Asm file for a dht11/dht22 temp/humid sensor.

Ok I feel like I've hijacked the thread now! Never thought about the resolution of ntc's - but I'd be needing adc's with those. Re code sample - project shelved in early design stage - I'd worked out operation, principles and protocols, some block diagrams, very sketchy code outline, that's about it. No code or hardware yet. But if/when I work on it again I'll no doubt be looking for help here and post results I thought the Dallas devices were more expensive than the dht11's, but apparently not.

<<hands thread back>>

throbscottle said:

Ok I feel like I've hijacked the thread now! Never thought about the resolution of ntc's - but I'd be needing adc's with those. Re code sample - project shelved in early design stage - I'd worked out operation, principles and protocols, some block diagrams, very sketchy code outline, that's about it. No code or hardware yet. But if/when I work on it again I'll no doubt be looking for help here and post results I thought the Dallas devices were more expensive than the dht11's, but apparently not.

<<hands thread back>>

Click to expand...

Hi,

Think you have to evaluate what sensors are suitable for your project.

The NTC are cheap but very limited in the distance from the micro, a few feet max.

The DTHT11 is quoted as 20 meters, though I would be cautious about that.

Both of them need one micro pin per sensor.

The DS18B20 and similar Maxim 1 wire devices can go over 100 meters and you can have many devices feeding on to that one wire back one pin of the micro.

What micro and compiler are you thinking of using ?

The only micro I have any knowledge of is low/mid range pic, and it's not much knowledge at that. The system is intended to be able to control motors to adjust the radiators, and be connected to a PC which would log data and provide centralised control (my son is the programmer of the family so he can do that bit!). So I was thinking each room could have a little card with sensor, little 8 leg micro, small battery, room control and a single wire port to communicate with the rest of the system. Possibly LCD if it gets very fancy. One of the cards would be master, with a micro with more i/o, and talk to each room card and to the PC. (**thinks** could actually daisy chain these and eliminate master) The communications would only take place every half hour or so, and each card would still control its radiator even if it was out of the loop. The really fancy version would be able to display room temperature, humidity, pipe soothing music and play with the children...

It all sounds a bit mad I know, but it's the best way I could think of to provide what he wants - individual room heating controls, controlled by a PC (the system could also eventually extend to control the boiler too) and giving good noise immunity, at reasonably low cost. I looked at 1 wire and decided it wasn't really suitable.

I know there are lots of issues with the system (like radiator valve position), but it is very early days with it and it's currently shelved anyway, as is pretty much everything interesting

(My initial reaction was to tell him to fit thermostatic radiator valves, but he has and he's not happy with them. Meh.)

So anyways, micro is Some Kind Of PIC and compiler would be mplabx, though I expect there are cheaper/more appropriate micros I could use, I just don't know what there is. (suggestions?)

Thanks for taking an interest

DHT11's would be ok for such an application, so would a mid level pc, such as the 16c series, and your system kept to a realistic level isnt too mad, in these days of contsant energy price rises its a good idea.
However if and when you get to the stage of pricing up suitable valves you might get a shock, motorised valves cost telephone numbers, even solenoid on/off valves are quite pricey.
You might be able to reduce the number of valves and control the system by having different flow rates through different rads and a priority system, ie some rooms are allways warmer and others may vary a bit depending on demand.
Dallas sensors seem to be reliable, if thats your main concern go for dallas, if you really want humidity then get a discreet sensor, you can throw a simple circuit together with a 555, philips humidity sensors come with a datasheet that includes a simple effective temp compensated circuit that uses a few comps and a logic chip.
I think thats what I'd do, however the dht11 is enough accuracy wise for humidity, I'll use a ds18b20 to measure the temp.

Wow - thanks for the good advice! I had been thinking along the lines of a rubber band drive to modified caps on the normal radiator valves with rotary encoders to get the position - I hadn't actually got as far as looking at what is already out there as my default thought process is design/build everything from scratch. Ho hum

(has a quick google... Oh yes, they are rather pricey aren't they :/ )

Hi,

Its sounds an interesting project but fraught with potential problems that may give you more headaches than you bargained for.

Also the old saying - why reinvent the wheel ..

Have never seen these RF valves before but they seem exactly what you need at a price far lower than you could ever hope to make your own.
https://www.wirelesslightingstore.com/lightwave-rf-central-heating-wireless-control/lightwaverf-wireless-radio-controlled-radiator-valve.html

Cool - looks like they would tie in nicely with my son's idea for a PC controlled setup.

Very smart and very reasonable, they'd be better powered from the mains, but that woudlnt be a problem for us would it.

Hi.

The aforementioned son here.

Thanks for the link to the radiator valves. I hadn't seen those before.

In order to work correctly a signal would also need to be sent to a relay located at the boiler in order to turn it on. Those radiator valves and the LightwaveRF system don't seem to provide a way to do that.

They are probably intended for systems where rads in infrequently used rooms can be shut off and the main rads not fitted with any electrically operated valves, meaning that the control is only partial rather than the whole system.
You can get usb or serial controlled cards with relay outputs, one of these might do the trick of the ancilliaries.

It specifically mentions wifi control (well, it says wif - you can add your own choice of last letter) and control from a web page (well, for their lighting at least - presumably this also applies to the rad valves), so I don't see why you couldn't equip a whole house with them and use cgi to do the control from the pc. If the valves will report back the temp, then there's your room sensor - if not it still simplifies the whole design since room sensor boards only phoning home now.

Oh look, they do a thermostat too. https://www.wirelesslightingstore.com/lightwave-rf-central-heating-wireless-control/JSJSLW910.html
I feel redundant now.

No matter what it is these days someone has made a gizmo for it.
Like you I'd go and buy the stuff for the job and find something else to build, doing it yourself would take way longer and probably cost double, sometimes you can make something and either save a fortune or have an item you wouldnt normally, in this case I think thats out the window.