EmonTX: Understanding inputs and outputs

I have the following config:

- emonTx V2

- Raspberry Pi 

- ready-to-go SD card image as OS

- clip-on current sensor CT

- AC-AC Power Supply Adapter - AC voltage sensor

- emonTx_CT123_Voltage firmware

--- emonTx_CT123_Voltage.ino

--- emontx_lib.ino

 

After playing with code and sacrificing several virgins to EmonTx gods it finally works!

Now I try to understand, what exactly works and why it works like this...

I have the following data in the log and try to interpret it.

 

10 - ID of node

18 - Watts used by CT1

0 - Watts used by CT2

17 - Watts used by CT3

(am I right?)

23931- Volts x 100

Looks logical and clear except that there is no power sensors installed and all CT1 - CT3 readings should be zero. To make it worse - CT1 and CT3 constantly changing.

 

OK, assume that device is not calibrated, but when I connect some (heavy, 200W) load to one of the inputs both 18 and 17 change to something like 22 or 27.. Or 9 and 18. No trend.

Additionally, the numbers constantly change, even without load.

So, what could be the problem here?

 

Robert Wall's picture

Re: EmonTX: Understanding inputs and outputs

The easiest place to pick up the numbers is via the serial monitor of the Arduino IDE when you have the programming lead plugged in to your emonTx. If you look at the sketch you're running on it, you should see two sections that you are interested in. The first is a data structure that tells you the variables sent to you RPi and the order in which they are sent. It should be something like:

typedef struct { int power1, power2, power3, Vrms; } PayloadTX;         // neat way of packaging data for RF comms
PayloadTX emontx;

The second is a series of "Serial.print(.....)" statements that send the data to the serial monitor.

Coming to the numbers varying, 200 W is not a heavy load. Your c.t is rated at 100 A which is roughly 24 kW, so 200 W is only 0.8% of full scale. The ADC is only 10 bits (and one of those is effectively a sign bit), and we expect  a few tens of "watts" of noise from the digital circuits in the processor. Here is some reading:

http://openenergymonitor.org/emon/buildingblocks/measurement-implication...
http://openenergymonitor.org/emon/emontx/accuracy
http://openenergymonitor.org/emon/node/1385
http://openenergymonitor.org/emon/node/2311
Trystan did some wonderful animated pages demonstrating how the errors vary here:  http://vis.openenergymonitor.org/dev/adc/noiseless, here: http://vis.openenergymonitor.org/dev/adc/noisy and here: http://vis.openenergymonitor.org/dev/adc/noise.mp4.

wbuilder's picture

Re: EmonTX: Understanding inputs and outputs

Thanks a lot. Pretty explanatory. (only "wonderful animated pages" produce 404)

 

Then, if we take that we have 512 (2^9) steps of 24kW, then 200W will take at least 4 of these steps...

 

And  how can I diagnose why there is no change in readings when I connect clamp to some ports?

And these are W, which are sent from EmonTX or some "Steps" of this 512/1024 scale?

Robert Wall's picture

Re: EmonTX: Understanding inputs and outputs

"Then, if we take that we have 512 (2^9) steps of 24kW, then 200W will take at least 4 of these steps..."

I think you are saying you can draw out the top half of the current wave as a set of 2-bit numbers. You should certainly see a 200 W load, but you cannot expect to measure it with any great accuracy. If you can't see a current, it is usually one of two things: you haven't pushed the plug on the c.t. fully home, or you've clipped the c.t. around both line and neutral.

Provided you have calibrated the emonTx, the numbers should be Watts for real power and 1/100 V for the voltage. You can always work through the sketch to verify that. If you have not calibrated it, the numbers will only be close.

(I get 404 for those diagrams too now - I'll ask Trystan what he's done with them.)

wbuilder's picture

Re: EmonTX: Understanding inputs and outputs

you've clipped the c.t. around both line and neutral

 

Sure, I did it! It is so logical, taking into account diameter of the clamp :) I think that it should be written in large red letters for those, like me, who think that electricity is sold in barrels on farmers' markets :)

 

I fixed this and now my readings are much closer to the reality.

 

Thanks a lot, you made my day! :)

Robert Wall's picture

Re: EmonTX: Understanding inputs and outputs

"It is so logical,..."

Not when you consider that the c.t. usually needs to go around a meter tail! Our USA constructors can't even use our standard 100 A c.t. because at half our voltage, they need twice the current for the same power, hence their cables are even thicker.

"I think that it should be written in large red letters for those, like me, who think that electricity is sold in barrels on farmers' markets :)"

I shall send the Marcolins to get you for heresy like that!  Actually, it is spelled out quite clearly in a few places - like the installation notes...

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.