• Category Archives Computers
  • Imagine a life with no computers……ahhhh……bliss…..

  • Your fridge has been hacked.

    I am not going to bother to even pretend to be surprised about this one….

    http://www.digitaljournal.com/technology/hackers-steal-gmail-credentials-by-hijacking-samsung-smart-fridge/article/441913

    Hackers have successfully extracted login details for Google’s Gmail email service from a Samsung smart fridge.

    The affected fridge is the Samsung RF28HMELBSR. The fridge is equipped with a built-in display that can be used to view, amongst other things, calendar details. The aim is to let the whole household plan their calendar online, putting an end to the scribbled notes found under fridge magnets around the world. Users have to provide the smart fridge with their Gmail details in order to use this feature.
    Unfortunately, the fridge fails to validate the SSL security certificates sent by the Google servers at login. This makes it open to man-in-the-middle attacks because the fridge never checks to see that it is actually connected to Google, although SSL is implemented and certificates requested.

    That pretty much sums it up.
    Don’t feel there is much more to add.


  • Solar Panel Efficiency

    A few blogs back, I mentioned that I was not getting as much power from my panels as I expected.
    My Dad emailed and suggested that given the warm temperatures I was seeing, he wondered if it that was the cause of the drop in output.
    He also asked about the epoxy coating I added to repair the panels.
    So, lets answer all his questions…. and ask a bunch more!

    First up. The epoxy.
    I tested the short circuit of the panels before I coated them. All were around 8.2 amps.
    I also measured their open circuit voltages, all were around 35 volts.
    After putting the epoxy on, they were exactly the same.
    In my mind, the epoxy is a non-issue.

    Temperature.
    What a can of worms……
    I found the panels spec sheet (such that it is) online.

    groov solar panel data sheet questions

    Ok, so lets break it down.
    The red oval is the panel I have.
    Now the numbers.
    1. The characteristics of the panels are listed at an irradiance of 800w/m2 (more on this when we get to point 4) and a temperature of 20°C.
    From my screenshot the other day, you can see that the panel is a LOT hotter than 20, so that might be a big factor. How much of a factor?

    2. The spec shows that the panel output is normally 45°C. This is closer to what I am seeing… but hang on, why do they give the rated output at 20°C, but the nominal to be 45°C? Which is correct? Why the two numbers?
    To make it ‘worse’, both are called the same thing…. “normal operating cell temperature”…. I am confused, how can both of those numbers be normal?

    3. I read this to show the de-rating to be -0.4% per °C. So for every deg C over either 20°C or 45°C, I will lose 0.4% of my output?

    4. Irradiance of 800w/m2. This is a number that is the amount of sunshine (all colors) that is shining on 1 square meter. (My panels are 1.6 meters square).
    As it happens, the weather station at Opto has this value since it has a solar sensor built in.
    My co-worker, and recovering software engineer, Mary wrote me some code to pull the number out of a text file on its FTP server and thus I have it in my PAC Controller and thus groov.
    We are now trending irradiance on the same trend as the panels output.
    groov solar irradiance
    You can see that we only added the data point in the past few days, and that the marker is showing around the magic 800 mark.

    So, bottom line. While I have no idea if it is 20°C or 45°C, I am collecting data and I am confident that I have all the necessary bits to work this out.

    Pretty sure it will involve me standing on the roof with a garden hose water cooling my panels while collecting their data.

    (Thats enough out of you lot!!).


  • Apple and Cisco

    This blog is for Gary and Dan. The rest of you are excused…..

    Ok you two…. what in the world is this about?

    http://www.apple.com/pr/library/2015/08/31Apple-and-Cisco-Partner-to-Deliver-Fast-Lane-for-iOS-Enterprise-Users.html

    I am going to have to dig a little deeper to get why this matters or what it means for the future.

    Apple and Cisco are also working together to make iPhone an even better business collaboration tool in Cisco voice and video environments, with the goal of providing employees with a seamless experience between iPhone and their desk phone.

    With Apple’s support, Cisco will deliver experiences specially optimized for iOS across mobile, cloud, and premises-based collaboration tools such as Cisco Spark, Cisco Telepresence and Cisco WebEx in order to deliver seamless team collaboration and reinvent the meeting experience.

    This bit I kinda get. Seamless transition from iPhone to deskphone to PC (Mac) while on a FaceTime call.
    Cute, but hardly groundbreaking. Sounds like they have Skype (and thus Microsoft) in their crosshairs on that one.

    Cisco is empowering countries, cities, industries and businesses around the globe to move faster in order to keep pace with digital transformation and the move to the Internet of Everything (IoE).

    Is this what Apple is really after? A byte of the IoE apple?

    I am lost on this one.

    Sub Note. The press is trying to link last years Apple -> IBM mashup with this Ciso thing….

    I still am lost.


  • Inverter efficiency

    Just a quick word on the solar inverter efficiency number you saw in the last blog groov screenshot.

    The code behind the number is pretty simple.
    gti_efficiency = (gti_output_power / panel_1_watts) * 100;
    gti_wasted_power = panel_1_watts - gti_output_power;

    This gives me both a percentage and a wattage.
    The wattage is of interest because the heat sink on the unit gets really hot
    (As you can see in the screenshot – the air temp was 31 and the heatsink was 60.)

    The slight twist to the code caught me off guard (I’m getting old and rusty!). At night, the ouput of both the inverter and solar panel is zero. For some reason the computer was not happy about dividing by zero.

    My solution, wrap the above code in more code;
    if (panel_1_volts > 15) then
    gti_efficiency = (gti_output_power / panel_1_watts) * 100;
    gti_wasted_power = panel_1_watts - gti_output_power;
    endif

    So, yeah, if it’s dark (or really really really overcast) don’t bother doing the code.
    For now, that seems to work.

    It’s oddly overcast today, a welcome respite from the heat we have been having, but not the best for solar system testing. At the moment we are only getting around 50 watts from the inverter….. and 87.1% efficiency.


  • More solar data

    Sharing is caring right?
    Got more of the system hooked up today (Sunday).
    Course it took waaaaaay longer than its allotted time, I am not happy with the performance, we have zero automation as yet (since it’s a work in progress I refuse to spend time coding when it still could (will) all change) and well… then there is the wiring…….

    20150830_092400

    I think we can just be happy that a) I did a better job wiring Garys RV and b) that I don’t do this for a living.

    Ok, so we have some Opto hardware, the key to it all really.
    Some solar power coming down the wall.
    Some batteries (crash test dummies, not the real deal yet).
    The grid tie inverter on the lounge next to the Opto.
    A multimeter.

    Then, all the lovely data ends up in groov.

    groov solar gti and batteries charging

    This is the current messy screen for it all.
    Below this section is a bunch of trends, you don’t need to see them yet, but they help me track whats happening over time.
    The one trend you can see is the power pulled by the house from the grid and the power pushed into the house from the grid tie inverter.
    I am pulling around a kW and am pushing between 100 and 200 watts. Pretty lame.
    Panel 2 is charging the battery. Negative power is charging, positive power is drawing from the battery. This way I can keep track of watt hours (amp hours) added and watt hours taken – well, thats the plan.
    Panel 3 is doing nothing.
    Looking at the trend, you can see the orange line go from 100 to 200 watts, for that duration, I had panel 3 switched into the grid tie.
    Note that I do the gti efficiency calculation. According to the math, it’s pretty good. The heatsink gets hot (you can see the temp to the left), I am going to have to do something about that (Like bolt a real heat sink to the back of it), and given the heat it pushes out, I think my math is wrong, other than that… its a pretty number.

    So yeah, I can ‘force’ the gti to work harder, but don’t see why I should…. Panel 1 is only outputting half its rated power.
    Then again, so are the other two. Its like I have 120 watt panels, not 255 watt panels. Watch this space because when I test short circuit current, I get 8 amps, so I know it can deliver the amps, but the voltage seems to be fading more than expected. (And yes, I have very thick wire coming from the panels, so thats not the issue).

    Right now, I have to press on with other honey-do jobs, so have left the batteries charging off panel 2 and panel 3 boosting the gti.