Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)

Categories



Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

PLEASE NOTE: A download link for Volume 1 will be sent to you by email and Volume 2 will be sent to you by post as a book.

Buy individually or both books together. Delivery is free!


powered by Surfing Waves




Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.




    • CommentAuthorjms452
    • CommentTimeMay 4th 2021
     
    Posted By: Jeff BP.S. Apologies to jms452 for hijacking the thread.


    no problem - I think I've got my answer and will report back on the process in a bit.

    Random GBF thread highjacks have been one the ways that I've learnt so much about building better houses and quite a lot about renewables over the last decade! :bigsmile:
    • CommentAuthorSteveZ
    • CommentTimeMay 4th 2021
     
    In the spirit of hijacking the thread, I was interested by the figures from other 4kWp arrays,especially why Jeff B's seem so low. I have a split array of 3kWp facing due East and 1kWp facing due South, both at ~30° with a Solaredge setup of inverter and individual panel Optimisers. Since 2012 the annual totals are as follows:
    2012 3.34 MWh (missing January)
    2013 3.62
    2014 3.71
    2015 3.71
    2016 3.60
    2017 3.40
    2018 3.55
    2019 3.67
    2020 3.77
    Given the orientation, I am pleased with the results.
    Re Jeff - does your micro-inverter setup have a reporting system in case of individual inverter failure? I ask because in theory my system reports such failures, but the one Optimiser failure I have had I found by accident as I logged in to the layout and noticed a blank report on one panel. i might have noticed the drop in the whole system output, but I doubt it. The unit was replaced under warranty, as they are covered for 25 years!

    Maybe the figures will be of interest to others thinking about which system to use.
    •  
      CommentAuthordjh
    • CommentTimeMay 4th 2021
     
    Posted By: SteveZdoes your micro-inverter setup have a reporting system in case of individual inverter failure?

    FWIW, my Enphase system does. It went off earlier this year and alerted me to low output from a couple of panels, suggesting I get somebody to fix the system. Fortunately a quick look showed those panels were covered in snow, so I didn't. :bigsmile: Still, at least it's nice to know the system works. Also 25 year warranty.
    • CommentAuthorphiledge
    • CommentTimeMay 4th 2021
     
    Interested in how your faulty optimiser got sorted out. Like you we only spotted a faulty unit when logging on to the monitoring portal. Our system uses optimisers integrated into the panels so Solaredge sent a standalone optimiser which has now been fitted. Its been paired with the inverter ok but is still showing zero generation on the monitoring site. Ive sent Solaredge the new optimisers serial number but not heard back from them.

    Do you know the detail of how yours got sorted??
    • CommentAuthorSteveZ
    • CommentTimeMay 4th 2021
     
    Hi Phil - my system has individual Optimisers, so it seemed to be a straight swap of components and the readings are OK. I can't remember if the optimiser ident had to be reset by Solaredge or maybe it just worked straight away. I contacted my original system installer, Sungift Solar, and they sorted it all from one call.
    • CommentAuthorphiledge
    • CommentTimeMay 4th 2021
     
    Thanks for the update. Maybe Sungifit updated the ID as I think they have access to installers apps etc.

    My installer wasnt interested in the repair and wanted £70 plus vat plus scaffold to fit the warranty replacement.
    • CommentAuthorJeff B
    • CommentTimeMay 4th 2021
     
    Posted By: SteveZ
    Re Jeff - does your micro-inverter setup have a reporting system in case of individual inverter failure? I ask because in theory my system reports such failures, but the one Optimiser failure I have had I found by accident as I logged in to the layout and noticed a blank report on one panel. i might have noticed the drop in the whole system output, but I doubt it. The unit was replaced under warranty, as they are covered for 25 years!

    Maybe the figures will be of interest to others thinking about which system to use.


    I have Enecsys microinverters. There is no "automatic" warning system but the software allows me to check the status of each inverter as/when I feel like it! I usually check once a week. There have been several failures in the 10 years I have had the system but all ok for the last 2 years or so. Since joining in this thread I am wondering if there might be something in the inverter firmware which is inhibiting the PV output, maybe capping the output for some reason? The panels are rated at 180W but I'm sure I have never seen an output higher than 150W.

    The software is not the original stuff supplied by Enecsys as this died when the company went bust (it required an Internet connection to the website, something which I was not happy with from the beginning as I wanted a standalone system). I now have some bespoke software which is very sophisticated - I only use a fraction of its capabilities!
    •  
      CommentAuthordjh
    • CommentTimeMay 5th 2021
     
    Posted By: Jeff BThe panels are rated at 180W but I'm sure I have never seen an output higher than 150W.

    What are the microinverters rated at? (or what is their model number, maybe?)
    • CommentAuthorJeff B
    • CommentTimeMay 5th 2021
     
    Posted By: djh
    Posted By: Jeff BThe panels are rated at 180W but I'm sure I have never seen an output higher than 150W.

    What are the microinverters rated at? (or what is their model number, maybe?)


    The microinverters are SMI-200/G83. Maximum output 190W.
    • CommentAuthorjms452
    • CommentTime5 days ago
     
    Hi All,

    Next step in transferring deemed FiT export to metered fit export.

    Octopus require confirmation that the DNO has been fully notified by the installer - This is what I received from the DNO (uk power networks):

    'We have checked our records and cannot find any reference to your property address. It is possible that the installer did not notify us at the time of installation. This installation should have been notified to UK Power Networks using the EREC G59 form as the capacity is above 3.68kW, whereas if these panels were installed today, the EREC G99 form would be required. Therefore an application cannot be made using either the older G59 form (now obsolete) or G99 forms.

    Given that your PV Installation has been connected to our network for over 8 years, UK Power Networks are agreeable to it remaining connected to our network but cannot provide retrospective approval.'

    The the thing I'm confused by is that while we have 4kWp panels our inverter is 3.6kWp so I'm not convinced the DNO is correct. What is the capacity of a PV installation the total rating of the panels or the rating of the inverter (i.e. the maximum the system can ever provide to the grid)?
  1.  
    Sorry to hear that.

    I'd understood it is based on the installed rating of all the panels and batteries including V2G, so you need a G99 if more than 16A of panels and batteries are installed. *Unless* you have a device that actively monitors the combined export connection and rapidly and reliably disconnects stuff if it ever exceeds 16A, a bit like a smart circuit breaker. That device might be built into the inverter. The model of inverter must be tested ('type approved') to meet the disconnection safety standards, and then you make a G98 application to let them know which model you are using.

    I'm sure you should be able to make a new G99 application, as you would if you were say adding more panels or a battery, but I don't know how that would affect your FIT.

    Edit to add: Given the tiny risk to the network if someone installed 4kW instead of 3.68kW, possibly they just turn a blind eye, but can't give written approval.

    Further edit: my mistake, if the inverter is type-tested to comply with G98 (max power output is limited to 3.68kW , among other requirements) then the installer notifies the network using G98 Form B. So G98 for ≤16A and G99 for >16A.
    • CommentAuthorEd Davies
    • CommentTime5 days ago
     
    Posted By: jms452What is the capacity of a PV installation the total rating of the panels or the rating of the inverter (i.e. the maximum the system can ever provide to the grid)?
    For FiT-scheme purposes it's the rating of the panels; for the DNO it's the rating of the inverter, the panels are none of their business, so maybe they've misunderstood something about your installation.

    For less than 3.68 kW [¹] it should have been certified to G83. Can't remember off the top of my head whether that's G98 or G99 now - I'd assume from their comment it's G98. For G83 they don't need to give approval (unlike for higher powers/currents). At most they can tell you shut it off for a while so they can fix their network to deal with it.

    Practically speaking, would that response from the DNO be sufficient for Octopus?

    [¹] Strictly speaking, 16A/phase but that works out as 3.68 kW at 230 V.
    • CommentAuthorjms452
    • CommentTime5 days ago edited
     
    Posted By: Ed DaviesPractically speaking, would that response from the DNO be sufficient for Octopus?


    I've submitted it so we'll see ...

    In parallel also highlighted that we have a 3.6kW inverter to the DNO and asked if their statement is correct in this case.

    our installers seemed pretty good so if they never notified it makes you wonder what else was going on in the wild west!

    Slightly reassuring that the DNO seems pragmatic about it though!
    •  
      CommentAuthordjh
    • CommentTime5 days ago
     
    FWIW, I've never noticed before but I think I can see evidence of my system limiting output. Instead of the nice curve that I'm used to seeing I noticed that yesterday's output has a flat top
      Screenshot_2021-05-07 Emoncms - limited power output.png
    • CommentAuthorjms452
    • CommentTime5 days ago
     
    Posted By: djhFWIW, I've never noticed before but I think I can see evidence of my system limiting output


    Because we're dual aspect it's pretty rare than happens with us but I can see that we never generate more than 3.6kW because on a bright sunny day with clouds we occasionally jump up to 3.6kW between clouds.
    • CommentAuthorrevor
    • CommentTime5 days ago
     
    @ djh would be strange if your inverter started clipping the output does any historical data show it has always done this.

    If one wires one's own house you need a part P certificate from an inspector'. Anyone aware if similarly one can get an MCS approval for one's own PV install?. I had an installer lined up they said to source the stuff myself as it was gear they did not have knowledge of (it's MCS approved) and they would put their electrician on the job. They did the DNO G98/99 application which was approved for 6Kw and would then do the MCS for a fee. As it turned out 4 requests for the electrician to come and see me over a 4 month period fell on deaf years so carried on with the job and completed it myself with assistance on line from an expert in the model of inverter. I was originally not to bothered about the MCS but after charging the battery I find I am exporting a phenomenal amount to the grid. We are intentionally a low energy house so can only run the dishwasher and washing machine so many times. Not much point putting the immersion on (winter maybe) as have solar thermal as well so if PV performs so does thermal. The "installer" is gibing somewhat as to whether they can certify the install they may yet do, but looking for a get out of jail card.
    • CommentAuthorbhommels
    • CommentTime5 days ago
     
    Posted By: djhFWIW, I've never noticed before but I think I can see evidence of my system limiting output. Instead of the nice curve that I'm used to seeing I noticed that yesterday's output has a flat top
      http:///newforum/extensions/InlineImages/image.php?AttachmentID=7921" alt="Screenshot_2021-05-07 Emoncms - limited power output.png" >

    Do you think that is the panel output exceeding the inverter rating or could it be the inverter running into its thermal limit? As per the plot below, my SMA SB3000HT inverter starts to thermally derate when ambient gets over 45 degrees.
      SMAderating.jpg
    •  
      CommentAuthordjh
    • CommentTime5 days ago
     
    As I've said before, I don't have an inverter. My system uses microinverters, so there's some clever control software somewhere. I would guess the brains are in the Envoy gateway unit but I suppose it might be distributed amongst the microinverters.

    I don't think we're likely to have run into any thermal limits yet this year :bigsmile:

    Now I look at my logs I can see it's always done this limiting. I've just never noticed it before for some reason.
    • CommentAuthorbhommels
    • CommentTime5 days ago
     
    Posted By: djhAs I've said before, I don't have an inverter. My system uses microinverters, so there's some clever control software somewhere. I would guess the brains are in the Envoy gateway unit but I suppose it might be distributed amongst the microinverters.

    Sorry, perhaps I should have read the whole thread before posting. Anyway, with an inverter hitting a thermal limit the plot would probably not have such a (nice?) flat top.
    • CommentAuthorjms452
    • CommentTime5 days ago
     
    Posted By: bhommelsSMA SB3000HT inverter starts to thermally derate when ambient gets over 45 degrees.
      http:///newforum/extensions/InlineImages/image.php?AttachmentID=7922" alt="SMAderating.jpg" >


    That makes me very glad our inverter isn't in the loft!
    • CommentAuthorEd Davies
    • CommentTime5 days ago
     
    Posted By: bhommelsDo you think that is the panel output exceeding the inverter rating or could it be the inverter running into its thermal limit?
    It's quite normal for the PV panels to be higher rated than the inverter and for grid-tie inverters to limit their output to the power or current limit set in their configuration which will, in turn, be limited to the name-plate power output of the inverter.

    Early on in the FiT scheme there was a bit of worry about this as, technically, the limit is 16A/phase so for inverters which had a power, rather than current, limit this meant that you ought to limit the power to something like 3.45 kW (16 A at around 216 V) just in case the voltage was near the bottom of the allowable range but DNOs took a common-sense approach and allowed inverters limited to 3.68 kW (16 A at the nominal 230 V) on the grounds that if the voltage was lower than that then a little more current would be welcome support anyway.

    Off-grid inverters tend to be a bit different: they'll produce output greater than their rated value for a short while until they get too warm at which point they'll shut themselves down. This is handy for things like starting motors.
    • CommentAuthorJeff B
    • CommentTime4 days ago
     
    Posted By: djhAs I've said before, I don't have an inverter. My system uses microinverters, so there's some clever control software somewhere. I would guess the brains are in the Envoy gateway unit but I suppose it might be distributed amongst the microinverters.

    I don't think we're likely to have run into any thermal limits yet this yearhttp:///newforum/extensions/Vanillacons/smilies/standard/bigsmile.gif" alt=":bigsmile:" title=":bigsmile:" >

    Now I look at my logs I can see it's always done this limiting. I've just never noticed it before for some reason.


    djh: I had no idea how complex the electronics are in an Enecsys microinverter until I took I took one apart. It had died and I discovered that moisture had got past the rubber gasket and caused corrosion on the PCB. The PCB looks more complicated than the motherboard in my Dell desktop!

    It seems that the firmware can be reprogrammed by the end user but I have never dared to tinker with the multiplicity of parameters that I can summon up in the software and change if so desired! As an end user all I need to know is that each and every microinverter is working and supplying a sensible looking amount of wattage to the generation meter.

    Having read your post about the flattening of the output curve with your system I decided to take a closer look at the log files that the software makes for each inverter and use the data therein to do some graph plotting. It is not as sophisticated as your system as the data is recorded as txt files which I then have to export to EXCEL to work on. The microinverters send info to the Gateway and then to the PC every minute which amounts to quite a mass of data every day. I deliberately collected data on the last almost completely blue-sky day we had, which was on the 7th. Using Excel I calculated the half-hourly average DC output from the PV panels and the AC output from each inverter and then used these data to plot the graphs. I was surprised to see that every microinverter followed the same curve regardless of its position on the roof despite the fact that the array is SE facing - I would have expected to see some differences as the sun traverses west and the angle of irradiation changes. Attached please see the plot from one inverter, M1. The graph confirms what I knew anecdotally that the maximum output per panel is around 150 watts AC, equating to 3kW for the whole array of 21 panels.

    As I suspected there is a marked drop-off in output after about 2.30pm as the sun has moved to the SW by then. I think the dip at about 12 noon must have been caused by clouds, as all 21 inverters show the same trend. It begs the question, would it be worth moving some of the panels to the SW facing roof to capture the afternoon/evening sun? Alternatively I could leave the existing array alone and add new PV panels on the SW roof as the system as it is never gets close to a 4kWp output but then I would fall foul of the FIT rules by making a modification to the system!
    • CommentAuthorphiledge
    • CommentTime4 days ago
     
    I think that if you split your array to be part SE and part SW youll not generate anything extra but will just spread the generating time over a longer period. Youll likely not gain anything in annual generation but will have a lower level of power over a longer period. That may be useful depending on your usage/time of usage.

    Looking at your graph you hit a peak of around 170watt about 12 which shows your panels are able to generate almost to spec. Is this the time of day when the sun is square on to your panels?? Its just a guess but maybe with 10 year old panels they dont tolerate less than optimum conditions as well as newer panels do?? Also dont forget that panel output is quoted at standard conditions that allow comparison between panels but may not equate to the real world output. Our JASolar panel spec sheet quotes the rated output of 250watts at standard conditions but also gives an output at "normal operating conditions" which 183watts!!!
    •  
      CommentAuthordjh
    • CommentTime4 days ago
     
    Posted By: philedgeOur JASolar panel spec sheet quotes the rated output of 250watts at standard conditions but also gives an output at "normal operating conditions" which 183watts!!!

    Our Sunmodule Plus SW 250 mono black data sheet has two performance specifications:

    (1) PERFORMANCE UNDER STANDARD TEST CONDITIONS (STC)* *STC: 1000W/m2, 25°C, AM 1.5
    Maximum power P max 250 Wp

    (2) PERFORMANCE AT 800 W/m2, NOCT, AM 1.5
    Minor reduction in efficiency under partial load conditions at 25°C: at 200 W/m2, 100% (+/-2%) of the STC efficiency (1000 W/m2) is achieved
    Maximum power P max 183.3 Wp

    Sufficiently similar to your figures that I suspect a similar reason. :bigsmile:
    • CommentAuthorJeff B
    • CommentTime4 days ago edited
     
    Posted By: philedgeI think that if you split your array to be part SE and part SW youll not generate anything extra but will just spread the generating time over a longer period. Youll likely not gain anything in annual generation but will have a lower level of power over a longer period. That may be useful depending on your usage/time of usage.

    Looking at your graph you hit a peak of around 170watt about 12 which shows your panels are able to generate almost to spec. Is this the time of day when the sun is square on to your panels?? Its just a guess but maybe with 10 year old panels they dont tolerate less than optimum conditions as well as newer panels do?? Also dont forget that panel output is quoted at standard conditions that allow comparison between panels but may not equate to the real world output. Our JASolar panel spec sheet quotes the rated output of 250watts at standard conditions but also gives an output at "normal operating conditions" which 183watts!!!


    Yes, between 12 noon and 1.30pm (at this time of year) is indeed when the sun is "square on" to the PV panels.
    PV panels are known to produce less output over time so yes this is probably a factor in my case as the panels are just over 10 years old now.

    I think you are right about moving some of the panels, as you say all that will happen is that the system will generate less in the morning and more in the afternoon, with a net balance exactly the same as now. Added to that there would be a cost associated with that work and I'm pretty sure OFGEM would not be happy if the arrangement did manage to generate more power as I am on the highest tariff rate!

    Edited note: I have Shuco PV panels and the spec says 80% of nominal output after 25 years.

    Also of course the panels are definitely affected by temperature. The output on a sunny day in early spring will be higher than a sunny day in June or July. My microinverters log the temperature too so it would be possible to maybe try to correlate the data rather than just rely on anecdotal evidence.
    • CommentAuthorJeff B
    • CommentTime4 days ago
     
    Posted By: djh
    Posted By: philedgeOur JASolar panel spec sheet quotes the rated output of 250watts at standard conditions but also gives an output at "normal operating conditions" which 183watts!!!

    Our Sunmodule Plus SW 250 mono black data sheet has two performance specifications:

    (1) PERFORMANCE UNDER STANDARD TEST CONDITIONS (STC)* *STC: 1000W/m2, 25°C, AM 1.5
    Maximum power P max 250 Wp

    (2) PERFORMANCE AT 800 W/m2, NOCT, AM 1.5
    Minor reduction in efficiency under partial load conditions at 25°C: at 200 W/m2, 100% (+/-2%) of the STC efficiency (1000 W/m2) is achieved
    Maximum power P max 183.3 Wp

    Sufficiently similar to your figures that I suspect a similar reason.http:///newforum/extensions/Vanillacons/smilies/standard/bigsmile.gif" alt=":bigsmile:" title=":bigsmile:" >


    See also my reply to philedge above.

    I am glad that I took the time to extract the data from the log files, spurred on to do so following your publication of your graphs as I now have a handle on exactly how the system is performing. I don't appear to have the "capping" effect that you have as the outputs are struggling to meet the specification anyway.

    Interestingly I found that one (1 out of the 21) of the microinverters is behaving erratically in that it generates happily for a time then occasionally goes into an alarm mode where the output drops to zero for a variable amount of time (typically 10 - 20 minutes) and then carries on as normal! I would never have seen this had I not done in an depth analysis of the data in the log files! The system will flag up an alarm on the screen display but unless I am watching my PC screen every minute of the day I would not know.
  2.  
    Coming with not much prior knowledge here, so maybe a dumb question: how much is 'normal' for loss of DC power in conversion to AC? Jeff is losing about 10% of generation, across the whole range of powers, about 15W per panel at peak, assuming the measurements are accurate.

    That would be 300W across the whole array. Do the inverters have good heat sinks? Sounds like 45⁰C could easily be exceeded...

    G98 requires the installer to configure all the inverter settings to limit the total output from the house at 3.68kW. They should attach a printout of the protection settings to the DNO notification, the settings are supposed to be locked so the householder cannot modify them. Perhaps Jeff could get someone to check his configuration is still set up to best effect?

    The 'clipping' at 3.68kW should presumably be done by tweaking the MPPT, rather than by dumping the excess power as heat, so the DC-to-AC losses in Jeff's graph would not be due to clipping.

    The clipping function is supposed to be distributed with all the inverters clipping equally, rather than controlled centrally with signals sent out based on each string's performance, as that communication route would be difficult to type-approve.

    If V2G takes off, it sounds like everyone will need a G99 as the total export capacity will be >>3.68kW, with or without PV. I wonder if during the G99 process the existing PV inverter configuration could be tweaked, to increase the PV output to 4kW or whatever the panels can do, assuming the inverter can deal thermally with a bit more power?
    • CommentAuthorbhommels
    • CommentTime3 days ago
     
    Posted By: WillInAberdeenComing with not much prior knowledge here, so maybe a dumb question: how much is 'normal' for loss of DC power in conversion to AC? Jeff is losing about 10% of generation, across the whole range of powers, about 15W per panel at peak, assuming the measurements are accurate.

    That would be 300W across the whole array. Do the inverters have good heat sinks? Sounds like 45⁰C could easily be exceeded...

    I think the DC-AC conversion is what is quoted as "efficiency" which depends on the total voltage the inverter sees, see yet another attached plot for my SB3000HF. I expect this to be more or less the same for any (micro)inverter, give or take a few tenths of a percent.
      SMAeff.jpg
    •  
      CommentAuthordjh
    • CommentTime3 days ago
     
    Posted By: WillInAberdeenThe clipping function is supposed to be distributed with all the inverters clipping equally, rather than controlled centrally with signals sent out based on each string's performance, as that communication route would be difficult to type-approve.

    Do you have a link to wherever you read that, please? I'd like to understand exactly what the rules say. I'm having trouble how 'all the inverters clipping equally' is compatible with maximising power output up to the limit whilst also accounting for different shading conditions on different panels. I'm particularly interested in the microinverter case of course, as I have 16 inverters.

    FWIW, under Efficiency the data sheet lists three figures:

    EN 50530 (EU) efficiency 95.7%

    Static MPPT efficiency (weighted, reference EN50530) 99.6%

    Dynamic MPPT efficiency (fast irradiation changes,
    reference EN50530) 99.3%

    I don't know what they mean :cry:
  3.  
    Yes, I thought that too, if one string is shaded or failed then ideally the clipping on another string should be reduced.

    Would be even better if the export current from the house were measured, after household consumption had been taken off the, and the inverters adjusted accordingly.

    Requirements are in G98. It wants any safety communication between units to be over manufacturer-supplied plug/socket interconnection cables, as site-made connections cannot be type approved, or some such wording.

    https://www.google.com/search?q=ena+g98

    Jeff's inverter seems to be loosing a few % more power in conversion than those other examples.
   
The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   
Logout    

© Green Building Press