Isochronism Adjustment - Waltham Model 1892 AT&Co Premier Grade

Discussion in 'American Pocket Watches' started by grtnev, Oct 14, 2019.

  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  1. grtnev

    grtnev Registered User
    NAWCC Member

    Jan 18, 2009
    541
    143
    43
    Male
    Retired = Consulting Engineer
    Minden, Nevada
    Country Flag:
    Region Flag:
    I was reviewing THE RAILROADERS' CORNER article in the June 2000 NAWCC Bulletin, page 368, and came across the following:

    “Curiously, neither the original nor the Premier version of the AT&Co. grade was adjusted to isochronism. This made the Appleton, Tracy and Co. Premier the only one of the model '92 railroad grades lacking this feature.”

    This is an very interesting observation, at least to me, because I assumed that as time marched on the manufactures had introduced adjustment to temperature and isochronism on all “adjusted” watches and then added varying degree of positional adjustments depending on the grade of the watch.

    Why would Waltham have a watch adjusted to temperature, 3 positions (Model 1892, AT&Co grade) and 5 positions (Model 1892, AT&Co Premier grade) but not be adjusted to isochronism?

    Did they do this on any other grades as well?

    The pictures are of s/n 12019796, a hunting case Model 1892 AT&Co Premier grade that I have. From other posts, estimated HC production was from 300 to 600 out of a total production of 3000 or so (HC & OF) grade 1892 AT&Co Premier grade movements.

    Am I correct to assume that a total production of 300 to 600 would be considered “uncommon” as opposed to “rare”?

    Thanks,

    Richard

    Dial 1.JPG Movement 3.JPG
     
    miguel angel cladera and Kent like this.
  2. Dr. Jon

    Dr. Jon Registered User
    NAWCC Member

    Dec 14, 2001
    5,376
    178
    63
    Aerospace Engineer
    New Hampshire
    Country Flag:
    Region Flag:
    The underlying issue is what did they do to adjust for isochronism?

    This is tricky because to my knowledge no one tested for it directly and I have never seen a standard or definition of one for isochronism. NIST, when they were the Bureau of Standards did some testing on this in the early 20th century and Theo Gribi, in his book, Practical Watch Adjusting, measure it on some watches by setting them up to run with vary weights.

    I suspect the adjustment was in the form of an overcoil shape but perhaps someone knows.
     
  3. Kent

    Kent Registered User
    Gibbs Literary Award NAWCC Fellow NAWCC Silver Member

    Aug 26, 2000
    18,208
    1,374
    113
    Country Flag:
    This is why one should be careful in making assumptions. After making a few unwarranted assumptions and having them blow up in my face, I adopted the attitude that if didn't see it in (primary source) print (documents - watch company or dealer ads, catalog sheets, technical documents, employee magazine articles, etc.) then it wasn't necessarily so.




    Perhaps because Railroad Time Service Watch Rules didn't require adjustment to isochronism. I'm not sure that it had all that much affect on the timekeeping rate (+/- 30 sec/wk) Like Dr. Jon, I believe that it was accomplished by forming the shape of the Breguet hairspring's overcoil to produce concentric (or, at least, nearly concentric) contraction/expansion of the hairspring as the balance rotated. I suspect that this was quick and easy as the hairspring's overcoil shape would seem to be the same from one watch to the next, but then I've tried to do it. Perhaps the expense was in the checking.


    Well, the only other major model 92 railroad watch for which I have a description is the grade No. 845 and a 1911 catalog shows that it too was adjusted to isochronism. It's possible that the Royal, Riverside (and a few others) and some private label watches like the New South Wales Railway watches lack this adjustment.




    Richard, you have a very nice watch!

    The production quantities of the Premier version of the AT&Co grade may never be known. They were mixed in with runs of regular AT&Co grade watches and runs of AT&Co Premier grade watches noted in Serial Numbers With Description of Waltham Watch Movements, Waltham Watch Co., Waltham, MA, 1954, (commonly referred to as "The Gray Book") had regular AT&Co grade watches mixed in (causing some watch dealers to claim that they were "Unmarked" AT&Co Premier grade watches - my viewpoint is that if they aren't marked "Premier", they're not AT&Co Premier grade watches).

    Page 101 of Complete Price Guide to Watches, No 33, R. Gilbert, T. Engle and C. Shugart, Tinderbox Press, Mount Pleasant, SC, 2013 (the most popular watch price guide on the market) lists the below ratings of quantities. Although the assignment of 'Stars' elsewhere in the book may be open to debate (or in some instances, is demonstrably incorrect), the definitions used are as good as any in establishing the meanings of terms such as rare, scarce, etc.

    Star Ratings

    No. of StarsTermQuantity
    *****Rare1-25
    ****Scarce100
    ***Very Few350
    **Sparse1,000
    *Uncommon2,500

    1902_Sep-10_Waltham_AT&Co_Premier.jpg 1903_Nov_Model_92.jpg 1911_Young_&_Co_18S.jpg
     
  4. Kent

    Kent Registered User
    Gibbs Literary Award NAWCC Fellow NAWCC Silver Member

    Aug 26, 2000
    18,208
    1,374
    113
    Country Flag:
    The above should read: "I suspect that this was quick and easy as the hairspring's overcoil shape would seem to be the same from one watch to the next, but then I've not tried to do it."
     
  5. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    349
    83
    Watchmaker
    Baltimore
    Country Flag:

    While Isochronism is most easily checked with modern instantaneous measurement instruments, it may have been checked by counting beats against a known regulator (5 beats per second). All it means is that the oscillator maintains its frequency at low and high amplitudes. But you have a point, what was the quantitative standard that was used to make the claim.

    I know I am happy with a difference of 10 seconds per day. But how that would be determined by counting beats is beyond me. Maybe the limitations of the method (beats in 5 minutes??) determined the presence of isochronism. If you could not detect a difference in counting beats for 5 minutes, then you would say the rates are equal?

    Interesting.
     
  6. Dr. Jon

    Dr. Jon Registered User
    NAWCC Member

    Dec 14, 2001
    5,376
    178
    63
    Aerospace Engineer
    New Hampshire
    Country Flag:
    Region Flag:
    I looked up what Exelsior wrote in his book on watch adjusting. The booke mentione several methods which control teh motive power but the one he advocates is comparing successive 12 hour periods of the 24. He compares the rates in the first 12 with those in the second. His is vast oversimplification but this is the basic idea.
     
  7. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    349
    83
    Watchmaker
    Baltimore
    Country Flag:
    Yes, of course, duh.... This is even what Omega specifies in its working instructions. When adjusting, I use the timing machine at 180 and full amplitudes (depending on watch 250 to 300 degrees). With the 12 hour period you are basically averaging across a continuous range of amplitudes (integration?? been 50 years). Fine for final performance but not all that helpful for initial adjustment.

    I once could not for the life of me recall why lathes are called WW. Thank you for reminding me of the 12 hour periods method. Totally forgot.

    The interesting thing I have found is that certain watches (high grade) only have about 20 degrees amplitude difference in vertical and horizontal positions (some Hamilton 21 and 23 jewel watches) while others can have 60 degrees (Hamilton 17 jewel). This is why isochronism becomes the first factor and is very important. The loss in amplitude compounds poise errors. Heavier balances with stronger springs are less impacted by position as might be expected.

    Interestingly, some watches max out at 230 degrees (Elgin A/C clocks in particular). This may well reflect an attempt to reduce the spread of amplitude variation. Dan Fenwick once told me something that I could not compute at the time; that amplitude is not end all and be all it is made out to be. At the time I wrote it off as Omega propaganda related to the coaxial but Dan has always been known as telling it straight.

    I can now see there may be intentional design goals involved, but I have no documentation to support that possibility.

    The generally accepted explanation for high amplitude is there is less natural escapement error. But I am now open to the possibility that some designers valued a reduced amplitude range more highly.

    Thanks for bringing this up.
     
  8. grtnev

    grtnev Registered User
    NAWCC Member

    Jan 18, 2009
    541
    143
    43
    Male
    Retired = Consulting Engineer
    Minden, Nevada
    Country Flag:
    Region Flag:
    #8 grtnev, Oct 15, 2019
    Last edited: Oct 15, 2019
    Kent,

    Thanks for the Waltham ads - I’ve had copies of these for a while but never noticed that in all 3 ads that the AT&Co grade is described as adjusted to temp and positions only and not to iscochronism.

    Without knowing any better, I would assume that the balance wheel / hairspring assemblies had to be the same for all 1892 grades and consequently all balance wheel/hairspring assemblies would be pretty consistent in term of poise, isochronism, and initial position adjustment.

    Perhaps since the AT&Co grade was initially adj 3P, the isochronism checks, as described by Dr Jon & Dewey, were not done. But why?

    Were there any other Waltham “adjusted” grades where isochronism wasn’t checked?

    Thanks again for all the good information.

    Richard
     
  9. Jerry Treiman

    Jerry Treiman Registered User
    NAWCC Member

    Aug 25, 2000
    5,951
    792
    113
    Geologist - California Geological Survey
    Los Angeles, CA
    Country Flag:
    Region Flag:
    Apparently quite a few. Just looking at the 1892 model, the Royal grade was "adjusted" with no mention of isochronism and, surprising to me, even the Riverside grade was only adjusted to 3 positions and not isochronism (per a 1900 Waltham brochure - The Perfected American Watch). The Riverside grade was more highly adjusted later in other sizes.
     
  10. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    349
    83
    Watchmaker
    Baltimore
    Country Flag:
    This is worth understanding. We were taught that isochronism is the very first thing you check after the watch is serviced (and demagnetized). We were told, and I fully accept, that given there is dropoff in amplitude when the watch is vertical, there was little point in adjusting to position until the watch was isochronous.

    I am perplexed.
     
  11. Ethan Lipsig

    Ethan Lipsig Registered User
    NAWCC Gold Member

    Jan 8, 2006
    1,758
    516
    113
    Pasadena
    Country Flag:
    Region Flag:
    Me too. I thought a hairspring's isochronism is achieved through its proper design, e.g., having a Breguet overcoil, rather than a watchmaker's meticulous adjustments. What would a watchmaker adjust to improve the isochronicity of a properly designed hairspring in good condition?
     
  12. johnbscott

    johnbscott Registered User
    NAWCC Member

    Feb 25, 2007
    341
    163
    43
    Male
    Country Flag:
    The design and quality of construction of the movement should minimise the variation of balance motion as the mainspring winds down or the movement is placed on edge. Adjustment for isochronism then proceeds to produce a gaining tendency for the shorter arcs of balance motion to compensate for the ever present energy losses that then assume a greater proportion of the stored energy and, therefore, that tend to cause a losing rate. The books on watch adjustment are full of strategies for achieving gaining rates for the short arcs and those strategies do include hairspring shape adjustment.

    The Breguet overcoil separates functions by allowing the centre of mass of the main part of the hairspring to be stay central whilst enabling independent creation and control of lateral forces at the balance staff that tend to influence isochronism.
     
    Ethan Lipsig likes this.
  13. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    349
    83
    Watchmaker
    Baltimore
    Country Flag:
    Ethan,

    In a freesprung timepiece there is very little and even then it is a matter of empirical experimentation. This is why many competition took months to prepare. Mainly it is about adjusting the terminal curves. See Jendritski.

    In watches with regulators you the adjustment is the spacing of the regulator pins. The bigger the space the larger will be the difference between full amplitude (when the spring this the pins for a longer period) and low amplitude (where the spring rests against the pins for a shorter period).

    This is the FIRST adjustment. If for some reason the positional rates indicate the spring is hitting one pin more than the other and the spacing has to adjusted then you start all over again.

    The one size fits all " the pins most be as close as possible to the spring" rule is invalid.

    For one thing, it is not quantifiable.

    This may be of interest:

    https://www.historictimekeepers.com/documents/Watch%20Adjustment.pdf

    What I am perplexed about is whether Waltham actually adjusted watches to position without verifying they were isochronous or if their advertising assumed it was present in adjusted watches.
     
    Ethan Lipsig likes this.
  14. Ethan Lipsig

    Ethan Lipsig Registered User
    NAWCC Gold Member

    Jan 8, 2006
    1,758
    516
    113
    Pasadena
    Country Flag:
    Region Flag:
    Very interesting and helpful, Dewey.
     
  15. Chris Radek

    Chris Radek Registered User
    NAWCC Member

    Apr 13, 2014
    421
    132
    43
    Male
    Watchmaker (now accepting new customers)
    Lincoln, NE, USA
    Country Flag:
    If I had no timing machine and wanted to measure the isochronism curve, I would wind the watch (for instance) every 2 hours for 24 hours, note the error (giving the error in s/d of the first two hours of wind), reset it, wind the watch every 4 hours for 24 hours and note that new error minus the 2 hour error (giving the error in s/d of the second two hours of wind), and so on. This seems easily done but time consuming. With experience and knowledge of the general shape of the curve, I bet you could get reasonable knowledge about the state of the adjustment with only about 3 data points (answer in 3 days).

    I'm suspicious of a technique of just checking after 12 hours. Surely the curve is rarely linear - the interesting things may all happen in the first hour...?
     
  16. grtnev

    grtnev Registered User
    NAWCC Member

    Jan 18, 2009
    541
    143
    43
    Male
    Retired = Consulting Engineer
    Minden, Nevada
    Country Flag:
    Region Flag:
    #16 grtnev, Oct 16, 2019
    Last edited: Oct 16, 2019
    I am also perplexed by what Waltham may have done regarding checking isochronism on their "Adjusted to Position" watches. Just doesn't seem to make any sense to make positional adjustments if the watch is not isochronus.

    Thank you to Dewey for his insights and providing the link to his article: "Watch Adjustment". Referring to that article: "The pivots, balance spring, balance and collet location and mainspring all determine how equal the watch rates will be when nearly fully wound and after 24 hours. This is called isochronism . Isochronism means that the balance will complete its rotation in the same amount of time at high amplitudes (full power) and at low amplitudes (low power). Amplitude is how watchmakers measure the extent to which the balance rotates. If it rotates 270 degrees, it is said to be at high amplitude (which is standard for a freshly serviced watch) and 180 degrees is considered low amplitude. The importance here is that there is very little a watchmaker can do to improve isochronism........a knowledgeable watchmaker can adjust the regulator pin spacing to improve isochronism........Their only functions are to adjust the effective length of the balance spring and isochronism."

    I also appreciate Kent's approach to historical research regarding assumptions: "This is why one should be careful in making assumptions. After making a few unwarranted assumptions and having them blow up in my face, I adopted the attitude that if didn't see it in (primary source) print (documents - watch company or dealer ads, catalog sheets, technical documents, employee magazine articles, etc.) then it wasn't necessarily so." The primary source, print documents in this case indicate that Waltham model 1892 AT&Co grade was adjusted to temp and positions, but not to isochronism.....and to my mechanical engineering mind that simply does not compute. Not at all saying that is not reality, but like others, I am perplexed and confused.

    Is there any documentation from any American manufacturer (Elgin, Hamilton, Illinois, Waltham) that describes how during factory assembly, the isochronism of a watch was to be checked and or adjusted? Or was the isochronism of a watch just assumed to fall within a known variation based on manufacturing and assembly tolerances of multiple reference balance and hairspring assemblies in test watches?

    Richard
     
  17. Kent

    Kent Registered User
    Gibbs Literary Award NAWCC Fellow NAWCC Silver Member

    Aug 26, 2000
    18,208
    1,374
    113
    Country Flag:
    Richard:

    Although this might be the case, I wouldn't make this assumption. They're probably the same within a run of a grade, but after that, I'm not so sure.
     
  18. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    349
    83
    Watchmaker
    Baltimore
    Country Flag:
    I know from the service literature Hamilton had 6 combinations of balance screws and balance spring. The higher grade watches had the heaviest balances and also seem to have the least drop in amplitude from horizontal to vertical (which should not be confused with isochronism which is assessed in one horizontal position). I have not bothered to record the instantaneous rates of the various Hamiltons at high and low amplitude. In hindsight, this may have been shortsighted. But my expectation would be that the higher grade watches with the heavier balances possess the better isochronism.

    I would suspect Waltham marketing simply presumed the reader would understand adjusted watches were isochronal. They may not have appreciated the value of specifying everything. Remember, early Hamilton 16s watches simply said "adjusted" on the movt.
     
  19. johnbscott

    johnbscott Registered User
    NAWCC Member

    Feb 25, 2007
    341
    163
    43
    Male
    Country Flag:
    I agree that position adjusted watches are isochronal. After all, that is the goal of adjustment - equal overall rates in positions (when all errors are balanced out) despite variations in balance arc. If a watch is wound in the morning, as recommended, any reduction of balance arc will be positional during the day (when carried) and due to reduced mainspring tension during the night (when lying). Winding in the morning is recommended because the assumption is that winding at night would allow undesirable contraction and over-tightening of the compressed mainspring as cooling occurs, when the watch is taken from the pocket.
     
  20. Tom McIntyre

    Tom McIntyre Technical Admin
    Staff Member NAWCC Star Fellow NAWCC Ruby Member Sponsor

    Aug 24, 2000
    81,704
    1,278
    176
    Male
    retired SW dev
    Boston
    Country Flag:
    Region Flag:
    It might be important to remember that Waltham claimed all the hairsprings at this late date were "hardened and tempered in form."

    My friend Pat Caruso had the job of reworking watches that did not meet specification and he was incredibly skilled at producing Breguet overcoils by hand and adjusting them with a pair of tweezers and a cork block. I believe he measured isochronism by adjusting the amplitude with the amount of power from the mainspring. If the rate was the same at minimum and maximum, it was good to go. I am pretty sure that was also the standard line test. I do not believe that Waltham used the fancy electronic timers that Hamilton did.

    Rework was piece work even after the company started to do away with that and go to strictly hourly wages. Pat was very fast and happy. :)
     

Share This Page