Hello,
I know the instructions say not to do it. But that will simplify my installation.
I have two different PV, with 2 different MPTT, I wan’t the sum. So I have two negative cable which I’d like to have the sum. Is if I put the CT around these two cables I’ll have the sum ?
Thank’s
Sorry for my english, i’mf frensh… 
Are your two PVs on the same phase, or on different phases? If both are on the same phase, then yes you can do that, and the CT will measure the sum (or the difference if the currents flow in opposite directions). If the PVs are on different phases, then you will measure the vector sum of the currents, which will not give you the answer you want.
I don’t think they say that. They say, for a normal installation, don’t thread both line and neutral through the CT, but if you look on the “EmonTx - Use in North America” page, you’ll find diagrams that show exactly as you are suggesting.
(And your English is better than my French, so don’t worry.)
Hi Kepon
Do you mean the negative cables of the d.c. side of the inverters? If so you won’t be able to measure it with CTs, you’ll have to use hall effect sensors instead.
If you do mean the a.c. side of the inverters, then Robert’s advice is spot on.
A lot of people think L & N mean Live and Negative, but that’s wrong, it is Line and Neutral. (And both Line and Neutral are “live” conductors unless completely isolated.) I took “Negative” to be a mis-translation.
Thank you to all
To be clear I’ll do a drawing (for I am not sure it is ok with the history of DC, AC ect …)
I would like my measurer consumption electric (AC, 230V) & my production PV total (2x2PV : DC 48V, 8A), here is a diagram of my setup:
To simplify my life I would have liked to buy the “emonPi - Solar PV” with just 2 CT and therefore do like this :
It’s ok ?
I am sorry, but you cannot do that. As Stuart wrote, a current transformer will not work on d.c. (direct current), therefore you cannot use CTs anywhere on the d.c. side. The only place that you can use a CT is on the 230 V a.c. side of your inverter.
To measure the direct current, again as Stuart wrote, you must use a Hall effect transducer. The reason is the Hall effect device can sense a static magnetic field, but a transformer needs an alternating magnetic field to work.
If you search the old forum, I think you will find several threads about measuring on the d.c. side. Try these: http://openenergymonitor.org/emon/node/837, http://openenergymonitor.org/emon/node/10337, http://openenergymonitor.org/emon/node/1995 and there might be more.
Also look at http://uk.farnell.com/lem/hx-03-p-sp2/current-transducer-3a-pcb/dp/1617419?categoryId=700000005923 for measuring the PV output current.
And more bad news: You cannot use the LEM modules with an emonPi, because the emonPi can only connect to a CT. You will need to have also an Arduino, make your own analogue input circuit board using the Hall effect devices, and connect that using a serial cable connection to your Pi.
This could become a very complicated project.
Ok no worries I now understood
Thank you for these expications
as suggestion you could build a simple cable adaptor with a arduino amp sensor and a cheap arduinos to run them all. ie http://www.ebay.ca/itm/5A-Range-ACS712ELC-05B-Chip-Current-Sensor-ACS712-Module-for-Arduino-/201669183400?hash=item2ef46b87a8:g:dQcAAOSwiCRUgYha this one a 5 amp version but you should be able to find 10 or 20 amp version suitable for your panels … I am doing something similar to this but i have a PWM installed to control the output of my panels if need be. But i use the pwm to drive the amp sensor and a esp8266 which sends mqtt data to my orange pi via wireless connection, but you could use a suitable voltage reducer of some form instead of the PWM. ( also the esp822 i use has a battery connector and charge controller on them so i have option of using battery if I need 24 hr connection - esptoy)