Friday 28 December 2012

Arduino and Sky box innards

 Well as its Christmas I've had a bit of time on my hands so thought I'd start on a project a friend of mine, Adam thought of. He's key developer on TVHeadend and liked the idea of replacing the Sky+ box. That's been the drive behind TVHeadend, XBMC, etc. So why not go a step further? Take that old Sky plus hardware/case and re-use it for the XBMC PC with fully functioning front-panel.

I've had a few Sky+ boxes lying in my garage for a while now (Faulty capacitors, or just old hardware so they just sit there), so I thought I'd rip one open and see whats going on. Firstly we know the insides of the Sky+ box,

its 1 main motherboard, a PSU board and hard-drive. Then a ribbon cable feeds the front panel. So HDD comes out, and I pull apart carefully the front panel. Luckily they're made to disassemble, so a couple of screws at the side and the front panel then can unclip with gently prising the locking tabs all round it.

This revealed the front panel innards, and a 2-part circuit board with a lot of circuitry on it.

   

Not what I was expecting, I was thinking more along the lines of either I2C or a simple LED cluster that fed back to the motherboard. I powered up and started to probe the pins to see what I was getting. Unfortunately I don't have an oscilloscope so I couldn't see any data/signalling, so just went on voltage initially. Unfortunately all I found was ground, power and standby-power. Pretty disappointing. I then tried to find out what the LEDs that light up the Sky playback circle (The sequence of LEDs that go round in a circle when playing back video, etc). Unfortunately as I don't have an active SKY card in, even the demo won't display this sequence, so again I've had to give up on this. So I've decided to pull the circuit board out, and do my own thing! Fit a cluster of LEDs in the back and just feed those back to my Arduino board.

The idea is that I'll plug the arduino into the XBMC/Openelec box using USB. I can then write a serial communication language that will send status to the arduino that will then display on the LEDs. On top of that I'm going to cut out part of the front panel of the Sky box and put in a backlit LCD screen, that will show current channel, now/next, something like that (Not fully decided). Last night I got as far as mounting the LEDs behind the facia and started to check they were aligned (tricky!). Next will be to start wiring that to the arduino to see if I can get the 'animation' of the LEDs looking anywhere near OK.

If anyone has done any more interfacing with the Sky+ front panels or figured out the control lines, etc, please let me know! I'd love to get the original circuit board running somehow.

 

 

Thursday 20 December 2012

Borri UPS for home use

Around a week ago my APC Smart UPS SC420 decided its battery needed replacing (again!). I only seem to get around 2-3 years out of mine which started me thinking, why on earth do I get such short life from something 'relatively' expensive. These units come in at around £120 and batteries around half that. For 420va that doesn't seem too great, as all I do is power 2 machines (desktops with a few add-ons) and a network switch/wifi. Mainly to protect the computers from nasty spikes and unexpected shutdowns more than running without power (since the internet router is in the house, not connected to a UPS).

However, I wanted to replace the UPS to keep the protection up, mainly as the power in my area seems to spike and brownout quite a bit and I've had a lot of things go pop as a result of this (X10 kit hates this kind of choppy power, hence most of my X10 home control stuff going bang). So I duly went to APC and looked up the replacement battery. It came out at around £50 after postage, etc. So I then looked at the APC upgrade wizard, see what they would offer me for a trade-in, as I particularly like the idea of trading in old hardware to get new for a bit of discount and I feel I've done the environment some good with them having batteries inside, etc. For trade-in I was disappointed. I could get a replacement unit for around £100, thats with the same 420va limitation, which since I run that at around 80-90% load isn't great (and I suspect why I go through batteries so often). So looking at the next up capacity 450va, this was £151 for really very little difference, or the 550va model for around the same price. Not very impressive.

So I started doing my usual answer to everything, google for recommendations or alternative brands, Borri kept coming up as they appear to do all sorts of UPS and generator equipment from SOHO right up to datacentres, etc, and looking quite promising I looked at the B60 range. These started at 800va and went up to 2000va, and shopping around for prices. I was pleasantly surprised, the B60-800va version would be £71 delivered, etc. So thats approximately double the rating for a nice £71.

The next issue before ordering, was it linux compatible? For years I've used APC units and used the ever popular opensource APCUPSD to shut down servers when batteries neared depletion, etc. Firstly Borri provided a download of linux software. An excellent start. Checking the software it wasn't what I'd hoped, it was a java-based server/applet that basically talked to the ups over USB and presented output via a flash-based web-ui on a custom java-tomcat url on your machine. OK, not the end of the world. I did more digging and thought of NUT for linux. This looked to support it, so worth giving it a go.

UPS arrived next day and it's quite a nice looking unit, really clear permanently lit LCD on the front telling you input volts, output volts and current status (Battery level in bars, current load level). I plugged it into my laptop to play around with and get an idea of the java tomcat web interface. It actually was quite neat, just a pity it was flash and didn't seem to have a raw data port/method of extracting data from it otherwise I'd have stuck with it.

Installing NUT and setting it up for the Borri wasn't too bad, it's partially supported according to NUT using the blazer_usb driver, but this doesn't give you battery level/capacity and time to depletion. No real issue, I can survive!

I've yet to add my load of machines to it and see how it copes, but so far its looking pretty good. If I figure out the battery level/capacity part of it I'll be very happy with this solution.

 

Sunday 9 September 2012

TVHeadend

 Quite a while since I did a post, and things in the TVHeadend and XBMC world have moved on quite a bit. TVHeadend is now being actively developed, primarily resolving the EPG issues that were around in the past, but then also moving onto loads more improvements. Therefore things have moved to a development environment, so if you want to try out the latest you can on github.com/tvheadend. Or via the original www.lonelycoder.com site where you can get debian packages and official releases. Hope you like all the updates!

A few people have also asked me about the dev stuff I've done, I've got a few testing areas for TVHeadend at my own github (http://github.com/andyb2000/) so take a look, and drop me a note if you want to try anything out as I'm always looking for new testers, or join the TVHeadend lot on IRC freenode.net #hts

 

Saturday 14 July 2012

XBMC, TVHeadend and watching live TV

 As some of you may have followed in earlier posts I've been replacing our old Sky satellite setup with my own setup here at home. Well over the past month or so its improved considerably so I thought I'd write up where I'm up to with the whole setup and how the family are getting along with it.

Firstly the problem, Sky is expensive and to be honest the majority are repeats and most of the channels we frequently watch are FTA (Free-to-air/satellite) in the UK, so dropping Sky was logical. So I then had the Sky dish and the problem of how to feed it into the house setup. Firstly the front-end. I've used XBMC since the early days on the original XBOX thanks to Bunnie Huang and his amazing books, etc. (Google it!). So I'm happy with XBMC, so setting that up on a small footprint PC with quiet fans and HDMI out. I bought a cheap Nvidia card with HDMI out (Cost around £35), as these cards can do hardware decoding, this is important as it takes the load off the PC's CPU and very important when your machine isn't the most powefull on the planet. So with that in the PC (It's a slim-line Dell so fitting the Nvidia was a pain, the case doesn't quite close properly, but its hidden out of sight so not a problem). The HDD in the XBMC machine isn't too important, big enough to hold the XBMC install is enough as you won't really be storing anything on it. Save your disks for the 2nd bit of hardware! Networking is also important, make sure its cabled up right and if possible use Gigabit network cards+switches as it'll speed the whole system up (Although for HD tv you can get away with 100Mbps as HD tv streams are 35Mbps (at worst) dropping down to 6Mbps.

Install XBMC: Simply download their Live CD XBMCUbuntu, burn that and run it. There is the option to install, choose that and wait whilst it flattens your HDD, partitions, installs, etc. Now a word of caution. Make sure your graphics card, hardware, etc, are all plugged in at install, as the installer does a really good job of installing the right components if it finds the right hardare. It'll save you a world of pain later! So now you should be booting XBMC from HDD and getting things looking reasonable. You can tweak this as you need. Insert a MS media centre remote for remote control/IR functions (Pretty much any will do, I found the older silver ones work better than the newer branded editions (HP badged, etc, I found didnt work reliably). There should be little else you need to do.

To get TV functions into XBMC you need the latest packages, Pulse-Eight provide these in an easy script, SSH to your XBMC machine, then run these commands:

       cd /tmp; wget http://packages.pulse-eight.net/ubuntu/install-xbmc.sh; chmod +x install-xbmc.sh; ./install-xbmc.sh

Thats all one one line to make copy/paste easier! What this does is grab the install script from Pulse-Eight, set it to executable and run it. This adds a repository into Ubuntu with the relevant updates, upgrades and does it all for you. When its done, just reboot and you should have XBMC again but this time it will have additional TV functions hiding. This is all we do with XBMC for now.

(NOTE: XBMC has a few problems with standby, resuming usb devices at the moment, this appears to be an Ubuntu issue and not XBMC, so if you set your settings > power savings to standby when power pressed and after resuming you find it won't make the IR remote work you've got the bug. There are loads of forum posts about it. I've tried them all and with no success as yet unfortunately).

If by chance your IR works after resume then you'll want to be able to wake the XBMC machine from its sleep by the remote (So a simple power button press on the remote will wake it). If thats the case you want to look at adding the following to /etc/rc.local:

 

echo "USB0" > /proc/acpi/wakeup
echo "USB1" > /proc/acpi/wakeup
echo "USB2" > /proc/acpi/wakeup
echo "USB3" > /proc/acpi/wakeup
echo "USB4" > /proc/acpi/wakeup
echo enabled > /sys/bus/usb/devices/7-2/power/wakeup
 
That last line you need to find the correct numeric value (Instead of 7-2). Do it just by typing lsusb, look for your IR device and the numbers you need are the BUS and DEVICE. So in mine above I had:
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 002: ID 0644:0200 TEAC Corp. All-In-One Multi-Card Reader CA200/B/S
Bus 002 Device 003: ID 046d:08cc Logitech, Inc. Mic (PTZ)
Bus 007 Device 002: ID 045e:006d Microsoft Corp. eHome Remote Control Keyboard keys
 
So the last entry is my IR remote, take the 2 numbers 007 002, remove the 0's and you have the 7-2 needed for the wake command. This will allow your USB devices to wake the machine when its in sleep.
 
 
 
Now onto the Satellite part of things. TVHeadend. This is best ran on a separate machine. A few people run it on the same machine as XBMC but this appears problematic and you will also need to leave the machine on, or do some trickery to wake it at the right time to handle recordings, so we're skipping that and using a dedicated machine. The other advantage is you can stick this out of the way (In my case, my modified Garage) so its on all the time. This means you don't need to use quiet fans, etc. So purchase some Satellite/Terrestrial cards. I use 2 satellite cards (DVB-S DVB-S2 are Satellite), and one terrestrial (DVB-T). If you're buying satellite cards then I'd strongly recommend to get DVB-S2 (And not the S1 cards), because some channels in the UK now are only available on the more advanced S2 streams, so to future-proof I'd go DVB-S2. I'd also recommend against combination cards (2 receivers on one card, one sat one terrestrial, etc), as in some cases Linux won't be able to use the different elements separately and you'll end up restricting what to do.
So I installed one DVB-S card, one DVB-S2 card and one DVB-T card (All PCI as USB is also problematic in some cases). I did a normal/basic Ubuntu server installation (No graphical interface needed, but if you're a novice it won't harm). Get it all up to date and then get ready for the other installations.
 
Next I created a storage area for recordings. On my network I have a NAS (Network Attached Storage) so I used this, however for these instructions I'll show you how to make your TVHeadend server into a basic NAS for TV use. The reason I did this was that it allowed all machines on the network (Regardless of if they ran XBMC or TVHeadend integrated) to view and playback recorded material. On my network at home I used NFS as the majority of my network is Linux-based. So you need to install the nfs server to your TVHeadend machine:
apt-get install nfs-kernel-server
After install you need to create your storage area, so do something like:
mkdir /home/recordedtv
We'll use that to save all our TV to. So you need to export that from NFS:
echo "/home/recordedtv *(rw,sync,no_subtree_check,no_root_squash)" >> /etc/exports
This adds the folder to the export for NFS with a few parameters, WARNING: we are letting anyone that connects full read/write access without restrictions. Only do this if you trust your network, otherwise read up on safer NFS settings.
 
Now we need to install TVHeadend. At the moment there is some active development which is providing amazing changes to how EPG and the main TVH system works, this work is done by Adam Sutton who has made some great updates to TVH. I've been very lucky in being able to beta-test a lot of code and provide input to its development, so I'm very grateful to Adam who has humoured me all along! So at the moment this would be the branch I would recommend over the main tree or the debian packages. So it's a little more involved compiling but its worth it! So here are a few simple instructions to get you started:
apt-get update
apt-get install build-essential git 
cd /usr/src
mkdir tvheadend
cd tvheadend
git clone https://github.com/adamsutton/tvheadend.git
git pull origin initialscan
You should now have a full copy of the latest dev code. Go into that folder (cd tvheadend) and you can then start compiling. I can't remember if there were additional dependencies for TVH so if configure fails you will need to add the dependencies. If you're unsure pop a comment/tweet over to me and I'll help out and update this doc.
 
cd tvheadend
./configure --prefix=/usr --release
make
 
That'll give you a working TVH binary to use. Only a couple more steps to go now! 
Create a user for TVHeadend to use:
adduser tvheadend
You can then use that to run TVH, etc. So change to that user and do a simple setup task. This is because at present the work is in development. So carry out:
mkdir ~/.hts
mkdir ~/.hts/tvheadend
cp -Rp /usr/src/tvheadend/tvheadend/data/epggrab ~/.hts/tvheadend
All this does is creates the TVHeadend config folders, and also copies the libraries needed for EPG grabbing to work.
I tend to run TVHeadend it in a screen process so you can see if any errors occur, so you can do this:
screen (Press enter on the information screen that appears)
cd /usr/src/tvheadend/tvheadend
./build.linux/tvheadend -s -d
 
That'll fire up TVHeadend. Hopefully it'll run and got exit with an error. If so you can now swap to the webui and start to configure it.
The rest of the config is fairly standard, in the new EPG grabber tab make sure under the OTA section you enable OpenTV: Sky and EIT grabbing. The channel name/number/icon is experimental so you may wish to leave this off for now. Continue on to the TV Adapaters section and setup the rest of your TVH setup as normal. This is documented elsewhere and I've documented under my old XBMC section so take a look there is you're totally new to TVH.
 
Good luck, and as always please do get in touch if you'd like a bit of extra help with your setup.
 

 

Wednesday 13 June 2012

UK Cookie Law

 UK Cookie Law Compliance.

Well the ICO have required all EU websites that use cookies to apply a compliance method on the 26th May 2012. What my views are on this? Well if you want to accept cookies, your browser will accept them. If you don't you set your browser to NOT accept cookies and you're done. But apparently that isn't the 'right' way according to the ICO's techs, so all website authors, operators, etc, have to do something a bit different. In most cases you have to block the cookies until you get confirmation from the user. However these have been updated and now gives us the option of implied consent which gives us a little simpler option to take here. We'll implement this on the website here, basically by visiting the site you are accepting that we will use cookies to assist in the website experience, however if you do not login or provide account details then this information will simply be used for statistical analysis and production of website usability.

The above is basically our cookie policy for this website, the website is a not-for-proft, hobbyist site and so the use of cookies are simply for user statistics, or where a user registers for easing their use of the site/login/administration.

 

(PS: You'll also notice at the bottom-right of the website my cookiecontrol compliance, this is a nifty little free add-on that helps you with the compliance, its from  http://civicuk.com/cookie-law/index if you want one!)

Monday 11 June 2012

Arduino - Cheap wifi connectivity

Building a cheap ethernet connection for my Arduino. I needed to relay data from my Arduino back to my home server. The location
makes wired prohibitive, but I also don't have CAT5 cabled to the location I had the Arduino in. I then looked at the various
shields available, the Ethernet shield wasn't too expensive at around £20, but this would mean wiring, and the location in the kitchen
was particularly difficult to get wiring to.
 
I then looked at wifi shields, these came in at around £85 and way too expensive for just a little hobby project, the in-between options
of using Zigbee interfaces, which is a slightly cheaper option, but I was concerned about the distance and complexity of setting
this up with my server.
 
I therefore started posting to mailing lists and trying to find a way of doing a wifi-bridge cheap, or USB over wifi. All I needed was
to get the usb/serial data out from my Arduino to my server so that I could graph it, monitor it, etc.
A kind chap on an ISP mailing list brought forward the TP-Link TL-WR703N suggesting that this device can be flashed with OpenWRT
and then act as a wifi bridge. This sounded interesting, and you can get them on eBay for less than £15 inc delivery. This was sounding
good.
 
The TP-Link TL-WR703N has 3 physical ports (RJ45, USB host and power (via a micro-usb port)). It also has integrated WIFI. So my
idea was to connect the wifi up to my house wifi, then use the USB port to monitor the serial output from Arduino. Ideal!
 
The TL-WR703N arrived, all in its chinese writing and unfortunately its chinese web admin interface. Luckily for me, the OpenWRT
site for this device was very detailed and told me exactly what to click to upload OpenWRT.
Then OpenWRT site at http://wiki.openwrt.org/toh/tp-link/tl-wr703n Had all the details to flash it, very simple and quick.
Once flashed with OpenWRT a little more tweaking was needed. I've generally not done much with OpenWRT in the past, I stick to DDWRT
which has a nice web-gui, so OpenWRT involved SSH'ing to the unit and setting the rules correctly for what I wanted.
 
I did a bit of reading on the OpenWRT recipies site on what I needed which was to be a client on a mixed WPA2-Personal AES 
encrypted wifi with MAC protection.
The basic method is follow the outline of a MASQUERADE CLIENT on OpenWRT and adjust accordingly. So my /etc/config/network looked like:
 
config interface 'loopback'
option ifname 'lo'
option proto 'static'
option ipaddr '127.0.0.1'
option netmask '255.0.0.0'
 
config interface 'lan'
option ifname 'eth0'
option type 'bridge'
option proto 'static'
option ipaddr '192.168.1.1'
option netmask '255.255.255.0'
 
config interface 'wan'
option ifname 'wlan0'
option proto 'dhcp'
 
(Note that I left LAN on the 192.168.1.1 setup, this meant that if the WAN or dhcp failed I can plug into it and still admin the thing)
 
/etc/config/wireless was a bit more tricky:
 
config wifi-device 'radio0'
option type 'mac80211'
option channel '1'
option macaddr '8c:21:0a:d8:7b:57'
option hwmode '11ng'
option htmode 'HT20'
list ht_capab 'SHORT-GI-20'
list ht_capab 'SHORT-GI-40'
list ht_capab 'RX-STBC1'
list ht_capab 'DSSS_CCK-40'
 
config wifi-iface
option device 'radio0'
option mode 'sta'
option ssid 'MYSSID'
option encryption 'psk2+ccmp'
option key 'MYPSKPASSWORD'
 
To find these settings you issue 'iwlist scan' and look for your SSID and the other settings. So in my case I set the channel of 1.
The encryption of psk2+ccmp (This refers to wpa2 with AES encryption). After doing that you can then do 'ifup wan' and 'wifi' which
should then get things working. Except in my case it didn't. I checked, I had the MAC address in my wifi router, all looked fine but it
wouldn't join the WIFI. The settings in OpenWRT are a little hazy so I assumed this was the problem and spent a while here, but in the end
I found turning MAC protection off temporarily on my DDWRT allowed it to connect, enabled it, saved and it all works each time now,
so unsure if I hit a bug in my DDWRT on my wifi access point or something, but for future, debugging with MAC protection turned off
is a lot easier!
 
Now I had it connected I needed to get info from the Arduino over the USB port. I plugged it in hoping for the /dev/ttyACM0 to be
created, but no luck. I got the USB insert message:
[   10.020000] usb 1-1: new full-speed USB device number 2 using ehci-platform
So it looks like the kernel driver doesn't exist.
Luckily the OpenWRT group have a packaging system that includes a lot of drivers and add-ons, so using the package management to install:
 
opkg update
opkg install kmod-usb-acm
 
Reboot, and when I plugin the arduino, success, I see:
[   10.020000] usb 1-1: new full-speed USB device number 2 using ehci-platform
[   10.160000] usbcore: registered new interface driver cdc_acm
[   10.160000] cdc_acm: USB Abstract Control Model driver for USB modems and ISDN adapters
[   10.210000] cdc_acm 1-1:1.0: ttyACM0: USB ACM device
 
And the /dev/ttyACM0 is there. Last trick is how to get a couple of lines of data from the serial connection and SSH/SCP it up
to my server. This proved a little tricky, as first of all getting just a couple of lines of data from a constantly flowing serial
terminal, then the scp function itself.
 
To get just a couple of lines of serial I firstly tried the stty trick "stty -F /dev/ttyACM0 9600 -echo -onlcr time 50 min 0" and then
cat /dev/ttyACM0 which will read data for a specific amount of time. This was very unreliable and often hung the /dev/ACM0 interface.
So I looked to alternatives. I then tried a constant/background "cat /dev/ttyACM0 > /tmp/serial.log", periodically from cron reading
it and zero'ing it. Problem here was the filesize growing and causing the device to hang due to the limited ram on the device. Also when
/dev/null'ing the /tmp/serial.log the cat then failed and a kill/hup to the cat caused the ACM0 to hang again! So no good there.
Finally I tried "head -n 12 /dev/ttyACM0 > /tmp/arduino.log" So this output the last 12 lines from the serial console to a file.
(Overwriting the file each time keeps it low filesize). Then I can parse this for the info I need, and then scp it over. Except scp
doesn't exist as the ssh is just dropbear! So again to opkg and install the ssh client tools so I get the scp binary. Generate an
id_rsa key and then I can scp my output over: "scp -i /etc/dropbear/id_rsa /tmp/arduino.log root@192.168.xx.xx:/tmp"
Success! I can now parse this with my graphing tools.
 
Enjoy, and thank you to the TP-Link TL-WR703N for a cheap and nifty solution! (PS: Got any direct questions? See me on twitter: www.twitter.com/andyb2000 )

Wednesday 9 May 2012

Arduino garden helper

 Over the recent bank holiday weekend we decided to plant a few vegetables in the garden. Nothing too radical, a few Onions, Peppers, Chilli and re-plant our little strawberry plant. We invested in one of those small polythene tunnels too, so help give them a bit of a boost start and see if we can get the plants happy despite the miserable cold weather we've been having. To this end I finally thought up a use for my Arduino I've had kicking around. So I started to think about a soil moisture and poly-tunnel temperature sensor type array that would feed back to the Arduino and tell us when it wanted watering, and also what the highs and lows are for temperature.

So I set about making a soil moisture sensor. My basic background in electronics helped and I knew the simplest method is to use a resistance type technique, so provide +DC to a probe, and them measure the resistance between that and a second probe. This should be easy for the Arudino as it has a few analogue inputs that I can use. The second input was temperature, so I used a temperature sensor I'd previously bought from oomlout.co.uk (TMP36) which has a nice range of -40oC up to 125oC so perfect for this. It's supplied in a standard semiconductor package so first I had to make it a little more weatherproof. Soldering up the 3 wires to a long length of cable (burglar alarm cable, which is my fave cable for general purpose as its 6-core and pretty reasonable in terms of weathering and loss of signal). I then coated each of the exposed wires with a bit of heatgun glue, followed by electricians tape round each leg, so each were electrically insulated and also hopefully sufficiently weather proofed. (Not sure how well that would survive totally exposed to the elements, but as this was going in the poly-tunnel I thought it would be fine).

Next was to build the soil moisture sensor. There are a few different ones on the net, so I went with one that seemed a common way of doing it. Set two probes (in my case 2 brass screws) in a piece of electrically non-conducting packing foam. So this lets the screws push down into the earth whilst keeping the rest of the screws insulated. At the top I then wire wrapped the alarm wire around the screws and added a few blobs of solder for good measure. A bit more electricians tape finished them off.

Here is the sensor that I came up with. You can see the TMP36 semiconductor sticking out the top, and the screw probes out the bottom.

 

After the probe was sorted, I ran it round outside into the poly-tunnel:

 

So the final part was the circuit into the Arduino, and also the code to make it function. The breadboard I had already connected to my Arduino had the LCD attached so I left that to give constant output on the Temperature and soil moisture, so all I had to add was the connections out to the remote probes. I found an interesting issue, at first the soil moisture values varied wildly, even when the probe wasn't in the soil, until I realised that this was because nothing was 'pulling' the input to ground, adding a quick resistor into the setup so that it pulled the input to 0 when it didn't have any other value sorted that out and I got values depending on the resistance being presented. At first I'd connected the soil moisture probe to +5ve and the other probe to my analogue input. One problem I'd found reported quite frequently online was about electrolysis of the metal probes (where electricity passing through them caused corrosion to take place). To counter this I moved the moisture probe to an output pin, and so I'll only switch on current to the probe when I wish to measure it, hopefully reducing the amount of electrolysis taking place.

A few iterations of code were required to get the reading showing anything sensible, and de-bouncing the inputs was essential. I found 4 iterations of the sensor read loop followed by an average calculation stopped a lot of the large fluctuations taking place and gave me what appear to be stable readings, and when compared with the temperature read-out from the car, the outdoor temperature looked around the right values. Soil moisture though is still a bit of an issue as I need to self-calibrate the sensor by checking the soil and deciding when it gets too dry to set that as the lower threshold.

I'll post the final code up here shortly as I'm still fine-tuning it, and also hopefully connecting my Arduino up so I can access the readings remotely and graph them using my rrd graphing system I use for other things around the house (heating, electricity, etc).

Here are a few images of the Arduino in its plastic (indoor) enclosure in operation. The Green LED lights up when the soil moisture drops below the triggered limit, and the Red LED lights when the readings are being taken. Debugging is also output via the serial port.

  

 

Aruino code + schematics to follow!

 

 

Update 10/May/2012 - Here is the current Arduino code I've been using, sorry about the messy code!

<pre>

// AndyB 7-may-2012 Outdoor temperature and soil moisture circuit

// ---------------------------------------------------------------

// include the library code:

#include <LiquidCrystal.h>

#include <EEPROM.h>

#include <Time.h>

#include <TimeAlarms.h>

 

// PINS

//   2 3 4 5 11 12 - LCD

//  9 = watering LED

// 10 = soil power pin

// 13 = Data collection LED

//  8 = LCD display light

//  6 = LCD display button light up

 

// initialize the library with the numbers of the interface pins

LiquidCrystal lcd(12, 11, 5, 4, 3, 2);

int sensorPin = A0;       // Temperature sensor

int soilMoisturePin = A1; // soil moisture

int sensorValue1 = 0;

int addr = 0;

int SoilValue1 = 0;

float temperatureC = 0;

const int buttonPin = 7;     // the number of the pushbutton pin

int buttonState = 0;         // variable for reading the pushbutton status

const int ledPin =  13;      // the number of the LED pin

const int waterPin = 9;      // the LED for watering

const int soilPowerPin = 10;  // Pin to power up the soil moisture sensor

const int lcdPowerPin = 8;    // Pin to light up the LCD

const int lcdButtonPin = 6;  // Switch to light up the LCD for a few loops

 

int lcdlightloop=0;

 

int logloop = 0;

 

int debug = 0;  // Debug setting

 

void setup() {

  Serial.begin(9600);

  if (debug == 1) {Serial.println("Program initialised");};

  pinMode(ledPin, OUTPUT);  

  pinMode(buttonPin, INPUT);

  pinMode(waterPin, OUTPUT);

  pinMode(soilPowerPin, OUTPUT);

  pinMode(lcdPowerPin, OUTPUT);

  pinMode(lcdButtonPin, INPUT);

  // set up the LCD's number of columns and rows: 

  lcd.begin(16, 2);

  digitalWrite(lcdPowerPin, HIGH);

  // Print a message to the LCD.

  // lcd.print("       oC");

  // write a 0 to all 512 bytes of the EEPROM

//  for (int i = 0; i < 512; i++) {

//    EEPROM.write(i, 0);

//  };

  // do not reset the EEprom, find the last value and set the pointer, avoids power resets loosing old data

  if (debug == 1) {Serial.println("reading EEPROM");};

  for (int i = 0; i < 512; i++) {

    byte tmp_val = EEPROM.read(i);

    Serial.print("EEPROM ");

    Serial.print(i);

    Serial.print(": ");

    Serial.println(tmp_val, DEC);

    int tmp_val_int=tmp_val;

    if (tmp_val_int == 0) {

        addr=i;

        if (debug == 1) {Serial.print("EEPROM empty at ");};

        if (debug == 1) {Serial.println(i);};

        break;

    };

  };

  digitalWrite(soilPowerPin, LOW);

//  sensorValue1 = analogRead(sensorPin);

//  float voltage = sensorValue1 * 5.0;

//  voltage /= 1024.0;

//  temperatureC = (voltage - 0.5) * 100;

//  SoilValue1 = analogRead(soilMoisturePin);

//  Serial.println("Back from reading sensors");

//  logdatatoeeprom();

  if (debug == 1) {Serial.println("Setup completed");};

  // Alarm.timerRepeat(9000, logdatatoeeprom); // every 15 minutes

  // Alarm.timerRepeat(15, logdatatoeeprom);

  digitalWrite(lcdPowerPin, LOW);

}

 

void loop() {

  if (debug == 1) {Serial.println("Starting main loop");};

  // check the button status

  buttonState = digitalRead(buttonPin);

  if (buttonState == HIGH) {     

    // Button pressed - Reset all of the EEPROM values

    if (debug == 1) {Serial.println("Button pressed - resetting EEPROM");};

    digitalWrite(ledPin, HIGH);  

    // write a 0 to all 512 bytes of the EEPROM

    for (int i = 0; i < 512; i++) {

      EEPROM.write(i, 0);

    };

    Serial.println("EEPROM ERASED");

    addr=0;

    logdatatoeeprom();

    digitalWrite(ledPin, LOW);

  };

  // check for LCD display button

  if (digitalRead(lcdButtonPin) == HIGH) {

      // LCD button was pressed

      lcdlightloop=1;

  };

  if (lcdlightloop > 0) {

      lcdlightloop += 1;

      digitalWrite(lcdPowerPin, HIGH);

  };

  if (lcdlightloop > 20) {

    lcdlightloop=0;

    digitalWrite(lcdPowerPin, LOW);

  };

  if (debug == 1) {Serial.println("Button checks complete");};

  // set the cursor to column 0, line 1

  // (note: line 1 is the second row, since counting begins with 0):

  digitalWrite(ledPin, HIGH);

  if (debug == 1) {Serial.println("Read temp sensor");};

  

  // READ SENSORS HERE

  temperatureC = readinput_temp();

  SoilValue1 = readinput_soil();

  

  if (SoilValue1 < 40) {

    digitalWrite(waterPin, HIGH);

  } else {

    digitalWrite(waterPin, LOW);

  };

 if (debug == 1) {Serial.println("Display to LCD");};

 Serial.println("Temp: ");

 Serial.println(temperatureC);

 Serial.println("Moisture: ");

 Serial.println(SoilValue1);

  lcd.clear();

  lcd.setCursor(0,0);

  lcd.print("temp: ");

  lcd.print(temperatureC,1);

  lcd.print(" oC");

  lcd.setCursor(0,1);

  lcd.print("soil: ");

  lcd.print(SoilValue1);

if (debug == 1) {Serial.println("Done writing LCD");};

    // we write to the eeprom every 10 iterations

    logloop=logloop+1;

    if (debug == 1) {Serial.print("Testing for logloop ");};

    if (debug == 1) {Serial.println(logloop);};

      if (logloop > 9) {

        logdatatoeeprom();

        logloop = 0;

      };

  digitalWrite(ledPin, LOW);

  if (debug == 1) {Serial.println("LED off");};

  Alarm.delay(10000); // wait ten second between clock display

  if (debug == 1) {Serial.println("Delay set");};

}

 

void logdatatoeeprom(){

  digitalWrite(ledPin, HIGH);

  lcd.setCursor(15,1);

  lcd.print("E");

  if (debug == 1) {Serial.println("logdatatoeeprom started");};

  if (debug == 1) {Serial.print("writing to eeprom ");};

  if (debug == 1) {Serial.print(addr);};

  if (debug == 1) {Serial.print(" : ");};

  if (debug == 1) {Serial.println(temperatureC);};

  EEPROM.write(addr, temperatureC);

  if (debug == 1) {Serial.println("back from writing to eeprom");};

  digitalWrite(ledPin, LOW);

  digitalWrite(ledPin, HIGH);

  digitalWrite(ledPin, LOW);

  // advance to the next address.  there are 512 bytes in 

  // the EEPROM, so go back to 0 when we hit 512.

  addr = addr + 1;

  if (addr == 512) {

    addr = 0;

  };

  if (debug == 1) {Serial.println("end logdatatoeeprom");};

}

 

float readinput_temp() {

  // Read inputs and de-bounce, so read 3 times and take an average

  if (debug == 1) {Serial.println("Readinputs (Temperature) function started");};

  float temp1 = 0;

  float temp2 = 0;

  int soil1 = 0;

  int soil2 = 0;

  int lpcount=0;

  

  while (lpcount<6) {

    sensorValue1 = analogRead(sensorPin);

    float voltage = sensorValue1 * 5.0;

    voltage /= 1024.0;

    temp2 = (voltage - 0.5) * 100;

    temp1 += temp2;

    if (debug == 1) {Serial.print("Temp loop :");};

    if (debug == 1) {Serial.print(lpcount);};

    if (debug == 1) {Serial.print(" : ");};

    if (debug == 1) {Serial.println(temp2);};

    lpcount=lpcount+1;

  };

  temperatureC = temp1/6;

  if (debug == 1) {Serial.print("Temperature average: ");};

  if (debug == 1) {Serial.println(temperatureC);};

  if (debug == 1) {Serial.println("Exit Readinputs (Temperature)");};

return temperatureC;

};

 

int readinput_soil() {

  if (debug == 1) {Serial.println("Readinputs (Soil) function started");};

  // Switch power pin on for the soil sensor

  digitalWrite(soilPowerPin, HIGH);

  float temp1 = 0;

  float temp2 = 0;

  int soil1 = 0;

  int soil2 = 0;

  int lpcount=0;

 

  if (debug == 1) {Serial.println("Read soil moisture sensor");};

 

  lpcount=0;

  while (lpcount<4) {

    soil2 = analogRead(soilMoisturePin);

    soil2 = (soil2 / 2);

    soil1 += soil2;

    if (debug == 1) {Serial.print("Soil loop :");};

    if (debug == 1) {Serial.print(lpcount);};

    if (debug == 1) {Serial.print(" : ");};

    if (debug == 1) {Serial.println(soil2);};

    lpcount=lpcount+1;

  };

  SoilValue1 = soil1/4;

  if (debug == 1) {Serial.print("Soil average: ");};

  if (debug == 1) {Serial.println(SoilValue1);};

  digitalWrite(soilPowerPin, LOW);

  if (debug == 1) {Serial.println("Exit Readinputs (Soil)");};

return SoilValue1;

}

 

</pre>

 ----------------------------------------------------------------------------------

Update: Tuesday 15th May 2012

I'm now graphing the output as I've hooked the Arduino up to a laptop via USB, below are the graphs (updated every hour):

    

 

Saturday 5 May 2012

XBMC, TVHeadend, MythTV, streaming to your TV

 I thought it was about time I wrote a bit more about what I've been up to with our TV setup at home. Firstly a bit of background, take a look at the previous posts on my XBMC setup, but that is now a bit out of date as I've found this isn't a set and forget sort of setup, so I can see why a lot of people won't bother with this day-to-day. But I'm persistant and wanted to get my setup as good as possible so we've got the best of several options.

Firstly FreeSat with the superior HD channels, then Freeview (via Aerial) as this still has some channels not on FreeSat, then PVR functionality for pause/rewind of live tv, finally playback from the network media drive. In addition I've also got a want to stream iPlayer, ITV, 4od and 5 via their online stream facilities.

So originally I had TVHeadend running (via their git repository) with a custom patch for adapter priorities. That patch allows me to 'weight' which adapter I want to be used, so the rule was, use the Freesat adapters, unless the channel wasn't available (or signal was too low). This worked great, but there are problems with the integration to XBMC (The frontend). The biggest problem was that the PVR functionality of pause/rewind didn't work, and the author of TVHeadend/XBMC PVR plug-in really doesn't have any interest in developing this as he has no need for it. That causes me a bit of a headache, as that component lets the system down quite a bit. There are also problems with TVHeadend in that it relies on XMLTV to grab TV listings, as the EIT grabber (Over the air/satellite) barely works at all, and again nobody can see to improve that at the moment. So these two items let it down quite a bit as it means we get dodgy entries in our TV Guide, so its not reliable, and we can't pause. Another huge bug, and I believe this is an XBMC problem, is that if a channel doesn't have an EPG entry, or the entry is deemed as 'invalid' by XBMC it actually hides the channel. Now this causes channels to disappear and re-appear due to bad EPG data, this is something I've asked numerous times on the forums on how to correct myself in the code and re-compile but nobody seemed to know the definitive answer, or didn't want to help. This also frustrated me!

So I thought, time to move to MythTV using XBMC as the front-end. Myth does have pause/rewind functions, has EIT working fully and a few other nice things so this seemed a good option to switch to. Installing Myth-backend is a pain as although they have debian/ubuntu packages, you have to guess which ones you need to run the backend (mythtv-backend, mythtv-web and a few others). Once installed onto my headless server that has my 2 FreeSat and 1 Freeview cards in it I needed to set it up. Here was the first problem, its a GUI that you run within an Xserver (a desktop). This is a headless server so I can't do that, so have to run it over an X forwarder to my laptop, so its a bit laggy but works. The mythtv-setup is HORRIBLE! It's clunky, and looks like its been designed to be used from a TV remote, but even that it would be horrible to use, so I didn't have a good experience there. The logic of channels, channel inputs, hardware adapters and channels is also a little confusing. So you first setup your capture card, thats pretty straight forward, the system finds my 3, you have to setup Diseqc and tell it just to use an LNB (even for the Freeview) which is weird as it seems to be a requirement rather than an option. Once they're all setup you have to create a video source. This is straight forward as I want to use EIT, so select that, and thats done. Now the bit that I've still not fully got my head around, Input Connections. You set these to match your channels, so on mine I've created 2, one for FreeSat and one for Freeview (I have no idea if this is right, as it doesn't seem 100% correct!), so I went in, scanned for FreeSat DVB-S first (this takes about 10 minutes), then for DVB-S2 (another 10 minutes). Then create another Input Connection for Freeview, and scan for channels on there (About 10 minutes again). Wow that takes a long time! Now you can go into channel listings. This is mostly pointless, as the interface is so unusable you have about 200 entries in a list and it expects you to click into them one-by-one and disable the ones you don't want, correct incorrect text, etc. I would be on forever with that, so opted to leave that as it is for now and just have a stupid number of channels visible. At this point you can save and exit.

Now you need to compile cmyth with XBMC. cmyth is a fork of XBMC as the developer has re-written some parts of XBMC to handle the Myth setup correctly. So I grabbed his sources and started compiling, hitting many incompatability problems along the way as debian packages were missing, I was running an old version of Ubuntu, so had to upgrade the distribution first. That took longer than I hoped, and if you do a custom compile, don't forget to put your ./configure command options in to make all your add-ons work. My line ended up as './configure --prefix=/usr --enable-vdpau --enable-static --disable-shared --enable-sdl --enable-rtmp' Doing this, a make and install got us up and running and I could see the EPG loading. Going in however I had about 10 BBC1's, 10 BBC2, etc. Basically what it had done is found each copy of the channel on the FreeSat devices, Freeview channels and just put each one into the listing, it didn't merge any! So that meant back to the MythTV listing setup (Also each time you go into myth-setup you have to shutdown the daemon, which disconnects all clients stopping TV from working again!). I tried again but found the channel editor impossible. Luckily I spotted the MythWeb add-on, went into that and found you can edit the channels through the web interface, so I went through removing channels I didnt want. But what to do about the duplicate BBC1? A bit more reading revealed that you should give these the same channel number. So find all the BBC1's in your listing and set them to a channel number (e.g. I used 101 to mirror the Sky numbering system). I've done that and sure enough it looks much better.

The main problems I have now (and not solved):

  • EPG loading time in XBMC is very slow. It can be up to 5 minutes and its not a good idea so start watching TV till this is finished as if you do that it seems to corrupt a lot of EPG entries. This is a pain as we all have to sit watching a loading bar before we watch TV now!
  • After selecting a channel, it takes almost a minute before the channel engages and starts streaming
  • Pressing anything like EPG, pause, etc, also freezes the playback for a few seconds as the box 'thinks'
  • Channels without proper EPG still don't appear in the EPG line-up, so again this is a big problem!

So after all of these, I'm unsure if I should revert back to TVHeadend, as there are a few too many bugs and problems that are frustrating, and I'm not sure if the gains (Pause TV) is enough to warrant it. I may re-do all of it over the weekend back to TVHeadend and see if I can correct the few bugs remaining there myself!

 

Thursday 19 April 2012

FTTC Install - What's involved

 

 

Firstly, what is FTTC, Fibre To The Cabinet is a service that BT is rolling out throughout the UK to provide stable, fast broadband to home users without requiring additional equipment to the customer premises (* ignoring BT VDSL modems). The service to the client premises is still delivered over the standard copper pairs used for PSTN provision, in the same way ADSL and ADSL2+ are, and it uses the same basic technology of frequency shift to 'shift' the data bits into a higher frequency range. The separation of this at customer premises takes place within the BT master socket, which is replaced during the installation of FTTC by a BT engineer. At the other end of your line things change quite dramatically.

In ADSL/ADSL2+ the BT copper pairs go from your master socket to a junction box in the street (the street cab), this then can pass through other street cabs/junction boxes (as mine did) and then eventually ends back in a BT exchange on a Main Distribution Frame (MDF). From here your line goes into an ADSL line card which is part of the DSLAM and provides the mo-dem part (modulation, de-modulation). This then turns into a 'regular' ethernet-like connection and will be fed back to BT Wholesale via their fibre connections to the exchange.

In FTTC the street cabinet is upgraded and takes a fibre connection from the exchange. The street cabinet then has the fibre-to-copper hardware inside it and this then uses the copper pairs to your premises to transmit the broadband. This reduces the distance the copper pairs are transmitting the data signal, and therefore provides much higher quality and therefore potential speeds. It is this key change that provides the improvements to the broadband provision, stabilises the connection and generally offers a higher quality service. The downside is that BT have to 'blow' fibre to the street cabinets, and upgrade the cabinets to support the active (i.e. powered) equipment to do the work (which was previously in the exchange).

The actual install to the end-user is pretty painless (your mileage may vary!). In my instance my current ADSL2+ service dropped around 20-30 minutes into the engineer visit window that had been scheduled (BT engineer had phoned confirming his attendance at the alloted time), and shortly afterwards he turned up to change the faceplate and do his work. The install was pretty quick. Removal of the existing BT master socket. Installation of the new socket. Connection of test equipment the BT engineer could then verify the signal quality and the potential speed of the circuit. Then installation of the BT supplied VDSL modem, and finally checking that I could authenticate over PPPOE.

 

The new installed BT master socket with integrated VDSL socket:

 Top socket is the VDSL socket, and the one parallel to the screws is for your PSTN phone (Which I don't use)

This is a snap of the tester the BT engineer uses, it shows the line sync and status on there and gives you your first indication of what your speeds should be like, etc.

 

This is the BT Openreach modem that the BT engineer supplies and for now is required to be installed. This isn't a router or anything smart, its a very simple modem, the LAN1 port presents a PPPOE connection for you to handle authentication, etc, to your ISP. All other ports are blocked off and not in use on the device.

Note this is the newer BT VDSL modem that is now being supplied in most installations. Earlier models were common to overheating, and had also had their firmware cracked open so the average user could log into them and view line stats, tinker about, etc. BT therefore started to replace their shipped/installed modem (generally cited as due to the overheating fault). Also note that most installs BT will attach this modem to a wall to help with cooling/circulation. You generally should NOT run it sitting on a carpet like you see in my pic! But as you can see its not powered on in the pic. The unit does get warm pretty quickly so beware of this!

An attempt at a close-up of the sockets:

My phone didn't do a very good job of this picture. Basically the ports are: DSL in, LAN1, LAN2 (blocked off), RESET, BBU (blocked off).

 

You would then plug your router in supplied by your ISP into the LAN1 port and wait for it to auth up.

Since I like to tinker, I went a little further and have removed the BT modem and installed my own, this allows me to view line stats, and generally have a closer view of what is going on over the VDSL connection. I'm using the vigor 2750n which I've found to be a very capable little router. Setup is a little more in-depth than defaults so here are the settings you'll need if you're doing the same.

Firstly, firmware update. I'm running 1.5.2_Beta so you need that or newer (Older ones don't always sync up correctly and will sit with their DSL light flashing constantly).

Then go through the Quick Start Wizard, this steps you through the usual stuff. When it comes to WAN, DO NOT CHOOSE DHCP (This is a little confusing), you want to choose PPPOE, as it will then give you the chance to put in your login and password provided from your ISP.

You then need to login to the web interface again, choose WAN and Multi-VLAN. In here you need to enable Multi-VLAN setup and in the WAN VLAN ID box put in 101. This is a BT wholesale FTTC setting, so I can only comment on this install. You might need something different if you are with a LLU or other provider.

Save it all up and you're good to go. By default the Vigor will do NAT, DHCP and all the other usual stuff. In my case I get a /30 allocated and run a server behind my connection so I want to remove NAT and do routing. This is pretty straight forward on this router too, just go into LAN and General Setup, and in the "For IP Routing" enable that and enter in your public IP address allocated by your ISP along with your subnet mask. Disable DHCP (not needed) and save that. This will then setup your router to route! One thing to bear in mind though, the router will still answer on 192.168.1.1 on the LAN and will still do NAT for this range, you can tweak this and lock this down further in the setup but most Vigor docs recommend against this for some reason.

I hope this has helped you setup FTTC or at least given a bit more info. If you've got any questions feel free to comment/drop me an email.

 

Wednesday 15 February 2012

Breakinguard - How old is it now!

  As some of you may know, a long time ago I had a need to block persistant password attackers from various servers I managed (Working for a relatively large domain hosting company at the time), and because iptables didn't exist, ipchains wasn't everywhere I wondered what to do. I was learning Perl primarily through automating jobs at my job, and so I decided to write a little script that would 'watch' certain logfiles for attempts at gaining entry via repeated password or dictionary attacks. If the attack tried several attempts within a certain time-period, and from the same source IP, then flag it up, alert and then block using one of a few techniques. Not fool-proof, not perfect but it did the job.

Perl had a handy module called File::Tail that does exactly the same (sort of) as doing a tail -f /var/log/blah in the linux shell, so just run that in a script, watch for certain patterns and act accordingly. At the time the blocking action was either add an ipchains block rule, or add a route entry pushing the offender to the loopback interface. OK so this wasn't going to stop them DOS'ing us, etc, but it kinda did the job, and at the time internet connections weren't huge, if I remember rightly we had a 4Mb line or something like that, that was for a large domain hosting company, none of this 50Mb to the home, fibre or gigabit connections! So it did the job nicely.

Well, I started to install it on more and more servers that I looked after, or helped to look after, people asked about the script and slowly I started to pass it around, just by email at first. So I then decided to package it up and distribute it. My first bit of 'freeware'/GPL software was born! The software existed quite some time before it, but the official publication date was 5/5/2005 as that was when the package was created, however the release date that you'll see most other places was 27/7/2005 as that was the date that I released it on sourceforge.net

The original code is still up there, but don't look! It's horrible perl code, badly written, but it does the job!

Just looking on sourceforge.net at the site http://breakinguard.sourceforge.net/ you'll also see very poor HTML design (something I can still never shake off, my design eye is terrible!) The release up there now is dated 08/01/2008 version 0.11 and hasn't changed a great deal since the original to be honest, a few filters and a few nicer tweaks and thats about it! I belive it is still used in some places, but to be honest its had its day.

Now, a lot of people would use simple iptables, as you can do the same natively such as:

</p>
<p>iptables -A INPUT -p tcp --dport 22 -i eth0 -m state --state NEW -m recent --set</p>
<p>iptables -A INPUT -p tcp --dport 22 -i eth0 -m state --state NEW -m recent --update --seconds 60 --hitcount 4 -j DROP</p>
<p>

Which does the same thing really, so I think my software is dead or dying! If you do still use it, do pop a comment on or a mention as I'd appreciate it. Unfortunately I doubt I'll be doing any further updates to it, but will keep it live just for historic reasons rather than anything else!

 

Wednesday 1 February 2012

Central Heating twiddles

 So recently I've been noticing that the heating in the house is on what seems like a lot of time, the radiators aren't always as hot as I'd expect and generally the house is a bit chilly. Now this is during the coldest time of year where I live in the UK, so temperatures outside are dropping to 1 or 2oC overnight and during the mornings not getting far beyond that. With everyone in the house out during the day I'm unsure what it goes to during this time, but on evenings/overnight is when we notice it. So I started to watch how the heating was setup and then started tinkering!

We use a gas-fired central boiler (not combi), that has timed on/off schedules for heating and hot water. The hot water/heating switch is done by changing the valve setting from the pump circulation (i.e. when hot water is set to on it will divert some/all of the heating hot water circulation into the hot water tank). So my first observation is that when heating and hot water are on at the same time then surely this will halve the efficiency of doing one or the other jobs quickly.

Radiators in each of the rooms, the majority have TRV (Thermostatic Radiator Valve)'s fitted on them, and have a 0-6 scale for setting the desired temperature. Most of these are set to maximum all the time. Only two radiators do not have TRV's, these are both in the living room/dining room (It's one large area, no partition walls, etc). This is also where the temperature room thermostat is located.

There is a single room thermostat in the dining room, this is a digital/electronic temperature sensor running off two AAA batteries (annoying when they run out), that switches the heating circulation on and off depending on the ambient temperature.

The other little niggle is that the house is quite open-plan. The downstairs is a large living room/dining room, open to the staircase, and obviously up to the upstairs landing. Therefore the majority of heat (in my limited knowledge) will escape straight upstairs to the top of the stairs! Not ideal.

So, my first steps were to change the heating programmer, to get different on/off times for the heating and hot water cycles, and to prioritise heating by only switching hot water on after I've giving the heating a chance to warm the place up. So I opted for 3 on/off sequences during the day. First one for when we get up on a morning, on 5am-8am heating only. Then for when we get home, 3am-9pm. Then one to keep the chill off at 10pm-11pm. I went for hot water going on once daily from 4pm-8pm. So that seems to keep the heating happily running and not letting the house get too cold, as I think that was an initial problem, the house dipping way too cold, so taking the heating too long to warm back up.

The room thermostat was then set to 20.5oC as that seemed a reasonable temperature that is warm but not boiling. This I think might not be helping things, as I think this might need to be higher, but more on that later.

I then attacked the radiators and the TRV's, as I know I had one of these fail in the kitchen and jam closed, so that TRV was removed, so that radiator is just on full all the time. A few of the radiators needed bleeding, so did that for a few days to make sure we didn't have any problems with air in the system. I then turned ALL TRV's down so they weren't sitting on maximum. This was because a few websites I had read stated that running a TRV at maximum is bad for the valve and heating system (can't find the quote to cite now, if you know please comment/post back!). Then went round each of the rooms and changed the TRV depending on if the room should be hot/warm/cold, the majority of bedrooms I set on 5 (one off maximum), bathrooms a little below that and thought I'd leave it at that.

Unfortunately the heating still wasn't really doing what I wanted, the children were complaining of a cold bedroom during the night/morning, our master bedroom got red-hot at night and early mornings, and the rest of the house was warm! So that didn't quite work it seems. My next discovery were the next generation of programmable TRV's. These you can set times/dates and temperatures and the unit will vary that individual radiator based on this. I thought this would be ideal for the master bedroom, as during the day/early evening it was pointlessly heating the room. So I purchased a programmable TRV from a heating/electronics firm in the UK (I'm not going to name them here, as I found their service SHOCKING, it took 2 devices and me piecing the working parts from the two together to get a single working TRV. Not impressed!).

I ordered "Radio energy-saver regulator" which is a german-manufactured device that simply plugs into your standard TRV fitting, so no draining the system, etc, just unscrew the old TRV top off and this on. You program it either via its LCD screen, or you can also use the USB computer programmer, which I chose. You then set what time-periods to use, temperatures and set it going. This has made a huge difference to the master bedroom, although some further tweaks to the schedule are needed as at the moment its like living in the antarctic! But that could also be due to the overall heating turning off too soon. Obviously the TRV will only heat the radiator up when three conditions are now met. The main heating timer is on, the downstairs thermostat is set to heat and the programmable TRV decides that its schedule and room temperature say it needs to heat. Unfortunately it has no control over the first two conditions, so therefore if the sequences aren't matched, then this fails, and I suspect this is where I'm going wrong at the moment. I'm setting the TRV schedule to warm the room before we retire to the bedroom, but at this time the rest of the house is warm enough, so the thermostat on the heating has cut out, so although our TRV wants to heat the room, it can't as the heating isn't firing. So more tweaking is needed here!

 

And thats about where I'm up to. I need to do some more tweaking with programs and temperatures to get things right, but I'm happy that I have much more control and I believe I can see where the problems are on the system now!