meteohub.de

Is is possible to move this thread from 'Whishes' and create a sub forum for the cc128 in it's own rigth or maybe 'energy monitors'?

I've now got my CC128 connected to my AlIX 1.D as well as my Vantage Pro. One interesting point of note is that my Vantage Pro was assigned device dev/tty/USB0 and thus I assigned dev/tty/USB1 to the CC128. When I rebooted, all went wrong. The AlIX1.D seemed to look for the CC128 on USB0 and the Vantage on USB1. I've now selected the devices the other way round on the Weather Stations page and rebooted a couple of timeas and all seems OK. I do not know if the assignments of dev/tty/USBx to physical USB ports is random at boot up or not.

What would be good next is if Boris could create a data upload to either to Pachube or MakeHistori unless there is something more suitable you guys could recommend I upload to?

Why would we want this?.... because like Wundereground, when the CF card fails or I lose all the data here at home, it's still there out in the websphere for me to go and get back... also easy to see what's happening at home (yes the girls are using hair straightners again) via the web and of course it removes the need to have a PC running all the time.

regards

Mike

gibsonmb wrote:What would be good next is if Boris could create a data upload to either to Pachube or MakeHistori unless there is something more suitable you guys could recommend I upload to?

May I suggest Meteoplug?

http://wiki.meteoplug.com/

Using that you effectively have an off-site version of all your Meteohub data... :cheer:

That is part of the answer. The reason I like Pachube like Wunderground is that the graphs are already available and you are essentially just loading data into them. I've never got to grips with the graphing side of Meteohub. I've just not got the time adn it seems really difficult. I'd laso then have to be able to create web pages and load the graphs into them. In essence become a web programmer. I like to see the web as a 'means to an end' whichis why I like programmes like WD live. In fact it woudl be great if WD live would accept CC128 data from Metoehub just like it does with the the Davis data

gibsonmb wrote:...I'd laso then have to be able to create web pages and load the graphs into them. In essence become a web programmer.

With the Meteoplug route you don't need to create a web page, you can just call a URL at www.meteoplug.com.

This topic on the Meteoplug Forum has some examples:
http://forum.meteoplug.com/viewtopic.ph ... =7&start=0

For anybody thinking about using the CC128 with Meteohub, it works really well, is easy to set up and with Meteohub as well you can create all sorts of interesting analysis. You can even then send data to Pachube to switch things on and off based on the data if you connect to Pachube with something like an Arduino

gibsonmb wrote:For anybody thinking about using the CC128 with Meteohub, it works really well, is easy to set up and with Meteohub as well you can create all sorts of interesting analysis.

I second all that. It's a great tool!

Just one word of caution. The CC128 uses assumptions about voltage and power factor.

The voltage thing is not a big deal, but the power factor can lead to some odd effects.

Longterm our CC128 readings on the domestic (i.e. non-heating) circuits over reports by 114% compared to the meter, varying a few % either way from daya to day. On the heating circuits it's 90%, and pretty consistent. I believe that difference is because the heating circuits will be almost entirely resitiive loads (which have a power factor of 100%, which will be above the CC128 assumption), but the domestic loads have varying power factors, mainly in our case below that of the CC128 assumption.

Our overnight background load seems to have a particularly low power factor. As a result the CC128 over reports that by around 40%. I spent quite a while trying to trace that "phantom load" before I reached the conclusion that it was a matter of power factor!

So great tool, especially for seeing the "shape" of your power usage but the figures need to be treated as the approximations they are.

This problem is also not specific to the CC128, but applys in general to the cheaper power monitors that employ simple "toroid clamps". For a clamp based solution that can handle power factor you need something like a Brultech, which is a much more expensive item...

N.B. There is a new Current Cost model on the way which will offer an alternative type of sensor that claims 100% accuracy (I suspect this might involve some thing like a device that count the 'blinks' of the LED on the meter?).

Interesting, I thought the night readings where high and spent ages trying to track down what might be causing it. I knew about the PF inaccuracies of cheap current clamps but never really thought about it. MY metering is of course much more screwed up as of last week as I've now got 3KW of PV on the roof and of course becuse the clamps are based simply on the size of the magentic field, it does not care and is not able to tell which way the current is flowing.

What I've not had time to work out yet is what the E.on meter (3PH) blinks do when the current through it goes in the opposite direction (when I'm generating more than I'm using). I could of course put the Arduino on the LED with a sensor and small logging programme as I have a logging shield for it... amazing what you can do with some bluetack!!

I've also got a PV generation meter which has a blinking LED as well which is where we get the measuremets for the FiTs tarrif.

gibsonmb wrote:MY metering is of course much more screwed up as of last week as I've now got 3KW of PV on the roof and of course becuse the clamps are based simply on the size of the magentic field, it does not care and is not able to tell which way the current is flowing.

You beat me on that one. Our PV install is due next week , so I'm anticipating the same problem.

Actually I'm awaiting a visit from a neighbour who happens to be an electrician to discuss the matter. The problem seems to be that with the new regulations the PV is expected to attach to a way in the CU (consumer unit), and that means that there is no where for the CC128 to see "all usage regardless of where the power came from". An idea that has been suggested to me is to add in an extra small CU between the main meter and the main CU. The small garage/shower style CU just has 2 MCB, one that acts as a feed to the main CU and the other that is where the PV attaches. That way the conductors between the small and main CU provide a point for measuring "all usage regardless of where the power came from". Considering to the cost of a PV array, a CU plus a couple of MCBs would not add much to the overall project cost. I just have to see now if the electrician thinks that the idea is a good one...

What I've not had time to work out yet is what the E.on meter (3PH) blinks do when the current through it goes in the opposite direction (when I'm generating more than I'm using).

At an install I visited, their LED that normally flashes RED, instead shows steady GREEN when exporting.

I could of course put the Arduino on the LED with a sensor and small logging programme as I have a logging shield for it... amazing what you can do with some bluetack!!

Have you come across the CurrentCost development board?

I've also got a PV generation meter which has a blinking LED as well which is where we get the measuremets for the FiTs tarrif.

I'm just hoping that the weather is decent when the people arrive to fit our array. Last Friday we had gusts of up to 50mph!

skyewright wrote:Actually I'm awaiting a visit from a neighbour who happens to be an electrician to discuss the matter.

We decided that yes the arrangement described was possible, but there were probably better ways to spend money...

I hope to get a 12mm clamp on a conductor somewhere on the supply from the inverter to allow me to monitor generation, and then later we'll see what the new CurrentCost model (the "Envi R") comes up with for accurate reading direct from the meter.

Let us know the results when you get the Envi R. I'm going to try and write a litle prog for the Arduino that reads the flashing LED (1000 flashes per KWH)on the generation meter

gibsonmb wrote:Let us know the results when you get the Envi R.

I'll see how much they want for it first!

I'm going to try and write a litle prog for the Arduino that reads the flashing LED (1000 flashes per KWH)on the generation meter

That sounds like a good idea.

gibsonmb wrote:Let us know the results when you get the Envi R. I'm going to try and write a litle prog for the Arduino that reads the flashing LED (1000 flashes per KWH)on the generation meter

It looks as though the release date for the Envi R has slipped. Last month one of the CurrentCost pages had mention of a September release, but that page has now reverted to "coming soon...", so I decided to expand my RFXCOM system by adding an RFXMeter plus some RFXPulse modules. That means I'm now monitoring our import (i.e. bought in) and generation (from PV) meters using the principle you mention, i.e. counting the LED flashes.

That means my main use for the CC128 is now for measuring the heating loads. It is well suited to that; they have a pretty consistent power factor so it's feasible to use a calibration factor to convert the CC128 readings to values that are close to the actual metered amount, and a LED based system would only give me an overall total figure whereas the multiple CC128 sensors can give me a breakdown for the different circuits.