KWh meter for measuring generated electricity?

[tonnyc]

First, some background. I live in Indonesia and am not a licensed electrician. That said, I do have some basic knowledge. Enough to know that what I am doing is not up to code. I am going to install a 100 Wp system with a plug-and-play grid-tie inverter plugged into an outlet. I repeat, yes, I am aware that this is not safe. I am doing this for research purpose. I need to know how much electricity is generated and whether it is economically feasible to proceed with a proper grid-tie set-up later. I calculate that with a 220 Volt grid voltage and a 100 Wp panel, the maximum current generated will be less that 0.5 Ampere, which is well within the safety margin of the house's cable and outlet.<br><br>Because I need to figure out whether it's economically feasible to proceed with a proper (and larger) set up, I need to measure and record the amount of electricity generated by the system. On youtube I see people using a small digital multi-function KWh meter that basically looks like this:<br>

<br><br>From what I know, meters like that are usually used to measure the amount of electricity flowing from the outlet to whatever device that is plugged into the meter. Meanwhile, I want to measure the amount of electricity from the solar power system going into the outlet. So basically the reverse. Can I still use that sort of digital meter or is there a special meter needed?

PS: And if you are wondering why I'm asking a bunch of UK electricians instead of local ones, well, PV grid-tie installations in Indonesia is still at its infancy and there's no Indonesian equivalent of this forum (yet) and second, you guys are electricians.

 

[TedM]

Grid-tie inverters operate within specific minimum/maximum input DC voltages and I am not aware of one small enough to run off a single 100Wp panel. Probably the minimalist setup would be a 250Wp panel with a suitably matched microinverter. This would also allow you to measure the generation via the microinverter's reporting interface.

 

[tonnyc]

Grid-tie inverters operate within specific minimum/maximum input DC voltages and I am not aware of one small enough to run off a single 100Wp panel. Probably the minimalist setup would be a 250Wp panel with a suitably matched microinverter. This would also allow you to measure the generation via the microinverter's reporting interface.

Thanks. The grid-tie inverter I am looking at claims to have a range of 10.5-28V and a max power of 500W. The 100Wp polycrystalline solar panel has a max voltage of 17.1V. Since I plan on using just a single panel for testing and getting the hands-on experience, I am not using a microinverter yet, but that will change when I start using multiple panels. Since I am not using a microinverter yet, I have to find an alternative way of measuring and recording output. Can the energy meter I was asking about do it?

 

[Sunshine777]

No idea if those meters can work in both directions - try it and let us know! :biggrin:

Best case it'll work, middle ground it won't but will still be useable in a normal installation and worst case you'll fry it.

You could also rig up two cables, one with sockets at both ends and one with plugs at both ends. Then you can wire the meter in the correct direction but that plug to plug cable is dangerous in the wrong hands.

And not trying to be argumentative but... what exactly are you looking to prove? Grid tie systems are used all over the world - it would seem there's no reason that a grid tie system won't work in Indonesia. Or have I missed something?

 

[tonnyc]

I probably will give it a try and let you guys know.

I am not looking to prove anything. I am trying to measure actual output of the system as a whole and as separate components. There is a wide range of claims of efficiency both from the manufacturers and from actual users. Some inverters claim to be 98% efficient but some users report only 50% efficiency in actual use. Some others report 70% or 85%. Solar panel manufacturers claim a rated power of 100 Wp in ideal conditions, but some testimonials says expect only 75% or 50% or some other number. Actual insolation in Indonesia averages 1000 W/m2 with a noon temperature reaching 35C. I ought to be getting 93 watt out of that 100 Wp on a clear midday, but I want to check whether I actually get it.

Eventually all the data basically makes a business case on whether I should bother with solar power next year or if I should wait until the price of electricity goes up some more and the price of solar power systems go down some more.

As far as I can tell, nobody in Indonesia has done this sort of measurement. Most local sellers make assumptions by saying "well, you are rated for so much Watt-peak power, and we have X hours of mid-day-equivalent, so you will get Y KWH" and that's it. And the figure for X varies a lot from a conservative 5 hours to an unrealistic 10 hours. I can't rely on that, so I better spent a little money and time measuring actual output for now before deciding on whether to invest significant amount of money on a larger system.

 

[Sunshine777]

Thanks for explaining where you're coming from with your request.

Have a look on pvoutput.org - there seem to be some systems in Indonesia that post their data. Indonesia | Map

Also have a look at PV potential estimation utility and enter your data for a prediction.

 

[tonnyc]

No idea if those meters can work in both directions - try it and let us know! :biggrin:

Best case it'll work, middle ground it won't but will still be useable in a normal installation and worst case you'll fry it.

You could also rig up two cables, one with sockets at both ends and one with plugs at both ends. Then you can wire the meter in the correct direction but that plug to plug cable is dangerous in the wrong hands.

And not trying to be argumentative but... what exactly are you looking to prove? Grid tie systems are used all over the world - it would seem there's no reason that a grid tie system won't work in Indonesia. Or have I missed something?

This is an old thread but I'm resurrecting it to report my findings.

First, I am able to use a WANF wattmeter to measure the output of the inverter without needing to make any modifications. Apparently not all AC wattmeter can do it, but the one I use doesn't care where the current comes from. I get voltage, wattage, and total Watthour produced.

My result was a disappointing average of 0.32 KWh per day over the month of February-March. This is due to two reasons. One is the rainy season. Most days were cloudy, and there were plenty of days where the inverter wasn't working during midday due to insufficient input. The other reason is inverter inefficiency. I measured (using a different meter) a peak DC wattage off approximately 95 Watt, slightly less than the panel's 100 Wp rating, and yet the inverter was only outputting 75 Watt at best. My inverter is rated for 500 Watt and yet I only have a single 100 Wp panel, so this may have something to do with it. I also noticed that the panel and inverter is producing less output during peak insolation (i.e., clear midday). This is probably due to heat. The weather here can easily reach 35C during midday.

I will give it some more time to see how things perform during the dry season, which starts this April.

EDIT: What I use. Link to photo at Upload the image directly to the thread.
Upload the image directly to the thread
The inverter plug goes into the wattmeter's socket.

 

[TedM]

Everything you report is to be expected.

Running the inverter at much lower (<20%) than rated Watts will give very poor efficiency. You may see if your inverter manufacturer publishes any data that shows the varying efficiency with varying Watts. You should be easily achieving >95% efficiency with a well matched panel/inverter setup.

 

[whlphil]

Hi Tonnyc, Sunshine777 recommendation of using the JRC website is a good one, the database now extends to Asia and its a reliable set of figures that will take into account your location from both an irradiation and climate perspective. The figures will be conservative and can be relied upon.

Ive been pricing some projects in Thailand recently and have used the figures as a quick measure, typically its predicting between 1300 and 1400kWh per kW of panels installed. From checks Ive done with people that have existing systems installed these figures would appear to be 10% conservative with 1500kWh/kWp being about average. These are AC kWh's delivered so allow for all the system losses.

The big thing that you have to account for is the time of day that those hours are generated and the time of day you use your electric. Also how much you can change your habits to match the PV production hours. If you size your system at 1 to 1 thats 1kW of inverter for 1kW of panel you should see an overall system efficiency of about 80-85% but thats based on available irradiance and the above figures have all that allowed already.

With regards to manufacturers you generally get what you pay for, go cheap and you run the risk of problems (some people are happy to put up with that others aren't, its your choice) one thing you need to remember is that you are at the moment looking at a performance figures for a snap shot in time and with PV you really need to be thinking long term averages.

I remember a German guy saying to me back when I started PV 'why are you trying to reinvent the wheel, PV just works why don't you just get on with it' of course I took no notice but he was right those overall average figures are achieved and it does just work. Of course the time of generation issue is still here but batteries really are just around the corner (less than 3 years)!