I live on a farm in Australia. We have an odd supply system: 2 Phase at 180 degree. (Actually 480v Single phase, with Neutral centre tap)
I guess looking at the information I will need 2 of EmonTx, to detect both voltages. I would also like to monitor my ‘Off-Peak’ circuit that is separately metered, but on one phase (so I guess a second Current transformer on that supply.
I also have Solar on both phases. I guess a current transformer for both those to?
My setup is a Solar Edge 5Kw inverter on each phase with Optimisers, & fed with 14 of 295W panels on the North, & 12 of 295W panels facing West. So total about 15Kw of panels (A very modest system!) LImited to 8.4Kw export. It is fitted with Solar Edge metering, but the software is limited to 1/2hr updates, & will not separate the phases.
So, if I set up 2 EmonTx boards, I guess I will need to make my own housing? Then I will need to modify the 3 phase software to 2 phase?
The house is about 25M away from the Shed, which is where the main board & solar installation is situated. Can I just load the Emoncms on a Pi for the base station/Web server?
ps: Early Summer, so my solar produces 60-80 Kw/day. Currently considering batteries.
So the same as in the USA, but at double the voltage. So most of the material on the “Use in North America” of Learn page will apply to you.
The first question there is, how accurately are the two voltages balanced? If the the answer is “close enough”, then not necessarily. Otherwise yes. But it looks as if you’ll need two emonTx’s anyway for the number of currents you need to measure.
Probably - but if the two currents are balanced, or balanced ‘good enough’, then one should do.
You’re not really 2-phase. A proper description of a two phase supply has the phases separated by 90°, yours is a “single phase, split phase” supply.
If you are having two emonTx’s, then I’d treat the two legs as two separate systems at 240 V each, and use the standard single phase sketch.
You can put two p.c.b’s in one housing, or you can have two off-the-shelf emonTx’s.
Indeed. A RPi with radio receiver is called an emonBase. First check your WiFi coverage to the shed, if it’s OK then the emonBase can be in the shed. If it’s not, then the lower frequency RFM radio in the emonTx will probably reach to the house, unless there are thick walls in the way.
Thanks for the input. From my gleanings, it seems I will need 2 of EmonTx boards, a 915Mhz RF link (For Australian Compliance) Can the EmonTx be supplied with a 915Mhz Board & antenna?
I will need to use an extension antenna cable to run the antenna outside the tin shed (about 6 feet, with waterproof antenna socket & bracket, which I can make/find locally)
This will link to an Emonbase in the house, which is fitted with 915M wireless link & wireless network link to server/web.
Is there anything else I need? (except of course current transformers, & voltage monitor transformer (locally sourced, as per forum)
btw, good to hear that I can fit 2 of EmonTx boards in one standard housing. I will only need one 915mhz board & Antenna though?
I didn’t say “standard housing”, because the shop-supplied emonTx case will only accept one pcb. I was thinking you had in mind sourcing an enclosure locally.
Unless you do something clever with passing serial data between the two pcb’s, and you’d have to develop that yourself (undoubtedly with help available from the forum), then it would be two cables and two aerials, unless @Bill.Thomson thinks a combiner will be suitable and is available.
From what I can understand, I will need 2 of EmonTx boards, that will be stacked together with stackable headers (for MartinR multi phase setup, 2 phase only: my setup is similar to a US setup). I take it from MartinR setup, there is a modified image in one Tx board. (Martin was using an earlier board in his 3 phase setup, but it appeared to be similar to the current Tx board. Of course, I will only require 2 boards, but the software should be the same.)
Also, being in Australia, I require 915Mhz radio to be compliant. (I can source 915 module & Antenna if n/a from GB). I can perform hardware mods, being a retired Tech.
The obsolete emonTx V2 lent itself readily to that set-up, the V3.4 doesn’t have stackable headers. I’d need to look carefully at the circuit diagrams to see how practical it is to do the same with the present board.
(If you download the free Eagle CAD, and the schematics - there must be a link in Resources, you’ll see what I mean.)
Thanks. Been there earlier.
So, I guess with the new board, it is communicate via the Uarts (Which was said earlier)
My issue is that I am not a programmer. How difficult would it be to merge MartinR code with current?
I really need 2 of EmonTx because I need current sensors on 2 inverters, 2 active phases, & one ‘Off-Peak’ circuit. I will also be using a battery system on one phase (upgrade planned soon).
I guess the other way to handle this would be to use both radio links (already mentioned before) & then use the Base to fiddle the data. I could still fit both 'Tx’s in a single case, & the antenna would need to go outside the tin shed anyway, so no big deal there. All I would need to do is change one bd address.
Normally, a combiner, or duplexer (if I’m reading Robert’s comment correctly) is used to
enable a transmitter and receiver to use the same antenna.
The main issue in this case, is preventing the RF energy from getting wasted in the transmitter
that’s not transmitting. Since the two transmitters operate on the same frequency, it would require
some type of switching “mechanism.” Doable, but not simple or inexpensive.
A duplexer works because the transmitter and recever operate on slightly different frequencies.
It’s a cavity filter with two very narrow passbands which keep the transmitter’s output away from the receiver as well as keeping the received RF energy from being wasted in the transnsmitter. In doing so, a duplexer enables the use of one antenna for simultaneous transmiting and receiving.
So the ‘best’ way is combine the data at source and use one transmitter. I wasn’t sure how lossy it might be trying to combine the r.f. signals - and even if it was possible, there would still be the problem with collisions.
I need to look at the diagrams and software. I think it’s doable in the software if the one with the transmitter is ‘master’, and it commands the ‘slave’ when to do the measurement and when to hand over the result - roughly the same as MartinR did it.
That is how I originally envisaged it. You’d need two cables going out and two antennas. In emonCMS it is easy enough to add (combine) the data from the two sources, the difficulty would be that over time, they would lose synchronism and after a few missing values because each transmitter was jamming the other, you might end up adding 10 s old data to new present data. That might not be a problem - you’re best placed to decide that.
Yes, I thought about that too. But doing that isn’t much different (in terms of what the 2 sketches need to do) as combining the numbers, and the latter means that powers that should logically be added at source can be, and sent in the one radio packet.
so, I need to have a serial cable between the 2 (Programming) Uarts, & have the master poll the slave, then transit all the data from one Radio module on the ‘Master’
That is pretty much what RobinR did on the 3 phase system. It seems like a good project to achieve because there must be others in the same situation (like the US people for instance)
Not for my case, but how would someone who wanted to monitor the 3 phases separately link their boards? Would it be easier to use an Arduino shield, then link the Arduinos? Or could we use ‘stacked’ Arduino shields, with the shields hacked to work together?