I looking for some sort of IP solution so that I can look at some temperature and humidity data via a web page whether I'm at home or away.

What I'm interested in is a couple of sensors for indoor temperatures, probably 2 upstairs and 2 down. 1 outdoor temperature, 1 indoor and 1 outdoor RH and 3 or 4 temp sensors on the hot water tank.

Came across this unit after a quick Google but haven't priced up a system.

http://www.itwatchdogs.com/product-detail-minigoose_ii-8.html

Anyone got a better suggestion?

My CurrentCost energy monitor does temperature (just the one) but can be connect via a bridge to a router and then viewed remotely. Saves having a PC on. Other method is to remotely view your PC with some VPN software if it is logging.

Posted By: cbatjesmond
I've got 10x OregonScientific THGR810 (http://www.oregonscientific.co.uk/cat-Weather-sub-Sensors-prod-Remote-Sensor-for-WMR100---WMR200.html) Their signals are received by a http://www.rfxcom.com/receivers.htm#80002

Me too on the rfxcom, though mine talks to an ALIX1D running Meteohub (http:/www.meteohub.de). For 1-wire I also have a 1-wire USB plugged into the ALIX1D, and an OWFS based custom plug-in to feed that data to the Meteohub application.

Posted By: cbatjesmondUnfortunately the THGR810s only have 10 possible ID settings so I can't measure anything else with them

Yes, but...
... you can set more than 1 sensor to a given channel. Most of the time that is fine, but periodically (i.e. it's pretty much predictable, once you get used to it) a pair of sensors on the same channel will conflict and during that time 1 or both will be unavailable (exactly what happens seems to depend on relative signal strength). After a while (typically a couple of hours) normal service resumes, till the next time.
I don't do this with the 10 channel sensors (of which I have only 2) but I do do it with the old 3 channel sensors of which I have 6! One of those is a sensor I don't want to have drop out, so I allow it a channel to itself, while 2 of the others share another channel and the other 3 share the remaining channel.
This happens because the time between transmits for each sensor is fractionally different, so the moment of transmission for 2 sensors will mostly be different, but will occasionally clash, and that's when one or both signals can be lost. The less often it happens, the longer the period of signal loss.
Useful to know if you can put up with occasional signal loss.
See also:
http://forum.meteohub.de/viewtopic.php?f=4&t=2453
http://board.homeseer.com/showthread.php?t=130500

Edit
N.B. The above is no use for those using these sensors with standard OS receivers, but the RFXCOM equipment is smarter...
End Edit

Posted By: cbatjesmondI can't find 433MHz sensors suitable for strapping to water pipes.

If you can still find them (they are no longer produced, I think?) the THC 238 sensors are good - they have a temp probe on a 3m lead.
I think I recall reading that some people have taken the 10 channel TH sensor apart and extended the temp sensor off board? I've not tried that but I did successfully (solder) repair the lead on a THC238 (the wires are very fine!).

Thanks for the feedback folks.

I'm afraid I wouldn't consider myself competent at implementing an open source solution so will probably have to pay the price for something that is plug and play.

The minigoose system which I linked to above isn't extortionate if I don't go mad with sensors.

It's filed in the "nice to have" category for now, I'll will just have to see if the budget stretches.

Posted By: JSHarris</cite................These arrived in the post this afternoon (having been opened by HMR&C by the look of things, including the inner packages)

Importing electronic bits JS; Your name and details will be on a list somewhere by now,

I've just tested it with a longer cable and it's fine, if anything very slightly more accurate with 30 metres of cable than with just the standard 1 metre.

The sensor is just a thermistor, which is around 12 k ohms at room temperature and has a negative temperature coefficient. There seems to be provision on the circuit board for selecting deg F, if anyone was so inclined.

Here are some photos of this sub-£2 unit:

<blockquote><cite>Posted By: SteamyTea</cite>Can it be linked to a simple (cheap) logger?</blockquote>

Not very easily, AFAICS. The thing seems to poll the sensor with a burst of ~1 kHz oscillations about every ten seconds, rather than use a plain resistance measurement. I only found this out when I tried to use 20 metres of coax cable I happened to have handy as an extension lead - the thing started giving crazy readings (around 85 deg C for room temperature). I looked at the signal on the sensor wires with the 'scope and discovered the burst of oscillations, presumably the capacitance of the coax cable was enough to throw the thing off.

It's because it only senses every ten seconds or so that the response seems sluggish, too, not that this should really be a problem for the sort of monitoring we're likely to do. They are designed to be panel mounted and my guess is that they are intended to be built in to things like shop refrigerated cabinets. They are a simple push fit into a rectangular cut out.

The current consumption is quite low, around 5 uA most of the time with a burst of around 20 uA when it senses, so an average consumption of about 7 uA. The two batteries are in parallel, so the unit actually runs on 1.5V. The LR44 cells have a capacity of about 100 mAh each (if they are decent ones), so two should run this unit continuously for at least 3 years. I've no doubt the cells it comes with are probably rubbish ones though, so may not do this well. There doesn't seem much point in fiddling around to run the thing from a bigger battery with this sort of battery life.

Overall I think these things are hard to beat for the price, if all you want to do is have a handy remote display of temperature. They're not super accurate, but good enough for keeping an eye on temperatures around the house or in a heating system.

<blockquote><cite>Posted By: owlman</cite>So JS was extending the cable a failure? What type of cable is fitted and do you think another type, other than your co-ax may work? Having half a dozen of these is cheap enough but only useful IMO if you can centralise them, or the readings.</blockquote>

No problem with extending the cable, as long as it's not coax.

<blockquote><cite>Posted By: JSHarris</cite>I've just tested it with a longer cable and it's fine, if anything very slightly more accurate with 30 metres of cable than with just the standard 1 metre.</blockquote>

That test was done with a couple of reels of ordinary thin hook-up wire (one for each conductor), so something like "figure 8" bell wire, or perhaps just a couple of conductors from a length of telephone extension cable, would probably work just fine. I have a roll of telephone cable somewhere, I'll dig it out and see if that works OK.

I've now powered up the rest of them and they all seem to display the same temperature within about 0.2 deg C with their sensors strapped together with a bit of aluminium foil, which seems pretty good to me, given their low price. It's a bit odd that they all ready very slightly high at low temperatures, but overall I don't think it's a significant issue. The sensors are pretty waterproof - I tested them by leaving them in a cup of water - so would probably tolerate being outside or in a fridge/freezer without a problem.

Five temperature sensors and digital displays, complete with batteries that will probably run them for a year or two, for less than £8 delivered has to be a bit of a bargain.

Two of them now have their sensors strapped to the flow and return pipes of my boiler, one has its sensor leading outside to show OAT. So far I'm happy with them, although they'd no doubt look neater if put in proper cases of some sort.

For what they are they seem pretty good. Now I need to get back to finishing the data logger and display...........

<blockquote><cite>Posted By: skyewright</cite>JS: Do you yet know what their supported temp range is?
With something vaguely similar I had a few years ago the display went to HHH at >50C.</blockquote>

I'll test it in the morning!

My calibrated thermometer only goes up to +50 deg C, so I've not taken them this high yet when testing. I can say that the one on the flow temp from my boiler has been over 50 deg C this evening, but only by a few degrees. The spec sheet (which I'd personally not trust much) says the range is -50 deg C to +110 deg C.

These: http://www.ebay.co.uk/itm/290431574781?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649 are the ones I bought that go up to +110 deg C (£1.58 delivered), but there are cheaper ones, like this one: http://www.ebay.co.uk/itm/Digital-LCD-Temperature-Tester-Thermometer-mit-Sensor-/290575175754?pt=AU_Gadgets&hash=item43a7a1504a that only goes to +70 deg C (but is only £1.43 delivered!).