Hairspring probably hitting, but then what?

[saswatch88]

2 hours ago, spectre6000 said:

When I get back to the bench, I'll pull the mainspring and measure it out and see how it compares to the spec on ranfft. 

YESSS! Do that so you can see if what’s in there is correct before you go killing yourself trying to get a weaker MS.

[spectre6000]

Oh. Interesting thought. 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Certina_304

The movement shows up on ranfft as being 17 jewels, and no doubt later (I'm assuming later...) versions were. Mine is only 16 jewels. Mayhap that's part of the problem. It's only one jewel, and in about the lowest impact location it can be, but it's not nothing.

Prior to service, I put it on the timegrapher, but didn't pay that close attention. It was running only in a binary sense (1/0 or True/False; no quality dimension implied). I think amplitude was on the order of 180°, rate was way off, but I don't recall how far or in what direction. 

I will definitely investigate what I have before I order anything. If it's to the spec I have (from ranfft, which is the only place I've seen anything remotely technical about the movement online), then I'll have to adjust. If it's not, then I have a spec to work with.

[saswatch88]

1 hour ago, spectre6000 said:

Oh. Interesting thought. 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Certina_304

The movement shows up on ranfft as being 17 jewels, and no doubt later (I'm assuming later...) versions were. Mine is only 16 jewels. Mayhap that's part of the problem. It's only one jewel, and in about the lowest impact location it can be, but it's not nothing.

Prior to service, I put it on the timegrapher, but didn't pay that close attention. It was running only in a binary sense (1/0 or True/False; no quality dimension implied). I think amplitude was on the order of 180°, rate was way off, but I don't recall how far or in what direction. 

I will definitely investigate what I have before I order anything. If it's to the spec I have (from ranfft, which is the only place I've seen anything remotely technical about the movement online), then I'll have to adjust. If it's not, then I have a spec to work with.

I dont think 1 jewel will make the amp go up that high but it is possible. The strength difference would prob be .01 greater so if your original MS is too spec then maybe try going .01 less in strength, again this is a band aid solution, if it is one at all....personally i would check the escapement. It is a good learning experience...i use to have tons of issues with balance and escapement then i purchased Fried’s book “the watch escapement” which is very good at learning how to diagnose issues and remedy them, some of the practices are old school but the overall principles remain the same...i also purchased the tools needed, staking set, roller removers, collet removers, tuning fork, shellac, pallet warmer, etc. One thing i have not done yet is adjust HS length and vibrate them....it really is tedious work that keeps you tight in the shoulders, i just cant do it but, hairsprings can always be measured purchased, although quite difficult but math is easier than sitting their sweating trying to attach a stud to the HS.

but either way now that i can identify escapement issues i see past repairs done poorly in these older movements more often than not....banking pins and pallet jewels being number 1 on the list....the movement has screw type banking pins so they will be easy to adjust, get a very strong magnifier or loupe like 5-10x and observe the drop and lock. The escape tooth needs to travel about 2/3rds of the way across the surface of the pallet jewel, tooth short escapement will lock to far and overbanking will occur. It doesn’t seem to have done this before but a through cleaning and possible over-oiling or not using correct oils could have caused it, or if you cleaned the pallet in alcohol the pallet jewels could have moved out of position.

[Klassiker]

8 hours ago, spectre6000 said:

Serious question on the beat error front: is 3.8mS really worth getting all that worked up over? Obviously, 0mS is the goal, but seeing as how this is the first watch I've touched with the intent to make an improvement rather than just get some time in with the tools, I'm not sure 3.8mS is worth the risk.

I agree. The methods suggested to correct the beat error are all high-risk for a non-expert, especially without the correct tools. I would leave it alone for the time being and concentrate on measures to reduce amplitude, and maybe coincidentally help with other problems you are seeing (poor isochronism, uneven traces).

8 hours ago, spectre6000 said:

When I get back to the bench, I'll pull the mainspring and measure it out and see how it compares to the spec on ranfft. If I'm to do what I intend to do with this thing, I need to order a crystal, so I guess that means I'll be ordering a mainspring to boot... Looking forward to notching a win!

This is a good idea. Even if the mainspring appears to be the correct spec after measuring, you will have learned something, and a replacement (same or slightly weaker) is not expensive. Another good source for mainspring data is Watchguy UK. He lists the Generale Ressorts suggestions, and for Certina 304 it is the GR5499. These are readily available.

 

 

Edited June 28, 2020 by Klassiker

[saswatch88]

5 minutes ago, Klassiker said:

I agree. The methods suggested to correct the beat error are all high-risk for a non-expert, especially without the correct tools. I would leave it alone for the time being and concentrate on measures to reduce amplitude, and maybe coincidentally help with other problems you are seeing (poor isochronism, uneven traces)

It is high risk but we should encourage him to learn how to work on the escapement. It’s risky I know I have been there I was always so nervous to mess with balance or escapement but I finally got tired of not doing things right and I got the tools and learned. Many more running movements in my drawers now. 

[Klassiker]

Certainly no harm in looking at banking pins and pallet jewels, but I would change other things first.

As a relative novice I am always conscious of limited time (and budget) and the need to focus on the low risk / high return actions first. If the pallet jewels are out of position, I would buy a replacement and swap it before trying to reset the stones.

[saswatch88]

Certainly no harm in looking at banking pins and pallet jewels, but I would change other things first.
As a relative novice I am always conscious of limited time (and budget) and the need to focus on the low risk / high return actions first. If the pallet jewels are out of position, I would buy a replacement and swap it before trying to reset the stones.

I agree as a novice I would do the same thing I would just find a donor movement and replace parts as needed. In this particular case he may not be able to find a suitable donor. Eventually he’s going to have to learn either way because for example I work a lot on Waltham Elgin movements and sometimes I get a movement that is a very high grade and I can’t use parts from a lower grade so I am left with no choice but to deal with what I have in front of me and I’m sure the OP will find himself in similar situations in the future. But he should definitely still try to diagnose the problem at least then if you can find a replacement part then great. But the point is just diagnosed the problem. Diagnosing is where the experience comes from Sexting with a solution comes later


Sent from my iPhone using Tapatalk

[spectre6000]

OK! Long day out in the canyon with a pickaxe, shovel, and pry bar slingin' dirt and rocks, and I'm beat... Muscle fatigue is bad for steady hands...

I unscrewed the hairspring stud screw, and the stud is really stuck in there... I could not get it out, and didn't want to risk damage to the hairspring. I was able to sight the impulse jewel between the banking pins, and it's clearly off to one side. I was not able to see any marks on the balance staff though, and I figure I'm going to leave it alone for now. There will be future opportunities with much more significant beat error with which to acquire this skill.

I measured out the mainspring, and we have a winner! 1.75mm X 0.11mm X ~265mm

GR5499 works out to 1.90 x .095 x 320 x 10 (what's the 10? Standard hook end?)

ranfft has 1.80 x 9.0 x 0.09 x 315mm (what's the deal with two middle numbers?) 

GR5231 looks like another viable option at 1.8 X 9.0 X 280 X 8 (what's the 8?) for a slightly decreased power reserve, but closer overall

So, the mainspring that came with the watch is between 3 and 5 strengths too strong, it's 1-3 sizes too narrow, and 50-55mm short. All manner of wrong.

[saswatch88]

OK! Long day out in the canyon with a pickaxe, shovel, and pry bar slingin' dirt and rocks, and I'm beat... Muscle fatigue is bad for steady hands...

I unscrewed the hairspring stud screw, and the stud is really stuck in there... I could not get it out, and didn't want to risk damage to the hairspring. I was able to sight the impulse jewel between the banking pins, and it's clearly off to one side. I was not able to see any marks on the balance staff though, and I figure I'm going to leave it alone for now. There will be future opportunities with much more significant beat error with which to acquire this skill.

I measured out the mainspring, and we have a winner! 1.75mm X 0.11mm X ~265mm

GR5499 works out to 1.90 x .095 x 320 x 10 (what's the 10? Standard hook end?)

ranfft has 1.80 x 9.0 x 0.09 x 315mm (what's the deal with two middle numbers?) 

GR5231 looks like another viable option at 1.8 X 9.0 X 280 X 8 (what's the 8?) for a slightly decreased power reserve, but closer overall

So, the mainspring that came with the watch is between 3 and 5 strengths too strong, it's 1-3 sizes too narrow, and 50-55mm short. All manner of wrong.

you want to get exactly what is on ranfft. Especially the height and length. That is very important since the main spring cannot take up more than half of the radius inside the barrel and if the height is too high you won’t be able to close the barrel lid Or have very high friction and rubbing which will give you very low amplitude. Strength You should get the exact number but if it’s .05 off then I think you will be OK but you really should get the exact spring and I’m sure you can find it somewhere.

 

Ranfft uses metric sizes height x strength x length

1.8mm height the 9.0 is height size in dennison measurement. The .09 is strength which is a 12 in dennison and length is 315mm which is about 12 1/2” dennison size.

 

So metric = 1.8 x .09 x 315

Dennison = 9 x 12 x 12 1/2

 

The 8 at the end of the one GR I believe is the barrel size. I don’t worry about that since you already have the exact measurements of the spring that should be in the Barrel.

 

 

Sent from my iPhone using Tapatalk

[spectre6000]

https://www.watchrepairtalk.com/topic/15316-what-kind-of-crystal/

[anilv]

Since it runs ok in all positions excepte dial down I would check the balance jewels/pivot on the mainplate. Another thing is check that the screw on the pallet fork bridge is correct as a thicker one can interfere with the balance.

I'm sorry I may have missed something but hairspring issues and overbanking would have an impact in all positions?

Anilv

[spectre6000]

It's rebanking DU and DD, just more consistently DD. I'm 99% positive the issue is an incorrect mainspring. It's 3 strengths too strong (among other out of spec details). 

[JohnR725]

3 hours ago, spectre6000 said:

It's rebanking DU and DD, just more consistently DD. I'm 99% positive the issue is an incorrect mainspring. It's 3 strengths too strong (among other out of spec details). 

Personally when changing mainsprings I always like to measure what is in the watch then look at the reference material see what it says and make sure everything is on the same page. For reference material I have the first link I would suggest downloading a couple of things like BestFit This will give you two separate files here named BestFit 111 Part 1.pdf & BestFit 111A Part 2.pdf. Another thing to download would be the Generale Ressorts Complete catalog named GeneraleRessorts.pdf. Then I snipped out images of both of them. The bestfit's nice because it usually give you additional information like what the base caliber areas.

Then the second leg is interesting site all sorts of nifty technical information. If you go down the list the mainspring of click on that ill give you some additional mainspring specifications.

https://watchguy.co.uk/cgi-bin/files

http://cgi.julesborel.com/cgi-bin/matcgi2?ref=KF_304

 

[spectre6000]

Interestingly, those two seem to disagree... The first excerpted image states a width of 1.90mm, and the second says 10 Dennison. 

I ordered:

Metric Width: 1.9 mm, Metric Strength: 0.09 mm, Length in Inches: 12.5
Dennison Width: 9, Dennison Strength: 12

Note the 1.90mm width and the Dennison Width of 9... 

Similarly, one states 0.095mm strength, the other 12 Dennison, and the spring I ordered is 0.09mm and also 12 Dennison... 

Meanwhile, I can't find a chart showing direct correlation between Dennison and mm... What's up with this Dennison scale?

They also disagree with Ranfft at 1.80 x 9.0 x 0.09 x 315mm...

Regardless, pretty sure I got the right spring. When it shows up, I'll get the little guy back together and report back.

Additionally, the J.Borel page is probably the first thing outside of Ranfft I've seen on this movement. Watchguy doesn't have anything for this movement either.

Edited June 30, 2020 by spectre6000

[saswatch88]

Interestingly, those two seem to disagree... The first excerpted image states a width of 1.90mm, and the second says 10 Dennison. 
I ordered:
Metric Width: 1.9 mm, Metric Strength: 0.09 mm, Length in Inches: 12.5
Dennison Width: 9, Dennison Strength: 12
Note the 1.90mm width and the Dennison Width of 9... 
Similarly, one states 0.095mm strength, the other 12 Dennison, and the spring I ordered is 0.09mm and also 12 Dennison... 
Meanwhile, I can't find a chart showing direct correlation between Dennison and mm... What's up with this Dennison scale?
They also disagree with Ranfft at 1.80 x 9.0 x 0.09 x 315mm...
Regardless, pretty sure I got the right spring. When it shows up, I'll get the little guy back together and report back.
Additionally, the J.Borel page is probably the first thing outside of Ranfft I've seen on this movement. Watchguy doesn't have anything for this movement either.

You definitely ordered the correct mean spring. I always go by ranfft whenever or wherever i can. It is common to see springs with same height and length but different strengths for same caliber movement. A spring for 7 jewel will be stronger than a spring for 15 jewel, and 15 jewel stronger than 17 jewel, if the particular movement caliber is offered in different grades. I have never heard of a main spring with different heights. Height 10 vs height 9 could mean a world of difference in some barrels but none in others so it could be possible.

You can find a conversion chart on the net. Its all based on region when it comes to movements and parts. Iam usa so i always go by dennison.


Sent from my iPhone using Tapatalk

[saswatch88]

Sent from my iPhone using Tapatalk

[spectre6000]

That conversion chart is too small to read, and to generic for a reverse search AND google doesn't like the correct spelling of Dennison... Any chance for a link or better resolution?

[JohnR725]

37 minutes ago, spectre6000 said:

Meanwhile, I can't find a chart showing direct correlation between Dennison and mm... What's up with this Dennison scale?

 

Mainspring - DennisonChart.pdf

[spectre6000]

Huh... That is an odd unit(?) of measurement... I'll have to share this with my buddy that works with the guys at NIST. He'll get a kick out of it.

[JohnR725]

5 hours ago, spectre6000 said:

Huh... That is an odd unit(?) of measurement... I'll have to share this with my buddy that works with the guys at NIST. He'll get a kick out of it.

I think you'll find the PDF at the link below interesting.

http://www.nawcc-index.net/Articles/Watkins-MainspringGauges.pdf

 

[spectre6000]

You may have figured me out in terms of what I find interesting. Excellent read! 

So it's not that Dennison is non-linear/symbolic, it's that it's not a universally agreed upon measure. I guess I'll continue to stick with metric measurements.

[JohnR725]

1 hour ago, spectre6000 said:

So it's not that Dennison is non-linear/symbolic, it's that it's not a universally agreed upon measure. I guess I'll continue to stick with metric measurements

Because of the resourcefulness of watchmakers adopting a variety of measurements  it makes ordering parts  interesting. Especially depending upon where you live  like in the US  it's a little more challenging to order a mainspring. Metric is universally the safe thing to use but  still handy to have Dennison chart handy  because the mainspring package might come with Dennison  measurements. Then up until recently  the length was always in inches.

Mainspring ordering for American pocket watches then things get really interesting. All the measurements typically get used inches universally for the length. Width and strength either metric or Dennison or both. Depending upon whose system was used to package up the mainspring they will have different numbers versus the factory numbers. Then companies like Elgin  if you look up their part number  it will usually tell you the strength but  in inches..

[saswatch88]

11 hours ago, JohnR725 said:

 

Mainspring ordering for American pocket watches then things get really interesting. All the measurements typically get used inches universally for the length. Width and strength either metric or Dennison or both. Depending upon whose system was used to package up the mainspring they will have different numbers versus the factory numbers. Then companies like Elgin  if you look up their part number  it will usually tell you the strength but  in inches..

American pocket watches use Dennison measurements, other companies who make Mainspring like Samson for example will use both measurements on their packages

[spectre6000]

OK... I got the new mainspring in. As best I can tell, it's the original specification. I was confident it would fix what ailed me, but all it's done is create an even deeper mystery... Here are the timegrapher outputs in each position. Sorry for the dirty screen. It doesn't look nearly that bad either in person or at the angle I'm sitting at, or maybe it's something about the camera focal depth... Who knows.

Dial up (amplitude too low to register):

Dial down:

Pendant up:

Pendant down:

Pendant left:

Pendant right:

Dial up, the amplitude drops so low, the timegrapher just loses it, and can't complete a full trace. Dial down, the amplitude is so high that it rebanks, but with the beat error (which seems more variable than it should be), it was visibly only rebanking in the one direction at one point (not pictured). The pendant positions are similarly variable. The performance is so variable, that it's like I could wear it, and it would just average out to 0 S/d.

I don't know what to make of such extreme variability. The only thing I can think of at the moment is that it's completely knackered. Anyone have any ideas to point me in the right direction?

[Klassiker]

Something has definitely changed on the balance (jewels, lower staff pivot) or hairspring between your first measurements and now. Check the hairspring for flatness or dirt.

Edited July 17, 2020 by Klassiker