Explanation of Escapement

Discussion in '400-Day & Atmos' started by KurtinSA, May 16, 2019.

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

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    ...again! I struggle with this whenever I have a problem clock and I have one in front of me. It's a Kundo miniature that I just can't get to run.

    In my limited understanding, I found that the drops weren't right, so I following the guidelines on Page 47 of the guide and bent the peninsula down. It looks like the drops are now even and about the amount that I've been noticing on other similar clocks. So, good there.

    But now that I look closer, it seems that the locks are too deep. The EW turns counter clockwise when viewed from the back. The lock on the exit pallet is definitely too deep. On the entrance pallet, I think it sort of looks OK.

    So, looking at the table, for normal drops and exit pallet locks that are deep or very deep, I see a couple of "Move down" and one "Move up" for the exit pallet. That's got me stalled...I don't know what to do.

    But my question. Say the exit pallet has a deep lock and the entrance pallet has a shallow lock (one of the choices in the table), how does moving the exit pallet down fix that situation? What happens with the geometry and/or escapement so I can get my mind around this? I know the pallets are aligned in this larger circle, but I just don't understand how this works.

    Thanks...Kurt
     
  2. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    #2 KurtinSA, May 17, 2019
    Last edited: May 17, 2019
    OK, I think I might have figured part of this out. I've been thinking all wrong on this. I thought that by lengthening the exit pallet, it would somehow change the lock on the exit pallet. What it will do it seems is to increase the lock on the entrance pallet. But in my situation, it would probably created too much lock on both pallets.

    So, that moves me to a new situation on the chart. In that case, the action for deep lock on both pallets is to raise both pallets. Now that makes sense. Once I get the locks even, it's a matter of moving the pallets the same amount until the proper lock is achieved.

    -- edit -- I went away and thought about that. Seems like I should just shorten the exit pallet until it equals the lock on the entrance pallet. No sense doing things in two steps when one might do it.

    Kurt
     
  3. victor miranda

    victor miranda Registered User

    Jan 13, 2017
    316
    29
    28
    computer programming stuff
    Baltimore
    Country Flag:
    Region Flag:
    I am not sure who has credit for inventing the escape.
    the device makes a lot of sense to me and I find it is not hard to adjust...
    I can't imagine I'd have ever invented it.
    buuuut it ain't my fault that it don't run... IF
    there is no power at the escape wheel
    the teeth are curved away from the entrance and exit locks
    the EW is out of round
    the pallet lock faces are an even curve into the impulse faces

    that list can get long.

    because of how the pallets are connected, one never plays with them separately.
    the lock face has to catch and hold on each EW tooth. both pallets. I only watch one side at a time.
    while just right is correct, too deep is far better than the opposite.
    Get the escape to lock reliably.
    I test with a screwdriver pushing the pin slowly and watching the pin/verge jump over to the
    other lock. it takes time to test all 15 teeth through both sides of the verge.

    then move the fork down till the escape will run(with any luck,the clock also....)
    any flutter is pretty sure to be caused by fork height.
    a small flutter might stop the clock just note that the clock is running at that point.

    victor
     
  4. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    It's a Graham deadbeat escapement, I guess Mr. Graham invented it! ;)

    The clock seems to have good power in terms of how the escape wheel is snapping from lock to lock. The clock ran 24 hours without the motion works. The over swing was not great but there was some, maybe 20 degrees. When I put the motion works on, the clock won't run.

    I was thinking that the lock is too deep on one side. If it's too deep, it takes power to drag the tooth off of the pallet. So, I wanted to reduce the amount of lock. I'm looking for just enough lock to prevent flutter, then with the over swing, the tooth rides up a bit more on the pallet.

    Kurt
     
    tracerjack likes this.
  5. victor miranda

    victor miranda Registered User

    Jan 13, 2017
    316
    29
    28
    computer programming stuff
    Baltimore
    Country Flag:
    Region Flag:
    hi Kurt,
    if you can get the escape to run for a day, you are close enough.
    it can absorb a lot of time to get closer to perfection from there.
    Keep in mind that your base level can adjust the escape verge.

    so if you are trying to get the clock to clock along
    the motion works are the problem to sort out....
    those are often far harder to find and fix than the escape.
    in some ways all you need to do is find enough power to run the motion works.
    However, since the spring is only so big it is better to find the reason the added gears and weight
    of the hands stop the clock.

    it sounds like you have a functioning escape.

    ok that said you are trying to get more swing into the pendulum?
    you will need a sharp set of eyes.

    start at the two pivots for the escape.
    remove any slop you can detect while allowing the arbors to spin.

    make sure the EW is concentric ( a bent arbor can cause no end of problems.)

    the verge should have a clear line where impulse face and lock face meet.
    lets call it the edge.

    you know how to get it in beat
    your pallets should each be adjusted (if you have that kind of verge)
    to catch a lock face with a small amount of distance from the edge into the lock face.
    yep, the other side has an effect on this... from what you say you can do this.
    To get the pin straight up, I adjust one pallet to hold against the EW in the middle
    of the impulse face and then move the other pallet for proper lock.
    that is the verge 'centered.'

    make sure the fork does not grab the pin, check it at the extremes of the fork range.
    lower the fork as needed to find the run.

    you will need to check for good kick, and you want that same kick on both halves of the swing.
    I often play with the base level for this to see if it is a close adjustment.

    you pay attention to swing angles.
    the swings must be symmetrical from running center.

    if you have a 360 degree swing, center is 180.
    the start of the impulse should be close to the same
    degrees away from center. just note it... If you are able... I often guess this
    as part of setting beat... because this check will tell you mostly the condition of the EW
    for bent teeth and out of round.

    victor
     
  6. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    Victor -

    My focus is on over swing...to some degree I don't really care about total swing...without enough over swing, the clock stops.

    I'm going to have to think about what you said:

    I can see what you're saying, but don't know how to do that. I've never been able to leave the pallets loose enough so I can make adjustments with the anchor in the clock. I have to see how the escapement is working, take out the anchor, loosen the pallets, adjust, then retighten before I put it back in the movement. I suspect if they're not tight enough, all kinds of stuff will happen as my clumsy fingers try to check things out.

    Kurt
     
  7. victor miranda

    victor miranda Registered User

    Jan 13, 2017
    316
    29
    28
    computer programming stuff
    Baltimore
    Country Flag:
    Region Flag:
    hi Kurt,

    You look at some problems far different than I do.
    short answer: lower the fork for more over-swing.
    get that far, and I'll share how I made a longer suspension hanger to avoid flutter.

    If your escape is working, let it run.
    try watching it to see how far from ideal it is.
    you may well have it set correctly. (in fact it sounds like it is.)

    how to adjust the verge?
    with no known measurements and from parts on a bench?
    put just the verge and EW into the clock plates.
    do your best guess of distance from verge where it holds the pallet to the closest EW tooth
    when the pin is straight up. Set the pallet lock edge at that measurement.

    This is very likely to be a little deep into the EW. but it is a start.
    always take the verge out to adjust it. for this part you should only need to
    have the verge and EW in the plates.
    the fact is that once you get the verge set you should not need to tinker here anymore.
    (I think you are already done with this part...)

    check that your pallets reach the lock edge at about the same pin angle.
    or that the pallets are near the center of the impulse face when the pin is straight up.
    it is far more important to have the EW lock so I adjust the pallets to 'will lock.'
    Then I check for pin straight up near center of impulse face.
    If your clock is like the ones I have tinkered you will need to average that centered and straight up.

    To have even over swing timed from EW striking a lock face, your EW has to be even in its tooth spacing.

    a repete; to get more overswing, lower the fork.
    it may come with complications to attend, save dealing with them for later.
    victor
     
  8. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    Victor -

    I wish I lived closer...I'd come by and get some mentoring. I haven't digested your input, but it appears you have mastered a hands-on way of getting things set up. I'm still struggling with all of the various parameters and getting them to work together.

    Thanks...Kurt
     
  9. MartinM

    MartinM Registered User

    Jun 24, 2011
    2,893
    68
    48
    Male
    Medical Insurance Systems Analyst
    El Dorado, CA
    Country Flag:
    Region Flag:
    I know one thing that over complicated my envisioning of how it was all supposed to work was looking at where the anchor pin, pallets and EW teeth were in relationship to each other after the drop occurs. All that's really telling you is what the previous EW tooth's contact relationship looked like.
    Once I learned the important part is to get the EW tooth just beginning to impulse when the pin is just before center, it became much easier to figure out what needs tweaking.
     
  10. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    Martin -

    I learned that configuration here on the forum within the past few months...I fully understand why having the tooth on the first part of the impulse face with the pin vertical is a must...that's one of the first things I check for. But after that, it becomes muddied for me.

    In this clock, the pin was operating completely to one side before I began to adjust the escapement...so that needed to be fixed. I first measured the amount of exposed pallets and found them to be unequal. I found the average between the two and adjusted the pallets accordingly. It certainly helped, but I think it still needs tweaking. The drops were unequal to so I bent the peninsula a bit to a point where I'm pretty comfortable with the drops. But the locks aren't right, so that's where I sit.

    I did the fork lowering tests...I got it as low as it would go but to no avail. The clock just doesn't have a lot of power resulting in marginal amount of over swing. It will run without motion works, but soon as I add the motion works, it won't run. I have to relook at the meshing of the minute arbor with the motion works and then address the deep locks on one pallet.

    I say that like I know what's going on...we will see!

    Kurt
     
  11. victor miranda

    victor miranda Registered User

    Jan 13, 2017
    316
    29
    28
    computer programming stuff
    Baltimore
    Country Flag:
    Region Flag:
    Hi Kurt,
    please give your self some (a lot of) credit for knowing how to handle an escape.
    you give me every indication you are doing things right with it.

    trust yourself. You are close on that part/area of the clock.
    An escape can't make power that is not delivered to the EW.
    I have a clock that will not tolerate any sort of level that does not include a little leaning back.
    one slice of shirt card board under the front feet is all the difference from runs good to dead stop.

    so look for clean, burnished pivots and no slop... and a properly lubricated spring next...
    If you have a Schatz with the fixed verge watch that one for a bit and adjust it.
    it will give you a sense of why the pallets do not usually need adjustments.

    and yeah the motion works can eat enough power to stop a 400 day.

    victor
     
  12. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    I spent some time with the pallets, moving them around and keeping in mind the requirement for the pin to work equally left to right relative to center.

    Things must have paid off as the clock has been running for nearly 24 hours and ended up with good over swing, around 45 degrees. I was impressed when I started it...I rotated the pendulum until an EW tooth dropped and let the pendulum go. It was in beat and it actually built up rotation from that humble beginning. That's usually a good sign.

    I just installed the motion works. I didn't find any issues when just running the motion works with the minute arbor input. So, I'll see how it goes from here. Fingers crossed!

    Kurt
     
  13. MartinM

    MartinM Registered User

    Jun 24, 2011
    2,893
    68
    48
    Male
    Medical Insurance Systems Analyst
    El Dorado, CA
    Country Flag:
    Region Flag:
    Good on you, Kurt!
    Noe just follow Johnny Cash's instructions and go "One Piece at a Time".
     
  14. KurtinSA

    KurtinSA Registered User
    NAWCC Member

    Nov 24, 2014
    2,923
    67
    48
    Aerospace Engineer (Ret.)
    San Antonio, TX
    Country Flag:
    Region Flag:
    Thanks! I'm still not believing in what I'm doing...I have to let the clock run for 24 hours with good movement before I can move to the next step. But if I'm fortunate enough, I can finish this up tomorrow and clear the bench for something else. I'm tired of looking at this thing.

    Kurt
     

Share This Page