Archived Forum

Hi Hardwic,

I like your LCD display, the graphs are very cool. I see your device has some very powerful waveform analysis features. Its very impressive what you have done with an Atmel 128! Do you plan on making your design and code open-source? 

In my opinion having the data automatically uploaded online to somewhere like emoncms/patchube/google power meter makes the data much more easily accessible and can give real-time graph feedback. 

I think a complete domestic monitoring systems should have both a real-time display for the living room / kitchen (will need to be wireless) and online access to data for in-depth analysis. 

Thanks for posting.

All the best, Glyn. 

Hi Glyn
The main reason I am apposed to having the realtime upload to a web interface is you have to run a PC/Mac to do the work.
This is self defeating because the PC will waste power. I have used 86% of the Atmega's 128 resources and can't accommodate ethernet. I am not that keen on wireless either because I want reliability foremost and speed. Cost is another major factor.

I think adding RS-232 login would be the easiest if this feature was really required. For instance I run a Mac Mini server for security monitoring so logging data would come at no extra cost.

Google power meter is actually a bit of a joke because they only have 10 minute resolution. I am supporting sub-second logging. You really need this if you want to analyze the data and detect the signatures of devices.

If you look at the graph on this page you will see what I mean: http://chtech.co.za/chtech/Energy_Monitor.html

As regards open source, I am still not sure what to do. Certainly I want more people to contribute and more people to use 1. I am producing all boards in modular format as they are generic MCU/SD/RTC boards. They will certainly be available to people who want to develop their own AVR project. I am not sure if it's necessary for everyone to spend money on originating boards and assembling them.

As regards the software, I have spent a year working on it and would like to recover some investment. Then again this is not about money but wanting to make a contribution to society. I must still think this through. Ideas would be welcome.

Regards Clinton

realtime display

http://www.youtube.com/watch?v=it5RJYZYswk

The colour changing bezel is not working correctly but it will soon.

Hi Clinton,

I agree, its defiantly not ideal to have a PC running 24/7. Have you looked into using the Atmega328? It has double the memory of the 168. You might be able to accommodate Ethernet with the additional memory.

Your energy monitor is very impressive but is obviously aimed at a different market than domestic energy monitors. For the majority of users a few readings per minute is adequate to provide an indication of their power consumption. I think your energy monitor could be very useful for people doing more in-depth analysis into energy consumption. I agree the 10min interval offered by google power meter is pretty bad. Google have actually recently killed off google power meter. 

Obviously weather or not to open-source your work is your choice but in my opinion the benefits of increased development, exposure and collaboration outweigh the potential disadvantages. Limor Fried (Lady Ada) from Adafruit Industries had written an interesting piece on open-source hardware and addresses many concerns. Read it here: http://www.ladyada.net/library/openhardware/license.html

In your case its seems to me that your potential customers are energy analysts and auditors. These people will want to buy a working unit and probably wont mind paying for a useful, well designed and reliable piece of kit. I doubt that many of your potential customers would want to download source code, hardware designs and build their own monitor. In other words open-sourcing your design should not reduce the potential profit you could make selling your product.

Best of luck and keep up the good work!

Glyn.

Hi Glyn

The thinking behind the detail I want is maybe the customer does not have the understanding of what they need to do but if you collected the data, uploaded it to a central database and then had an analyst advise and refer you to good savings vendors, there could be an interesting business.

Personally having 1 figure is not good enough but then again I am a techie :)

Oddly enough I spoke to our Fluke agent last week and he said this unit would be nice to have installed in his own home. So what he was saying is there are features in this you can't get in even a very top end product.
This is quiet encouraging.

My target is to sell the unit for under $180 so Ethernet will need to be an optional extra.
I may just add RS232 for the budget version just in case, after all the board does have a UART.

I agree open source collaboration would be amazing as it would help grow the reporting functionality drastically.
Did I mention the board has 4 CT interfaces? This is to collect the main feed and 3 either circuits.
I still need to do the code and a good example where collaboration would be good.
As you say very much more an analysis tool than a number on the wall.

If people were keen, I would be quiet happy just build the boards and feed this into a much bigger effort.
Not sure if everyone wants to build and design the boards or just get on with coding.

Clinton

The display looks great! it must have been quite a challenge to write the code for it!

Its great to see that you have made use of the AC theory page I wrote on your site. All our documentation is licences under the GNU Free Documentation Licence which gives anyone the right to: copy, redistribute, and modify our work as long as all copies and derivatives are made available under the same license and are attributed to their original authors. I would appreciate if you could do this. The same thing goes of course to all the openenergymonitor hardware and software, we have licensed these under GNU General Public Licence v3 and Creative Commons Attribution-ShareAlike 3.0 Unported License as detailed here: http://openenergymonitor.org/emon/node/4.

Thanks

Trystan

Fair enough Trystan, I will quote you as the original source but in the same vain I don't see you quoting yours:)

I don't think you can lay claim to the the theory but the presentation was neat so I used it and will credit you as such.

Clinton

 Hi Clinton,

I can only congratulate you for the neat work you're doing, but I must join Trystan in asking you to share your knowledge. Crediting trystan and Glyn is not enough, the license requires to credit and share the work.

I understand that you spent precious time coding for it, but believe me, sharing will benefit to everyone including you.

Example: there are many people who are developping ethernet connectivity for the arduino with the chip ENC 28J60: Jeelabs, Nanode, etc. and thanks to the efforts of these people all together this solution is now more complete than the official arduino shield, including DNS and DHCP.

Regards and bravo again for your work.

Amin

Hi Amin
I agree with both of you and yes if everyone would freely contribute we would see significant benefit to all.
I am however conflicted by a few issues, firstly my project is not based on any Arduino project. All hardware is developed and paid for by my self. Code is winavr gcc.

The only thing in common with this project is I read an ADC and calculate RMS values accordingly. Thereafter how this all works with multiple timers/RTC/GLCD/SPI etc is the big trick.
My plan is to find a commercial funder to produce this product in volumes so it can be used to make a difference. Open source right now will eliminate this option. If that fails, I will release my work to achieve the same objective and right off my investment.

I also feel that I need to make a contribution to the community, not just this one but all.

The commercial interest thus far looks non existent so this could be a product of my own aspirations enjoyed by few...

I must point out, Trystan's document is published without copy right or author reference in a pdf simply accessed by googling "ac theory"

Clintion

Thanks Amin.

Clinton: 

my work certainly should be referenced so that what is writen can be checked up on, its not the best practice I know. So my leg to stand on while asking you to credit is somewhat tenuous ;) The pdf found from google is covered by the openenergymonitor documentation licence as the licence as I understand covers all documentation on the site. Anyhow lets put this to the side.

On the subject of wheather to open source, its up to you, I would of course very much recommend it, Amin's example is a great example of its benefits. Take a look at this page too: http://openenergymonitor.org/emon/node/112

All the best

Trystan

I think this puts things into better perspective:

http://www.youtube.com/watch?feature=player_embedded&v=UF8uR6Z6KLc

 Hi Clinton,

I see and understand your point. We're all on the same side: Trystan Glyn, you and me. We all want to make some money with this.

If you want to protect your work you need to patent it, otherwise the "funder" you're looking for can just steal your design, or the manufacturer, or anyone will buy it and reverse engineer it. If you dont patent it you dont have any legal rights. So just make sure that you have the time, the money and the patience to patent it, or simply play it open.

We've been there before, believe me :)

Amin

Just to step back a bit and let me fill you in on my thoughts.
Firstly the reason I developed the unit is because there was nothing that could do the job to my satisfaction.

Once I had the unit, I found the data incredibly helpful and thought this would be fantastic if large amounts of people would do the same as me and reduce their consumption.  My target is to product the unit for under $180. There is not much margin for me but the objective is to have more people use 1 and as quickly as possible.

Open source software meats the same objective.

I am comfortable releasing the code. Not so comfortable with the hardware for a few reasons.
The boards were designed as modules which can be used for any project/product.
I have invested more then $5.5k in tools that allow me to design and assemble these boards. This is just a quick sum of what I remember but it's most likely more. I don't see the point in people doing the same, they just need to buy what I have and get on with using improving the code.
Does this sound reasonable?