Re: [Hampshire] Another Raspberry PI question.

Top Page

Reply to this message
Author: James Courtier-Dutton
Date:  
To: Hampshire LUG Discussion List
Subject: Re: [Hampshire] Another Raspberry PI question.
On Feb 5, 2013 6:55 PM, "Tim Brocklehurst" <timb@???>
wrote:
>
> On Tuesday 05 Feb 2013 18:12:24 Bob Dunlop wrote:
> > Anyway whilst it is possible to build isolated inputs that
> > understand two or more levels and relay them through to the
> > micro as an analogue signal to measure with and ADC I bet the
> > setup is ticky. I'd suggest builing two digital inputs to allow
> > you to monitor white and grey directly and ignore the orange.
>
> Interfacing an ADC is not hard. The MAX 146 communicating over SPI is

pretty
> simple (I've done it and I'm happy to share the circuit diagram and code).
>
> The further question that springs to mind is do you actually have to

monitor
> AC at all? Typically, there is a number of thermostats around the house.

As I
> understand it, these are just bi-metallic switches. Do these still switch
> correctly at a lower voltage? Or can they be replaced with something like

the
> DS18B20 1 wire temperature sensor? These are cheap and simple to

interface.
>
> At which point your controls are the valve(s), boiler and pump.
>


The 1 wire therm is a good idea. The fewer moving parts the better.
With therm sensors rather than on/off at particular temp, i can make the
heating system better. I.E. When in heating hours, heat normally, when out
of heating hours, make sure the system does not freeze. I don't need to
adjust the heating with a stat dial on the wall, i can use a smart phone
app.
I still wish to detect 240V AC because that will help with fault finding.
The system could do analysis for me and tell me which part has failed. I
think the ADC would be good for that.
Example diag:
1) Turn burner on and send heat to the HW. If the HW temp does not rise a
bit, something is wrong in that part.
2) If the burner is on but the HW temp is still falling. The sensor is Ok,
but something else is faulty.
3) If different power is sent to the diverter valve, does the orange output
change as expected. Points to working valve or not.
4) Measure current and volts to the pump. A pump should always draw a
predictable amount of current when on. If not, faulty pump.
--
Please post to: Hampshire@???
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--------------------------------------------------------------