RF for sensor reading and valve actuation

[NJ Roadmap]

Currently my PIC-based lawn irrigation controller is all hard-wired to the environment - moisture sensor, temp sensor, rain sensor and solenoid valve wired directly to the PIC board. I'd like to make this wireless to rid the project of all the wires which would have to run from the kitchen to the garden, but I don't have a clue about how to go about it as I don't know ANYTHING about RF microelectronics!

My obvious idea is a PIC at each end with some form of serial RF communication inbetween. I'd like the main PIC indoors for easy user control (LCD+keypad on main board) and a smaller PIC outside for valve actuation, data collection and transmission to the main PIC. Some form of RF-modulated serial communication between the two is what I expect will be easiest unless someone has a better idea?

So basically what I'm proposing is this:

LCD+keypad -> User Interface PIC ----RF-----> Field PIC -> Valve + sensors

What I've got underlined is what I need to build. The rest already exists. I'm open to buying rx/tx modules and interfacing them with the current setup and the Field PIC, but would prefer to build the rx/tx circuit from scratch if it isn't too much hassle as this would allow me to keep costs down. Any ideas?

Edit: I have also realised that the Field PIC would need a decent supply to power the sensors and valve. That can be taken care of.

 

[Nigel Goodwin]

Try reading my tutorials then, which include everything you need! - and it's fully legal!.

 

[NJ Roadmap]

Thanks Nigel, am having a read-through now. So far I've gathered that it's probably cheaper and better if I buy the modules. Farnell quotes £6.65 for a module from RF solutions UK:

https://uk.farnell.com/jsp/endecaSearch/partDetail.jsp?sku=1200970

What do you think?

 

[Nigel Goodwin]

That's exactly the same one as used in my tutorials!.

 

[NJ Roadmap]

Sweet, thanks!

 

[NJ Roadmap]

I've only started looking into this properly today as have been busy of late, and I realised one major problem: the transmission needs to be bi-directional because the sensor data needs to sent to the UI PIC from the Field PIC, and the valve-control data needs to be sent from the UI PIC to the Field PIC. This means I would need two sets of RX and TX modules, which would double the cost (from £13 to £26 approx.). And I would also have to use two different bands I'm guessing. Unless there are hybrid RX/TX modules available?

p.s. I'm a bit confused by how RF Solutions quotes their modules as 'hybrid' transmitter and receiver modules when they aren't. Am I missing something here?

 

[Nigel Goodwin]

I've only started looking into this properly today as have been busy of late, and I realised one major problem: the transmission needs to be bi-directional because the sensor data needs to sent to the UI PIC from the Field PIC, and the valve-control data needs to be sent from the UI PIC to the Field PIC. This means I would need two sets of RX and TX modules, which would double the cost (from £13 to £26 approx.). And I would also have to use two different bands I'm guessing. Unless there are hybrid RX/TX modules available?

You can get transceiver modules which work both ways, but it's only a simplex link - which is all you really need. You could do it with seperate modules, and switch them on and off as required, which is essentially what the transceiver does.

p.s. I'm a bit confused by how RF Solutions quotes their modules as 'hybrid' transmitter and receiver modules when they aren't. Am I missing something here?

What do you think 'hybrid' means?.

 

[NJ Roadmap]

I know I know

I was just wondering what your thoughts are on inverting the RS-232 signals before transmission and doing it the crude way? I'm running a bit short on time so am looking for a faster solution to this (RF was not part of my original plan for the device, so it's a bonus really and I can't to spend too much time on it). I'd imagine that the transmission distance will be about 5 metres max. with a wall inbetween. So the received signal *should* be fairly clean.

 

[Nigel Goodwin]

It's crude, but 'may' work well enough - but why not just use the Manchester routines from my tutorial? - it's there, it's free, and it works!.

 

[NJ Roadmap]

I guess I could do that. Would you happen to have the C-code for it? I've never really bothered with asm!

 

[Nigel Goodwin]

I guess I could do that. Would you happen to have the C-code for it? I've never really bothered with asm!

Why would I have C code for an assembler program?.

Presumably your C compiler allows you to insert assembler sections?.

 

[NJ Roadmap]

Yes, but I'd rather understand how you do it than just copy-paste asm. I'm not lazy you know!

p.s. I knew it was a long shot, I guess I'll try and decipher the asm!

 

[Nigel Goodwin]

I didn't write it, I just modified it for my tutorial, and give a link to the original source. But the transmit side is trivial to do, it's the receive side which is more complicated.