Jump to content

Welcome to Smart Home Forum by FIBARO

Dear Guest,

 

as you can notice parts of Smart Home Forum by FIBARO is not available for you. You have to register in order to view all content and post in our community. Don't worry! Registration is a simple free process that requires minimal information for you to sign up. Become a part of of Smart Home Forum by FIBARO by creating an account.

 

As a member you can:

  •     Start new topics and reply to others
  •     Follow topics and users to get email updates
  •     Get your own profile page and make new friends
  •     Send personal messages
  •     ... and learn a lot about our system!

 

Regards,

Smart Home Forum by FIBARO Team


Please add support for Aeotec ZW095 (Gen 5 HEM)


AutoFrank
 Share

Recommended Posts

@morpheus75  I agree with @AutoFrank. An alternative explanation might be this. Not updating values is something that happens to my RGBW in input mode as well, randomly, this makes me think that it's an HC issue, not really device specific. I usually end up reconfiguring the device (soft or hard). What happens if you manually poll the device? I'd say: if it doesn't update, then go for the "reconfigure" approach...

Link to comment
Share on other sites

  • Topic Author
  • Some data from overnight

     

    Please login or register to see this attachment.

     

    Also my Gen 2 and Gen 5 are tracking closely on the same cable

    The Gen 2 seems to react a bit slower

    The time series is 1 sample per minute

     

    Please login or register to see this attachment.

     

    Link to comment
    Share on other sites

    Thanks for including the (k)Wh graph. I'm crunching the numbers and I'm writing a (nice) comment. It'll take some time, but I'll get there. ;-)

     

     

    Link to comment
    Share on other sites

  • Topic Author
  • 1 minute ago, petergebruers said:

    Thanks for including the (k)Wh graph. I'm crunching the numbers and I'm writing a (nice) comment. It'll take some time, but I'll get there. ;-)

     

     

     

    Thanks @petergebruers

     

    I'll run the numbers again later when I get some more sunlight on my panels and see how the story evolves

    I think I'll have enough data by 3 or 4 pm and will try and change p2 = 3 (negative only) on one Grid clamp to see if can get any negative reporting

     

    Link to comment
    Share on other sites

    2 minutes ago, AutoFrank said:

     

    Thanks @petergebruers

     

    I'll run the numbers again later when I get some more sunlight on my panels and see how the story evolves

    I think I'll have enough data by 3 or 4 pm and will try and change p2 = 3 (negative only) on one Grid clamp to see if can get any negative reporting

     

     

    OK. I'll wait a bit then. I have a partial story, but I think it makes more sense if I get more date.

    Oh, please change the legend of the "kWh" graph lines to "Wh". Now it's 1000 x too high. So then you have "W" and "Wh" on the same graph.

    Link to comment
    Share on other sites

    ive just added the 'energy' reports to the graphs so hopefully will start seeing data.

    today is an overcast cloudy rainy day so will not see much solar.

    the energy report for billing was way off the mark so i changed parameter 2 to 1, the algebraic sum so will see what i get in terms of instantaseous readings and cumulative enegy readings.

    the solar power and house power now seem to be working. from some reason the global polling doesnt seem to be working on my HC2, as i had a UBS which showed breach but when i know it wasnt so i manually polled it and it went back to safe.

    the last time it did breach was a few hours earlier and poll time is set to 10 mins, so that lead me to believe that my HEM were not being polled either.

    i have set the polling to periodic intervals within the device itself to 120s and now seem to be getting more accurate results.

    lets wait and see

     

    also @AutoFrank what do you use to make graphs of your data? with thingspeak can only do 1 graph per data, i liked the multiple lines on 1 graph visually see the usage/generation.

    Link to comment
    Share on other sites

  • Topic Author
  • @petergebruers

    @morpheus75

     

    Some more data and a 'stretch' theory

     

    All the data but Wh instead of KWh

     

    Please login or register to see this attachment.

     

    secondary axis used for Wh to better show variations in both trends

     

    Please login or register to see this attachment.

     

    and here is the stretch

     

    It got sunny here and and I noticed that when grid (gen 5 device) went to zero my Grid 2 device (Gen 2) continue to read .. so I though that what if these readings were really zero so I flipped the Grid 2 dataset to negative when the Grid was zero and here is the result

     

    Please login or register to see this attachment.

     

    interesting ....

     

     

    Link to comment
    Share on other sites

    I've had a quick look at the top diagram and I'd say it correctly shows "grid consumption". In the afternoon, you can see it levels and at times even slopes downwards (so pv output exceeds house needs -> your meter "turns backwards"). Is this this what you expected? Do you agree/disagree with me?

     

    Edit: I mean: the green line.

    Edited by petergebruers
    Link to comment
    Share on other sites

    @AutoFrank 

    link to new thread re thingspeak

     

    also data seems to be ok for me now. the energy graphs do seem to add up:

    house usage is 7 units

    solar gen is 5 units

    billed is 2 units

     

    also the instantaneous power usage graphs are matching up aswell now:

    current house usage 1000w

    current solar gen 300w

    current billed 700w

     

    here some graphs again:

     

    Please login or register to see this link.

     

     

     

     

     

     

    Link to comment
    Share on other sites

  • Topic Author
  • 56 minutes ago, morpheus75 said:

    @AutoFrank 

    link to new thread re thingspeak

     

    also data seems to be ok for me now. the energy graphs do seem to add up:

    house usage is 7 units

    solar gen is 5 units

    billed is 2 units

     

    also the instantaneous power usage graphs are matching up aswell now:

    current house usage 1000w

    current solar gen 300w

    current billed 700w

     

    here some graphs again:

     

    Please login or register to see this link.

     

     

     

     

     

     

     

    Thats good news @morpheus75

    How are your HEM's setup 

    How do you can calculate total consumption ?

    What happens if you generate more than you consume

     

    I'm still not sure...

    its a very good day here and this is what I'm seeing from our PV app

     

    Please login or register to see this attachment.

     

     

    and my own data (two axis v 1 axis) 

    mistake in the chart p2 = 1

     

    Please login or register to see this attachment.

     

    Please login or register to see this attachment.

    All seems okay in the positive side but not sure when it turns negative

     

    its still sunny here so I'm going to change p2 = 3 (negative only) and see if it records negative space (excess back in to grid) better

    I'll let it run for the rest if the night

     

     

     

     

     

     

     

     

     

     

     

     

    3 hours ago, petergebruers said:

    downwards (so pv output exceeds house needs -> your meter "turns backwards"). Is this this what you expected? Do you agree/disagree with me?

     

    @petergebruers

     

    yes, I think that makes sense ....

    I can see now why you wanted the Wh values as well....

     

    I can see that also on the latest chart I attached..

     

    Its a pity that we can't see that at the watt level. I don't think its a sampling issue as for this this afternoon it stayed close to zero (and presumably below) for a long period of time....

     

    ... but I think we are making progress 

     

     

     

     

    Edited by AutoFrank
    Link to comment
    Share on other sites

    @AutoFrank I know you posted more recent data, but my reply to post #42 was almost finished so I'll post it anyway. Unfortunately I don't have time today to analyse your more recent graphs...

     

    At first glance, anyone coming here might conclude that the Wh graph is difficult to analyze, while the W graph seems more ... natural, easy, wysiwyg? I'll try to demonstrate why I think it is quite the opposite, at least when you are talking about power consumption, solar production and injecting power into the grid. And comparing all of those.

    It's just a matter of learning to read the Wh graph...

     

    Because (k)W = instantaneous = "power at a point in time" it jumps like crazy. For instance, when you're sleeping, they grey line jumps up and down all the time, for instance at 22:42 you have a 3700 W peak. Did you turn on the espresso machine, or did you boil one last kettle to make some hot soup?  You can also see some cyclic pattern during the night, about every 30 minutes W varies between 400 and 700. Interesting, what is happening here? Between 7:00 and 8:30 the grey graph shows a lot of activity and a higher baseline consumption. All very plausible, but I'll now compare this to the Wh graph.

     

    The Wh graph (yellow line, for "grid" consumption) is some much more... relaxing, don't you think so? And yet, EVERYTHING is there! Let's find the "peak" at 22:42. Look at the slope of the yellow line. I can't see a dent, so it's as if this very powerful kettle (let's assume you were boiling a cup of soup) didn't consume any power, at least not at this scale. Is this possible? Assume the kettle needs 1/2 minute @ 3000 W to boil the water. That's 1/120 part of an hour, so you need 3000 W / 120 = 25 Wh. When you look at the graph, you'll appreciate that there is not enough vertical resolution to show this little bump. Well, I counted the resolution and I think you can just see a 20 (ballpark) W difference. I don't think it is bad, that the large peak isn't that visible on the yellow graph, if you are trying to economize. The grey graph may have suggested that you have high power consumption, and that you need to do something about it. But the yellow graph tells you that you won't save any money, because the power consumption was high, but short and insignificant compared to the other parts of the graph. Note: if you sample the grey one faster than once per minute, say once every 10 seconds, you'd get a better idea of the duration of the peak, but then you would have more peaks and troughs, so it wouldn't look prettier. Note 2: it's not because you don't see the 25 Wh on the yellow graph, that the meter didn't see them. The meter will have added to it's total Wh, it's the data processing (and the graph resolution) that makes it 'disappear'.

     

    Lets look at the yellow line between 22:00 and 6:45. I laid a piece of paper on it and it's almost flat. So this gives me the idea that we can calculate how much standby power is needed. The Wh graph goes from 4000 Wh to 4500 Wh = 500 Wh. The time difference is 7,66 so that is 65 watt. So what's in it? The fridge, Fibaro modules (0,5 W each), pump, some lights, stereo, TV... Let's move on to the next time segment.

     

    Between 7:00 and 8:30 the grey graph shows a lot of activity, and again I prefer the yellow line to reason about power consumption. You see a 15 minute sharp increase. Then, the line continues almost like before, to start rising again at 7:55. After that, the slope is almost horizontal, but that'll be my next point. Now let's look at the solar graph. Blue line = solar produced. I estimate you get 200 W in the morning. The Wh line starts to move after 6:30. Solar output doesn't jump like the power consumption of the house.

     

    One remaining question is... How does the solar output influence the total? If we had the three graphs (grid, solar, house) then we would be able to do the math. So I'll have a look at the next set of data, maybe tomorrow.

    Link to comment
    Share on other sites

  • Topic Author
  • 1 hour ago, petergebruers said:

    @AutoFrank I know you posted more recent data, but my reply to post #42 was almost finished so I'll post it anyway. Unfortunately I don't have time today to analyse your more recent graphs...

     

    At first glance, anyone coming here might conclude that the Wh graph is difficult to analyze, while the W graph seems more ... natural, easy, wysiwyg? I'll try to demonstrate why I think it is quite the opposite, at least when you are talking about power consumption, solar production and injecting power into the grid. And comparing all of those.

    It's just a matter of learning to read the Wh graph...

     

    Because (k)W = instantaneous = "power at a point in time" it jumps like crazy. For instance, when you're sleeping, they grey line jumps up and down all the time, for instance at 22:42 you have a 3700 W peak. Did you turn on the espresso machine, or did you boil one last kettle to make some hot soup?  You can also see some cyclic pattern during the night, about every 30 minutes W varies between 400 and 700. Interesting, what is happening here? Between 7:00 and 8:30 the grey graph shows a lot of activity and a higher baseline consumption. All very plausible, but I'll now compare this to the Wh graph.

     

    The Wh graph (yellow line, for "grid" consumption) is some much more... relaxing, don't you think so? And yet, EVERYTHING is there! Let's find the "peak" at 22:42. Look at the slope of the yellow line. I can't see a dent, so it's as if this very powerful kettle (let's assume you were boiling a cup of soup) didn't consume any power, at least not at this scale. Is this possible? Assume the kettle needs 1/2 minute @ 3000 W to boil the water. That's 1/120 part of an hour, so you need 3000 W / 120 = 25 Wh. When you look at the graph, you'll appreciate that there is not enough vertical resolution to show this little bump. Well, I counted the resolution and I think you can just see a 20 (ballpark) W difference. I don't think it is bad, that the large peak isn't that visible on the yellow graph, if you are trying to economize. The grey graph may have suggested that you have high power consumption, and that you need to do something about it. But the yellow graph tells you that you won't save any money, because the power consumption was high, but short and insignificant compared to the other parts of the graph. Note: if you sample the grey one faster than once per minute, say once every 10 seconds, you'd get a better idea of the duration of the peak, but then you would have more peaks and troughs, so it wouldn't look prettier. Note 2: it's not because you don't see the 25 Wh on the yellow graph, that the meter didn't see them. The meter will have added to it's total Wh, it's the data processing (and the graph resolution) that makes it 'disappear'.

     

    Lets look at the yellow line between 22:00 and 6:45. I laid a piece of paper on it and it's almost flat. So this gives me the idea that we can calculate how much standby power is needed. The Wh graph goes from 4000 Wh to 4500 Wh = 500 Wh. The time difference is 7,66 so that is 65 watt. So what's in it? The fridge, Fibaro modules (0,5 W each), pump, some lights, stereo, TV... Let's move on to the next time segment.

     

    Between 7:00 and 8:30 the grey graph shows a lot of activity, and again I prefer the yellow line to reason about power consumption. You see a 15 minute sharp increase. Then, the line continues almost like before, to start rising again at 7:55. After that, the slope is almost horizontal, but that'll be my next point. Now let's look at the solar graph. Blue line = solar produced. I estimate you get 200 W in the morning. The Wh line starts to move after 6:30. Solar output doesn't jump like the power consumption of the house.

     

    One remaining question is... How does the solar output influence the total? If we had the three graphs (grid, solar, house) then we would be able to do the math. So I'll have a look at the next set of data, maybe tomorrow.

     

    Firstly @petergebruers, thanks for taking the time to analyse and apply your knowledge and experience

    It is much appreciated 

     

    The summary above makes a lot of sense and matches the house activity and I can now see why Wh is more valuable. 

    Ill increase the resolution on the data and see if we can extract more insight...

     

    I had two reads and think i'll need to print out the graphs and work through it...

     

    I have one question and perhaps @morpheus75 can weigh in here.

     

    With the current positioning of my clamp I'm not sure if I can measure the the required data .

    Solar in - This one is okay, its on the AC side of the inverter and reading accuratly

     

    My concern if whether the other two clamps are on the main feed into the house (the grid side of the main isolator in my consumer unit) can read the Grid consumption, excess back to grid - I'm not sure about...I think I can extrapolate what the house is consuming from the data if it is accurate

    @morpheus75  clamps are much better positioned but I don't have way of positioning mine in the same way 

     

    Do you think that I can measure what I am looking for with the position of my clamps

     

    I'll also take a look at the data with p2 = 3 and see if I can see any negative data.....

     

    Thanks again to you both in trying to help understand this one ...

     

    Edit

    Question....

    When I set p2 = 3 (showing negative data only) would I expect to see negative data or just positive data that has the same absolute value as the negative equivalent ..... 

     

     

    Edited by AutoFrank
    Link to comment
    Share on other sites

    @AutoFrank

    My HEM have been setup to measure 2 main values. actual house consumption irrespective of source. and actual solar contribution.

    the 2 clamps on each of these Solar and House provide that data.

    the billing clamp is the problem one. 

    also my clamps are on the actual meter tails 25mm cables with the outer grey sheath taken off as these clamps were quite small.

    the energy data jumps all over the place currently set on parameter '2' ie not to measure reverse flow, but it never goes to zero.

    i have checked my meter when i am producing more energy than i am using and would have expected a zero billing reading but no.

    also my Power meters on all 3 dont seem to be adding up properly and sometimes just freeze even though the the V and A are ok.

    so im going to soft reconfigure them now, before i changed the polling interval on the device itself.

     

    Also setup wise whenever i installed solar system i always installed them as per my orig diagram so it was independent of the house system and also for this reason of measurement. without seeing your full pics of setup cannot advise how to change position of clamps.

    theoretically how you have positioned your clamps to measure on grid feed in both directions means calculations are needed and 1 clamp should always be reading zero as you have set your parameter to read positive flow only.

     

    i may get rid of my billing clamp and just use the house and solar clamps only and also then just use a calculation for the billing amount. also probably use the V and A values only as they seem to be updating properly and perform another calculation to get power value. 

    i will continue to monitor to find best way.

     

    @petergebruers nice explanation of the graphs..

     

    *** UPDATE ***

    before soft config i have decided to reset the stats on the HEM also taken meter reads aswell.

    now i can compare measured kWh units against actual metered units and see what we get....

    in theory house units used will be equal to solar gen units PLUS billed units.

    lets see

    ive reset data on thingspeak aswell to match

    Edited by morpheus75
    update
    Link to comment
    Share on other sites

    2 hours ago, morpheus75 said:

    (...)

    the billing clamp is the problem one. 

    also my clamps are on the actual meter tails 25mm cables with the outer grey sheath taken off as these clamps were quite small.

    (...)

    i may get rid of my billing clamp and just use the house and solar clamps only and also then just use a calculation for the billing amount. also probably use the V and A values only as they seem to be updating properly and perform another calculation to get power value. (...)

     

    @petergebruers nice explanation of the graphs..

     

    Thank you, I'm glad you liked my post.

     

    If you say 25 mm and "grey outer", I'd say you are clamping on a cable that contains more than one wire. Unless you mean you have aluminium 25 mm2 single wires? Nah, that's possible but that size would be used for a cable coming from the street, not a wire from your meter to the RCDs and breakers... Between meter and distribution board, you'd typically have 10 mm2 copper in Europe (63 A main breaker).

     

    Anyway, the clamp has to fit around a single wire. The clamp has to close properly, otherwise the meter will show lower values than the actual current. Even one-tenth of a millimeter will cause the meter to be out of spec.

     

    If you still read a few tens of mA when there is no current, maybe you have a stray magnetic field nearby. Could happen if you have an iron core transformer a few cm away from the clamp (and this will also be orientation dependent).

     

    If you do Amps x Volts you will get "apparent power", not "power" (what they charge money for). If you test this with a halogen lamp, both values will be equal. If you test with a Philips Warm Glow LED lamp, Volts x Amps will be about 1.10 times actual power. If you do this with a very capacitive load (some cheap LED lamps come to mind) the V x A value will be twice the power...

    Link to comment
    Share on other sites

    @petergebruers

    in UK the incoming mains spec is 25mm single core cables.

    from my diagram all cables on the Henley Block are single core 25mm double insulated cables.

    so the clamps are on the single Live cables.

    we have 100A as our incoming fuse but most households never go above 60A as in the fuseboards we use 63A RCD.

    i understand about what you mean about apparent power and so will wait and see over the next day or so what my readings are.....

    Link to comment
    Share on other sites

    1 minute ago, morpheus75 said:

    @petergebruers

    in UK the incoming mains spec is 25mm single core cables.

    from my diagram all cables on the Henley Block are single core 25mm double insulated cables.

    so the clamps are on the single Live cables.

    we have 100A as our incoming fuse but most households never go above 60A as in the fuseboards we use 63A RCD.

    i understand about what you mean about apparent power and so will wait and see over the next day or so what my readings are.....

     

    I see, yes, 100 A. I see. So 25mm squared, single cable. Sorry, I forgot about the schematic you posted!

     

    I try to learn about UK stuff, and this is my youtube hero regarding regulations and equipment: John Ward

     

    Please login or register to see this link.

     

     

    Link to comment
    Share on other sites

  • Topic Author
  • 2 hours ago, petergebruers said:

     

    I see, yes, 100 A. I see. So 25mm squared, single cable. Sorry, I forgot about the schematic you posted!

     

    I try to learn about UK stuff, and this is my youtube hero regarding regulations and equipment: John Ward

     

    Please login or register to see this link.

     

     

    @petergebruers

    @morpheus75

     

    I'm still logging the data with p2=3 on one of my incoming clamps. I had a quick look at the raw data and I think it's less useful than p2=1 but I'll need to graph later.

     

    I'll probably end up tracking the KWh for relative increase or decrease over a period of time (~15 mins) and use that as an indicator that we are exporting electricity to the grid. 

     

    Will let you know 

     

     

    Link to comment
    Share on other sites

    i extracted data from thingspeak and amended the billed kWh readings to normal rather than the 100kWh

    general trend is matching but there are a some anomalies.

    we can see that the house usage goes up and the billed usage follows the same trend but it sits below depending on how much solar is generated so shows me that solar is doing its job.

    on the kWh chart solar unit is matching house units and billed units showing as zero which is as expected. 

    gonna have to play with the settings to get consistent data reporting.

    on billing clamp parameter 2 is 2.

    the others are default settings

    Please login or register to see this attachment.

    Please login or register to see this attachment.

    Link to comment
    Share on other sites

  • Topic Author
  • @morpheus75, @petergebruers

     

    I'm still collecting the data and the mist is starting to clear... a little

    p2 = 3 doesn't do anything for my setup

    I took a look at the physical arrangement and I won't be able to get the clean setup that @morpheus75 has as my PV was retro fitted and the power come in through my garage feed..

    I think I can extrapolate from the data I can get though which is good news

     

     

    Here is a snapshot from today (my my Wh data for the grid feed was all the same data value so I left it out) 

     

    Please login or register to see this attachment.

     

    Even though we are spreading the use of the hungry appliances we still have an excess

    Curiously though I noticed that my panels seem to be topping out at 2.2 kW which is odd since I have 10 x 260 watt panels and my installed expected 2.6 or 28 peaks on a good day like today.

    I did check my PV app and it's confirmed that the best its done is 2.2 kW

    .... so this is a question for my installer

    I also overlayed the PV app chart to mine and its a bit smoother with not many peaks so either the data collection is either every 5 mins or they smooth the graph for display purposes. There is a MQTT api that I haven't tapped in to yet,

     

    I'm back with p2=1 for the next 24 hrs to gather more data

    I also took note of my meter reading and reset my HEM units to start to correlate to actual billed values

     

    I'm also going to take a look at my overnight usage ...my equipment racks consume about 120w overnight so I can probably cut some of that and we have 2 freezers and 3 fridges running.  We have no natural gas or oil so it's all electricity usage so I've also been looking at our heat pump activity. Currently the UFH is off at this stage but the system is setup for 24 hr hot water priority which means that as soon as we use any hot water the heat pump replenishes it. (so hot water used for showers in the morning is replenished immediately) I can adjust this so it only heats at given times of the day which I'll look at next.. 

     

    I'll update again tomorrow

    ...

     

    _f

     

     

    Edit

    I found the PV data export, is sampled every 10 mins and seems close to my data collected every minute

    Please login or register to see this attachment.

     

     

    Edited by AutoFrank
    Link to comment
    Share on other sites

    Join the conversation

    You can post now and register later. If you have an account, sign in now to post with your account.

    Guest
    Reply to this topic...

    ×   Pasted as rich text.   Paste as plain text instead

      Only 75 emoji are allowed.

    ×   Your link has been automatically embedded.   Display as a link instead

    ×   Your previous content has been restored.   Clear editor

    ×   You cannot paste images directly. Upload or insert images from URL.

     Share

    ×
    ×
    • Create New...