Mainspring alignment issue

[LittleWatchShop]

Working on this Russian pocket watch today.  I put the mainspring in without realizing that the tab must not only align with the cap but the barrel itself.  It is not.

Should I remove the mainspring and try again, or is there a clever method of "winding" it into place without damaging the arbor???

[oldhippy]

You don't want to do any damage to the spring or the barrel. I would start again. 

[LittleWatchShop]

51 minutes ago, oldhippy said:

You don't want to do any damage to the spring or the barrel. I would start again. 

I did again and it is impossible to align with wound spring with the barrel and punch it in.  The only thing I think can work is to wind it by hand (no winding tool).  That way you can get the tab started in the right place. 

[Watcher]

One technique that I use on IWC barrels is to provide a stop for the bridle so that the tab will be in the right place. A hand held screwdriver is sufficient. Then wind the spring from the ratchet wheel side until the outer turn slips up to the stop and so into the slot. You need three hands or the barrel must be held in a clamp. My friend in OZ hand winds but I find this even more difficult.

[LittleWatchShop]

OK, I got lucky.  What I did was to wind the spring up to the tongue so that the tongue was still outside of the winder housing sufficiently.  Then I was able to orient the barrel to align with the tab.  Then very carefully, I pressed the spring into the barrel.  I don't think this technique is repeatable...like I say...I think I got lucky.

Here is the winder I used for this exercise.  My other winders were too small.  Always wondered why Dad had so many winders...now I know.  Had never used this one before today.

[JohnR725]

preferred way of inserting any of the T end mainsprings is to wind it into the winder. but leave a little bit out then you insert the winder with the little bit stuck out into the barrel rotate until the T part is over whatever it goes into. Then push it in and hold it in there with the end of the biggest screwdriver you have or the backside of the tweezers you have to hold it in place. Then you can push the spring out if you don't hold it in place it will almost always jump out and no you can't manually rotated and get it to magically go in place.

[nickelsilver]

What John describes is textbook and proper. What happens with smaller calibers is proportionally you can't leave enough outside the winder to do it without distorting stuff.

On that size barrel I try to get the spring in so the T is a bit behind the slot, from the viewpoint of outside end of spring. Then, holding the barrel firmly to a bench block with a leather faced stick (leather buff, but a clean one), you can pull the end of the spring along with brass tweezers until the T drops in. Going the other way doesn't work,  you have to "pull it around ", not push back.

[LittleWatchShop]

1 hour ago, JohnR725 said:

preferred way of inserting any of the T end mainsprings is to wind it into the winder. but leave a little bit out then you insert the winder with the little bit stuck out into the barrel rotate until the T part is over whatever it goes into. Then push it in and hold it in there with the end of the biggest screwdriver you have or the backside of the tweezers you have to hold it in place. Then you can push the spring out if you don't hold it in place it will almost always jump out and no you can't manually rotated and get it to magically go in place.

Yup this is almost exactly what I did.  I did not try to hold the T in place.  Just aligned it and then pushed the mainspring in.  Like I said...I think I got lucky.  Next time, I will add the screwdriver element into the solution.

[LittleWatchShop]

The watch is all together with terrible amplitude.  Gives me a new problem to solve and develop some skills.

[JohnR725]

5 hours ago, LittleWatchShop said:

The watch is all together with terrible amplitude

you're not going to share a picture of the timing machine? We love pictures of timing machine printouts

[oldhippy]

8 hours ago, LittleWatchShop said:

The watch is all together with terrible amplitude.  Gives me a new problem to solve and develop some skills.

Have you got the correct mainspring? 

[LittleWatchShop]

The watch is all cleaned and put back together.  I put it on the timing machine which automatically detects beats and it settled on 18,000 but showed losing 600 seconds a day.  It was late, so I put a minute hand on it to monitor overnight.  Seems to be keeping accurate time, so I think the timing machine is fouling up on the beat test.  I can enter manually, but there are many possibilities.

In my journey, I am beginning to appreciate the different watch designs and which ideas are good and the ones that are bad.  This watch was particularly helpful at illustrating bad ideas.  In particular, the top jewel plate (one shown here on the balance) that has no impression in the plate to secure its location is a bad idea.  Once in place, perhaps it is fine, but setting it is a real pain because it slides around as you attempt to set the screw.

[LittleWatchShop]

9 hours ago, oldhippy said:

Have you got the correct mainspring?

Not sure how I could know.  I can use the mainspring equation to determine if it is the right length, but there is no way to know about strength.  I don't have any Russian friends.

[JohnR725]

1 hour ago, LittleWatchShop said:

Not sure how I could know.  I can use the mainspring equation to determine if it is the right length, but there is no way to know about strength.  I don't have any Russian friends.

 

it's amazing what you can do with the search engine online. Or just going the first site and entering in numbers like 3602 and disregarding everything is physically too small.

Looks like from one other reference and are not giving you that it's probably an Elgin mainsprings C are limited with what you're going to do. Mainly because of the T brace type end.

oh and the last link as usual watchweasol found the tech sheet heads at the very bottom.

 

5 hours ago, LittleWatchShop said:

I put it on the timing machine which automatically detects beats and it settled on 18,000 but showed losing 600 seconds a day.  It was late, so I put a minute hand on it to monitor overnight.  Seems to be keeping accurate time, so I think the timing machine is fouling up on the beat test. 

this is where it be nice to have on oscilloscope and see what the waveform looks like that's making the timing machine unhappy.

Then it appears to be that you may have an over coil hairspring?

Then one other little thing instantaneously on the timing machine it looks like it's going slow by 600 seconds but what is it doing overnight at the very end? Is the problem with the timing machine telling you at that second that you have a on its doing this and you still didn't give us a picture to look By the way of the timing machine. But that doesn't necessarily mean that's actually what the watches going to do unfortunately.

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Molnia_3602

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Molnia_3602_H3_5_2_UF

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Molnia_3602_H3_5_Chel

https://17jewels.info/movements/m/molnija/molnija-3602/

 

 

 

[LittleWatchShop]

44 minutes ago, JohnR725 said:

it's amazing what you can do with the search engine online

OK, I register being thumped.

44 minutes ago, JohnR725 said:

Then it appears to be that you may have an over coil hairspring?

Yes.  I removed it so that I could properly clean and oil the jewel.

44 minutes ago, JohnR725 said:

Then one other little thing instantaneously on the timing machine it looks like it's going slow by 600 seconds but what is it doing overnight at the very end?

It kept good time overnight.  I took a look at the waveform on my oscilloscope and I confirm that it is 18,000.  The timing machine may be picking up some other noise or just screwing up...dunno.  But I think that the timing machine makes a determination of beat rate and then only samples in those time windows.  I played around with gain to see if I could clean things up.  No luck.

Edited July 8, 2021 by LittleWatchShop

[JohnR725]

the reason I asked about the over coil is you need to make sure the regulator pins are as close as possible together. Flat hairsprings they can be spaced out or supposed that little the spacing over coils are supposed to be as tight as you can get them as long as they hairspring will still slide.

Then the reason I was curious about the oscilloscope was it be nice to actually see it I wasn't thinking about the frequency I'm just curious about what it looks like.

this is where some of the software machines beat out the physical in that there's other ways to look at things and then you can see if there's a noise issue or a pickup issue or something else.

On the other hand the reality may be that these are just noisy movements they're not necessarily known for making super precision watches and noisy movements can be issue for a timing machine that's expecting a little quieter ticking sound.

[LittleWatchShop]

1 minute ago, JohnR725 said:

Then the reason I was curious about the oscilloscope was it be nice to actually see it I wasn't thinking about the frequency I'm just curious about what it looks like.

I will take a picture later today.  Unfortunately, I have to dedicate the next couple of hours to my real job!  Not as much fun as this.

[LittleWatchShop]

Here is one screen shot

[LittleWatchShop]

2 hours ago, JohnR725 said:

make sure the regulator pins are as close as possible together

They appear to be.  I observed this yesterday without appreciating why.

[JohnR725]

34 minutes ago, LittleWatchShop said:

Here is one screen shot

my bad I forgot something? All the timing machines why see anything that resembles our oscilloscope has had a filter applied to it all the lower frequencies are gone below 1000 cycles which would clean this up considerably hopefully.

[LittleWatchShop]

34 minutes ago, JohnR725 said:

my bad I forgot something? All the timing machines why see anything that resembles our oscilloscope has had a filter applied to it all the lower frequencies are gone below 1000 cycles which would clean this up considerably hopefully.

My circuit does some bandwidth limiting.  It is not the one everybody talks about on this forum...it is my uber simplified design.  Regardless, I am a little surprised the timing machine has a problem with it...I will study it.  Mysteries of the universe...

[JohnR725]

I don't suppose you have an actual timing machine? It be nice to see what the graphical display looks like.

 

I find that work with their nifty $10,000 witschi machine yes I think that is about the price between the timing machine in the automatic microphone. In a case the oscilloscope is interesting because anytime you get weird timing stuff you to look at the oscilloscope and see if you can try to figure out the problem. A lot of times will see amplitude variations because the locking isn't nice something and you'll see of the timing machine can't make a decision as to where locking occurs and it keeps moving that around it shows you on the oscilloscope where it thinks locking is occurring. So you'll see the numbers fluctuate between a low number in a high number.

Because this is a not super nicely made watch as a guess if all of the ticking sound is not nice is not really a technical term but if it's not nice the timing machine may have issues or probably will of issues picking it up and giving you a good number or a decent looking graphical display.

6 minutes ago, LittleWatchShop said:

It is not the one everybody talks about on this forum.

then that's fine because I don't use the circuit that everyone else uses on the form either.

[Malocchio]

On 7/7/2021 at 10:16 PM, LittleWatchShop said:

I did again and it is impossible to align with wound spring with the barrel and punch it in.  The only thing I think can work is to wind it by hand (no winding tool).  That way you can get the tab started in the right place. 

I just had this same problem. Zenith pocket watch and mainspring with hole in the end and tab near the end, "Special Bridle" as it's called. The end was impossible to align while winding by hand because the end moves when the coils are wound in the barrel. I finally managed to do it so that I left the the end of the spring a couple mm behind the tab in barrel wall and after I had wound a couple of coils in I pulled from the arbor end until the coils slipped against the wall and the spring hole snapped in the barrel tab. Then I push the spring against the barrel wall so that the spring tab went in the barrel hole. I hope someone understood