Archived Forum

Google is your friend

http://community.pachube.com/node/65

http://www.cerebralmeltdown.com/programming_projects/Arduino_Kilowatt_Meter/index.html

and 48 300 more results

Yes I have heard of google, I'm looking more for  information relevant to this project and experiences that anyone may have in implementing it with an emonTx

Don't rule out CT's their the best option even if you have to open up a panel (carefully!). Obviously it would be entirely at your own risk, consult an electrician if you are in doubt.

 I'm pretty sure that it would be illegal to have a 'device'  behind the board wires coming out to the front (despite no electrical connection being made). I doubt I'd find an electrician who would sign off on it.

Electricity suppliers are rather particular about what they allow in the meter box, despite it belonging to the property owner!

the cables from the meter still go inside your house, I say track them and clamp on a CT from the inside, not at the or around the meter to avoid unnecessary explanations. This is exactly what I have done, I placed my CT inside the breaker box, it is inside my house

Breaker box and meter box are one and the same. Not sure how practical that would be to monitor after the meters as I'd have a lot more lines to monitor.

From one meter, it will split out to (I think) 3 power circuits and 1 lighting circuit plus a circuit for the stove/oven. From the second meter I would only have the off peak (hot water) but from the third, theres at least three circuits for storage (space) heaters and a couple of heat pumps. Some cables run through the roof, others under the floor. Messy.

In my house the meter box is outside, the "fuses box" or breaker box how i call it is inside, that may have caused the misunderstanding

Actually I do have a small sub-board inside with the light and (most) power CBs. Hot water and heating CBs are in the main box outside.

The main reason I want to read from the meter directly is that each meter is on a different tariff which makes cost calculation easier.

Gee, I guess I wouldn't know anyone who has mucked about with this in the past...

Thanks for the link, every bit of information is useful.

I'd really like to get my hands on an old electricity meter so I could do some testing of the best way to mount sensors so they can work reliable through the glass window of the meter. Shouldn't be too hard seeing there's a trend to replace the analogue ones with smart meters.

Check Ebay. When the utility replaced my analog, disc-type meter with a flash Itron Centron (thereby ruining my default solar net metering system), I bought an identical electronic meter at auction for $19, shredded it up into little pieces so I'd know how it worked and performed other diabolical experiments on it's innards. Eventually, it expired during a moment of experimental inattention, and I went back and bought a case of four more so I could continue.

No sign of a local meter on eBay. The energy authority that owns them, probably scraps them.

In my case, three meters, I'd thought of making a 'special' emontx that just monitored those three inputs and nothing else (and reporting them obviously :-) )

I'm going to try and get and old (broken) meter from the suppliers so I can work with it on the bench. I only need it  to be able the turn the disk,

My meter is similar. I will try and take some pictures of the meter and the gadget to keep the reflective object sensor in place to give you some idea of my setup. If I remember correct the code I used was based on the link the code tries to detect the transition from dark to light by using a comparing window of samples. The one thing I have noticed in my setup is that it is easy to miss one revolution if the adunio is busy sending data to the server and the dark to light transition happens. Thus the need to calibrate things.

I see your meter has 266 and 2/3 revolutions per kWh. I have changed my setup allows for entering the meter reading as per the meter. The server then works out a factor based on the meter reading and the virtual reading and applies this to calculate the kWh virtual reading. It has proven to be over and under by approximately 5 kWh per day in my case. Some days it reports the same value as the real meter and other days it just looses it totally. On average over a month it pretty accurate and works for my application. 

I've made some progress in that it seems an electrician can and will install current transformers between the meter and the load - for a price of course. This will be more accurate and allow more frequent updates than waiting for a disk to rotate.

I've ordered some CTs from Seeed, I just hope they fit behind the switchboard panel.

Next problem will be getting a signal from the emontx to the emonbase (when I get them ordered). The meter box that has to contain the CTs and emontx is metal!

The CTs that I've seen described simply clip over the incoming cable so there's no need to disconnect the original wiring.  From this thread, I get the impression that a CT may work better if the 1000+ turns are wrapped directly around the incoming cable, rather than around a 1-turn intermediary conductor (like 2 links of a chain rather than 3). 

If an Arduino can get the necessary input from the '3-link' split-core CT approach, why go to the trouble/risk of interfering with the original wiring?

Could a simple/cheap '3-link' CT be made from a screw-type carribena (or a beefy length of copper with the ends firmly pressed or soldered together) and a simple bobbin of fine copper wire?

Robin

If you're thinking of building your own CT's, you need a highly permeable ferrous core, copper won't work and mild steel will work only poorly. Most clamp-on CT's will use laminated cores of high silicon steel for highest efficiency.

There are some posts on this site from a user who was building their own CT's, winding the coils and such. Search around a bit and see if you can uncover that information if this interests you. I think his designs were through-the-coil type, but might be adaptable to split-core type modification.

Also see:
http://en.wikipedia.org/wiki/Current_transformer
http://en.wikipedia.org/wiki/Transformer

Thanks for that advice.  As I only intend to measure current at one place, I will no doubt go for a proper CT from somewhere.  I do recall the term 'ferrite core' from long ago, but hadn't really thought through what it meant.  Sorry, early days ...

In my case, the problem in connecting  of the CT is a legislative one rather than technical.

Unless you're a licensed electrician, it's illegal to perform any work  on a fixed electrical installation i.e. house wiring.

Hefty fines and, if someone if seriously injured or killed,  a jail sentence.

Geoffs, I appreciate that the laws are different in Aussie to the UK, but I'd be surprised if it was illegal to clamp on a CT to a power cable. (even if it's behind the meter). 

There would be no physical contact with the installation, you are just measuring induced current. 

If it were illegal, how can effergy and others market CT connected power monitors in Australia?

Paul

Paul,

Unless you have an older installation, all wiring in meter boxes is located behind the board on which the meters, circuit breakers are mounted. To access these requires that the board is opened (it's usually hinged) which can only be done by a licensed electrician. If a meter reader or some other inspection found CT's connected, they would want to know who did it and if not by a suitable authorized person, then they MAY choose to disconnect your supply even if the CTs are connected between the meters and the load.

In Australia, it's not illegal to market monitoring devices such as those sold by Effergy, there's just no requirement on the sellers part to ensure that they are correctly installed only that the devices have the relevant certifications for emissions and insulation. As these devices are either battery powered or from a plug in power supply, getting these certifications is easy.

It's a stupid situation but one we're stuck with. I'm sure there are lots of people here who have installed there own CTs without having any issues at all but considering the possible consequences, I'm not willing to risk it.

There are no statistics to prove that DIY wiring is inherently dangerous but unlike New Zealand which has identical electrical wiring rules (AS/NZ 3000:2007), we're not allowed to do DIY wiring where as a New Zealander can. BTW, this incudes fitting a 3 pin plug to an appliance.

hello, I am certain of such topic and it was part of our current discussion in electrician course. Tapping a connection in a CT is illegal and is considered prohibited to users. It can cause several damage to the panel that would result to destruction. The best way to fix it is to consult an electrician expert. I was actually enrolled in an electrician school in Texas, currently on my second year in the course.

Cecilia,

There might be local rules where you live that prohibit any interaction with the installation, but in general clipping a split-core c.t. around a - in strict terminology - low voltage cable can by itself neither damage nor destroy a panel, and unless damage is caused to the insulation or elsewhere in the process, any effect on the installation will be immeasurably small. However, if the c.t. not short-circuited, connected to its burden or otherwise protected and the conductor is carrying current, then dangerously high voltages may appear on the secondary winding and there is a risk of electric shock to the operator and damage to the c.t.