Archived Forum

Are you using an emonTx or emonTx shield? The emonTx has a processor on-board and three c.t. inputs, the Shield does not have the processor and has 4 c.t. inputs. They work on different voltages so some components have different values and the calibration is different.

You say what the wrong numbers are, but what are the right numbers?

Are your three c.t's all the same?

If you change the c.t.'s around, is it the third c.t. or the third input circuit that is wrong?

Robert, could he be wrongly using the sketch for the new v3 emontx, ie the one with increased  sensitivity on CT3, without the correct burden resistor fitted? Just a thought.

Paul

It's an emonTX

If i place the 3 CTs on the same Wire, CT1 & CT2 will return for example 100 while CT3 will return 1200, 1127, 1143... very different.

CTs are all the same, baught with the emonTX from openenergymonitor.com
I tried swaping CTs, CT3 is still troubling so imo it's not a faulty sensor, perhaps the 3rd circuit.

But i looked carfully to the CT3 circuit, the 3 resistors and the capacitor... all checked with metter, solders are good for me.

Can you check again R16, 18 Ω. Unplug the c.t. then measure in-circuit: put one meter probe on Port 2, AIO1 and the other on the junction of R15, R16, R17 & C11  You should read 18 Ω.

Is the voltage at Port 2, AIO1 approximately 1.35 V d.c.?

Can you measure a very small a.c. voltage? If you can, you should measure approximately 10 mV across R16 per 1 A of current.

And the sketch you are using is https://github.com/openenergymonitor/emonTxFirmware/tree/master/emonTx_C... ?

(Paul: yes, it does sound like an o/c burden!)

R16 => 18 homs

Port2, AIO1 => R15,16,17 & C11 = 18 homs

Btw port2, AIO1 and GND, i read 2.2vdc

Yes, i'm using the right sketch.

Update: Metering Port2 Ai01, on C11, got 18 homs on 1 pin, nothing on the other.... normal ?

Metering C11 btw his pins, got 592 homs

CT1 and CT2 unplugged: Feed gives 0
CT3 unplugged: Feed gives 253

"Port2, AIO1 => R15,16,17 & C11 = 18 homs"  That is correct, so R16 is correct and soldered properly. That was my first thought.

"Btw port2, AIO1 and GND, i read 2.2vdc"  That is wrong. It should be V_CC/2 = 1.65 V, so it seems you have a problem there.

"Update: Metering Port2 Ai01, on C11, got 18 homs on 1 pin, nothing on the other.... normal ?
Metering C11 btw his pins, got 592 homs"  I am not clear what you are measuring there.

"CT3 unplugged: Feed gives 253"   We expect a small power from digital noise from the processor, but not that much. That is wrong.

Do you measure 3.3 V on the battery connections? 

Can you get and load calypso-rae's test sketch http://openenergymonitor.org/emon/sites/default/files/RawSamplesTool_4ss_2.ino_.zip

[Wrong sketch - see Calypso_rae's post below]

This should give you the voltage wave and the current wave from CT3, on the serial monitor. The voltage wave will be a straight line, and he current wave should look a little like http://openenergymonitor.org/emon/node/1705#comment-8535  - the txt file - (but you should have a full wave, not half !) - the important point is the current wave should lie in the middle. If the voltage you read (2.2 V) is correct, it won't be, and that is a fault. Post the result.

For the other c.t's, you must change one line

byte currentSensorPin = 1;

to either 3 (CT1) or 0 (CT2).

RW: Can you get and load calypso-rae's test sketch http://openenergymonitor.org/emon/sites/default/files/RawSamplesTool_4ss...      This should give you the voltage wave and the current wave from CT3, on the serial monitor. The voltage wave will be a straight line, ...

This tool will only run if AC voltage is present at input A2.  It "syncs up" to the voltage waveform, then records the voltage and current waveforms for three consecutive cycles of the mains voltage.  If no AC voltage is present, this sketch gives unpredictable and/or meaningless results.

My other standard test sketch should reveal whether all three CTs have similar sensitivities:

•  MinMaxAndRangeChecker.  An improved version of the above tool, this one records raw ADC values for the first four of the Atmega processor's analog inputs.  Posted on 20/04/13 at http://openenergymonitor.org/emon/node/2269, this tool is available at http://openenergymonitor.org/emon/sites/default/files/minMaxAndRangeChec...

I seem to have the same issue. Did you solve this one? How?

This thread is quite old, and I don't think Teo has been back since. I don't think it is likely he will see your question. Tell us about what you have and what you have measured, and why you think it is wrong.

My problem:

I was using CT1 and CT2 befor without problems.

Yesterday I've installed new solar panels and wanted to monitor them with CT3. However when I connect CT3 (and reset EmonTx) the input value in EmonCms is 19455, when I expect something between 0 and 400.

CT4 gives 3547, while nothing is connected (don't remember if it was 0 before).

I've tried changing the probe from CT1 and CT2 to CT3 (and the other way around), but same result, the probes are working.

CT1 and CT2 still seem to work correctly.

I haven't measured anything yet, I'll try this evening/weekend and put the Arduino monitor on the EmonTX. Hopefully this will shed some light on my issue.

There's little I can say until you come back with more data. Here are some points to check:

It is an emonTx v3 pre-assembled from the shop.
You are using the sketch that was already loaded (emonTxV3_RFM12B_DiscreteSampling.ino - prints: "emonTx V3 Discrete Sampling V1.2" at start-up).
Can you measure the value of the burden resistor. With the CT unplugged (first take it off its cable, unless under no load), measure between the socket sleeve and GND - you should read close to 22 Ω (CT1-3) or 120 Ω (CT4).
With the CT connected (but not on a cable), do you measure about 1.65 V d.c between the plug tip and GND (CT1-4).
Are you sure you are looking at the right number in emonCMS. By default, the V3 sends power1, power2, power3, power4, Vrms, temperature always and in that order, voltage is × 100 and temperature is × 10. When sending to Serial, only the inputs detected are sent, but in the same order.

Robert, thanks for the clear explanation.

I have a pre-assembeled emonTx v3 with pre loaded sketch.

I've taken it out of the switch board an all measurements were ok. Connected the arduino serial monitor and the only odd thin was that CT4 was always detected even though no CT was connected. When connecting CT3, it was correctly detected.

I've reïnstalled the EmonTX and everything works again! No idea what was wrong for two days.

Thanks anyway for all the help.

Regards

CT4 being detected points to a faulty socket. The "break" contact on the plug tip grounds the input, (instead of it being at 1.65 V) and that's detected at start-up. It's that contact that isn't working. It won't affect operation, except that the serial output will report the CT4 values when nothing is connected. If that causes a problem, you could solder a temporary (until you need CT4!) link from the plug tip contact to GND (or insert a spare plug with the tip contact brought out and wired into the GND terminal of the terminal block).