Avoiding data clashes when monitoring multiple flats in the same block

Hi,

For a research project some of us at University College London are going to be monitoring some heat pumps in neighboring flats in a block. Trystan helped us develop a monitoring kit for this which includes an emonTH transmitting to an emonpi2 for each home. If the flats are very close together, what is the best way of making sure that each emonTH sends its data to the right logger? Is it to use different channels for each flat, and if so how many potential emonTH channels can we have on the emonpi? Thanks a lot!

Hi @JennyCrawley - if you have multiple THs transmitting to multiple Pi2s which may overlap, the easiest method would be to assign a unique ID to each TH. Without any firmware modification, you can select 4 IDs by adjusting the slide switches on the TH2 board:

Then, on each emonPi, you change the interfacer definition to match the ID for the TH it is associated with. The base ID is 23, and then you add the value of the two switches as a binary value: pushing the slide towards the ON position will add the number marked on the slide position. For example, sliding both to ON makes the ID 26.

Is that enough for you? If not, you’ll need to modify the TH firmware a little but it’s not difficult :slight_smile:

1 Like

To which I add, you can have 1023 possible nodes on 256 possible networks - but this doesn’t imply no problems! What it does is apply an unique code to each emonTH, so that all other messages can be screened out by its emonPi. ID=5 is used by the emonPi itself.

But another problem remains. All transmitters (emonTHs) work on the same radio channel, so it’s possible for two - or more - to transmit at the same time, and if that happens, each scrambles (jams) each others data and the receiver will reject it and the data is lost, for both. The solution, again not difficult to implement but it again requires a software change, is to reduce each transmitter’s power as far as possible consistent with the data getting through to its receiver, in the hope that when the signal reaches more distant emonPi’s, it will be attenuated enough by the building structure to not be detected.

2 Likes

Thank you so much, Angus! We are going to do what you suggested. We might need a few more channels so are going to get the firmware updated too.

1 Like

Thanks a lot, Robert! We think we’ll be OK regarding the scrambling but it’s something we will keep an eye on.