hi everyone, just asking for a bit of advice on the best way to wire up the “soon to arrive” DS18B20 sensors to my system.
i have 3x EmonTx v3.4’s in my Fuse box ( monitoring every circuit seperately) so i need to run a cable or 6 from the fuse box to the ceiling and hence to each room.
should i daisy chain 1 wire for all 6 sensors and just hook into the screw terminal of 1 of the EmonTx’s, or bring each individual sensor down with 3 into the screw terminals of each, and terminate the other 3 with RJ45’s?
or some variation on this?
the run for the individual would be approx 2 x 21m, 1 x 19m, 1 x 15m, 1 x 8m, 1 x 3m
the daisy chain would be approx 40m when counting down and back up walls from sensors.
if i ran all individuals and had 1 on a screw terminal and 1 on a rj45, would i need to modify the sketch of each Emontx, or would they just work?
Whether the emonTx would “just work” depends on the sketch that you’re running. I haven’t got enough sensors to test the latest CM sketch, I think that’s why it’s set to accept 3 only. The earlier ‘discrete sample’ sketch (which works in a completely different way, and if you’ve had your emonTx’s since before mid-October 2019 will be what you use) accepts 6 by default.
But one caveat: the power available to the emonTx from the a.c. adapter is limited by design, so you probably will need a 5 V d.c. USB power supply for that one.
Also note that the power to the sensors using the screw terminals is switched, for those using the RJ45, it isn’t.
thanks for the info, i cant recall what sketch i have( and not sure how to find out), I played with it many moons ago to fix the AC voltage calibration.
i would be on the old firmware as this was all purchased back in 2017, and i havent done anything since the initial calibration.
i power the 3 TX’s from the USB plug and just use the AC for voltage monitoring.
I don’t know what the ESP8266 passes on to emonCMS, but it might just be worth looking at your emonhub.conf. The data structure that’s sent has changed over time, with luck it might be identifiable.
If you do find you have those, and you have the old (not CM) sketch, you should be OK if you permanently power them from 3.3 V, rather than using the emonTx’s switched power (labelled DS18B20 PWR).
I don’t really feel qualified to advise on the best way to interface them to the emonTX kit. Most of my work has been on tuning the cpu side of things for maximum robustness. I’m happy to share where I ended up there, but it’s not directly applicable to interfacing them to the emonTx. I can report that 1wire is surprisingly robust once you start tweaking things a bit. I’m away from base at the moment - enjoying Qld’s recent lockdown relaxations, but I left that set up described in the other thread running. So far:
The two buses are headed towards a million conversions between them, and so far not a single soft error. Bus0 is the horror topology described in this post.
As Robert mentioned, if you post your UIDs when you get them, we might be able to tell you what to look out for.