I then pop’d in a cheepy offbrand maplins Gbit switch I had bought for my mums house 2 years ago that i now had spare. It worked a treat. So i have to dig up my soldering Iron and order up a batch of the capacitors to try this out and might have functional switches to use/sell on if I or someone else needs em down the line :)

Good on you Chris,, like you say, saving on devices going in to landfill. it’s worth saving them.

I’ve got the caps ordered, will report back.

What I’ve learned is that most of these devices have failed due to voltage surges sometimes, amazingly, created by thunder storms. I suspect the one you repaired has suffered another transient surge and did not fail due to some failure of your repair work.

The one I repaired is happily still working.

“Good morning. There is an encryption layer that you need to interface to be able to add devices to read the meter, you will need to contact the power company for them to add the device.

Thank you
Technical support”

Have you made any more discoveries into how such a system might work?

Another question I had is more specific to the RaspiGateway project. I’m using Occidentalis v.2 on Adafruit’s fantastic WebIDE which allows you to clone directly from Git Repo’s. I was able to clone the RaspiGateway Repo but have not had any success in installing “pySerial” or any of the other library extensions.
When I send “python -V” it returns “Python 2.7.3rc2″ but when I type ” cd /h
ome/tinaja/downloads/” this is what it returns-

“webide@raspberrypi /usr/share/adafruit/webide/repositories/gateway_raspi $ cd /h
ome/tinaja/downloads/
bash: cd: /home/tinaja/downloads/: No such file or directory”

Is this because I’m trying to access from within the WebIDE or am I running a newer version of Python? Or is it something else?

Thank you, and let me just say I appreciate what you are doing here greatly!

Yeeeees, just changed the 2 1000uF (the brown) the 470uF was black and seemed to be good
my GS608 V2 came back to life
thanks

All the code in my projects lives in the gateway and at the moment, all python. The python script reads the serial port, polling for data from the various xbee radios, bundles up the data and sends it out to web services like cosm, thingspeak, open.sen.se.

My sensor boards do not have an arduino or microcontroller. They have a voltage regulator, an xbee radio, and up to 4 sensors each. At some point I might need a microcontroller, but I’m trying to keep the code centrally located in the gateway device. It’s clearly not the only way to go, but it’s my way for now. Keep as much of the logic in one central location/device within the house.

I struggled with communicating both ways with the xbee. But I did finally make it work. Look up the link to github on this site. There are some python samples for controlling a servo.

Alternative to apache/php: I recall seeing a python script that acted as a web service but can’t remember where. Perhaps a google search will find it.

Not sure If that makes sense, to clarify 1. Can my server communicate both ways with my wireless xbee cards/arduinos?

2. Is the best way to send commands through apache/php?

3. Should tasker work to fill a field with a variable in my website that my server can use?

I guess that your probably not familiar with tasker I’m just wondering if that sounds like it would work. Sorry for taking up so much of your time, thanks very much.

Adam

You mentioned using the alarm on your android. If there is a way to trigger a web url call from your alarm app, this would be excellent. Perhaps Tasker is the app which can provide this functionality.

The next thing to hook up is a web server on the raspi, like apache. You could then write a service using php (here’s the coding) to respond to the call from your android alarm trigger. This php code could trigger an output on the raspi GPIO pins to flip a relay, which in turn, turns on a lamp, or starts the coffee pot.

Here’s an interesting article that might help with the web service approach:
http://www.raspberrypi.org/phpBB3/viewtopic.php?f=44&t=15257
Other search results: http://goo.gl/xkEdO

You can find variations on a theme at the Adafruit Learning system site here:
http://learn.adafruit.com/category/raspberry-pi
http://learn.adafruit.com/adafruit-16-channel-servo-driver-with-raspberry-pi

The Adafruit Cobler kit will probably help get you started.
http://learn.adafruit.com/adafruit-pi-cobbler-kit

I used a GPIO board that was ideally suited to accessing the serial output from an xbee radio. The Cobbler connection is a general purpose device that I think will be useful for your project.

At some point I plan to remove the hard coded references to the xbee radios and read that from an xml config file.

regards
Marc

You responded to my post related to Nest’s new termostat launch.

My linkedin name is bjarne hadland!….looking forward to communicate with you.

All the best!

Bjarne

Glad it helped, Michael.

Greats from Germany

Michael

I’m really interested in getting wireless to work. I couldn’t find a total tutorial
In how to do this. Do you have one.

Respectfully,

James K Smith, Sr

That way we filter out any IP traffic, or powerline module traffic, that might be travelling over our house power wiring; in or out, and which might be tracked by the “powers” that be.

I use X10 modules, there are devices that allow people to use the house power wiring to transmit IP service throughout the house.

Filter required!

It seems the ideal approach for me would be a bunch of your low cost nodes reporting energy usage to a master PICOwatt device. That master PICOwatt would send the data to a portal site where users could review the statistics.

I built a tweet-a-watt (http://www.ladyada.net/make/tweetawatt/) as a proof of concept. With parts, each tweet-a-watt costs about $45. It seems that by optimizing the manufacturing of this kind of device it could be made quite inexpensively.

After that it’s just a software challenge.

I’d love to talk to you about this when you get a chance.

as you can see by my email I have a vested interest here – so I’m sure you’ll take my comments with a grain of salt.

Your point about price is a good one and moving forward we will be able to reduce the cost of the device significantly – but I wanted to point out a couple of things -

PICOwatt does keep a database of energy usage onboard – and it is downloadable in spreadsheet format or you can view it in table form or graphically on the PICOwatt.

If you have more than one PICOwatt on your Home Area Network, they automatically recognize each other and all access and control is then routed through a “master” PICOwatt which acts as a “whole house” monitor website – eliminating the need for buying an expensive home energy monitor.

PICOwatt also lets you set timers to take advantage of things like off peak pricing if available – and if real time rates were internet accessible from your utility, PICOwatt could easily leverage those to reduce operating costs.

There’s also a facebook app that let’s you access PICOwatt from outside your HAN.

In development right now are low cost nodes to complement the PICOwatt. These are dumb nodes that facilitate power measurement and switching off and on, but leave all the data recording, timer set points, web pages, etc. to the “master” PICOwatt. We expect these to be in the $20-30 range.

Regards,
Peter

Very cool!! Once again you are the trailblazer, the cyber Pied Piper, so to speak.

By the way, I especially like the modern, high tech furniture that comes with the system.