Are you ready for a challenge?

Hello I am new to the forum and after looking for a short time did not find what fits my needs.

I would like to build a circuit that can detect a pendulum passing by, measure the period of a pendulum, and calculate the pendulum’s length. When the length calculation is completed the circuit will display the length accurate to within one tenth of a centimeter. This will require a three digit display.

The periods of these pendulums will vary in length from 0.7 seconds to 1.5 seconds. The shaft of the pendulums as it passes by the detector portion of the circuit will vary in width from 0.5 inches to 1.5 inches. The depth of the pendulums will be about 12 inches. To better understand this, imagine a 1”x12” board swinging. I was thinking along the lines of using a photocell to detect the shaft passing the detector but I am open to other methods as long as the method does not touch the pendulum or require anything to be attached to the pendulum.

This description of a pendulum may help you if you are not familiar with the terminology. When a pendulum is displaced from its resting equilibrium position, it is subject to a restoring force due to gravity that will accelerate it back toward the equilibrium position. When released, the restoring force combined with the pendulum's mass causes it to oscillate about the equilibrium position, swinging back and forth. The time for one complete cycle, a left swing and a right swing, is called the period. A pendulum swings with a specific period which depends on its length. Therefore one period will be the time it takes for the pendulum to pass the detector two times.

Requirements:
1. All circuitry should be powered by an commonly available power soured
a. Power source can be AC, DC, or AC/DC Converter
b. The availability of either AC or DC would be beneficial
2. An On/Off button to apply power if the unit is off
3. The On/Off button will remove power if the unit is on
4. An Visual indication that power is on
5. No power off timer circuitry is required
6. A Reset up button that resets all counters and timers
7. The Reset button should stop any circuitry
8. The Reset button should not remove power
9. A Start button
10. Timing will not initiate until the pendulum passes the detector after the Start button is pressed
11. Timing will stop after a predetermined number of periods has been detected
12. A switch or button should be available to select the average of 1, 5, or 10 periods for measuring the length of a single period.
13. An audible indication will be sounded each time the pendulum passes the detector
14. A visual indication will be displayed each time the pendulum passes the detector

If all I need is a detector and a microprocessor that I program that is fine. I would also like to see the bells and whistles on the circuit. I would like to put all this circuitry in a plastic case so the smaller the better.

Help me if you can.

M10SINV

Also the weight distribution of the pendulum will be over its entire length. The pendulum will be suspended at some point other than one end. The period will be dependant on the Center of Mass of the Pendulum relative to the axis point and not the entire length of the pendulum.

M10SINV

M10SINV said:

Also the weight distribution of the pendulum will be over its entire length. The pendulum will be suspended at some point other than one end. The period will be dependant on the Center of Mass of the Pendulum relative to the axis point and not the entire length of the pendulum.

M10SINV

Click to expand...

hi,
Interesting, whats your location.?

I am just outside Atlanta, Ga in Marietta. If you are familiar with the area I am not too far from Kennesaw Mtn.

M10SINV

M10SINV said:

I am just outside Atlanta, Ga in Marietta. If you are familiar with the area I am not too far from Kennesaw Mtn.

M10SINV

Click to expand...

hi,
I dont know your area, the reason for asking the location is that often we advise posters on designs but they are in country's that cannot get the components.

I guess you shouldn't have a problem in the States.

Is the application for a clock project.??

The application is for an idea of mine. I may want to patent it some day so I would prefer not to go into too much specific detail. I hope you can understand that. I am not trying to be rude.

I have an object that I am swinging from a fixed axis. I can time its swing for 10 periods with a stop watch and divide that by 10 to get the time for one period, but this is not as accurate as I would like for it to be. Every 10 periods I will be off at least a tenth of a second or so from one sequence to the next. I would prefer an electronic circuit that could improve on my timing by eliminating the reaction time of the user.

I am going to have to leave home for a little while. I am going to my Grandson's school for lunch. It is Grandparent's Day at school. This will be my second year. I can't wait to sit in those little chairs at those short tables. LOL I will be back this afternoon.

Hi, this is like an order and throwing a challenge. Not fair, I personally feel.

I am not sure why you think this request is unfair. The description for this specific forum is, "Are you building an electronic circuit, or want to? Submit your electronic questions and get assistance from our experienced members help." The best way I felt to submit my questions was to put it in the form what the requirements for the circuit would be. I am not challenging anyone to figure this out for me just to get me headed in the right direction. I am more than willing to do what it takes to gain the knowledge to do all the real work myself.

If you don't want to assist me in finding the answers to my problem that is perfectly okay with me.

M10SINV said:

The application is for an idea of mine. I may want to patent it some day so I would prefer not to go into too much specific detail. I hope you can understand that. I am not trying to be rude.

I have an object that I am swinging from a fixed axis. I can time its swing for 10 periods with a stop watch and divide that by 10 to get the time for one period, but this is not as accurate as I would like for it to be. Every 10 periods I will be off at least a tenth of a second or so from one sequence to the next. I would prefer an electronic circuit that could improve on my timing by eliminating the reaction time of the user.

I am going to have to leave home for a little while. I am going to my Grandson's school for lunch. It is Grandparent's Day at school. This will be my second year. I can't wait to sit in those little chairs at those short tables. LOL I will be back this afternoon.

Click to expand...

hi,
Looking over the maths I have a couple of points of concern regarding the application of the pendulum formula's.
One is the air damping factor of the leaf type pendulum and the other is the variable width of the pendulum, especially as you want the period relative to the centre of rotation.

Have you done any preparative work on how you would see the LED detector/s mounted.?
I would guess at least a couple of LED IR paths would be needed to try to cover the pendulums width effect upon the timing.

ericgibbs said:

...is the air damping factor of the leaf type pendulum and the other is the variable width of the pendulum, especially as you want the period relative to the centre of rotation...

Click to expand...

Good question, I do not believe the air dampening will make any appreciable difference as the object being swung is relatively aerodynamic and heavy enough to keep the object moving. If allowed to swing freely the air friction and the friction of the pivot device will probably stop the object and return it to it equilibrium position with a few minutes. But long before that ever happens I will have taken all the measurements I need.

ericgibbs said:

...Have you done any preparative work on how you would see the LED detector/s mounted.?
I would guess at least a couple of LED IR paths would be needed to try to cover the pendulums width effect upon the timing

Click to expand...

No I have not, but I do not think mounting would be an issue. If all the circuitry could fit into a compact box the box could be supported many ways. It would be okay if everything were placed in the box and the box was hand held. I think I am only going to need one detector path, as all I need to accomplish my task is to capture the time the leading edge of the pendulum passes a specific point. That point would best be located directly below the pivot point but could be placed anywhere in the path of the pendulum as long as it is close enough to captre the pendulum passing.

By the way, I am not sure if it makes any difference or not but the formula I will be using to determine the pendulum length is L = gTT/4PiPi (where L = length, g = gravitational force (980 cm/secsec,) T= period, and Pi = Pi)

I have assumed that this project will require the use of a microprocessor and I been assuming I would have to program that myself.

M10SINV said:

By the way I am not sure if it makes any difference or not but the formula I will be using to determine the pendulum length is L = gTT/4PiPi (where L = length, g = gravitational force 980 cm/secsec, T= period, and Pi =Pi)

Click to expand...

hi,
Thats the formula I was using.
Looking at the maths for the MCU would it be problem for you to work in inches.?

Also would a small square of self adhesive reflective tape stuck along the bottom edge of the pendulum, in line with the pivot point be a problem.?
If its OK I would consider a reflective IR LED path, this would cancel out the width/thickness of the leaf.

I would greatly prefer to work in metric units of cm and grams. If all the circuit did at the end is display the time of a period in hundreths of a second (like 1.25 seconds) that would be fine and I can compute the length. If the circuit could give me thousandths of a second that would be better (like 1.248 seconds.) All I really need is the stopwatch function to eliminate human reaction time so when different users use the application the final result will be the same.

I am going to be working with a relatively light object with close tolerances and placing anything on the pendulum will throw off the length calculation. The absolute worse place it could be placed in on the bottom of the pendulum as that is were it would have the most level and create the greatest error. If the piece of tape were to weigh less and a half a gram it could throw off the period calculation by as much as a tenth of a second and that throw the length off.

I was thinking of using a photocell and focusing a light source directly on the photocel. When the pendulum passes by the photocell I have my trigger.

In this circuit's most simplistic form all I need is a stopwatch with an electronic eye for a trigger. Once the user is ready to make the measurement he presses a start buttom and the apparatus takes over from there.

M10SINV said:

I would greatly prefer to work in metric units of cm and grams. If all the circuit did at the end is display the time of a period in hundreths of a second (like 1.25 seconds) that would be fine and I can compute the length. If the circuit could give me thousandths of a second that would be better (like 1.248 seconds.) All I really need is the stopwatch function to eliminate human reaction time so when different users use the application the final result will be the same.

I am going to be working with a relatively light object with close tolerances and placing anything on the pendulum will throw off the length calculation. The absolute worse place it could be placed in on the bottom of the pendulum as that is were it would have the most level and create the greatest error. If the piece of tape were to weigh less and a half a gram it could throw off the period calculation by as much as a tenth of a second and that throw the length off.

I was thinking of using a photocell and focusing a light source directly on the photocel. When the pendulum passes by the photocell I have my trigger.

In this circuit's most simplistic form all I need is a stopwatch with an electronic eye for a trigger. Once the user is ready to make the measurement he presses a start buttom and the apparatus takes over from there.

Click to expand...

hi,
If I follow correctly, in the simple form, just display the pendulum period in say 'S.SS' and let the user calculate the length of the pendulum, is that correct.
If so a MCU is not essential, just a discrete logic circuit with a 3 decade display would do the job.

Depending upon the material of the pendulum a LED IR emitter/detector could bounce off the pendulums edge.

Here is a perfect example of how this circuit could be used. Let's assume you have a slot car track but instead of the normal track you have a figure 8 track and the start / finish line is at the intersection in the center of the track. With this circuit I could place some light sensing device under the track. The driver is allowed to place their car on the track and warm up. When they are ready, the timer presses a start buttom to see how long it takes the car to go around the figure 8, but the circuit does not start timing until the slot car passes over the sensor. Because the sensor is in the intersection the car will need to pass over the sensor two times. When the car passes over the sensor the second time the timer is stopped. At this point the time is displayed or the running time could have been displayed all along and stops when the car passes over the sensor the second time.

ericgibbs said:

hi,
If I follow correctly, in the simple form, just display the pendulum period in say 'S.SS' and let the user calculate the length of the pendulum, is that correct.
If so a MCU is not essential, just a discrete logic circuit with a 3 decade display would do the job.

Depending upon the material of the pendulum a LED IR emitter/detector could bounce off the pendulums edge.

Click to expand...

Some of the pendulums will be reflective some won't. It would be ok to place the reflective tape in a location near the center of mass but not on the bottom. It would be better not to have any attachments to the pendulum even it means the cost of the timer goes up. I could use a stopwatch to determine the approximate period and calculate the length to the center of mass and place my tape. Other than that you are spot on. Look at my example of a slot car on a figure 8 track maybe that will help you some.

Hi,

If the pendulum including the arm is symmetrical about its center and we know the mass we can use a different method to calculate the width, especially if the end of the pendulum has circular cross section. In this case we would only have to find the place of the exact end of the swinging action, then calculate the rest knowing the distance to the center of the pivot point too. The reason this could work is because there is a differential equation for a pendulum which describes its motion, and if we can describe the motion of the center of the pendulum end and we also know the maximum swing position then we can calculate the width by subtracting.
The various setup distances would be easy to measure, but as i said though, the mass would have to be known so you would also have to weigh the pendulum, and if it wasnt symmetrical about its center then the calculation would be more difficult.

That said and assuming you dont want to do it that way then here are a couple other ideas...

1. I would think two thin beams of light being broken by the pendulum end part could be used to measure both speed and width by noting the timing between breaks. If you want to use one beam then you have to do some more involved calculations.
2. A simpler way would be to have the sources and detectors of the two thin beams of light on arms that can move side to side. The circuit would logically AND both detectors so that both detectors light would have to be broken at the same time to get an output pulse. One or both of the arms are then adjusted until a pulse is detected, and then moved very slowly apart until the pulse just goes away. At that point the distance between the sensors is the same width as the pendulum end.
The entire circuit might be made from a single AND gate, possibly with a latch to latch the pulse, unless you have a logic probe that detects pulses and you can use that to detect the presents (or absence) of a pulse. Once the one or two arms are adjusted as above, the distance can be measured with a ruler.
Total parts would be one ic chip, such as a NAND gate using one gate to detect and two more cross coupled to latch. Of course two beams of light would be needed as well as two sensors.

since the calculation part is done by you, the rest would be simple as i feel. what you may need is a stop watch that can be reset/start and stopped by a logic input. an existing watch also can be modified to serve the purpose. if you expect an accuracy of 3 decimals, it may require to design the stop watch. otherwise you can count the time for more than 10 oscillations and take the average. the trigger can be designed for any number of counts.

for IR, if weight of a reflective tape would make errors then what about keeping the surface shiny? (by polishing)

since you are working with light weight frictional forces may influence in your reading especially air friction. you may use a formula for distributed load too.

MrAl said:

Hi,

If the pendulum including the arm is symmetrical about its center and we know the mass we can use a different method to calculate the width, especially if the end of the pendulum has circular cross section. In this case we would only have to find the place of the exact end of the swinging action, then calculate the rest knowing the distance to the center of the pivot point too. The reason this could work is because there is a differential equation for a pendulum which describes its motion, and if we can describe the motion of the center of the pendulum end and we also know the maximum swing position then we can calculate the width by subtracting.
The various setup distances would be easy to measure, but as i said though, the mass would have to be known so you would also have to weigh the pendulum, and if it wasnt symmetrical about its center then the calculation would be more difficult.

That said and assuming you dont want to do it that way then here are a couple other ideas...

1. I would think two thin beams of light being broken by the pendulum end part could be used to measure both speed and width by noting the timing between breaks. If you want to use one beam then you have to do some more involved calculations.
2. A simpler way would be to have the sources and detectors of the two thin beams of light on arms that can move side to side. The circuit would logically AND both detectors so that both detectors light would have to be broken at the same time to get an output pulse. One or both of the arms are then adjusted until a pulse is detected, and then moved very slowly apart until the pulse just goes away. At that point the distance between the sensors is the same width as the pendulum end.
The entire circuit might be made from a single AND gate, possibly with a latch to latch the pulse, unless you have a logic probe that detects pulses and you can use that to detect the presents (or absence) of a pulse. Once the one or two arms are adjusted as above, the distance can be measured with a ruler.
Total parts would be one ic chip, such as a NAND gate using one gate to detect and two more cross coupled to latch. Of course two beams of light would be needed as well as two sensors.

Click to expand...

I forgot to mention that #1 above can be done without human intervention, but #2 requires the human to adjust the position of at least one of the arms. #1 requires a micro controller to do all the work yes, while #2 does not. Both require two sensors and one or two light sources.

MrAl said:

Hi,


1. I would think two thin beams of light being broken by the pendulum end part could be used to measure both speed and width by noting the timing between breaks. If you want to use one beam then you have to do some more involved calculations.

Click to expand...

when we consider a full oscillation then the interrupting edge would be the same. i don't think any additional calculations are required in this case.