Looking for sensor suggestions

[imlost]

Hi,

I have to measure a time interval in which a steel rod drops ( i think about a 1 foot, I'll know the exact measurements tomorrow and its a linear drop ). I have been looking into several "types" of sensors. However i been unable to find more detail/quality information about each type.

The types I've looked into so far are:
Laser Triangular Sensors
Laser Interferometer
Photo Reflective Sensor

I'm unable to put the sensor above or below (double check the below part again but i was told no already). So i have to mount the sensor on the side of the falling rod or at an angle below the falling rod.

Any suggestion or comments would be appreciated! (also not sure if i posted this in the right spot so bare with me!)

 

[Reloadron]

How is the rod released? Any signal available to you from the release mechanism (electrical signals)? When the rod drops does it produce any sort of a "bang" at bottom? I have done some time interval measurements similar in nature, in my case I was able to use an accelerometer to measure the rod bottom (stop) by catching the "bang" produced by the rod at rod bottom. Also used a laser interferometer to measure the bounce the rod does at rod bottom.

Ron

 

[Gary B]

So why not do it the easy way? Just drop it through two coils of wire a known distance apart. When the steel rod goes through it the inductance will rise. Since you wind both coils the same the change will be the same in both units as the rod enters. Just pick a threshold point and you should be able to measure the time between the two coils reaching that point. Since you have time and distance, you can calculate the speed over the measurement interval. Since the speed of an object falling through a vacuum is known (32 feet per second per second) I assume that you are trying to measure the drag on the rod falling through some fluid. This method will allow you to put the rod in a glass or plastic pipe and still measure its speed even if you can’t see it.

 

[imlost]

I'm going to try to get a diagram of what exactly i have to set it up on.

The rod that drops is lifted from rest by a magnetic up to a specificed height. It has a maximum height but the height it is dropped from may vary.

For instance the maximum height is 26 cm, however it could start at 15/21/23 cm etc.. height.

will i be able to configure the induction method if the height varies? Didn't even think about approaching it that way I'll have to read more into it!

The rod that is dropping is also being accelerated by spring that are attached to it. So as the magnetic drags it up to x height (maximum being 26 cm) it also compresses the spring creating additional acceleration for its fall.

The rod being dropped is actual being caught by a dash-pot. It would be nice to measure the bounce the rod does at the bottom so if a suggestion for a inexpensive interferometer would be nice.

As for the dropping mechanism, i can look at that for more info but i'm unsure if its mechanical or electrical. All i know is that they push a button and the magnetic releases the rod. So i could assume that is is electrical switch.

As far as the bang goes, I think it produce a little vibration but i'm not sure how significant it is because ideally they want to use data from my sensor to be able to configure the dash pots to approach a critically damping state. ( i think that's why they want the sensors haha, they just told me they want a start and stop time, and if i can a couple more points so they can graph it on excel or something)

Thanks for the reply so far!

 

[Gary B]

Mount your two coils on a single form, and then just move the whole sensor up and down to get a reading at any point along the path. Also, if the rod retains any residual magnetism through the drop, a pickup coil at the bottom should register a voltage related to the bounce. The resulting waveform should tell you quite a lot about what is happening at that point. Indeed, you could even induce a magnetic field with the sensor coils once you have your speed reading. I see all sorts of interactive possibilities.

 

[Reloadron]

I'm going to try to get a diagram of what exactly i have to set it up on.

The rod that drops is lifted from rest by a magnetic up to a specificed height. It has a maximum height but the height it is dropped from may vary.

For instance the maximum height is 26 cm, however it could start at 15/21/23 cm etc.. height.

will i be able to configure the induction method if the height varies? Didn't even think about approaching it that way I'll have to read more into it!

The rod that is dropping is also being accelerated by spring that are attached to it. So as the magnetic drags it up to x height (maximum being 26 cm) it also compresses the spring creating additional acceleration for its fall.

The rod being dropped is actual being caught by a dash-pot. It would be nice to measure the bounce the rod does at the bottom so if a suggestion for a inexpensive interferometer would be nice.

As for the dropping mechanism, i can look at that for more info but i'm unsure if its mechanical or electrical. All i know is that they push a button and the magnetic releases the rod. So i could assume that is is electrical switch.

As far as the bang goes, I think it produce a little vibration but i'm not sure how significant it is because ideally they want to use data from my sensor to be able to configure the dash pots to approach a critically damping state. ( i think that's why they want the sensors haha, they just told me they want a start and stop time, and if i can a couple more points so they can graph it on excel or something)

Thanks for the reply so far!

This is vaguely familiar to a test that I work with including the spring, only the rods are much longer and drop a much greater distance. Including the spring.

There are several ways to go about it as has been mentioned. Let us know if the magnet lifting the device (rod) is an AC or DC magnet. If you could find a way to start your timer function when the magnet field collapses or when power is removed from the magnet (allowing for field collapse) that would be the Start command to your counter. For a Stop command I use an accelerometer that picks up the bang at rod bottom. I amplify the accelerometer out and use the signal to trigger a comparator at a given set level. I measure the time interval. With the distance you mention and the spring I would venture a guess you are looking at very quick intervals of several milli-seconds.

Bounce testing is done with a Laser Interferometer but those units we use are pretty expensive. I think Micro-Epsilon makes the ones we use. The rods are fitted with a fixture to allow good LASER bounce. More to it but that is the general idea.

Additionally either test could be recorded using a paperless chart recording method. That in addition to a display of time interval for the drop time.

If this is a one time test then you want to go inexpensive but accurate. If this will be ongoing test of parts and product then you can place more effort into your test fixtures. The fixture(s) will play a big roll in how this is setup and done.

Also as Gary B points out two coils with their centers spaced equidistant could likely work well for the Start & Stop signals. Depends on the alloy of the rod but we use similar for another test.

Ron

 

[imlost]

I'm pretty sure the magnet is a DC magnet.

Just to clarify really quick the rod rest upon a circular steel platform which is what the magnetic will pick up then move up to x height and drop. So when its dropped the rod rest upon the steel platform but the platform will be dropping. Hope that make sense and doesn't change methods being suggested! Trying to get access to solid works so i can give a better picture.

This device will be used for ongoing testing.

My question with the two coils on a single form is how will the sensor know how high the magnetic will rise? For instance say when they start the test and it drops the first time then rises to a different height the second time, how will the whole sensor know where to reposition it self properly? Would i just put like a magnetic on the sensor device it self to align with the steel platform which the rod resting upon?

I won't be able to manual reposition the device because in the end it will be going inside a container with no easy access and must be self sustaining.

You are correct Ron to assume it is a several milli-second interval that i will be trying to time. For the start time, I'll ask if i can pull a signal into my micro controller when the magnetic is released or just setup an inductor near the magnetic in order to detect when the magnetic field disappears, assume the field being generated is significant enough to be detected.

I'll read into the accelerometer, the concept of it makes it a very viable option. Do you know if it will pick up vibration if the the rod is dropped like 5-8 cm from the ground?

As for the Laser Interferometer, the problem i have with it is positioning, if i can't put it below or above the dropping rod/platform then how would i be able to detect the bouncing?

I plan to time and store data inside a micro-controller. Then pull data out via a computer and display the total time on some digital display unit. This part I believe is pretty straight forward. However, is it feasible for a time interval of a few Milli-seconds?

Sorry if I'm asking silly questions that should be obvious, I have very weak background in sensor and magnetic!

Thanks for the reply!

 

[crutschow]

One way to measure the passage of the rod through a coil is to apply a DC current through the coil with a resistor to limit the current. When the rod passes through the coil it will disturb the magnetic field which should produce a measurable voltage blip across the coil and resistor. Just detect the blip by AC coupling the coil voltage to an op amp amplifier and then to a comparator. With a little experimentation you should be able to select a comparator trip point that will reliably detect the passage of the rod. The more turns of coil wire, the more sensitive the device will be.

 

[Reloadron]

I can tell you how I would start as to measuring rod bottom time, and rod bottom time being the time it takes the rod to drop from when the field of the magnet collapses to when the rod hits the plate it crashes into. You have a time interval measurement needing a start & stop command. First the start, I would think about using a hall effect sensor (assume DC Coil magnet). Place the sensor in close proximity to the magnet. Allegro makes a line of hall effect sensors that act as switches and can be found here. Nice thing about this is that it will switch when the field collapses which should be when the rod is released. Another option would be to grab a signal from the switching when the magnet power is removed. Downside is it will not take into consideration the current decay in the coil unless you get fancy. A hall effect may be the best bet or something for you to consider. I would likely start with considering one of those methods as a start.

For the stop signal I would again consider an accelerometer. Since the term accelerometer has become real popular I'll attach an image of the type I am talking about. The ones in the image actually have a screw hole tapped into the bottom and one in the images has a magnetic base attached. The connectors on them are also a screw connector, here in the US a 10-32 screw but there is a metric version I am sure. I happen to use Endevco (Meggit) which can be seen here but others make the same thing. Truth be known and for what you want maybe even a cheap guitar pickup would work attached to the impact plate. When the rod drops and hits the plate there will be a bang and that bang is all we care about.

I amplify the output of the accelerometer (more I think about it a cheap electric guitar pickup may work as well) and eventually run it to a comparator circuit. When the bang happens it triggers the comparator and I use that to drive a 7474 ancient flip flop. The circuit is really simple (well pretty much).

The bottom line is I have two
Type D Edge Triggered Flip Flops using the preset and clear. The outputs of those flip flops give me the negative slope triggers I use for start and stop of my counter.

This is how I do it for my applications and it works for me. There are likely a dozen ways to do this so be open. Both Gary B and Crutschow may have ways to implement their thoughts. You mention acceleration as a result of a spring so the velocity of the object (rod) dropping won't be a constant. I like the coil ideas.

<EDIT> When I said electric guitar pickup I did not mean a $1,000 USD pickup, I had in mind a $5.00 USD mic as used on a kids guitar. I just looked at some pickups and thought I would clarify after the sticker shock wore off. </EDIT>

Ron

 

[Gary B]

I don’t think I understand the sensor positioning issue. Do you want to measure the speed at a particular point below the release point as opposed to an absolute position above the final resting point? I was just thinking about something like a plastic tube the rod would drop through and you manually clamp the sensor to the outside of the tube at the position you want the measurement to take place. You could use a slightly larger tube with slots cut in one end and a hose clamp to squeeze it down to hold. Wrap two coils around the moveable tube and build what amounts to two metal detectors.

 

[Reloadron]

Per Gary B's questions and just so we are on the same channel my understanding is you are not concerned with the velocity but the time interval between rod release and when it hist the bottom, the plate it falls on. Also if or when you have any information on the rod dimensions that would help so we aren't envisioning two different pictures. The more information that can be provided as you get it the better.

Ron

 

[colin55]

Does the rod interrupt a beam of light?

 

[imlost]

Hi,

Sorry for not responding for a while, I wasn't able to make schematics (i will do that once i get the proper access to the programs to make the diagrams). However i was able to just take picture which will help.

The Starting position of the drop will vary. The max drop height is 26 cm, however it can be dropped from 20 cm or 15 cm etc.. the drop height is random.

The first picture shows the rod at rest on the bottom being supported by the dash pot it will hit when dropped from the top. The black thing at the top of the rod is the magnetic that will be lowered to pick up the rod from rest. The magnetic will drag the rod up to x height and drop it.

The second picture is the view from above the rod. Sorry about the resolution, The base of the platform is the area i can use to setup the sensor.

I cant put anything in between the supporting beams. The beams allows the magnetic to go up and down so if i put a sensor there it will block the magnetic path.

Now back to the previous suggestion of a accelerometer that will sense vibration. I don't think i can use this because 4 of these mechanism will be placed adjacent (the metal bases will be touching) to each other so how would i be able to differentiate the vibration from one mechanism to the other one sitting next to it.

I'm going to look into using an inductor type sensor to detect when the rod hits the bottom.

As a side note I will refer to the dropping object as a rod in order not to confuse how our discussion has be proceeding. However as you can see in the picture the rod is connected to a flat disc which is then displaced by the magnetic.

The bottom line is i need to find the time it takes from start to stop at any height with out me readjusting the sensors each time. I've considered just splicing into the signal that will turn off the magnetic as my start time. Additional data that i would like to be able to get is the distance traveled within total time at various interval. For example say my start/stop timer measures 4 ms to drop say 20 cm, I'd like to be able to determine displacement in the first 2 ms and the last 2 ms. The reason why i can't use a simple analytical analysis is because of the spring that is pushing downwards with gravity therefore trying to use sensors. The additional part can wait first I'd have to be able to determine a reliable method to measure start/stop time.

Another thing I'll toss out there is this mechanism once fitted with the sensor will be put inside a giant container. They only pull this mechanism out like twice a year for maintenance thus the reason why i can't reposition the starting sensor by hand.

Hopefully this helps clarify things! As always thanks for the responses

 

[crutschow]

How about using a reflective light sensor to detect the rod? That way you could mount the LED light source adjacent to the sensor and don't have to interrupt anything.

 

[imlost]

that is what i was considering to do initially but wouldn't i need multiple sensor to make sure it can see exactly how high it the rod rises before releasing? or are you referring to when it drops have the sensor detect the moment it hits the dash pot because that seems like a good idea too.

 

[crutschow]

that is what i was considering to do initially but wouldn't i need multiple sensor to make sure it can see exactly how high it the rod rises before releasing? or are you referring to when it drops have the sensor detect the moment it hits the dash pot because that seems like a good idea too.

It depends upon whether you are measuring its speed or the time it takes to drop. It's not clear to me which measurement you want.

To measure the speed you would need two sensors. If you need the height, then you would need multiple sensors. If you need the time, then you could use one sensor at the bottom and detect the time from the release of the magnet holding the rod until the bottom sensor detects it.

 

[imlost]

It depends upon whether you are measuring its speed or the time it takes to drop. It's not clear to me which measurement you want.

To measure the speed you would need two sensors. If you need the height, then you would need multiple sensors. If you need the time, then you could use one sensor at the bottom and detect the time from the release of the magnet holding the rod until the bottom sensor detects it.

I need the time it takes dropping from a random height to the bottom.The height can vary after each drop. I will need the initial height it drops from (if they don't have a sensor that tells that) and the time interval during the drop. Eventually i would like the speed/velocity.

Now that i think about it what i think i might do is have a IR sensor on the bottom to tell me when the moment the rod hits the ground as you said.Then I'll just use the same signal that is used to turn off the magnetic (which initiates the drop) for the start timer. Does that sound feasible?

 

[crutschow]

Now that i think about it what i think i might do is have a IR sensor on the bottom to tell me when the moment the rod hits the ground as you suggested.Then I'll just use the same signal that is used to turn off the magnetic (which initiates the drop) for the start timer. Does that sound feasible?

That's basically what I said in the last sentence of my previous post.

If you need the initial height, then that requires an additional sensor(s).

 

[Reloadron]

I need the time it takes dropping from a random height to the bottom.The height can vary after each drop. I will need the initial height it drops from (if they don't have a sensor that tells that) and the time interval during the drop. Eventually i would like the speed/velocity.

Now that i think about it what i think i might do is have a IR sensor on the bottom to tell me when the moment the rod hits the ground as you suggested.Then I'll just use the same signal that is used to turn off the magnetic (which initiates the drop) for the start timer. Does that sound feasible?

That sounds feasible as long as your stop signal is right across the dash pot so you get the true bottom. By the way, the dash pot is acting like a shock buffer correct? If so does the dash pot move? When the rod strikes the dash pot does it collapse any? The fact there are several of these things in close proximity could screw with using an accelerometer to catch the bang. Bummer on that note.

Ron

 

[imlost]

That sounds feasible as long as your stop signal is right across the dash pot so you get the true bottom. By the way, the dash pot is acting like a shock buffer correct? If so does the dash pot move? When the rod strikes the dash pot does it collapse any? The fact there are several of these things in close proximity could screw with using an accelerometer to catch the bang. Bummer on that note.

Ron

Yes it does act like a dampener that's why I wasn't sure if the accelerometer would work out so well although it was a good idea . Eventually i have to figure out how far down the dash pot is pushed down but, that's a problem for later.

@crutschow
I'll have to determine the height as well using sensors it seems. I was thinking using the magnetic it-self as a point of reference then perhaps use a IR that measure distance by -10 to 10v. I'll just have to think on where to position it to make that work.

Thanks for the advice!