If humidity control is an issue with non-forced air distribution, there are humidifiers that mount in a wall opening, are really a duct mount, but with own fan and a grill for the finished wall side. They need a closet or utility room for the other side of the wall and require 120 volt, water supply (1/4" tube), and a drain, possibly a condensate pump for that. Installed one for daughter last winter, works fine with 2 story house. Bob wrote: > Hi, > > Thanks for all the replies. I have got some good pointers to work with. > > My thoughts now are: > > 1. Each room to be controlled will have a PIC controlled radiator valve from > here: > http://www.heatmisershop.co.uk/Heatmiser-In-Line-Wax-Valve-p-16203.html. > This is a wax valve important if you fit to a bedroom and want silent > operation. A local PIC will control this valve directly by monitoring the > room temperature. > > 2. An RF link to be used as to wire will require substantial carpet lifting > and decoration. Will investigate the cheapest RF method. > > 3. Central controller will constantly update each room with the on/off state > and max and min temperature range. This gives me the ability to only heat > the rooms that are in use (a boost button to switch on a room will also be > designed). > > 4. Central controller will monitor outside temperature and adjust the on/off > times to ensure room reaches temperature at the correct time. It will also > control the Boiler and water thermostat. > > This will take me some time I will start by controlling 1 or 2 rooms and > expand from there. But designing it will only be one hurdle, training the > family to shut doors will be the hard part! > > Thanks, > Bob > > > > > -----Original Message----- > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of > Walter Banks > Sent: 18 September 2008 5:39 pm > To: Microcontroller discussion list - Public. > Subject: Re: [PIC]: Using PICs to control Domestic Central Heating > > I didn't lay out the boards. The antenna was a multiturn > rectangular coil about 1.25 inches by 2.5 inches with a cap > across it. When I got involved they had a transmit layer > running on three different processors and had some reliability > numbers for that layer as a proof of concept. > > I recoded the transmit and receiver layers and worked on the > application, application protocol and error recovery software. > (So far back it was in asm, but I did write the assembler that > was used. :) > > For me they violated every rule I knew about RF (not much) > but it did work remarkably well. There was a test store near > my house that were using still using the prototypes a few years > later. > > BTW the company was Telepanel. The work I am referring > to has been public for at least twenty years. > > w.. > > > > > > > > Bob Blick wrote: > > >> Making a good antenna seems like the hardest part - a big ferrite >> loopstick? >> >> Cheerful regards, >> >> Bob >> >> On Thu, 18 Sep 2008 11:52:27 -0400, "Walter Banks" >> said: >> >>> Bob, >>> >>> The original implementation was done on a Sanyo 5800 series CMOS >>> 4 bit processor that was used for watches and small PDA's and in >>> this case store labels. the processor was used because it was low power >>> and an internal gate array that was designed to drive LCD segments. >>> It will take some looking for the implementation notes here, its here >>> we moved and the old mental index doesn't work here. The basic >>> implementation sync'd the serial data with the carrier frequency. The >>> carrier was 32.768Kz. I can't remember the data rate carrier /8 or 16 >>> maybe / 32. The data was manchester encoded so the carrier did a >>> 180 deg shift on a data transition. The links were half duplex >>> bi-directional with a range of about 300 feet in an electrically noisy >>> environment . >>> >>> The transmitter consisted of two output pins on the micro that >>> always complimented each other to drive the tuned circuit that >>> served as an antenna. Data was added by exoring 11 with the >>> port mask that was being output. >>> >>> Input used the same two pins and a software detect of the phase >>> shifts . >>> >>> The system had a low data rate and enough redundancy to be >>> quite reliable. It also had a layer of error correcting in the protocol. >>> >>> When I get a chance I will pull notes. We used a similar system >>> on a wireless transmitter that reported data out of an instrumented >>> tooth brush (forces and direction information for a research project) >>> >>> >>> >>> w.. >>> >>> >>> >>> >>> >>> >>> Bob Blick wrote: >>> >>> >>>> On Wed, 17 Sep 2008 17:26:49 -0400, "Walter Banks" >>>> said: >>>> >>>>> The Tx/Rx had everything in software. >>>>> >>>>> Yep bit banged RF >>>>> >>>> That is way cool. I'd love to actually do something like that. Is >>>> > there > >>>> any documentation or more info around? >>>> >>>> Thanks, >>>> >>>> Bob >>>> >>> -- >>> http://www.piclist.com PIC/SX FAQ & list archive >>> View/change your membership options at >>> http://mailman.mit.edu/mailman/listinfo/piclist >>> >> -- >> http://www.fastmail.fm - Send your email first class >> >> -- >> http://www.piclist.com PIC/SX FAQ & list archive >> View/change your membership options at >> http://mailman.mit.edu/mailman/listinfo/piclist >> > > -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist