Hi,
I would like to count my ants when I take my queen from the fridge and she hopefully start breeding workers 
The idea is that their nest will be connected to the outworld with narrow tubes with only one ant being able to go through. At first I wanted to measure (infra)red light coming from a LED on opposite side and detect them as a decrease of intensity. They should be OK with it because it is said ants (as other insects) cannot see red light (but they can see near UV light).
Now I have found this article counting bees. As I understand it you put some metal around the tube (fine, ants love dark places) and measure capacitance of the metal. It should work on the same principle as touch sensors. Passing ant should change capacitance of the sensor. There are many things I don't know: can capacitance of such setup be so large that Arduino could reliably measure it? Can one ant (much smaller than a bee) make Arduino-measurable difference? What about energy requirements compared to LED solution (it would be nice to have the whole thing battery powered)? Is it possible the passing ant will be somehow affected by the field?
The spring is near so I will have to take her out of hibernation and it would be nice to have some recording device working from the beginning so I will see increase in activity as the colony grows. I think this is quite strange project (I have found a few similar on internet but none with reported success) but if anyone have some experience I would be grateful for any advice.
Thanks
My sense is that this is a project that will involve considerable experimentation to get a working solution. It's made harder since ants, the horrifying swarming little bastards that they are, are wont to crawl over eachother and move very close to eachother in tight quarters. I suspect that reliable counting will be quite difficult - though a number that approximately reflects ant count wouldn't be as hard.
As far as I'm concerned, any number of ants greater than 0 is too many.
You need two counters to differentiate the ants entering and exiting. Counters themselves need to be small enough and close enough to not get confused when two ants pass them one right after another. This physical size issue may be hard to solve.
Furthermore, you can get the total count only if all your ants exit the habitat at once, which will not happen in normal conditions.
Just out of curiosity, how old is your queen, and what species?
Years ago, I kept a few L. Nigers, but couldn't find the time to build them decent habitats, so I let them go in my garden. Food supply was also an issue. My family would throw me out in the street if I started a maggot farm, and tbh, I don't like them myself either, even though I know they are perfectly clean and not dangerous. They are just yucky.
I believe the queen is L. Niger, I found her under a stone last fall. She laid eggs and small larvae hatched but they didn't grow up. I was told she should hibernate for at least 3 months so she is hibernating now
I know I can hardly count all ants but I wanted to measure their acivity - counting ants outside their nest. But I didn't realize I am not able to reliably recognize if the ant is going out or in. Or has anyone some idea how to manage it? I am afraid two counters would be fooled too often.
Since ants often travel "in convoy" - nose to tail - I think you are going to have extreme difficulty actually counting them.
Even flow cytometry could be tricky, though you might just be able to distinguish head/ thorax/ abdomen sequences.
You could have two tubes, one for entry and one for exit, with hinged flaps that prevent the ants from using the wrong one. The micro-engineering of those flaps, though, is something I wouldn't want to undertake.
Henry_Best:
You could have two tubes, one for entry and one for exit, with hinged flaps that prevent the ants from using the wrong one. The micro-engineering of those flaps, though, is something I wouldn't want to undertake.
Or perhaps your local vet could inject an rfid chip in each ant, then you won't need tubes and flaps, but a very sensitive rfid reader. (Not to mention, reeeeeaaaaalllllly small chips.)
edit... apologies to OP for my flippancy, but as a Monty Python fan of over 40 years standing I couldn't help but see the Pythonesque nature of this thread, reminiscent as it is of the "Michael Ellis" sketch of MPFC Episode 41 (S4E2).
Flypaper will work on ants.
Lay the paper on the ground, let the ants get on the stickey paper, count the ants, throw the paper away.
Job done.
.
Henry_Best:
You could have two tubes, one for entry and one for exit, with hinged flaps that prevent the ants from using the wrong one. The micro-engineering of those flaps, though, is something I wouldn't want to undertake.
I have already thought about it. I believe one-way tube would be quite simple to make. Many plants use insect traps in form of narrow tube with hairs tipped in one direction - in this direction the insect may go nearly freely. In the opposite the hairs are blocking it's way. It shouln't be so difficult to make something like that. I am afraid the ants won't be happy with it - as far as I know they prefer going home the same way they left following pheromones. I believe the need for another way would confuse them. But I am no expert, I may try it later if (when?) I will be desperate.
LarryD:
Flypaper will work on ants.
Lay the paper on the ground, let the ants get on the stickey paper, count the ants, throw the paper away.
Job done.
Poor ants thrown away with your ugly paper. It will never happen! I will have them for noble purposes - such as trying if they can became addicted to morphine
Smajdalf:
as far as I know they prefer going home the same way they left following pheromones.
Precisely.
I think some form of flow cytometry would be the most appropriate. Probably using capacitive sensing, though they may be averse to the electric field (as indeed, they may be to any sensing technology).
looks like you are in uncharted waters.
I believe that a video counter can match patterns at your speeds and get counts for you.
you may be able to create two tunnels that are like a figure 8, but not connected in the center. that would put them close enough (hopefully) to the scent trail. I suspect ants normally find rocks or twigs they walk around. this would be no different
if you are doing this in a cage, you could split the cage, sugar water or whatever on each section, tubes out on both sides. if they hate one or favor one, it would show by activity. figure 8 with hairs, capacitance in another, video in a third, rfid chips in another.. and so-on.
to add to Jimboza's idea, force them to get an ant license for your pet ant Eric.
one another part, you could put in multiple tubes, one with no sensors, the other with your sensor of choice for that area, laser or IR sensors so they walk past the sensors, capacitance counter, etc. again, if they hate one tube with sensors, , they will favor the tube with no sensors. if they are not bothered, they will use them all equally.
you mentioned hairs as one way control. well the hairs have to move. wire up sensors to them.
I understand that someone is fireing up their electron microscope this week, maybe you can get some tiny tweezers and some scope time......
as a general observance, it is rare that people get it 100% on the first try. Edison said his laboratory workers found 3,000 ways to not make filaments for light before he was successful in moving Davy's and la Rue's work forward to make a practical and cost effective and mass producible product based on the work of those who invented electric lamps some 40 years earlier.
dave-in-nj:
to add to Jimboza's idea, force them to get an ant license for your pet ant Eric.
How did you know his name's Eric?
JimboZA:
How did you know his name's Eric?
well, his pet dog is named Eric
his pet cat is named Eric
his pet fish is named Eric
his pet fruit bat is named Eric
his pet bee is named Eric... well. it's really half a bee....
A one... two-- A one... two... three... four...
Half a bee, philosophically,
Must, ipso facto, half not be.
But half the bee has got to be
Vis a vis, its entity. D'you see?
But can a bee be said to be
Or not to be an entire bee
When half the bee is not a bee
Due to some ancient injury?
Maybe the solution is an ant detector van....