EmonTx v4 is arduino or custom board?

hi all, I’m new and just making my way around the parts of this project.
Is the EmonTx v4 (GitHub - openenergymonitor/emontx4: emontx4) a custom board that cannot be purchased on its own at the shop? It has to be purchased as full emontx unit?

I’m interested in using an arduino, so then I would need the shield (emonTx Arduino Shield SMT - Shop | OpenEnergyMonitor) and an ordinary arduino uno right?
The shield looks to be emontx v2, so there is no arduino v4 I take it? What would be the difference between v2 and v4 then?

If I bought the shield above and used my at home arduino uno and bought a 30A CT sensor, would that work as a power monitoring station (apart from the rasp pi side to receive the data of course), or would I be missing something still? What arduino program would I flash onto my uno in this case?

I want to try a diy unit because the ready made emontx is not in my budget. In Africa the currency conversion from pounds + shipping and taxes makes it out of range.

Thanks. Any info or anyone who has bought the CT sensor and shield and gotten it to work accurately, I would appreciate the advice.

Welcome, @h23934 to the OEM forum.

Yes, that will be correct, when the emonTx V4 becomes available.

Yes, that is correct.

All the emonTx range upto but excluding the emonTx V4 (which is not yet available) use the Atmel ATMega 328P processor, as does the Arduino Uno. The present emonTx Shield uses mainly surface mount components, much the same as the emonTx V3. The emonTx V2 and the original emonTx Shield 1 used through-hole components throughout and was supplied as a kit to assemble and solder yourself.
If you use the SCT-013-030 c.t, it has the burden resistor built in and has a 1 V output; and so you will need to remove the burden resistor already soldered onto the Shield p.c.b. The external burden (on the p.c.b.) is needed for our standard SCT-013-000 c.t, which has a 50 mA output, not for a c.t. with a voltage output.

Measuring current only, what you are missing is accuracy. You will only get a “good guess” for the amount of power being consumed. For best accuracy, you also need to monitor the voltage. Not only does monitoring the voltage improve the accuracy because voltage varies according to the load on the system and what your neighbours are using, it also allows the software to accurately calculate the “real power” (which is what you pay for) and which, for some things, is very much less than you think from knowing the current alone. Read the theory in the ‘Learn’ section here.

Getting the best accuracy is as much about careful calibration as anything else. All the components we use, from the c.t. and a.c. adapter down to individual resistors on the circuit board have a manufacturing tolerance: the value can vary from the nominal by a few percent. Calibration removes the effect of that variation.

You would use one of the sketches for the emonTx Shield, from Github.

Thanks for the info, appreciated.

Ah okay. I take it this is it: emonTx Arduino Shield SMT - Shop | OpenEnergyMonitor and the connectors I will self solder. The resistors etc are already soldered on which helps a lot

Noted thanks!

Thanks, I will get a ac-ac adapter to monitor voltage too then. I can buy the uk plug and use a international plug to just plug it in, should work? We have same 220v as uk here.

Great, something like this right: emontx-shield/Shield_CT1234_Voltage.ino at master · openenergymonitor/emontx-shield · GitHub
So in theory I get the shield, solder the connectors on, flash the above to my arduino and fit on the shield and it should work? (having connected CT sensor noting burden resistor removal if required, and using ac-ac to get voltage measurement)

No, The Shield SMT (Surface Mount) Is the Shield 2.5. This is the original through-hole version, no longer available:

Yes that will give you real power on 4 channels on a single phase. You’ll need to change the current calibration constant(s) (to 30.0) because of the different c.t.

Just look carefully at the photos first! Some (all?) go on the “wrong” side of the board.

If you want to use a wired serial data connection to a Raspberry Pi on which you run emonCMS, you can cheapen it further by not having the RFM module for the Shield, then you don’t need a RFM69Pi on your RPi either. But it will cost running a cable between the two.

1 Like

Thanks. So my shopping list is going to be:

I will need rf because I need to have emontx and raspi talking wireless, I cannot run wire between them.

Thanks for this, I will do so if I end up going with the 30A sensor, most likely will be the case.

Noted I will triple check before starting the soldering, thanks for the warning

Thanks again for your help and time, I appreciate it

1 Like

That looks to be correct. Don’t forget power supplies - the a.c. adapter does NOT power the Shield, so you still need power for the Arduino and (obviously) the RPi. It’s said that the 7 V power input (Is it, I’m not an Arduino expert) and the on-board regulator is cleaner than using the 5 V USB, so you get better more accurate low power readings. (This as another argument for using the 30 A c.t. if your current draw doesn’t exceed 30 A.)

1 Like

Ah yes, thanks.

No idea to be honest, I’m also no arduino expert. I will probably use the 7v power input anyway, but noted too thanks.

It is for a water geyser in the roof. Geyser is connected to a 20A circuit breaker on mainboard so 30A sensor sounds perfect. I want to start out monitoring its power use, then later connect it to a relay to turn it on/off via app. Much later I want to even see if there is a way to monitor water level and have relay connected water inlet to eventually make it heat up the minimum amount of water needed for given time of day. Maybe heat up whole geyser before shower time, but rest of the day heat up half or even less water since whole thing not required

That’s the general consensus AIUI.
The Arduino board uses a linear regulator vice a switching-mode device.

1 Like