Weight triggered trap door

Discussion in 'The Projects Forum' started by Crickeatz, May 11, 2015.

  1. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    Hi,

    I have no electronics background but came across this forum as was wondering if anyone could help.
    I run a Cricket (the insect) breeding facility for the pet food industry. I'm looking for a way for the crickets to effectively self harvest without the need for the existing manual processes other farms use.
    Basically I'm looking to build a small platform on to which the crickets can crawl. When the total weight on the platform reaches a specific amount (number of crickets) then the platform itself drops down and the crickets fall into a collection bin beneath. The platform can then flick back up for the next group of crickets to crawl on to.
    I've been trying to find something designed for this purpose for a few days now with no luck so thinking I might need to build my own. Considering they have a brain the size of a grain of sand, these crickets are surprisingly intelligent and my other attempts to persuade them to do stupid things have failed - therfore any help and guidance most gratefully received!!

    Matt
     
  2. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,300
    880
    Trying to do this by weight could get really tricky: crickets weigh so little (a few hundred milligrams?) that weighing them would probably require an extremely delicate mechanism. Offhand, I don't know of any ready-made sensor that would do the job.

    How about an optical approach? Paint the platform white, place a light source a few inches above it (an LED, or multiple LEDs), along with a reflected-light sensor (a phototransistor or CdS cell), and trigger the trapdoor when the reflected light level drops below some threshold value (which you'd have to determine by experiment).

    My thinking is that approach would likely be a lot more reliable than trying to weigh the tiny little critters...
     
  3. Reloadron

    Active Member

    Jan 15, 2015
    963
    232
    Anything like this will always be a rough estimate. Using for example a trap door concept can be done by measuring the weight on the trap door less the door weight (tare weight). There will always be crickets along the lines of the door and while some drop, some won't drop. Guess it comes down to how accurate the drop weight must be?

    You would use a weight transducer and when a preset weight was reached the associated circuitry would trigger a solenoid for the drop. I also agree with OBW0549's thinking in that this won't be easy simply because the critters weigh very little and move around (jump) a lot. Heck I am even unsure how well a volumetric approach would be. Like saying how much does a soup can of known volume weigh when filled with crickets?

    Ron
     
  4. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,969
    744
    i suppose its either sound,light or weight,

    considering they weigh approx 0.5 grams each, so 100 critters is 50g, thats a small amount, you could use infrared or ultraviolet leds with detectors on the floor to detect them when the light level drops,trip the trap door.
     
    Last edited: May 11, 2015
  5. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,300
    880
    You could also use a light beam interruption approach: arrange a row of 3 or 4 phototransistors along one side of the trap door, and a corresponding row of LEDs on the other side, such that each LED illuminates its corresponding photo sensor. Space the emitter/sensor pairs so that any single cricket on the platform will block only one of the light beams, but multiple crickets will block several. Trip the trapdoor when more than one light beam is blocked.

    Another approach: put a vibration sensor (such as one of those piezoelectric film thingies) on the trapdoor, and use its output to listen for the pitter-patter of little cricket feet. When its output exceeds a certain level (or perhaps a specific number of pitter-patters per second), trip the trapdoor.

    I could probably think of other, crazier solutions, but that's enough for now...
     
  6. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    Thanks for feedback guys - definite food for thought regarding both approaches. Full grown crickets weight about 0.5 grams each and I'd be looking for the trap door to trigger when there were say 50 on the platform so 25 grams; maybe even more if they could fit on. Completely appreciate some along the door lines wouldn't drop in and others would jump out etc so the system doesn't need to be super accurate. Was basically looking to leave access to the platform overnight so hopefully over the course of 8 hours, enough of them will have been attracted to a blue LED and/or food source that the cricket pen will be more or less emptied.
    Please forgive my ignorance but given my electronics experience doesn't go much beyond fitting a lightbulb, are there any step by step guides out there which help with the physical fitting and building of equivalent systems? Would love to try both approaches but don't even know how to link a switch to a platform or how it would be 'triggered'?

    Thanks again.

    Matt
     
  7. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,300
    880
    I kinda doubt it. Without having any electronic design training or experience, my hunch is you'd be in way over your head. You're probably going to have to find a local EE to work with you on designing and putting together a solution, and that unfortunately involves $$$.
     
  8. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,515
    1,246
    Another approach would be reflectance. Illuminate the surface of a white trapdoor with a constant brightness, look at it with a focused photo receptor, and spring the trap when the reflected brightness decreases to a certain level. Experimentation determines the brightness level that equals approx. 50 crickets. Because the sensor is looking at average brightness over an area, individual bugs still can be moving around and not affect the reading.

    Depending on how mobile the crickets are, another way is to run them through very small chute and count them with a photo beam. Works for cattle...

    ak

    OBTW, heck of a question.
     
  9. joeyd999

    AAC Fanatic!

    Jun 6, 2011
    2,673
    2,713
  10. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    Thanks to all for your responses whihc were very helpfu. Agreed, it is a bit of a random one!! I'll try and get in touch with some local engineers to put your ideas into a working model.
    Thanks again
     
  11. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    I am inclined to think that a fully mechanical solution such as a tipping platform or tube may prove to be the simplest and most reliable method; the same principal as a Japanese tipping fountain.

    Monitor the process electronically (magnet and hall effect sensor) to ensure that the tipping is occurring at regular intervals and if it stops; drive a motor or solenoid to shake the mechanism and clear any debris/blockage and if that fails, after a few attempts, raise an alarm.

    Control the whole thing with a micro-controller or, if electronics is not your thing, a mini PLC/ Smart relay.
     
    Last edited: May 11, 2015
    cmartinez and AnalogKid like this.
  12. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    4,769
    969
    I'm sure this exists in virtually every food sorting/packaging plant throughout the world.
    Its fairly simple for anyone with some electronics/mechanical skills and there is probably an off the shelf system..
    Combine a PLC/Scale/Platform/Motor(or cylinder)

    Unfortunately there are way too many fat people in the world and a Google search for things like "weigh count tip sort" isn't bringing up good results on that..
    But I'd think contacting any company that designs food packing/sorting machines would get you to the right place.

    A purely mechanical counterbalance system could probably be accomplished too but I'd much rather have an electro-mechanical system as I'd think it would be easier and more reliable..

    Even a "2 pan balance" (scales of justice) might be able to work for you just need to have it drop down and tip over..
     
  13. izon

    Member

    Mar 17, 2013
    163
    5
    Hi. I've been reading the thread about this unusual project. I'm not real sharp in all areas of electronics but have dabbled into a variety of things and have use the kind members of this group to help me a lot. I don't have a lot of time to spare but possibly could take a shot at helping with some design of some optional ways of accomplishing your goal. Can you "private message" me and we can discuss it a bit more? I'm not sure how to do the PM so maybe you can try contacting me that way. Best regards...
     
  14. MrAl

    Well-Known Member

    Jun 17, 2014
    2,418
    488
    Hello,

    Some problems are more difficult than others, that's all there is too it.

    To weight small things pharmaceutical techs use precision weight scales. These are used to measure the internal organs of dissected subjects like mice. They are extremely accurate.
    The problem would be how to incorporate one of these things into your project. My guess would be to have them walk over the surface, then brush them off to the side with a kind of brush, or some other way you can think of. Having it fall open would be more difficult and probably a little too violent for the weight scale to put up with on a regular basis.
    Some cheaper weight scales can do grams and maybe tenths of a gram, but it would depend on how accurate the critter count has to be. For example, 25 critters plus or minus 1, or plus or minus 5, etc.?

    Another idea would be to find a way to convince them to walk through a narrow tube, which would force them into single file. You could then count the ones that pass through.

    Maybe a vacuum?

    Automation is wonderful when it works.
     
  15. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Is the objective simply to lure crickets to a method of exiting the pen, and thus emptying the pen? Or, are you trying to collect a certain number of crickets in a batch?
     
  16. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    The aim is to empty the pen completely. In a perfect world, I'd like to be able to set up for specific weight of crickets and once this is reached the device stops but appreciate this may be difficult so happy just to get them all out of there! The scales would only need to be sensitive to say 50 crickets (25 grams) and the trap door or whatever activate once that weight is reached. Wouldn't need something to activate for each individual cricket.

    Thanks again for your help all
     
  17. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,770
    1,103
    Could you just lure the critters out with some scent they can't resist?
     
  18. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    Getting them out isn't too difficult, its getting them somewhere they then can't get back in. They are surprisingly/annoyingly unstupid so refuse to just fall off ledges or into collection devices etc. I find myself well and truly outfoxed by these creatures so far!
     
  19. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    If you could get them to gather in a tube, e.g. 100 to 150mm plastic drainpipe, you could fire them them out the end, and through a one-way flap valve, into the collection container with a blast of compressed air. That way, they would have no choice in the matter:D. This method also has the advantages of very few moving parts and the absence of any powered door mechanism means it is unlikely to trap or injure the crickets
     
    Last edited: May 13, 2015
  20. Crickeatz

    Thread Starter New Member

    May 11, 2015
    6
    0
    Sounds like that approach would have potential. Is it possible to get air compressors to fire at regular intervals automatically?
     
Loading...