Archived Forum

I agree. The cost of the emon TH doesn't stack up against the ESP8266 with built in WiFi and as many DS1820s connected as you need or humidity and temperature sensors.

There's plenty of folks out there with this up and running with MQTT. Not open source as you rightly say but one hell of a lot cheaper.

I'm guessing that another reason that theres less chatter about the EmonTH is that the initial purpose of the site was to monitor electricity hence the title of the site and what has historically brought the majority of users here

I'm not aware of any off-the-shelf wireless thermometer/humidity sensor that uses the same radio as the OEM system.  While you can get wireless temperature sensors cheaper, you'd then need another base radio system to decode the signals, which would add to the cost.  If you are aware of such a beast perhaps you could add a link?

While the ESP8266 looks exciting it is a very immature technology at the moment. For example it is less than 2 weeks since Tuan PM’s MQTT library would not reconnect if the wifi disconnected (see http://tech.scargill.net/esp8266-mqtt-and-loss-of-wifi/).  It is also currently (no pun intended) far more power hungry.  Hopefully it will become more accessible by less hard-core programmers / "hackers" in future.  But given the issues that are often raised it isn't friendly enough for a significant percentage of people happily running Openenergy based systems today.

The good news is that this being an opensource project theres nothing to stop you designing a PCB based on the RFM12/RFM69 radio or designing a board around the ESP8266, getting it manufactured, finding a case to put it in, and doing it yourself. 

To me temperature monitoring and electricity or power consumption in total are closely tied together in most situation, and particularly so in domestic situations.

Many off the shelf wireless thermometer/humidity sensors operate on 433MHz, so it should be a matter of finding the correct protocol and then decoding the data stream. It is highly likely that most of these sensors use the same or very similar protocol and data stream.

While the ESP8266 looks good, as you say power consumption is the killer for this application. I have seen a solution where a power timer/controller was used, however by the time this was done, it was no longer a low cost solution.

FYI someone did reverse engineer the protocol of some off the shelf wireless thermometer/humidity sensors operating on 433MHz.  

See http://rayshobby.net/reverse-engineer-wireless-temperature-humidity-rain-sensors-part-1/

The biggest issue with 433Mhz devices are that they are almost all amplitude modulated signals a bit like Morse code. Not an issue in itself of course but an issue when you want to integrate it into a system which uses the Hope chip set which uses frequency modulation to my understanding.

If you search on the jeelabs site there is a hack to make the rfm12b work in some way.

There's also a good hack similar to the one above but using a protocol analyser -> https://tickett.wordpress.com/2012/06/27/more-433mhz-rf-hacking/

Also if you know the chip set in the device you are trying to emulate the manufacturer may have some software, e.g. the packet sniffer from TI I think.

Having said all of that you will end up with 2 parallel systems, the emon system using the rfm12b and another for all the other devices using an OOK style amplitude modulated signalling which would have to run on a Pi or another arduino.

Also while the 8266 devices use more power its still not a lot. Personally I'm not too worried about having these controllers running at an absolute minimum level of power consumption. I will be building one system to measure various temperatures on my heat bank and controlling the boiler and the CH pumps. So this will be taking over from my standard CH controller and a system of thermostats on the heat bank.

I guess if you want lots of remote temperature and humidity readings then the 8266 may be too power hungry.

Having said all of that you will end up with 2 parallel systems, the emon system using the rfm12b and another for all the other devices using an OOK style amplitude modulated signalling which would have to run on a Pi or another arduino.

This need not be the case. From my understanding with the introduction of the RFM69Pi, which should be able to support both systems. I am not sure if it will work if you already have the RFM12b on the emonTX though.

Hi Craig. I'd be interested to understand the spec but I can't see it supporting both at the same time, so in order not to miss transmissions from either an emonTX or emonTH and any temperature sensors using OOK, you'd have to make a choice.

Also, there's been another thread active on RF in the last couple of days where Paul pointed out that the soon to be released emonPi has provision for an OOK transmitter on the board alongside the rfm transceiver. But this is only for transmit, you'd need to rig up a receiver to read from off the shelf temperature monitors and get the code integrated.

Simon

Hi,

Just seen one of these: http://shop.energyhive.com/products/multi-sensor - looks interesting. It is intended to work with the efergy / energyhive hub. Have also found some info here: http://aussiehomeenergy.com.au/EnergyhiveMultiSensor.pdf - photo suggests an RFM12 or similar on board?

The above info indicates that it actually detects vibration, not motion (as in PIR movement detectors) as well as temperature and light levels, and that it runs off a CR2032 coin cell, which only lasts about 3 months - but it could be powered by 2 AAs in a battery holder, à la emonTH.

I may order one (£14.99) and have a play.

Mike

Energyhive was developed by Hildebrand and is anything but open source.  Some of the work was funded by an EU funded project called DEHEMS and some of the reports pay lip service to open-source but the only open-source files I can find are trivial rubbish.  The only semi-technical description I can find is http://luci.ics.uci.edu/websiteContent/weAreLuci/biographies/faculty/djp3/LocalCopy/05678451.pdf which mentions that the data collection used a 433MHz radio modem and that the base station ran OpenWRT.  As to the radio modulation etc., I found no info.

I'm hoping to get one, and see if I can link it to emonhub / emoncms.