I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way
to make this happen? I was thinking that resistance inserted between
the battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the
latter so much so that it keeps hitting the insides of the clock. It's
not a loud sound, but far from your typical "tick-tock" which is
desired. I'm thinking an added resistance might reduce both battery
drain as above and reduce the pendulum amplitude, but what about
frequency?

Thanks in advance for any help.

Bill

On 10/11/2012 10:22 AM, Bill Walston wrote:
I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way to
make this happen? I was thinking that resistance inserted between the
battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the
latter so much so that it keeps hitting the insides of the clock. It's
not a loud sound, but far from your typical "tick-tock" which is
desired. I'm thinking an added resistance might reduce both battery
drain as above and reduce the pendulum amplitude, but what about frequency?

Thanks in advance for any help.

Bill
First thing to do is state the problem.
"very long"
"much longer"
We talking minutes? years?

Can't tell from the link.
You make a pendulum slower by making it longer.
But google would have told you that in the first hit.
You make it more efficient by reducing drag.
You probably don't want to use a vacuum enclosure,
but you might be able to reduce bearing friction.
Then you "kick" it in resonance to reduce the battery drain.
The link doesn't say how the "kicker" works.
You reduce the energy "kick" to the point that the energy
in balances the pendulum losses at the amplitude you want.
Resistor might help the amplitude, but may not help the
battery drain. Unless the "kicker" is designed for it,
the amplitude will change with the battery age.

On Thu, 11 Oct 2012 13:22:02 -0400, Bill Walston wrote:

I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way
to make this happen? I was thinking that resistance inserted between
the battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the
latter so much so that it keeps hitting the insides of the clock. It's
not a loud sound, but far from your typical "tick-tock" which is
desired. I'm thinking an added resistance might reduce both battery
drain as above and reduce the pendulum amplitude, but what about
frequency?

Thanks in advance for any help.

Bill
Frequency is dependent on the length of the pendulum and the strength
of the gravitational field. To slow the pendulum down either increase
the length of the pendulum or reduce the strength of the gravitational
field. Either use a D cell or an external power supply to increase
the battey life.

PlainBill

Frequency is dependent on the length of the pendulum
and the strength of the gravitational field.

Point... As this pendulum is designed for a clock (I assume), wouldn't it
have a specified period appropriate for a clock?

On 10/11/2012 03:01 PM, mike wrote:
On 10/11/2012 10:22 AM, Bill Walston wrote:
I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way to
make this happen? I was thinking that resistance inserted between the
battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the
latter so much so that it keeps hitting the insides of the clock. It's
not a loud sound, but far from your typical "tick-tock" which is
desired. I'm thinking an added resistance might reduce both battery
drain as above and reduce the pendulum amplitude, but what about
frequency?

Thanks in advance for any help.

Bill
First thing to do is state the problem.
"very long"
"much longer"
We talking minutes? years?

Well, the former lithium 1.5V battery powered it for maybe 4-5 months.
I'd like to stretch this out to a year at least.

Can't tell from the link.
You make a pendulum slower by making it longer.
But google would have told you that in the first hit.
You make it more efficient by reducing drag.
You probably don't want to use a vacuum enclosure,
but you might be able to reduce bearing friction.
Then you "kick" it in resonance to reduce the battery drain.
The link doesn't say how the "kicker" works.
You reduce the energy "kick" to the point that the energy
in balances the pendulum losses at the amplitude you want.
Resistor might help the amplitude, but may not help the
battery drain. Unless the "kicker" is designed for it,
the amplitude will change with the battery age.

The circuit is a 2 transistor design with resistors, capacitors and of
course the coil. I didn't try to map it out, but it looks similar to
the 2 transistor pendulum circuits on the web.

I know we can change the frequency by increasing pendulum length, but
there's no room in the enclosure. Why wouldn't a simple capacitor
change in the circuit accomplish the same goal? Isn't the coil just
part of a resonant LC circuit?

Bill

Bill Walston wrote:

On 10/11/2012 03:01 PM, mike wrote:
On 10/11/2012 10:22 AM, Bill Walston wrote:
I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way to
make this happen? I was thinking that resistance inserted between the
battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the
latter so much so that it keeps hitting the insides of the clock. It's
not a loud sound, but far from your typical "tick-tock" which is
desired. I'm thinking an added resistance might reduce both battery
drain as above and reduce the pendulum amplitude, but what about
frequency?

Thanks in advance for any help.

Bill
First thing to do is state the problem.
"very long"
"much longer"
We talking minutes? years?

Well, the former lithium 1.5V battery powered it for maybe 4-5 months.
I'd like to stretch this out to a year at least.

Can't tell from the link.
You make a pendulum slower by making it longer.
But google would have told you that in the first hit.
You make it more efficient by reducing drag.
You probably don't want to use a vacuum enclosure,
but you might be able to reduce bearing friction.
Then you "kick" it in resonance to reduce the battery drain.
The link doesn't say how the "kicker" works.
You reduce the energy "kick" to the point that the energy
in balances the pendulum losses at the amplitude you want.
Resistor might help the amplitude, but may not help the
battery drain. Unless the "kicker" is designed for it,
the amplitude will change with the battery age.

The circuit is a 2 transistor design with resistors, capacitors and of
course the coil. I didn't try to map it out, but it looks similar to
the 2 transistor pendulum circuits on the web.

I know we can change the frequency by increasing pendulum length, but
there's no room in the enclosure. Why wouldn't a simple capacitor
change in the circuit accomplish the same goal? Isn't the coil just
part of a resonant LC circuit?


No. The resonance is mechanical, the coil just applies a little
energy to overcome friction and keep it moving. All changing the
circuit's frequency will do is waste the battery's life as heat.

Bill Walston wrote:


I know we can change the frequency by increasing pendulum length, but
there's no room in the enclosure. Why wouldn't a simple capacitor
change in the circuit accomplish the same goal? Isn't the coil just
part of a resonant LC circuit?
No, the RC or LC timer in the circuit does not set the pendulum swing
rate. it really can't, the pendulum is WAY too massive for a tiny,
battery-powered circuit. So, first, you need to fix the pendulum.

Second, you need to reduce the kick the circuit applies to the pendulum each
swing. That might be done by adding turns to the coil that provides the
kick. Or, the RC timer is leaving the coil current on too long, so
reducing the cap value might reduce the energy imparted, as well as preserve
battery life.

Jon

I recently purchased an electronic pendulum to be used in a clock I recently
digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which didn't
last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way to
make this happen? I was thinking that resistance inserted between the
battery and circuit might do the trick, but unsure.

b) The pendulum runs too fast and has too much amplitude, with the latter so
much so that it keeps hitting the insides of the clock. It's not a loud
sound, but far from your typical "tick-tock" which is desired. I'm thinking
an added resistance might reduce both battery drain as above and reduce the
pendulum amplitude, but what about frequency?

Thanks in advance for any help.

Bill

Cut out the sides of the clock to allow it full swing.
If this is a pendulum designed to tick at the correct rate then you
have to modify the cabinet structure to accomodate it not the other way
around.
So when you cut out the sides of the clock cabinet you will have a more
unique conversation piece. Put up a photo after you get it working.

On Thu, 11 Oct 2012 13:22:02 -0400, Bill Walston put
finger to keyboard and composed:

I recently purchased an electronic pendulum to be used in a clock I
recently digitized:

http://www.klockit.com/products/dept...sku-20071.html

The pendulum take a single AA battery. I used a lithium battery which
didn't last very long. I have several goals I want to try and implement:

a) I would like the battery to last much longer. What's the best way
to make this happen? I was thinking that resistan

Would there be any way to utilise a solar panel and a NiCad or NiMH
battery, or maybe a super capacitor? For example, you could
cannibalise a cheap solar garden light.

Excuse my ignorance, but how does your circuit know when to apply a
kick? If the kick were to come when the pendulum is on the rise, then
that would work against it. Therefore the circuit would need to know
when the pendulum has begun its descent and apply the kick at that
time.

Furthermore, if the pendulum's amplitude is growing too large, then
the circuit would need to refrain from kicking it until the amplitude
subsides, if only to conserve the battery.

Therefore ISTM that the circuit must be sensing the pendulum's
position, in which case you would need to adjust the sensor in order
to set your desired amplitude. Or am I way off?

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Excuse my ignorance, but how does your circuit
know when to apply a kick?

It doesn't.

Assume the pendulum is supposed to have a period of one second. You design
it to be a little bit longer, then make the driver circuit operate at
exactly one second. The pendulum will eventually sync with the driver.

Remember when TVs had hold controls? The principle is the same. The
most-stable operation is obtained when the oscillator runs a tiny bit slower
than it should, with the sync signals "kicking" it at the right frequency.

On Thu, 11 Oct 2012 12:10:30 -0700, "William Sommerwerck"
wrote:

Frequency is dependent on the length of the pendulum
and the strength of the gravitational field.

Point... As this pendulum is designed for a clock (I assume), wouldn't it
have a specified period appropriate for a clock?

The specific product he identifies is designed as an add-on for a
quartz movement and comes with a pendulum with an adjustable length
arm. I'm unsure of the correlation between pendulum length and
amplitude of the swing, but I would expect a shorter pendulum to swing
through a wider angle.

PlainBill

On Oct 12, 6:00*pm, wrote:
On Thu, 11 Oct 2012 12:10:30 -0700, "William Sommerwerck"

wrote:
Frequency is dependent on the length of the pendulum
and the strength of the gravitational field.

Point... As this pendulum is designed for a clock (I assume), wouldn't it
have a specified period appropriate for a clock?

The specific product he identifies is designed as an add-on for a
quartz movement and comes with a pendulum with an adjustable length
arm. *I'm unsure of the correlation between pendulum length and
amplitude of the swing, but I would expect a shorter pendulum to swing
through a wider angle.

PlainBill

If I remember correctly, a larger weight on the pendulum will also
slow it down.

William Sommerwerck wrote in message
...
Excuse my ignorance, but how does your circuit
know when to apply a kick?

It doesn't.

Assume the pendulum is supposed to have a period of one second. You design
it to be a little bit longer, then make the driver circuit operate at
exactly one second. The pendulum will eventually sync with the driver.

Remember when TVs had hold controls? The principle is the same. The
most-stable operation is obtained when the oscillator runs a tiny bit
slower
than it should, with the sync signals "kicking" it at the right frequency.



Human clocks are the same. Put a human in a cave, out of touch with the
outside world, and his natural day-length reverts to about 24.5 to 25 hour
days, requires the sun etc to sync him to 24 hour days

On Fri, 12 Oct 2012 20:55:56 -0700 (PDT), "hr(bob) "
wrote:

On Oct 12, 6:00*pm, wrote:
On Thu, 11 Oct 2012 12:10:30 -0700, "William Sommerwerck"

wrote:
Frequency is dependent on the length of the pendulum
and the strength of the gravitational field.

Point... As this pendulum is designed for a clock (I assume), wouldn't it
have a specified period appropriate for a clock?

The specific product he identifies is designed as an add-on for a
quartz movement and comes with a pendulum with an adjustable length
arm. *I'm unsure of the correlation between pendulum length and
amplitude of the swing, but I would expect a shorter pendulum to swing
through a wider angle.

PlainBill

If I remember correctly, a larger weight on the pendulum will also
slow it down.
You don't. The effective length of the pendulum (distance between the
pivot and the center of the mass) and the gravitational force are the
only things that affect the speed. Remember the experiment Galileo
did a few years before he turned his eye to the heavens.

PlainBill

On Fri, 12 Oct 2012 14:02:31 -0700, "William Sommerwerck"
put finger to keyboard and composed:

Excuse my ignorance, but how does your circuit
know when to apply a kick?

It doesn't.

Assume the pendulum is supposed to have a period of one second. You design
it to be a little bit longer, then make the driver circuit operate at
exactly one second. The pendulum will eventually sync with the driver.

Remember when TVs had hold controls? The principle is the same. The
most-stable operation is obtained when the oscillator runs a tiny bit slower
than it should, with the sync signals "kicking" it at the right frequency.

Thanks. I'm still trying to understand how it syncs, though.

I'm trying to envision an equilibrium where the kick comes at a
regular point in the swing. Does the pendulum sync so that it gets a
kick at the beginning of the downswing, or does it come at the end of
the upswing, or can it come at any point in the arc?

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Franc Zabkar wrote:

On Fri, 12 Oct 2012 14:02:31 -0700, "William Sommerwerck"
put finger to keyboard and composed:

Excuse my ignorance, but how does your circuit
know when to apply a kick?

It doesn't.

Assume the pendulum is supposed to have a period of one second. You design
it to be a little bit longer, then make the driver circuit operate at
exactly one second. The pendulum will eventually sync with the driver.

Remember when TVs had hold controls? The principle is the same. The
most-stable operation is obtained when the oscillator runs a tiny bit slower
than it should, with the sync signals "kicking" it at the right frequency.

Thanks. I'm still trying to understand how it syncs, though.

I'm trying to envision an equilibrium where the kick comes at a
regular point in the swing. Does the pendulum sync so that it gets a
kick at the beginning of the downswing, or does it come at the end of
the upswing, or can it come at any point in the arc?


I've only seen one electronic pendulum. it used a sensing coil to
detect the approach of the pendulum, and fired a pulse to add to the
stored energy.

Assume the pendulum is supposed to have a period of one second.
You design it to be a little bit longer, then make the driver circuit
operate at exactly one second. The pendulum will eventually sync
with the driver.

Remember when TVs had hold controls? The principle is the same.
The most-stable operation is obtained when the oscillator runs a bit
slower than it should, with the sync signals "kicking" it at the right
frequency.

Thanks. I'm still trying to understand how it syncs, though.

There is no "intelligence" to the synchronization process. Because the
pendulum swings at a different frequency with respect to the drive circuit,
it has a continually varying phase relationship with it. So, at some point
the kick will occur when the pendulum is near the energized drive
electromagnet. The cycle starts at this point.

On the next swing, the pendulum will be slightly "late". But if it's "close
enough" to the energized electromagnet, it will receive another kick. And so
on, and so on...

I'm trying to envision an equilibrium where the kick comes at a
regular point in the swing. Does the pendulum sync so that it gets
a kick at the beginning of the downswing, or does it come at the
end of the upswing, or can it come at any point in the arc?

I assume at the end of the upswing.

I've only seen one electronic pendulum. it used a sensing coil
to detect the approach of the pendulum, and fired a pulse to
add to the stored energy.

I'm not sure that would work correctly. You want the drive coil to fire at a
fixed frequency. Otherwise, the pendulum would be synched at its lower
native frequency.

hr(bob) wrote in message
...
On Oct 12, 6:00 pm, wrote:
On Thu, 11 Oct 2012 12:10:30 -0700, "William Sommerwerck"

wrote:
Frequency is dependent on the length of the pendulum
and the strength of the gravitational field.

Point... As this pendulum is designed for a clock (I assume), wouldn't it
have a specified period appropriate for a clock?

The specific product he identifies is designed as an add-on for a
quartz movement and comes with a pendulum with an adjustable length
arm. I'm unsure of the correlation between pendulum length and
amplitude of the swing, but I would expect a shorter pendulum to swing
through a wider angle.

PlainBill

If I remember correctly, a larger weight on the pendulum will also
slow it down.


from http://www.adam-hart-davis.org/odd_but_interesting.htm

Q. Why do the people who regulate "Big Ben's clock use adding or
subtracting old pennies to the weight on the bottom?
- John Gifford, London

A. Yes the period of pendulum is governed by its length and is independent
of the mass at the end. But if you add an old penny to the bob of the
pendulum of Big Ben, then you slightly alter its length; in fact you make it
a bit shorter, because the centre of mass of the bob is slightly raised.
Similarly if you take pennies off, you will in effect slightly lengthen the
pendulum. This seems an odd way to do it, but it is obviously very
convenient, and simpler than winding the bob up and down by a minute amount.

If I remember correctly, a larger weight on the pendulum will also
slow it down.

Nope. The period depends on the length of the pendulum and the acceleration
of gravity.

Physics courses often include units analysis. * Such an analysis shows --
without even performing an experiment -- that length and the strength of the
gravitational field are the factors that matter.

* That probably isn't the right term, but I can't think of what it is.

On Sun, 14 Oct 2012 02:32:28 -0700, "William Sommerwerck"
put finger to keyboard and composed:

I've only seen one electronic pendulum. it used a sensing coil
to detect the approach of the pendulum, and fired a pulse to
add to the stored energy.

I'm not sure that would work correctly. You want the drive coil to fire at a
fixed frequency. Otherwise, the pendulum would be synched at its lower
native frequency.

Perhaps there are two different design philosophies.

In one case the pendulum is tuned to a frequency of 1Hz and the
electronic circuit functions to compensate for losses due to friction
and drag.

In the second case the electronic circuit provides an accurate crystal
controlled time base and it keeps the pendulum synced to this time
base.

In other words, perhaps in the first instance the pendulum is driving
the clock movement, while in the second case the clock movement is
driving the pendulum.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

"Franc Zabkar" wrote in message
...
On Sun, 14 Oct 2012 02:32:28 -0700, "William Sommerwerck"
put finger to keyboard and composed:

I've only seen one electronic pendulum. it used a sensing coil
to detect the approach of the pendulum, and fired a pulse to
add to the stored energy.

I'm not sure that would work correctly. You want the drive coil to
fire at a fixed frequency. Otherwise, the pendulum would be
synched at its lower native frequency.

Perhaps there are two different design philosophies.

In one case the pendulum is tuned to a frequency of 1Hz and the
electronic circuit functions to compensate for losses due to friction
and drag.

In the second case the electronic circuit provides an accurate crystal
controlled time base and it keeps the pendulum synced to this time
base.

In other words, perhaps in the first instance the pendulum is driving
the clock movement, while in the second case the clock movement is
driving the pendulum.

What you say makes sense -- but the drive circuit will always compensate for
losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow, and
have the drive circuit force it to the correct frequency. This would also
make trimming the frequency a simple matter.

On Sun, 14 Oct 2012 13:49:46 -0700, "William Sommerwerck"
wrote:

"Franc Zabkar" wrote in message
.. .
On Sun, 14 Oct 2012 02:32:28 -0700, "William Sommerwerck"
put finger to keyboard and composed:

I've only seen one electronic pendulum. it used a sensing coil
to detect the approach of the pendulum, and fired a pulse to
add to the stored energy.

I'm not sure that would work correctly. You want the drive coil to
fire at a fixed frequency. Otherwise, the pendulum would be
synched at its lower native frequency.

Perhaps there are two different design philosophies.

In one case the pendulum is tuned to a frequency of 1Hz and the
electronic circuit functions to compensate for losses due to friction
and drag.

In the second case the electronic circuit provides an accurate crystal
controlled time base and it keeps the pendulum synced to this time
base.

In other words, perhaps in the first instance the pendulum is driving
the clock movement, while in the second case the clock movement is
driving the pendulum.

What you say makes sense -- but the drive circuit will always compensate for
losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow, and
have the drive circuit force it to the correct frequency. This would also
make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.
Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at it's native frequency. Most of the designs did nothing to
optimize drive current.

PlainBill

wrote in message
...
On Sun, 14 Oct 2012 13:49:46 -0700, "William Sommerwerck"
wrote:

"Franc Zabkar" wrote in message
.. .
On Sun, 14 Oct 2012 02:32:28 -0700, "William Sommerwerck"
put finger to keyboard and composed:

I've only seen one electronic pendulum. it used a sensing coil
to detect the approach of the pendulum, and fired a pulse to
add to the stored energy.

I'm not sure that would work correctly. You want the drive coil to
fire at a fixed frequency. Otherwise, the pendulum would be
synched at its lower native frequency.

Perhaps there are two different design philosophies.

In one case the pendulum is tuned to a frequency of 1Hz and the
electronic circuit functions to compensate for losses due to friction
and drag.

In the second case the electronic circuit provides an accurate crystal
controlled time base and it keeps the pendulum synced to this time
base.

In other words, perhaps in the first instance the pendulum is driving
the clock movement, while in the second case the clock movement is
driving the pendulum.

What you say makes sense -- but the drive circuit will always compensate
for
losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow, and
have the drive circuit force it to the correct frequency. This would also
make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.
Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at it's native frequency. Most of the designs did nothing to
optimize drive current.

PlainBill



Perhaps you could gate the output drive by a divide by 10 counter on the
clock ouput and only power kick every tenth swing of the pendulum. Would
conserve battery and perhaps less amplitude of swing

What you say makes sense -- but the drive circuit will always compensate
for losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow,
and have the drive circuit force it to the correct frequency. This would
also make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.

ALL pendulums oscillate at their native frequency. They can't help but.

Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at its native frequency.

You're missing the point. The pendulum presumably drives the clock gears. If
all you care about is "efficiency", switch to an all-electronic clock with
an LCD.

If you're going to power the pendulum electronically, it makes sense to have
a system that keeps the pendulum running at the "right" frequency. The
system I described allows the pendulum frequency to be tweaked without
mechanical adjustments.

In article ,
wrote:

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.
Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at it's native frequency. Most of the designs did nothing to
optimize drive current.

One trick I have seen used, is to use a magnetic pendulum, and place a
drive-and-sense coil immediately beneath the center of its swing. The
drive circuit senses the beginning of the inductive pulse generated in
the coil as the pendulum swings down towards it, and then sends a
drive-current pulse through the coil to magnetize it and attract the
pendulum magnet just before it "reaches bottom" in its swing.

There are all sorts of tricks you can play with this approach. You
can use a Hall-effect sensor in addition to the coil (separating the
sense and drive functions). If you use a coil, you can detect the
height of the pulse during the pendulum swing in one direction, and
use this as a way of estimating the pendulum's speed... if it's high
enough, you don't need to "kick" the pendulum as hard during the next
swing (don't drive it at all, or reduce the strength or duration of
the drive pulse).

I've seen little "desktop toy" pendulum systems, or "rotating wheel"
toys, which use this approach. They can be rather mysterious, if the
drive/sense electronics are concealed in the base... the pendulum just
keeps swinging, or the wheel keeps rotating, with no visible drive
force and no tick-tock sound.

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!

In article ,
William Sommerwerck wrote:

You're missing the point. The pendulum presumably drives the clock gears.

I believe that the Klockit pendulum system being described here, does
not use the pendulum to drive the gears. It's described on their web
site as a "pendulum case assembly", into which you insert your
(separate) quartz clock movement. It has a separate battery.

If this is what the OP was using (as I recall), then the pendulum
system is a purely cosmetic add-on to the clock. It plays no part in
the drive, or time regulation of the clock itself... those are
entirely the role of the quartz movement assembly.

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!

On Mon, 15 Oct 2012 09:21:25 -0700, "William Sommerwerck"
wrote:

What you say makes sense -- but the drive circuit will always compensate
for losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow,
and have the drive circuit force it to the correct frequency. This would
also make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.

ALL pendulums oscillate at their native frequency. They can't help but.

Only true if you add the qualifier 'if there are no additional
forces'. A great deal of effort is required to ensure this is true.
Did you ever wonder why the best Grandfather clocks use weights to
provide power?

In a happier time my wife decided we needed a real pendulum clock on
the mantle of our new (to us) home. She picked up one (made in Korea)
that required monthly winding. With a little care I could adjust it
so it was correct at the begining of the month and again at the end of
the month. The force provided by the spring changed depending on how
tight it was, this changed the force on the escapement, which changed
to force applied to the pendulum. The clock would gain time at the
beginning of the month and lose it toward the end of the month.

Look attractive? Yes. Sound good? Definitely. Keep good time - no
way.

PlainBill

Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at its native frequency.

You're missing the point. The pendulum presumably drives the clock gears. If
all you care about is "efficiency", switch to an all-electronic clock with
an LCD.

If you're going to power the pendulum electronically, it makes sense to have
a system that keeps the pendulum running at the "right" frequency. The
system I described allows the pendulum frequency to be tweaked without
mechanical adjustments.

What you say makes sense -- but the drive circuit will always
compensate
for losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow,
and have the drive circuit force it to the correct frequency. This
would
also make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.

ALL pendulums oscillate at their native frequency. They can't help but.

Only true if you add the qualifier 'if there are no additional
forces'. A great deal of effort is required to ensure this is true.
Did you ever wonder why the best Grandfather clocks use
weights to provide power?

No, I never wondered, because I knew why. And it's energy, not power, by the
way.

wrote in message
...
On Mon, 15 Oct 2012 09:21:25 -0700, "William Sommerwerck"
wrote:

What you say makes sense -- but the drive circuit will always
compensate
for losses, regardless of design philosophy.

Therefore, it makes sense to have the pendulum swing a tiny bit slow,
and have the drive circuit force it to the correct frequency. This
would
also make trimming the frequency a simple matter.

To me the way that makes the most sense is to completely uncouple the
two functions. The accuracy of the cheapest quartz movement is far
better than you can get with the most precise pendulum. The most
efficient pendulum is one that oscillates at it's native frequency.

ALL pendulums oscillate at their native frequency. They can't help but.

Only true if you add the qualifier 'if there are no additional
forces'. A great deal of effort is required to ensure this is true.
Did you ever wonder why the best Grandfather clocks use weights to
provide power?

In a happier time my wife decided we needed a real pendulum clock on
the mantle of our new (to us) home. She picked up one (made in Korea)
that required monthly winding. With a little care I could adjust it
so it was correct at the begining of the month and again at the end of
the month. The force provided by the spring changed depending on how
tight it was, this changed the force on the escapement, which changed
to force applied to the pendulum. The clock would gain time at the
beginning of the month and lose it toward the end of the month.

Look attractive? Yes. Sound good? Definitely. Keep good time - no
way.

PlainBill

Googeling electronic pendulum drive circuit yields a great deal of
information, including some designs that simply provide a boost to a
pendulum at its native frequency.

You're missing the point. The pendulum presumably drives the clock gears.
If
all you care about is "efficiency", switch to an all-electronic clock
with
an LCD.

If you're going to power the pendulum electronically, it makes sense to
have
a system that keeps the pendulum running at the "right" frequency. The
system I described allows the pendulum frequency to be tweaked without
mechanical adjustments.



Thats why fusees were invented, but could only compensate to a certain
extent

In article ,
William Sommerwerck wrote:
units analysis ... probably isn't the right term, but I can't think of what it is.

Dimensional analysis?

--
--------------------------------------+------------------------------------
Mike Brown: mjb[-at-]signal11.org.uk | http://www.signal11.org.uk

--- news://freenews.netfront.net/ - complaints: ---

units analysis ... probably isn't the right term, but I can't think of
what it is.

Dimensional analysis?

That's it! Thank you.

Efter mange tanker skrev N_Cook:
William Sommerwerck wrote in message
...
Excuse my ignorance, but how does your circuit
know when to apply a kick?

It doesn't.

Assume the pendulum is supposed to have a period of one second. You design
it to be a little bit longer, then make the driver circuit operate at
exactly one second. The pendulum will eventually sync with the driver.

Remember when TVs had hold controls? The principle is the same. The
most-stable operation is obtained when the oscillator runs a tiny bit slower
than it should, with the sync signals "kicking" it at the right frequency.



Human clocks are the same. Put a human in a cave, out of touch with the
outside world, and his natural day-length reverts to about 24.5 to 25 hour
days, requires the sun etc to sync him to 24 hour days

Not all people "run slow". Some "run fast". These are the people with
the nasty habit of waking up at 5AM pestering the other half until the
early birds go to sleep when the late risers want to have fun.

If you are locked in the office/shop/factory all the hours with
daylight, and never seeing the light of day during the winter, you can
get depressed; this can be reduced by, in the morning, looking into a
strong "wake-up-lamp" designed to have the spectrum of sunlight.



--
Husk kørelys bagpå, hvis din bilfabrikant har taget den idiotiske
beslutning at undlade det.