the amplitude reading.

 

[R2D4]

Cool experiment. Will they ever fully replicate a Rolex movement? That's what I want to know.

 

[Wiz]

I just did the same experiment with a 6497, taking him from 44K to 49K, and the rate changed by 5 seconds.

 

[14060 or 16610?]

I just did the same experiment with a 6497, taking him from 44K to 49K, and the rate changed by 5 seconds.

Interesting. Was that a 21,600 bph movement? Perhaps the lower-beat movements are more affected?

 

[Wiz]

Interesting. Was that a 21,600 bph movement? Perhaps the lower-beat movements are more affected?

Yes, it was a 21.6K, but I just did the same with a 18K and the difference was less than 1 second.

Weird.

 

[honeybear]

The other experiment to do is to keep the lift angle fixed, and disconnect and set up the measurement again, just to see how repeatable the measurements are with fixed parameters.

 

[14060 or 16610?]

I'm sure there is a mathematical answer. But that's where my interest ends. I think its safe to say that the effect is minimal, particularly if the lift angle is off by just a tiny amount.

 

[14060 or 16610?]

The other experiment to do is to keep the lift angle fixed, and disconnect and set up the measurement again, just to see how repeatable the measurements are with fixed parameters.

That is, in effect, what already happens each time the analyzer is run.

 

[dustin]

Good experiment.

Like the second reply, I always thought that lift angle should only affect amplitude and not affect rate and/or beat error. I'm actually surprised to see the amplitude change the way it did. I would think it would go down in one case and increase in the other. IE., at 49 degrees it should be significantly less (which it was not) and at 55 degrees it should be significantly higher (which it was). Perhaps there were other factors in play here also.

Just to clarify... my understanding is that lift angle = the number of degrees around the arc where the balance is in contact with the oscillator and that timegraphers measure the amplitude (total rotation of the balance) from the midpoint of that contact.

If I'm wrong about that, which I certainly could be as most of my knowledge is from reading on the internet, I would love to know the truth.

 

[14060 or 16610?]

Keep in mind analyzer readings are also dynamic. They are not absolute.

At the end of the day, if the lift angle is off by a degree or two I wouldn't worry too much.

 

[SSTEEL]

I know this is an old thread, but this thread came up in my Google search results when I too like many others before me were looking for lift angle data.

I also now are under the impression the lift angle is not that big a deal set at its default, 52 degrees.

For anyone else interested, here's a great video demonstrating how to easily calculate a watch movements balance lift angle...

[ame="http://www.youtube.com/watch?v=-Xgcck692js"]Classic Watchmaking: How To Determine The Balance Lift Angle Of A Watch Movement - YouTube[/ame]

I for one have to admit to being very new to timegraphers, only getting mine this week, and I was always under the impression that the lift angle was the physical angle of the watch on the microphone/stand. :facepalm:

I only learned earlier today that this is not the case after researching a little more, and of course, ending up here

 

[SSTEEL]

Oh and if anyone is interested, here's a comprehensive breakdown of the various balance lift angles by watch brand, and it's caliber..

http://pczw.uhren-mikl.com/downloads/gamma.pdf

Lift Angle is in the far right column under, Hebewinkel.

Here's an English version, which is more clearer...

http://www.awci.com/wp-content/uploads/watch-tech-guides/Lift-Angles//Wostep.pdf

Also, following on from the video I posted above, here's more detailed info on the balance lift angle, and related items of interest..

http://www.bmumford.com/mset/tech/tick2/index.html

The amplitude of a balance wheel can be calculated from the geometry of the escapement and the relationship of sounds within the tick. Specifically, the interval of time between the unlocking and the drop is used with the beat time and lift angle to calculate a value for the amplitude.

More mathematical breakdown...

http://books.google.fi/books?id=ZWq...CQ#v=onepage&q=watchmaking lift angle&f=false