Overwater timer for center pivot irrigation

Discussion in 'The Projects Forum' started by djreiswig, Jul 6, 2014.

  1. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    There is an off the shelf product available from the dealer to fit the bill, but it is pretty costly, so I thought maybe I could come up with a more reasonable solution, plus it's fun figuring these things out.

    The problem: A hydraulically driven center pivot irrigation system for watering crops circles the field at a very very slow rate. At times it can get stuck and remain in the same place watering until we notice and shut it down and get it unstuck. This not only wastes water, but slows down the rotation rate of the unit, causing the crops to not get irrigated in a timely manner.

    I am looking for some sort of sensor/circuit to detect when the unit stops rotating for a given period of time. I believe the off the shelf solution uses some sort of encoder and timer to determine this. The encoder runs against the vertical pipe that the unit pivots on. I believe the pipe is about 8" diameter. A larger "wheel" could be mounted on the pipe if necessary. It also has some LED's that flash to show when the unit is receiving pulses from the encoder. The time delay would need to be easily adjustable as the speed of the unit is adjustable depending on the watering needs.

    I have used a proximity sensor/relay and a plug-in timer to accomplish something similar with another piece of equipment, but it operated at a much faster RPM. I'm not sure how to do this with something that takes 18-36+ hours to make a revolution.

    Any ideas?
     
  2. crutschow

    Expert

    Mar 14, 2008
    13,000
    3,229
    What do you want this to do when a stuck irrigator is detected? Do you need a remote wireless monitor to go with it?

    My first though would be to drive a shaft incremental encoder with a toothed (timing) belt around the pipe going to a small toothed pulley on the encoder. You wouldn't need teeth on the pipe since there is such a large area for the belt contact. That should speed up the encoder enough so that it would give a reasonable number of pulses per minute. Alternately a small rubber pulley on the encoder driven by friction contact with the pipe should also work. Either way you would then need some logic circuit (most easily provided by a microprocessor) to monitor the pulses and decide when the irrigator is stuck.
     
  3. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    A good place to start, determine the number of pulses needed per revolution. It may not be as many as first thought.

    24P ~ 1/hour

    100P could alarm in 15 minutes or less.

    From that use pulses to reset a time delay relay.

    This may be the same as mentioned in your other project.

    That's how I generally program a missing pulse detector in PLC logic.
     
  4. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    Thanks for the replies.

    Would simply like to have a relay that switches. We could use this to shut down the system and/or turn on a strobe or something to indicate a stuck condition.

    I think a rubber wheel on the encoder is probably the way they are normally operated. Not really a way to do a belt, as the pipe is attached at the top and bottom. No way to get a belt around it.

    What would be the simplest/cheapest solution? I'm thinking a microprocessor could get expensive. Is it possible to build with some inexpensive IC's. I am capable of building circuits if given the proper schematics.

    Not sure on the number of pulses (or the range of rotation speeds for the center pipe) that would be needed. I'll have to check and find out for sure the shortest and longest circle times that we need.

    Not sure how to go from the pulses from an encoder to resetting a timer. I don't have much experience with encoders.

    I'll see if I can find some more specific info and post back.
     
  5. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    The logical solution is an encoder as you and crutschow mentioned.

    The reason I asked about the reaction time is that 100-200 pulses per revolution could be done easily with a simple prox sensor. (if run-out is not an issue)

    The timer method is simple, although may require more than on timer for "one shot" pulses.
     
  6. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    As I said, not sure on the pulses. I guess if you figured the 8" pipe went around once every 24 hours. Depending on the wheel size on the encoder, one could probably estimate how many pulses per hour.

    I'll try and find out how long it usually takes for the unit to make one revolution. Then we can better estimate the acceptable stopped time before alarming.
     
  7. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,172
    395
    Cut a 5/in timing belt and cement it around the pipe, about 25 in. Read the cogs with wheeled micro switch, hall sensor & magnet, magnet and reed switch or photo interupter. About 11.5 min per cog, at 24 hr / rev. 555 IC timer set for 15 min. If a pulse is not received within 15 min 555 output goes hi closing a relay.
     
  8. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,979
    744
    Well can you take some photos of the system so we can give better options.
     
  9. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
  10. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    Bernard,
    Sounds like you have it figured out. I'll try and find out actual times, so we can proceed.
     
  11. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,979
    744
    Does yours have wheels on,maybe you can monitor the rotation of these?
     
  12. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    It does have wheels, but if the wheels get stuck in the mud they will still be rotating. I need to sense if the main pivot pipe is rotating so I can tell if the unit is still circling.
     
  13. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    More info:

    The center pivot pipe is 8" diameter steel pipe. The unit makes a circle every 96 hours. I would like some adjustability built into the system if we decide to speed it up or slow it down some. What would be the shortest reliable stopped time to expect to detect?
     
  14. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
  15. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    I'm intrigued. How would this work exactly. Would this sense the earth's magnetic field and tell if the pipe is rotating? What about all of the steel around the area? Would that cause a problem?
     
  16. crutschow

    Expert

    Mar 14, 2008
    13,000
    3,229
    The better question is "what's the maximum time you would want the go before detection that it is stuck?"
     
  17. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    More info:

    There are several of these units of different lengths. The shortest one makes a circle in 24 hours and the longest one is 96 hours.

    I was asking what the shortest time would be that would still be reliably detectable. I know I can't say 5 minutes, because the center pipe might not rotate a detectable amount in 5 minutes. I would say an hour would be acceptable.
     
  18. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    If you can accept an hour, mount a 100+ tooth sprocket ( or previous mentioned belt) and a small prox switch to trigger on each tooth passing.

    Only a hundred points opens up to custom trigger points, such as machine screw heads around the center pipe.
    An encoder of course could give many more pulses per rev.https://www.sparkfun.com/products/1...R9OYCrXwFPFptafO3hRZSfI18rbXC4dFEDhoCjvXw_wcB

    This closing contact goes to an interval timer set to a small interval in relation to the <hour per tooth.

    The output from interval timer will be a "one-shot" from each tooth that resets a free running delay timer ~15min - 1 hour.

    If timer ever gets to total time with no reset = alarm.

    [​IMG]
    [​IMG]
     
    Last edited: Jul 7, 2014
  19. djreiswig

    Thread Starter Well-Known Member

    Aug 8, 2008
    73
    0
    First question, how does the timer circuit react if the switch stops "on" a screw head? Will it still time out?

    I think the encoder type system is what the OEM uses. I'm thinking that might be more reliable than the switches. There tends to be a lot of water in the area around the pipe so sealed components would be better.

    Is it possible to use an encoder without going the microprocessor route? I would also like to have some sort of LED feedback to tell me when the encoder is getting pulses, so I can be sure everything is working.
     
  20. tcmtech

    Well-Known Member

    Nov 4, 2013
    2,034
    1,641
    As I understand it most center pivot systems have the outer most section driving constantly and all the one between it and the center have simple switching systems that turn the drive motors on and off as the sections of line go in and out of alignment with each other.

    From that if one or more wheel drives cycles on or off for a more than a specific average of time the odds are that something is stuck.

    To me I would be looking at how often the section try and align themselves and for how long they normally have to move to catch up with each other.
     
Loading...