Roughly measuring the rate of millet seeds hitting a surface.

I hope you dont mind that I ask these kind of questions here on this forum. :-)

One of my current projects needs to measure the rate of small spherical items (millet seeds) hitting the surface and I was thinking of maybe a piezo wafer to generate small electrical voltages when each seed hits. I dont know if this is possible with a piezo wafer or not, but as far as I know, these things generate voltages when they are flexed in some small way.

Obviously the mass of millet seeds is pretty small and would not generate much voltage so I would need something very sensitive to changes in state.
Can anyone give me advice as to what kind of electronic component I would need to do this please?


The other thing I need to do is to generate a static electricity charge inside a hard-plastic bottle. Currently the only way I know is a Van de Graaf generator or physically rubbing something plastic/rubber on your jumper, but I am sure there is a way to generate this electrically somehow.

How would I do that electronically please?

We would have to assume a lot to give you proper help at this point. Can you give us more parameters? Range of the rate of impacts? What is the plate made of? What is it's dimensions? What environment is the system in? Etc...
Glad to help if I can.

By the way, if the environment is fairly ambient I would suggest that you will have a lot more flexibility with a low impedance coil suspended around a coin shaped rare earth magnet mounted to the plate. I have a design I used for a similar application a while back. I'll see if I can drum it up.

I presume you just need help with the sensor and have the digital processing side of things sorted out according to application already?

Hmm, sounds like a fun project... I think!

Can the base be modified? If so, I would think the easiest way to measure would be to replace the base with a funnel. Then you can narrow down the end nozzle to a few millimeters. From there, you have a few options. I would consider a laser to detect when the beam breaks and then counts. That could be built from a few beginners off the shelf electronic kits with a little knowledge.

Also, if the millet seed can be funneled into a small spot, you could make a very fine strip of say aluminum (like a diving board arrangement) stuck on a mercury type switch/counter.

As far a generating a static charge, I would look up circuit designs used to do that in laser printers. Again, depending on the application parameters, you may get away with yanking that part of the circuit out of an old laser printer.

Be warned: KNOW WHAT YOU ARE DOING WHEN DEALLING WITH HIGH VOLTAGE CIRCUITS!!!

originally posted by: Qumulys
Hmm, sounds like a fun project... I think!

Can the base be modified? If so, I would think the easiest way to measure would be to replace the base with a funnel. Then you can narrow down the end nozzle to a few millimeters. From there, you have a few options. I would consider a laser to detect when the beam breaks and then counts. That could be built from a few beginners off the shelf electronic kits with a little knowledge.

Also, if the millet seed can be funneled into a small spot, you could make a very fine strip of say aluminum (like a diving board arrangement) stuck on a mercury type switch/counter.

You don't really do electronics do you?

a reply to: CraftBuilder

Well pardon me for trying to help, I'll just leave you to your client then.

Talk about being rude, and with that attitude I'll leave what could have been an interesting thread on a downer.

originally posted by: Qumulys
a reply to: CraftBuilder

Well pardon me for trying to help, I'll just leave you to your client then.

Talk about being rude, and with that attitude I'll leave what could have been an interesting thread on a downer.

since you didn't answer the question I'll take that as a no. Not that I couldn't tell.
Ruder would have been to let you lead the OP astray.
Deny ignorance.

I'd go with a National LM10 or LM11 op-amp or any high gain low noise op-amp for that matter. Once you have a reasonable output in the 10's to 100's of millivolts range, you can amplify that with LF356's or any other form of Jfet or conventional op-amp. It depends on whether you need comparator output for counting (pulse driven) into a cheap processor, like a PIC or ARM, or if you need an analogue output that shows persistence in a bar graph display which could be driven by an LM3914. This is just off the top of my head...

Cheers - Dave

OP is this an educational exercise or are you actually going to be building one or more of these. If so, what level of fabrication do you have available? Are you designing around a one of or for large volume manufacture? Do you have access to exotic components or just hobby level components and equipment?

We could really use some more parameters.

a reply to: qmantoo
Your project description is short on details but I have a lot of experience with industrial counters (which can be used to count seeds), but they don't work the way you mention. The reason these industrial type counters are somewhat accurate is the seeds are lined up so they can be easily counted. If you're just dropping a mass of seeds on a surface, it will be difficult to get an accurate count, though you might be able to get an estimate if that's all you're after. Anyway watch this video and see if it helps at all, this is proven and reasonably accurate counting technology:

Seed Counting Machine


The counter will give you a good estimate of the rate of seeds striking the surface of the collecting container, even though there's no sensor on that surface.

Here is an inexpensive, stage one prototype of a cheap impact sensor I built. It should give you an idea of what I'm getting at. As you can see from the annotation on the coil it was wrapped with 400 turns of 30 gauge enamel wire resulting in a 9.8 ohm coil. This is far more turns than you will want. I was able to easily hear my finger rubbing the other end of a six foot long lab table just by connecting regular headphones directly to the filter circuit. One nice advantage with this system is that it is only receptive to directly coupled vibrations so there is no need to filter audio.

Right click on the image and choose "view image" to see detail. Sorry about the quick and dirty.

It will take me a bit to dig up the schematic for the board but hopefully by tomorrow morning...


The magnet

Posted on the OPs behalf:

Hi CraftBuilder,
I am grateful for yours and all of the replies I have received on this thread.

I am having great difficulties posting on ATS. New messages and new threads are fine but replies often do not work for me. Must be something about the way the ATS site is programmed or where I am located (China).

Anyway, I wanted to give more background information on this project. It is an idea I had to do with ghosts. www.abovetopsecret.com... I was wondering if ghosts would influence millet seeds in a static-electrically charged plastic tube or bottle. When I come close to the device, the seeds jump around. If the ghosts could influence the bottle with millet seeds inside then the millet seeds would jump around and fall to the bottom of the bottle as the charge changed or dissipated. I cannot use a funnel to collect them because I need the seeds to jump around freely without being constrained.

The 'ghostly activity' would be measured by an arduino(?) by the rate or activity of the seeds falling onto the base of the bottle. Normally, without any ghost being present, there would be no activity because there would be nothing to disturb the static elctric field. So.. it is not really about counting the seeds but monitoring their activity.

Any of this you can quote me if you like. If I could have posted it, I would have.

Thanks,
Q

Maybe consider this as a simple solution then:
Make the coil and plate section of the device I showed above. Place two rectifiers in parallel but opposite polarity across the output of the coil for over-voltage protection. Connect an 1/8 inch male headphone connector to the output of the coil and plug it into your computer's mic input. Use a free audio editing program such as Audacity to record the activity. Audacity has a visual graph that displays volume over time which will give you and activity log of events within the container.

There is also software that uses the computers mic input to record and do analysis on Geiger counter readings which is exactly the same type of output waveform that you are trying to record. I can't remember the name of the software off the top of my head but I will try and find it.

For software look up "Intune" and "pulse recorder/analyser" both by Marek Dolleiser. These will do the job nicely.
There are other fancier ones but I don't know if they are free or not.

a reply to: qmantoo

Given your material, very light in mass, I would think that an optical counter, or other non-contact means, would be the way you would want to investigate for your solution. For example, the density of the falling mass of seeds would be calibrated with experimentation by a "light" shining through/among the falling seeds. The ratio of density observed would be converted to the actual weight/volume of the seeds either individually or in a selected mass to provide your answer.

If the seeds were hitting one at a time, you could use a microphone of some sort and count impacts.

As far as the static charge goes, you could get a negative ion sprayer from someplace and just put the needle down inside the bottle.

You can get them out of broken down air cleaners, or find them on ebay. Something like this

My reply function is working again now this morning so I can thank you all for your very helpful comments and suggestions.

It seems as if I have basically two ways to go - light, & sound analysis. My electronics is not very good as I am a programmer by trade, so I think I will do some searching and experimenting along these lines and see what I can come up with.

I realise that anything which generates static charge is going to involve high voltages so I am not sure that anything like this would be suitable for the public - or at least need some serious safety testimg and certification in a final product.

At the moment I am not sure I want to go so far as producing this commercially, due to the likely low demand for these things. Maybe it is an interesting project which will stay at the prototype stage - depending on how well or whether it works at all. Either way, I will have learned a lot through the making of it.

Again, thanks to all.

If you are going the vibration route I suggest starting with the following circuit.


This is an extremely high gain version that you probably wont need.
You can try to power the circuit by taking advantage of the sound card's supplied biasing voltage with option A. I've never tried this though.
If the gain is too high you may have to implement a voltage divider at the output with optional capacitor B. This makes for a handy filter circuit too.
You can try substituting the op amp with another type but the 5532 has really good low noise performance in this application.


Here are some pics of the coil components.





The magnet mounted as it would be to the bottom of the experiment.


Cutting out the top and bottom coil forms from Styrofoam backing board.


Using a drill bit, paper and tape for the inside diameter of the coil form.


Making the inside diameters of the top and bottom coil forms the right size (a tight fit on the paper tube).


The finished coil form. I like to drip a lot of 5 minute epoxy into the form when I wrap the wire into it so that it creates a solid coil block that holds the form together too. This way I can rip all of the paper out of the form when it sets, and if I've chosen the right drill bit size it produces the smallest possible air gap between the coil and the magnet.


Wrapping the coil


The experiment upside-down, showing where the coil would be glued (or two way taped) to the bottom making sure there is an air gap all the way around the magnet/nuts. The foam board is the same thickness as the nuts so the coil is suspended at the same height as the magnet.


Notes:
Make sure your nuts and bolt are stainless steel so they do not dampen the magnetic field.

If you are using this on anything besides a low carbon (regular) steel plate I suggest shielding the coil assembly in a low carbon steel cup of some kind to avoid interference (hum) from nearby mains fields. Test the type of steel used as a shield to make sure a magnet is drawn to it strongly.

Tighten the magnet retainer nuts only hand tight (this is why there are two). The magnet is too brittle to tighten against with a wrench (spanner)