IWC Calibre 52 regulation

[FuzzyBuns]

I’ve had this pocket watch movement for years and finally gained the muster to strip and clean it.  Turned out relatively well but now to regulate it, I have a cheap phone Timegrapher app that says +76s with a beat error of 3.1.  The only lift angle info I could find was 44 but nothing definite.  I’d love to regulate it myself so if you guys have some tips please share.  I’m not adverse to taking it in but probably won’t be able to until winter 

[VWatchie]

If your app works decently I'd say a beat error of 3.1 is excessive. You would have to remove the balance and rotate the collet. I believe I read somewhere that a beat error of about 0.8 ms is acceptable for a watch like this. Watch Repair Channel has at least one video demonstrating how make the adjustment of the collet. Perhaps you'd want to practice on a scrap movement first. As for the rate I see this movement is fitted with a swan neck regulator, so hopefully you'll just have to turn the screw clockwise to slow it down.

[watchweasol]

Hi VWatchie is correct follow that advise and you will be ok.

[rodabod]

What amplitude are you getting? The beat error is relative to this figure, and 3mS is not necessarily “bad”. There are other parameters which are more important. Remember you need to regulate the watch across positions, so at least test it in pendant-up position as well as dial-up. 

[jdm]

23 hours ago, VWatchie said:

If your app works decently I'd say a beat error of 3.1 is excessive. You would have to remove the balance and rotate the collet. I believe I read somewhere that a beat error of about 0.8 ms is acceptable for a watch like this. 

Not excessive for a watch like that. Brand new quality watches are often shipped with 1.5 - 2 ms. As correctly pointed out above, other parameters are more important.

[VWatchie]

3 hours ago, jdm said:

Not excessive for a watch like that.

It may, as @rodabod puts it, not be bad, but in my opinion it is on the excessive side and can be improved.

3 hours ago, jdm said:

Brand new quality watches are often shipped with 1.5 - 2 ms.

In my book that's definitely unacceptable. I would expect the B.E. of brand new quality watch to be less than or equal to 0.3 ms. and for an old watch lacking a mobile stud carrier less than or equal to 0.8 ms.

[rodabod]

Well, you have to consider the reasons why you want to aim for a beat error of less than a particular value. Is it the ease of ability to “self-start”? Or is it the potential to increase overall balance amplitude. Or do you believe it will significantly improve timekeeping?

In my opinion, the reason why people are so interested in it these days is because it is now so simple to measure via timegraphers. Also the fact that the term has the word “error” in the title. And the fact that on modern watches it is so easy to adjust. You don’t hear people talk about it in the same way as positional-error which is usually far more difficult to adjust. 

[anilv]

3 hours ago, rodabod said:

 

In my opinion, the reason why people are so interested in it these days is because it is now so simple to measure via timegraphers. Also the fact that the term has the word “error” in the title. And the fact that on modern watches it is so easy to adjust. You don’t hear people talk about it in the same way as positional-error which is usually far more difficult to adjust. 

Very true, I usually set it in beat by sighting the roller jewel with only the balance installed in the mainplate. Unless you're aiming for chronometer grade standards thats usually enough. If you want to to get the beat perfect you also need to get the balance poised and make sure the pivots are prefect, otherwise you'll be running in circles as the beat will change depending on position.

Anilv

[praezis]

Glad to read the good points made here.

If you contrast the danger of damaging the valuable hairspring against having a moderate beat error (that means almost nothing), guess what my advice is? 

On 7/11/2020 at 1:45 PM, VWatchie said:

... a beat error of about 0.8 ms is acceptable for a watch like this.

Why?

Frank

 

[VWatchie]

2 hours ago, praezis said:

Why?

 

On 7/11/2020 at 1:45 PM, VWatchie said:

I believe I read somewhere that a beat error of about 0.8 ms is acceptable for a watch like this.

From what I've picked up I believe that it is a good thing in general to strive for a low beat error and I think the reasons for this are pretty much covered in @rodabod's somewhat, shall we say "defiant" questions:

12 hours ago, rodabod said:

Is it the ease of ability to “self-start”? Or is it the potential to increase overall balance amplitude. Or do you believe it will significantly improve timekeeping?

When working with watches the question of what is acceptable tolerances can always be debated. From what I've learnt and personally find to be reasonable tolerances we'd generally want to strive for a beat error less than or equal to about 0.3 ms for a watch with a beat corrector arm and less than or equal to about 0.8 ms for a watch without a beat corrector arm (I guess because it's more difficult and time consuming to exactly adjust a hairspring collect). I think of these boundaries as a rule of thumb and something to strive for, but if it can't be achieved it doesn't mean the watch isn't working well enough or that it is an actual "error".

BTW, here are the watch repair channel videos discussing how to adjust the beat error on watches both with a without a beat corrector.

[VWatchie]

2 hours ago, praezis said:

If you contrast the danger of damaging the valuable hairspring against having a moderate beat error (that means almost nothing), guess what my advice is? 

Hmm... my guess is that your advice is to accept a "moderate beat error"  As mentioned in my previous post it is not always easy to determine what is an acceptable deviation. That is, to determine when an adjustment is more reasonable than not. For a watch with a beat corrector arm it only takes a minute or so to get a near perfect beat (provided the movement in all other aspects is healthy) so in that case I wouldn't refrain from doing it. So far I never had to adjust the beat of a watch lacking a beat corrector, but before I do I will practice it on my old scrap movements until I've gained enough confidence to eliminate as much risk as possible.

[praezis]

VWatchie, thanks for your detailed answer to my short question ("Why?")!

But in rodabod's quotation I saw questions only, not reasons (sorry, rodabod )

I tried to give a founded answer here.

Frank

 

[VWatchie]

39 minutes ago, praezis said:

VWatchie, thanks for your detailed answer to my short question ("Why?")!

My pleasure...

39 minutes ago, praezis said:

I tried to give a founded answer here.

and thanks for the link to the thread covering this interesting topic!

[jdm]

34 minutes ago, praezis said:

VWatchie, thanks for your detailed answer to my short question ("Why?")!
But in rodabod's quotation I saw questions only, not reasons (sorry, rodabod )
I tried to give a founded answer here.

Frank, when a real watchmaker like you, with many hundreds if not thousands, of successful repairs under his belt spends his precious time to explain something here, that is just generous and very appreciated. And I think 99% of the "advanced hobbyists" will think and do just like a professional.

[rodabod]

Sorry for the repeated posts which literally just contain questions, but they are genuinely there to try to help people (not more experienced people) understand things. I think it’s the only way to understand things fully. 

What I often observe is that many watchmakers (including some here) do work “by rote”. And some newer people learn from those same people. I have spent years studying watchmaking (including going down the BHI diploma route) and ultimately most of understanding how things really work is through theory..... But that does not mean that you necessarily need to know the theory to be able to clean and assemble a watch by rote without understanding how it really works. But I was always told that if you want to take it to a higher level, then you need a real understanding. I still have a lot more to learn.

There is sometimes what I would call “noise” on this forum, and I do wonder if people would worthwhile spending £10 on a good watchmaking book.

 

[rodabod]

Going back to the original post, if people really do want to investigate the effects of beat error, then do just that. Take a watch with adjustable stud and try it for a couple of days at ~0mS and ~3mS error with the timing regulated to the same rate, and see how it affects the operation. 

[nickelsilver]

Hmm, when I was in school we were expected to get beat error on fixed stud watches to under 1ms; when I was assembling complications for one of the big houses here they had the same criteria. Now part of my work is SAV for one of the old prestige names, they also expect under 1ms. In 6 positions. And a lot of their watches are ultra thin or very small diameter movements. Whether or not there is a real observable performance issue between 1 or 2 or 3ms error I don't know- I just correct it and get on with it. If it's a personal watch then it doesn't matter (it won't hurt anything at least); if you are charging money for it I would say correct it, at least to 1.5ms.