Archive - Aug 2006

Having an 18 pin PIC in my possession I decided to limit myself to this microchip and to build the entire project around no more than 18 pins. Initial experiments with the internal PIC oscillator were unsuccessful as it loses or gains too many seconds each day in an unpredictable way. With 2 pins used up for an external oscillator, 2 for power and 5 for the hbridge that left 9 pins for sensors and display. I wanted to use a 7 segment LED for cost, space and simplicity and so ended up with 7 pins for the LED, 1 pin for the feeder door sensor and 1 pin for an input button. I wondered if I might charlieplex the display somehow to get more display from the same number of pins but this seemed beyond me.

replace image

When I set out with this project I bought a selection of PIC microcontrollers to play with. Mostly costing £1-£2 they are remarkable little devices. The original plan was just to operate the timer for the door with a small and simple device and so I chose the little 8 pin pic12f683 and used the internal oscillator to keep time. Unfortunately from experiments this lost or gained an unacceptable amount of seconds each day and despite fine tuning I realised I would have to use an external crystal oscillator for the desired accuracy. I also decided to use more pins to increase the project functionality. I managed to destroy a couple of pic16f648a and ended up with a pic16f88 which the project is based on. Just about any 18 pin pic would do (the pic16fxx would be a cheaper option) and converting the software to work on different pics is relatively easy.

This was perhaps going to be the trickiest part of the project. It seemed to make sense to use gravity to assist the dispensing mechanism and reduce the power consumption of the feeder. I thought of a number of different designs e.g. vertical screw, rotating drum (like a waterwheel), slider under a funnel, horizontal rotating chambers (I prototyped this with a Pringles can and yoghurt pots and it fell apart!) but all had problems. Essential to the design was minimal power use i.e. quick, low power motor or solenoid.

With all the individual components assembled it was time to start using the project.

It works but I am open to any suggestions on improvement.

As I have explained, this is a project designed and constructed without any specialist knowledge of electronics or mechanics/robotics. Information has been gathered from the internet and some books. The design has many inadequecies which I would be pleased for people to point out. I hope to incorporate any suggestions for improvement in to a future version.

insert pic here

I am pleased to announce the completion of phase II of the ultimate chicken house project. This comprises a digital timer to delay opening the chicken house door, an automatic feed dispenser to deliver grain/pellet feed at specified times and some indoor lights to make it easier to see if the chickens are in the house at night. It is based around a PIC microcontoller...

I had a surprise this morning when looking for a marrow to stuff. A small amphibian was resting inside one of the courgette flowers and it was well coated in pollen. At first sight I thought it was a common frog (Rana temporaria) but the warty skin and similarity to previous sightings make me sure it's a baby common toad (Bufo bufo). It got me thinking whether toads or frogs are flower pollinators anywhere in the world. I didn't manage to get an answer after a quick google around although I suspect there must be some species of perhaps tree frogs that are rainforest pollinators?