# Do clocks run slower as mainspring winds down?

Discussion in 'Clock Repair' started by bipid44, Mar 5, 2011.

1. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#1
I've been told that a clock will run faster with fully wound mainspring because the spring pushes the gears faster. The clock will slow down as the spring winds down and exerts less force.

I've also heard the opposite, that a clock will run faster as mainspring winds down because the lower tensions on the gears allows them to move faster. (Sort of like a person can walk faster when he isn't trying to walk faster.)

Both seem wrong to me. The pendulum should keep it accurate throughout the range of the spring until the power impulse is too small to keep the pendulum swinging.

The period of pendulum (the time to complete 1 oscillation) is only dependent on the length of the pendulum. Weight of the bob has no effect, other than inertial.

A pendulum with a period of 1 second will stay at that period unless the length is changed, for example by moving the bob up or down (and altering the length by moving the center of gravity).

How would the strength of the spring change anything? A stronger spring would impart more force to the pendulum, possibly increasing the amplitude but not changing the period. The weaker spring would impart less force, and when too weak to keep the amplitude large enough, the escapement stops, and then the pendulum stops as the amplitude gets smaller and smaller, but not slower and slower.

Am I missing something?

2. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
40,770
720
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#2
You're right in theory. If we could work with a perfect pendulum swinging in a vacuum with perfect power applied you'd be right on. However, there are so many things screwing up 'perfect' in a clock that all of that goes out the window Typically, a recoil type escapement will run faster under stronger power, and slower as the spring uncoils. The loss should be less than 5 minutes per week, so is minimal. Dead beat escapements have better rates over a wide range of power. There aren't any perfect clocks, unfortunately ... so we generally deal with what we have

3. ### Randy Johnson Registered User

Mar 30, 2005
767
1
18
Male
Self-Employed Clock Repairman
Country Flag:
Region Flag:
#3
Or, as I once heard an old timer explain to an elderly customer (who seemed to understand him perfectly)... "As that sprang runs down, the tick and the tock's gonna get closer together 'n speed up ye clock!"

4. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#4
I'm still missing something. The vacuum would not effect the period of the pendulum. Air or vacuum would change friction that will tend to stop the pendulum (smaller amplitude), not slow it down. And "perfect power" would better impart energy to counteracting frictional forces, but again would not change the period of the pendulum.

Temperature changes would effect the period, but the spring shouldn't.

Are these clocks using a pendulum? If they do, I can't see how the spring can effect the period.

5. ### Hayson Registered User

Mar 2, 2008
1,074
2
38
Country Flag:
Region Flag:
#5
Last edited: Mar 6, 2011
Bip.. Do your spring-wound pendulum clocks keep perfect time? If not, then the fact that they have pendulums which should THEORETICALLY keep perfect time, proves that other factors are in play. I'm not going to provide my opinions about what those factors are because I've done so before and been cried down by many others, each of whom has their own pet theory and I don't care to repeat the futility of that exercise. The one thing we (nearly) all can agree on, is that very few spring wound clocks approach the consistency and predictability of a good weight driven clock. Their performance can be enhanced significantly by using a long spring capable of running the clock for 2-3 wks but controlling the run time to 8 days by using geneva stops so that only the mid-range of the spring is used, but even so some variability is usually detectable. The upside? They're much more portable.

6. ### Scottie-TX Registered User Deceased

Apr 6, 2004
936
38
0
Retired
Mesquite, TX
Country Flag:
Region Flag:
#6
Please don't believe us. We're novices and theory is way over our heads. Get a springwound clock and observe it's behavior over a week of run for several weeks.

7. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#7
The phenomenon in question is called isochronal error. As has been mentioned, it occurs with spring driven clocks but not weight driven ones. The reason is that power delivered by gravity is essentially a flat curve, while power delivered by an uncoiling mainspring isn't.

A tightly wound spring at first releases its energy quickly, then settles down to a more or less constant rate for most of its period, and at the end when is it almost spent, it releases its energy more slowly. Geneva stops were invented to restrict the spring to the level part of its energy curve.

The period of a clock pendulum isn't constant, because it's not operating in a frictionless environment. That's why it requires an impulse from the scapewheel to keep going. The stronger the impulse, the wider the swing and the longer the period. As the old man said: the further apart the ticks and tocks, the slower the clock runs: the mechanism is only turning when the scapewheel is moving...between tooth release and capture. The longer between those events, the longer between the bits of movement in the mechanism, and the slower the clock runs.

Highly accurate clocks are designed to minimize every sort of error, including isochronal error. With ordinary clocks, the kind we're familiar with, the error is there, but slight enough not to notice. A slight loss at the beginning of the week's run is made up by a slight gain at the end of the run, with a fairly constant rate in between.

Or something like that.

bangster

8. ### Jay Fortner Registered User

Feb 5, 2011
5,023
24
38
Clockmaker/Watchmaker
Chiefland,Fl.
Country Flag:
Region Flag:
#8
Yeah, more like he said! I would watch the behavior of your clock over it's winding period. If your clock unloads it's time train during winding(you'll know this by the minute hand jumping back and to while you're winding it) set the rate a little fast so that it makes up the time it takes to wind it. You'll have good AVERAGE accuracy. I'm a big fan of the second rating nut, you use it to fine tune the rate.

9. ### cazboy Registered User

Apr 27, 2006
1,106
1
0
Retired!
Prescott Valley, Arizona
Country Flag:
Region Flag:
#9
Fascinating discussion. I am in the habit of winding my 8-day clocks twice a week - Sunday morning and Wednesday evening. That splits the week in half, and keeps the springs wound up to their upper half. That's just me, though.

10. ### Willie X Registered User

Feb 9, 2008
12,003
697
113
#10
bipi,

Most will run faster when fully wound, there are exceptions. Keeping a log on several clocks of the different categories of clocks would give you some good empirical info to go on.

The ones most notorious for going from fast to slow are in the heavy spring, recoil escapement, with light and short pendulum category.

Willie X

11. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#11
The speed of the clock is the speed of the escape wheel. I can't think of any way that excess power could cause an escape wheel to turn faster than the pendulum will allow.

But perhaps I'm overlooking something.

bangster

12. ### Charles E. Davis Registered User NAWCC Life MemberNAWCC FellowGolden Circle

Nov 6, 2000
909
6
18
La Verne, CA
Country Flag:
Region Flag:
#12
The problem of "fast . . steady slow" is mainly a problem with recoil. The fusee and deadbeat and weight driven all give better timekeepking.
The cause with the recoil escapement is that when fully wind it does not allow the pendulum to make a full swing. Thus the tick-tocks are very rapid. Runs fast. As the spring winds down the pendulum assumes its normal wider swing and that swing continues until the clock stops and needs to be wound. It moves gradually to the normal speed of the pendulum which is slower and then may gradually speed up as the pendulum swing dies due to loss of power and clock comes to a complete stop.

13. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#13

I'm not asking about a specific clock that I have, just a theoretical discussion.

My next step is reading up on isochronal errors. In addition, it would seem that clock pendulums are examples of damped harmonic oscillators, with step input. I think the system is an example of "under-damped" with the frequency being slightly different than the un-damped case.

Maybe the transmission of impulse from the escapement to the pendulum may not occur at the exact proper point in the cycle and changes the effect of the step input and making it more of a damping effect?

http://farside.ph.utexas.edu/teaching/315/Waves/node12.html

but I'm getting dizzy from trying to work my way through the info.

A working demo that I haven't played with yet, from Stephen Wolfram, is at
http://demonstrations.wolfram.com/DrivenDampedOscillator/

14. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#14
Hey Charles...

I'll stand corrected; but please explain the part about a tightly wound spring preventing the pendulum from taking a full swing. How does it do that?

And does a barely-wound spring, at the end of the run, cause the pendulum to take a more-than-full swing? How would it do that?

I need explained to, cuz I don't get it.

bangster

15. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#15
I think this is the answer, at least I like it, that the impulse is not imparted to the pendulum at the correct part of the oscillation of the pendulum.

16. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#16
Details?? How does excess power affect the point of impulse??

17. ### Dave B Banned

Jun 7, 2008
2,389
1
0
Civil Engineer; woodwind musician; clock repairman
Westminster. MD
Country Flag:
#17
That may sound good in theory, but I beg to differ empirically. My Seth Thomas Naples (I can't recall, offhand, which movement number - it is a round one, with loop end springs, and both springs winding counter clockwise) runs slow when fully wound, and gains when it is almost completely run down. I wind my eight day clocks before I go to bed on Sunday evening. The Naples looses as much as a minute and a half on Monday, but by the following Sunday, it agrees with my weight-driven clocks.

18. ### Willie X Registered User

Feb 9, 2008
12,003
697
113
#18
bipidie,

So you just asked the question to show off knowledge
(or information) you already have. You probably know that after you do that a few times on these list, people will no longer respond to your queries. In 'list language', when you troll they won't bite.

Willie X

19. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#19
Hi
There are several factors one has to consider.
1. Large swing on a pendulum will be slower ( most use
such a small swing that this doesn't have a lot of effect.
2. An impulse is suppose to be an instantaneous addition
of energy. This never happens in a practical clock. The entire
time that the escapement causes a push on the pendulum,
it is either speeding it up or in the case of the recoil,
slowing it down for part of the swing( both shortening the time ).
3. The method of the escapement will have friction. Such
as a dead beat. Part of the swing is dragging on the pallet.
This tends to slow the swing.
4. Friction, such as air friction or dead beat both slows
the swing, making it take longer and also reduces the
amplitude of the swing, making it go faster. For shorter
pendulum clocks, ampitude change is the bigger factor.
This may be where the faster at the end of wind comes from.

I would expect mantel clocks to run faster but grandfather
to run slower.

Triptin likes this.
20. ### Kevin W. Registered User NAWCC Member

Apr 11, 2002
21,671
214
63
I work at the Veritas Tools machine shop.
Country Flag:
Region Flag:
#20
I just know that the recoil escapement does not run as accurate over time as a dead beat.And they usually slow down in the summer when its hotter.My dead beats seem to keep the same time all the time.
I am not a expert in clocks, this is just what i have observed from my own experiences.

21. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#21
Dead beats are generally more accurate because they transfer
energy at the time of the swing when a change in volocity has
the least effect on the timing of the swing. That is when the
pendulum crosses at the highest speed and for the shortest amount
of time.
Anything like a recoil that changes the energy in the pendulum
near the end of the swing will cause a greater effect because
the pendulum is going slow and will be influenced by that
force longer.
Tinker Dwight

22. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#22
Let me ruminate on this bidness for a minute, see if I can clear my head.

It's a given fact that energy release is greater at the beginning of a spring's unwinding than it is later. The question is how that difference translates into the speed of a clock.

The Common Sense (or "Probably Wrong") analysis I gave earlier is this: Greater energy ==> stronger impulse to pendulum ==> wider swing ==> longer period ==> slower clock.

Charles says: Not so, at least with recoil escapements. With them, Greater energy ==> narrower swing ==> shorter period ==> faster clock.

I'm trying to figure out how greater energy to a recoil EW could retard the pendulum swing. And how lesser energy at the tail end of the period could advance pendulum swing.

44 suggests that the retardation is because greater energy causes the impulse to happen at the wrong point in the pendulum's arc. This would indicate that, as the spring runs down, the impulse point gets righter and righter, until it's so right that the clock runs slow (or something like that).

But that gives me no picture that I can understand. What EXACTLY does more energy to the recoil EW do, to retard the pendulum's swing?

Is it this? the recoil occurs during the pendulum's "run" after tooth capture. In effect, the pendulum pushes the EW backwards. More energy to the EW means more resistance to the push, meaning the pendulum doesn't "run" as far, meaning its swing in retarded from what it would be if there were less resistance (= less energy) at the EW. That's a picture I can understand.

But still, after the recoil is over, the EW is giving the pendulum a stronger impulse than it would have had if there were less energy to the EW...advancing the pendulum's swing. Why shouldn't the advance cancel out the retard? If Charles is right, then it doesn't. But why not? Here's where I run out of ideas and need help.

And a second issue has branched out here that needs some explaining: recoil vs. deadbeat. If the above account were the whole story, then deadbeat escapements ought to be immune to isochronal error. I've never read anywhere that they are immune: they are less prone than recoils, but it's still there. If that's not the case...if the deadbeat escapement actually eliminates all isochronal error, then I've learned something important that I didn't know.

But if deadbeats are also susceptible to isochronal error, the same question needs answered: How does greater energy release from the mainspring affect the speed of the clock? And here I'd retreat to my earlier Commmon Sense/Probably Wrong notion: stronger impulse ==> wider swing ==> slower clock. Probably wrong, but I need to know how.

bangster

23. ### Jerry Treiman Registered User NAWCC Member

Aug 25, 2000
6,050
1,019
113
Geologist - California Geological Survey
Los Angeles, CA
Country Flag:
Region Flag:
#23
There have been a lot of good comments about the effect of the escapement mechanism and how it might respond to power variation in the train. Two other factors in the clock construction affect a) the uniformity of power delivered by the mainspring and the b) isochronicity of the pendulum.

a) The fusee was an attempt to compensate for the varying power as the spring runs down. A properly designed and properly set up fusee can largely eliminate the power variation (hence its use in marine chronometers). Stopworks can also restrict the running of a timepiece to the central part of the power curve, but I do not know if these were used as much in clocks as they were in watches.

b) Most pendulums are not truly isochronal for different amplitudes of arc. The bob follows part of a circular arc which, if small, will have only slight errors with amplitude (circular error). To truly have an isochronal pendulum the bob must follow a cycloidal arc. This was effected by Christian Huygens in some of his clocks with a special suspension system (cycloidal cheeks and silk thread suspension, I believe).

24. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#24
That brings it back to the conflicting info I heard years ago, that clocks with run down springs run faster.

Sounds counter intuitive so it has a certain appeal.

Seems like a simple thing to check, but turns out hard to keep track for an undisciplined person.

I know what you mean.

Isn't the escapement to prevent an impulse of power going to the pendulum at the wrong time in the cycle?

And _IF_ the power impulse does effect the amplitude of the pendulum swing, by supplying the power either too soon or too late, and that slowed down the period, the clock should come to a stop. Each tick would limit the amplitude until it was too small to keep the clock running.

Back where I started. Confused.

Yes, I've heard that fusee work better, but are they correcting a clock that is running faster or slower as the spring winds down?

And while the pendulums aren't isochronal for differing amplitudes, will the strength of the spring change the amplitude? If the spring increases the amplitude, as the clock continues to run the pendulum would hit the side of the case. If the spring decreases the amplitude, the pendulum should come to a stop.

25. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#25
As I see it, the escapement is to prevent the clock from running down all at once; it allows the spring to run down a bit at a time. The escapement also serves to keep the pendulum going. The timing of the impulse is a function of the "beat", which is independent of spring power. Improperly synchronized impulses mean stoppage: "Out of Beat."

bangster

26. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#26
Last edited: Mar 6, 2011
I have a clock that I just changed the springs on and it ran for almost week fine and then on the 6th day it lost 1 and a half minutes. When I wound it back up it went back to keeping the correct time. I wonder if the clock having perfectly strait pivots with less friction and if the clock is setup properly as far as the beat of the clock being perfect has anything to do with it also? It seems it would make springs work harder if the clock was struggling to keep in beat.
If you put the wrong size springs in a clock on the strike side, does it change how hard the hammer strikes the gong? I am just asking. You guys know a lot more than me.
My vienna regulator has a weight for consistent pressure to the pendulum and it keeps better time than any of my spring driven, and does not fluctuate at all, like my spring driven i just changed the springs on did.
I wonder if I changed the size of the weight on my vienna regulator, if it would change how fast my pendulum swings?
Does anyone here know the answer to that?

27. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#27
Would one way to find the answer to his question as to whether springs cause a difference in the accuracy of a clock be to use a weight driven clock and change the weight of the clock, trying heavier and then lighter weight and see if that changes the speed of the pendulum? Wouldn't that be the same as changing a stronger gauge spring out for a weaker gauge spring?
If he is trying to find out the effects of power to the train and to the pendulum, that is what I would try.

28. ### Scottie-TX Registered User Deceased

Apr 6, 2004
936
38
0
Retired
Mesquite, TX
Country Flag:
Region Flag:
#28
Changing the weight of a weight driven clock will affect regulation.

29. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#29
Last edited: Mar 6, 2011
So if that is the case scottie, would the changing of a spring to a weaker or stronger spring affect the speed of a clock in the same way?
What do you mean by regulation?
Will a lighter weight slow it down and a heavier weight speed it up?

30. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#30
Klocken: Have you been following this thread at all? If you have, you should know that several of your many questions have already been answered.

Also...

Slowing down at the end of the run is due to isochronal error, discussed passim in the thread above. Read the posts.

A clock doesn't struggle to stay in beat. If it's out of beat it doesn't try to get back in. It just quits.

No. The hammer spring (if it has one) or gravity determines how hard the hammer strikes.

As stated above, weight-driven clocks are by their nature more accurate than spring driven clocks. The reason is isochronal error. Read the posts.

It would change how far they swing. That in turn was affect the rate: how fast or slow the clock runs.

bangster

31. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#31
I think that indeed goes by the name of "Circular Error" rather than "Isochronal Error", even though it does involve isochronicity.

Doesn't it?

bangster

32. ### Scottie-TX Registered User Deceased

Apr 6, 2004
936
38
0
Retired
Mesquite, TX
Country Flag:
Region Flag:
#32
KLOCKEN, by, "regulation" we mean the process of making a clock produce accurate time. If a clock runs fast or slow it needs regulated. An accurate clock is properly regulated.
A good example of a poorly regulated clock are the reproduction foliots of those made in the 13th C. They have only three wheels and increasing weight makes them run much faster - visibly and audibly faster.

33. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#33
Sorry, I didn't mean to get anyone fusster8ed.
I will be quiet.

34. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#34
Can someone who know's email me in private message and explain it to me. I don't understand this statement. Why does a recoil escapement not allow the pendulum to make a full swing when fully wound? I don't want to upset anyone, so if you could private msg me and tell me that would be great.

35. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
40,770
720
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#35
You're OK. The recoil escapement is 'charged' by backing up the entire train, which then pushes it back for the impulse. The spring determines in large part how much it can back the train up, and progressively weaker impulses result as the spring is depleted. You'd think that would speed things up because of decreased swing, but it does the opposite.
I'm curious about your inquiry, though. Most people who work on clocks just accept things as they are and don't really need or desire to fully comprehend the exact forces at work. I know I don't

36. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#36
Last edited: Mar 6, 2011
I like to know what makes things tick.

I am looking up all of these terms like Isochronal and circular error etc. One of my favorite show's growing up was Mr. Wizard.

The main question here was asked (do clocks run slower as mainsprings wind down?).

I am not sure whether I heard the answer because the terms used were complicated. So I am still not sure if clocks run slower as the mainsprings wind down or not.

I installed springs one time in a mantel clock and everyone said that the sound of the hammers was weaker and I never did understand what I did that caused the sound to dampen. I figured it was because I used a spring that was not strong enough, but according to some post here, I should have checked the spring on the hammer itself. I wish I would have known that at the time. I must have somehow loosened the hammer spring when it was apart, I guess.

37. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#37
I love learning new stuff. To me clocks are a blast to learn and work on. I want to know everything about them. They are fascinating to me. I wish I could go to school with guy's like these here because I would love to see them work and fix some of these complicated workings on these clocks. I guess i am just weird that way.

38. ### bangster Moderator NAWCC Member

Jan 1, 2005
19,279
328
83
utah
Country Flag:
Region Flag:
#38
See post #14 of this thread, and what follows.

39. ### bipid44 Registered User

Feb 6, 2011
32
0
0
#39

Not directed at me, but as threads get beyond a certain length, it's really hard to read everything and to understand everything. Posts that summarize every few pages are greatly appreciated especially if I'm reading an old thread.

40. ### Scottie-TX Registered User Deceased

Apr 6, 2004
936
38
0
Retired
Mesquite, TX
Country Flag:
Region Flag:
#40
Along with the strength of the hammer drive spring as well as those that don't have drive springs - loudness of strike is much dependent on gap between coil or gong and hammer that strikes it. More likely in your case is that perhaps when installing the movement, the gap between them became wider and so not as loud. Bending the hammer wire to make hammer closer to coil or gong would have made it louder too..

41. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#41
Last edited: Mar 6, 2011
I went back and read and tried to understand it. I just have so many questions that it would make you even more frustrated in trying to get me to understand. For instance.
1. If wound springs and weights are theoretically the same thing, then why are they not compared when trying to figure out if wound springs lose power as they unwind?
Would it be possible to build 2 clocks, one being a weight driven and one being a spring driven. In every respect making them identical except for the method used for driving their respective trains.
Making them both use deadbeats escapements etc. The only difference would be what type of power is supplied.
What would be the difference between the 2 clocks if you changed the amount of weights on the one and spring tension on the other?
Again I am asking that the pendulum be the exact same size and weight and everything identical. The only difference is the method of power used.
Is the reason that the weight driven clock is more accurate because the power remains constant and does not change?

Quote:
My vienna regulator has a weight for consistent pressure to the pendulum and it keeps better time than any of my spring driven, and does not fluctuate at all, like my spring driven i just changed the springs on did.

Bangster quote
As stated above, weight-driven clocks are by their nature more accurate than spring driven clocks. The reason is isochronal error. Read the posts.

Isochronal definition = recurring at regular intervals

So what does Ischronal error mean when talking about clocks?
What causes Isochronal error?

Thanks bangster for being patient with me. I am just trying to learn.

42. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#42
Last edited: Mar 6, 2011
So scottie, if I use a spring on the strike side of the train that is weak, what effect does it have on the sound and also the time of the strike? Would it make the train just run slow and not affect the force of the strike?

43. ### harold bain Registered User NAWCC MemberDeceased

Nov 4, 2002
40,850
143
63
Male
deceased
Country Flag:
Region Flag:
#43
I have to ask, what is a time strike spring?

44. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#44
Last edited: Mar 6, 2011
Sorry, I changed the spring on the strike side of the train of a mantel clock and I apparently used a weak spring and the train moved really slow and the strike sounded weak. Sorry Harold it is late and I am tired. Thanks for pointing that out.

Harold, is there a book or article that I can go and read that deals with Isochronal errors and circular errors? These guy's are so far advanced that i can't keep up. I am ticking them off!

45. ### harold bain Registered User NAWCC MemberDeceased

Nov 4, 2002
40,850
143
63
Male
deceased
Country Flag:
Region Flag:
#45
A very good book on pendulum theory, and much more, is The Modern Clock by Ward Goodrich. Although it was first written in 1905, it is still very relevant today.

46. ### Thyme Banned

Sep 18, 2006
3,948
2
0
metro NY area
#46
This is an interesting thread. I too, just accepted that it is how most spring driven clocks behave. Of my 30+ clocks, all except two are spring driven. Most run slightly fast when first wound, then level off by the end of the seventh day to within five minutes of the "real" time. But I am quite sure that a few do keep time steadily, without the phenomenon of speeding up then eventually slowing down over the course of a week.

Admittedly, I don't know which clocks have movements that are recoil, deadbeat, or whatever. The long drop ones are usually more accurate simply because of the longer drop.

Two observations: since not all clocks do the speed up & slow down routine to the same degree (or at all) I suspect it is caused by slight amounts of wear in the entire train. (After all, these clocks are mostly over a century old, and most have bushings and pivots that are OK but not 'like new'.) Another possibility (adding to the aforementioned) is in the realization that the springs are not new either, so they may be less potent than new springs would be. A good example of this is illustrated in one of them, an Ansonia long drop wall regulator that runs out of pep in the strike train on day six. It still tries to strike but hasn't enough impetus to hit the gong. That's a spring that has lost its vigor - and it becomes obvious on the final day of the run.

47. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#47
Last edited: Mar 6, 2011
I have that red book, I just haven't read it yet as you can tell. Thanks Harold. I appreciate all of you guys.
I just downloaded Mr. Labounty's pdf on lever's. I just fixed a clock that was striking the full hour on the half hour and it was a mess. I bent the levers wrong and it took me hours to figure out how to set it back correctly. You guy's helped me understand it and I finally set it back right. But I kept taking the movement apart and trying to set the levers and then putting it back together and I have been told that you can actually set the levers without taking the movement apart. I learned the hard way.

48. ### chimeclockfan Registered User NAWCC Member

Dec 21, 2006
3,812
139
63
Chime clock & gong studies.
WI
Country Flag:
Region Flag:
#48
Duh.

49. ### Richard T. Deceased Deceased

Apr 7, 2005
5,064
2
0
Country Flag:
Region Flag:
#49
Read and study...........you cannot expect to learn in a short period of time, something that has taken years of study and experience to fully understand. There are things related to clocks that many of us still do not fully understand. If you will, a weak area. It may be escapements or making a replacement rack and snail. (just examples).

If you are not already a member of the NAWCC, join, find a local chapter and a mentor to help you. Local chapters often have classes on various topics related to clocks/clock repair. Local chapters also provide the opportunity to be with like minded people and discuss problem areas or anything related to or not related to horology.

Best,

Richard T.

50. ### klocken Registered User

Feb 25, 2011
191
0
0
Country Flag:
Region Flag:
#50
You duh man!