Monday, May 13, 2013

DIY ESS - 1700W + pushed into my home grid

Captured this morning before leaving home, with 7 GTIs at full power : 1714W


  1. Hi Christophe

    An interesting project you have here!

    I have been planning on doing something similar albeit not as sophisicated as your setup. I was wondering why you went for multiple cheap GTIs instead of say a single 2+KW SMA Sunnyboy.....was it the challenge of stepping up the inverter input voltage to 200+ VDC? I'm still trying to figure it out....

    The other thing I'm curious about is the cost effectiveness of storing solar PV electricity in LiFePo4 batteries for nitetime use. Here in Queensland, Australia, we can sell generated PV back to the grid for $0.09/KWh and buy it for $0.28 (peak) [it used to be much more lucrative] however, based on the current price of LiFePo4 cells, it is more cost effective to sell the excess to the grid however the break even point is coming soon! I may have missed it but any chance of measuring and publishing the efficiencies of your system. i.e. Energy (KWh) feed into your system VS the energy out. AFAIK the Sun GTIs are about 85% efficient. What about your battery chargers? 85~90% efficiency? The old lead acid battery systems were terrible with over 30% losses. So, some figures would be interesting.

    Keep up the good work.

    Sunshine Coast, Qld, Australia

    1. Hi Chris,

      Thanks a lot !

      I used these SUN250G Grid Tie Inverters, because my set is based on a low voltage battery pack - 58V, now 24V requiring less cells, safer to handle - and these work great with it because their range is 10.8 to 30V
      They are also inexpensive and easy to get, stackable, work in //, etc
      Anyways not many options at this input voltage ... until SMA releases their New small SMA 240 ... see my recent post on this new option for my system (much more expensive thought, many 4 to 5 times the price, but probably a better efficiency)
      About the ROI, this is not at all what I am interested in :-), I have been experimenting these past two years and building from scratch this LiFePO4 based system, improving step by step, spent a lot of spare time, adjusting the programming in the Arduino board, and I am not an electrical engineer at all, more in the IT (SAP developer), but it really works , with solar since last week exactly, as predicted filling up my battery pack during the day, and helping, during the day, and emptying in the evening with cooking, lights, tv, A/C, etc
      My goal was to become less dependent on the grid, and I am pretty happy with the result
      I am finishing the detailed documentation and preparing to sell this base version of the DIY ESS Kit on my blog

  2. Hello again Christophe

    Since I'm also doing this from an interest point of view, ROI is not crucial but if my wife finds out that all the batteries, inverters, etc in the garage actually cost us more than they save us, she is going to send me back to work ;-(

    I do think the efficiency of whatever gets stored in those expensive little LiFePo4 cells is important [Unfortunately, I was only an Oracle techo and not a SAP consultant ;-( ]. If one generates plenty of PV power than losses charging batteries is probably not an issue, however, once stored, it is important to minimise losses on the way back to the grid [IMHO]. I'm looking at grid-tie inverters with >95% efficiency and I'm looking at how to minimise losses from the battery to the inverter.... 10% has got to be the absolute worst case!

    I'll have a serious look at the SMA 240 Microinverter. On first impressions they seem rather bulky for a microinverter! The one I'm considering is the Power One Aurora Micro-0.3-I-OUTD. This is a 300W model and has efficiencies of 95.5%. DC input voltage range is 25V to 60V so not suitable for your 24V battery but should be fine for 48V. Cost is around $250USD but you have to buy a wireless/WiFi monitoring device which is up near the $300 mark but it will handle up to 30 inverters. I guess the reason I'm scared of the Sun GTIs is that they are not certified here and there seems to be lots of failures discussed on YouTube. Whatever I use must be from the likes of SMA, Xantrex, Power One or Frontius because in our Nanny State it will still be illegal!

    Well I better get down and start watching your YouTube stuff, etc. as that is likely to be most informative.

    Thanks again for sharing this stuff

    Sunshine Coast, Qld, Australia

    1. Hi Chris,

      My wife does not really look at how much I spent on this, starting with the Prius Enginer LiFePO4 Kit in 2009, and this is a good thing :-) because it adds up over the last couple years especially, but the it is a passion I must confess ...
      About the efficiency, although the SUN250G spreadsheet mentions 92.5% peak, I do not think this is what I am getting, maybe partly because I run them at 26V or less and they might be rated for 30V, the maximum (??). But you are right about getting better quality ones, if they can perform much better, because the price difference is huge (I saw 240USD online); I will be looking at this new comer SMA 240-US (or -EU is there is a European version) .
      I saw that Emphase and other makers of small GTIs could be of interest too, and it is a good thing that are looking into these. I am not sure why you would need a WiFi monitoring device for them (is it to make them work? Without this they would not start ?)
      What is illegal in your state ? I di not get it ?
      Anyways, I just finished writing my full Detailed Kit Building Documentation, and now I will focus on the detailed Components List (including links and addresses to get them), and start buying Arduino boards to flash them with my program
      This first version of my Kit (Kit #1) will save more than a year to people willing to build their own ESS



Please Share this Post if you Liked it !

Please Share this Post if you Liked it ! Thanks !
Related Posts Plugin for WordPress, Blogger...