Notes from meeting 8/8/2017

[hamishmb]

Hi all, here's a brief summary of our meeting. I know I've forgotten some stuff, so please add anything I've forgotten below

Software Notes:

• (DONE) Don't use the PWM pin (BCM 18) in any code.
• (DONE) Max/min ideal levels for the sump - 800/400.
• (IN PROGRESS) Need to write the logic for controlling the levels.
• (DONE) Need to write a monitor thread for the float switch.
• (DONE) Need to write standalone software for the float switch.
• (UNDECIDED IMPLEMENTATION) Must be able to change reading interval at will.

Water wheel notes:

• We will raise the inlet by the sump slightly so the water can fall into the wheel - should make it spin more reliably.
• Penri has a 3D-printed prototype with sensors on it right now, but it working through some issues before we can use it.
• We want a sturdy, adjustable mount for the water wheel fixed by the sides of the sump.

Hardware notes:

• Terry will come in soon to fetch his router. As discussed below it will be left as is.
• He will also do a brief test using the 2 systems to check the setup is working (?)
• The RTC (real time clock) will be installed, and ntp server software on the main pi so they can both be in sync.
• Some other stuff I can't remember.

I think I got most of it, but let me know below if needed

Hamish

[TerryJC]

Hamish,

Mostly right:

• I think that the ideal max/min levels should be considered a starting point.
• I did say that I would increase security on the router, but then you said that you would prefer it if the AP was open and only the Pi secured. Which do you want?
• If I can grab the router during a dry spell, I should be able to test the sumppi with the RTC and a pretend buttspi by this weekend at the latest. I now have a stock of Vero pins but am now waiting for the battery for the RTC (I forgot that we didn't keep them as spares because of shelf-life). By then I should also have the transistors for the second Levels Translator, so I will fit them (although we don't need it yet).

Related to the last point, I will need to know the new GPIO pin allocation so that I can wire the Interface Board.

[hamishmb]

Maybe we should leave it as open for now - it's only temporary anyway.

[TerryJC]

OK.

[hamishmb]

The new pin allocation for the resistance probe is (BCM pins, low to high) 15, 17, 27, 22, 23, 24, 10, 9, 25, 11.

I haven't yet decided on pins for the float switch. How many pins would it require?

[TerryJC]

Thanks.

Only one pin for the float switch will be needed. Penri has made the circuit a simple micro-switch, so we connect one terminal to ground and pull up the other terminal at the Pi end. If needed, I'll do the pull-up by adding a resistor on the board, but for simplicity a RPi.GPIO Library pull-up will probably suffice.

[hamishmb]

Okay, I have allotted it BCM pin 8

How would we take a reading? Why pull the pin up?

[TerryJC]

OK.

To take a reading we would monitor Pin 8 and check for a High. When the float switch actuates, it will short GPIO pin 8 to 0V which the software will see as a Low.

The pull-up is a common device used wherever there are digital signals. When a digital pin is programmed as an input, its input impedance is quite high (of the order of 100s of K at least). If there is nothing connected to the pin, eg an open-circuit switch, then the pin voltage will 'float' and the level seen by the pin could be read as a High or a Low. Pull-ups (and Pull-downs) ensure that when the input is in the open state, the processor sees a defined level which is the opposite of when the input is active.

(If you search for 'Resistor Pack' you will see that component manufacturers even supply resistors packaged in 8s with one end commoned to make this easy for pulling busses up or down.)

[hamishmb]

So, I need to first configure the pin as an output so I can pull it high, then reconfigure as an input? And keep doing that each time I want a reading?

[TerryJC]

You pull up inputs not outputs.