S.C.E verification testing

I am currently testing outputs on an active two meg. system in Fontana, CA. Looking for information on the best way to measure amperage readings on individual strings of pv cells. Cells are six panels 72v , 1.03 amp output.

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: S.C.E verification testing

    Are you looking for just debugging and functional testing--so using a standard DC current clamp is "good enough"? Fluke, and others, make such probes. Assuming you have access to a single exposed wire that you can clamp too. If not--the answer may not be so easy.

    Or are you trying to validate power output and need the accuracy of a current shunt and a lab calibrated four wire current/voltage meter? DC current probes tend not to be very accurate. Plus, to measure power (what is usually the "important part") you need the string voltage at the same time (and solar irradiation levels and cell temperature and wind speed--because that is constantly changing)--so you are poking around some high voltage/high power circuits for each string in a 2 MW system--pretty time consuming.

    And then if the second option--I would ask exactly why you would need to measure the "internals" of a field deployed system (down to the string level) instead of doing "component level" verification in the "lab"?

    To a degree--a reference source (such as a calibrated panel+plus appropriate hardware) setup in the sun then compared to the output of the system as a whole (or at least each inverter subsection) using the internal or the external calibrated power meter logging setup.

    Sorry to be so questioning about your needs... From your initial question--I am not sure exactly what you are looking for.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: S.C.E verification testing

    If you just need the DC clamp-on meter, I have one and am in SoCal. I don't pass by Fontana too often, but one of the guys who works for me goes out to San Bernardino a lot, maybe he could stop by and let you use it. If you need fancy lab equipment, good luck, but I don't have any.
  • E.Pair
    E.Pair Registered Users Posts: 5
    Re: S.C.E verification testing
    BB. wrote: »
    Are you looking for just debugging and functional testing--so using a standard DC current clamp is "good enough"? Fluke, and others, make such probes. Assuming you have access to a single exposed wire that you can clamp too. If not--the answer may not be so easy.

    Or are you trying to validate power output and need the accuracy of a current shunt and a lab calibrated four wire current/voltage meter? DC current probes tend not to be very accurate. Plus, to measure power (what is usually the "important part") you need the string voltage at the same time (and solar irradiation levels and cell temperature and wind speed--because that is constantly changing)--so you are poking around some high voltage/high power circuits for each string in a 2 MW system--pretty time consuming.

    And then if the second option--I would ask exactly why you would need to measure the "internals" of a field deployed system (down to the string level) instead of doing "component level" verification in the "lab"?

    To a degree--a reference source (such as a calibrated panel+plus appropriate hardware) setup in the sun then compared to the output of the system as a whole (or at least each inverter subsection) using the internal or the external calibrated power meter logging setup.

    Sorry to be so questioning about your needs... From your initial question--I am not sure exactly what you are looking for.

    -Bill

    Bill yes we are trying validate actual output. we have used and do have a Fluke Dc clamp on amp probe this method was not accurate enough. we are currrently using a Fluke multi meter testing each string at combier boxes with an open circuit condition.not a short circuit. Edison Int. Questions validity.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: S.C.E verification testing

    If you are trying to validate each string is functional... then you really need to test with a load that will put the I*V curve in the proper operational area...

    Isc shows electrical connectivity. Isc shows that (roughly) all cells in series are producing voltage. These two tests will probably catch 80-90% of the common failures--but they do not tell you with 100% accuracy that the panels are working correctly.

    But Pmp=Imp*Vmp is the only way to garrantee that the panels are 100% functional.

    Imp is based both on current drawn, the amount of sunlight, and a little bit on temperature.

    Vmp is is based on current and cell temperature (and a bit on sunlight levels).

    To get a good idea of all of the interactions, here is a nice data sheet with the graphs drawn to show the major relationships (you can pretty take the graphs and change the scales to match your panel's ratings--the physics remain the same).

    My two cents--you have several ways to go here...
    1. Measure the string performance individually. Measure the array's Voltage and Current against some sort of known load (may need an electronic load where you can set some sort of profile). And, at the same time, you will have to log the current conditions with a reference panel or string.
    2. With a DC current probe, measure all of the paralleled strings and make sure that each is supplying 1/n of the current and that Voltage is somewhere near Vmp. Assuming that you are looking for a "bad" string--this is actually quit quick and accurate (enough).
    Problem is that almost any field method you choose is going to have inaccuracies--Variable sun angle, dust, smoke, humidity, wind, panel variation, aging, equipment inaccuracies, etc.

    It would be unreasonable to expect lab level accuracy / validation in a field test. Too many variables. However, it is reasonable to expect the customer to have some way of knowing the system is performing at 100% of expectations.

    Instead of me guessing (and causing confusion everywhere :roll:)--what is it that your customer wants measured/verified and to what level of accuracy? Does the system have some sort of monitoring built in (per inverter, per installation, to what level of accuracy, etc.)? Is there some sort of reference (solar irradiation monitor, reference panel/system, etc.) too? Is there a known/guaranteed relationship between the 2 MW system and the Reference (corrected for temperature, wind, etc.).?

    In the end, you may need to get an "expert" in solar design/qualification to setup your qualification/ongoing quality assurance for your systems to make the customer happy.,

    -Bill

    PS: The inaccuracies of the DC current clamp is probably on order of the "error bars" of solar irradiation and panel temperature (plus other variables)--if those are not taken into account too.

    Knowing one parameter accurately is not enough.

    Also, what is the specs. of each array... The 6 cells 72 volts 1.03 amps does not make sense to me.... I don't think you are doing that huge array with 75 watt panels are you?
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • E.Pair
    E.Pair Registered Users Posts: 5
    Re: S.C.E verification testing
    BB. wrote: »
    If you are trying to validate each string is functional... then you really need to test with a load that will put the I*V curve in the proper operational area...

    Isc shows electrical connectivity. Isc shows that (roughly) all cells in series are producing voltage. These two tests will probably catch 80-90% of the common failures--but they do not tell you with 100% accuracy that the panels are working correctly.

    But Pmp=Imp*Vmp is the only way to garrantee that the panels are 100% functional.

    Imp is based both on current drawn, the amount of sunlight, and a little bit on temperature.

    Vmp is is based on current and cell temperature (and a bit on sunlight levels).

    To get a good idea of all of the interactions, here is a nice data sheet with the graphs drawn to show the major relationships (you can pretty take the graphs and change the scales to match your panel's ratings--the physics remain the same).

    My two cents--you have several ways to go here...
    1. Measure the string performance individually. Measure the array's Voltage and Current against some sort of known load (may need an electronic load where you can set some sort of profile). And, at the same time, you will have to log the current conditions with a reference panel or string.
    2. With a DC current probe, measure all of the paralleled strings and make sure that each is supplying 1/n of the current and that Voltage is somewhere near Vmp. Assuming that you are looking for a "bad" string--this is actually quit quick and accurate (enough).
    Problem is that almost any field method you choose is going to have inaccuracies--Variable sun angle, dust, smoke, humidity, wind, panel variation, aging, equipment inaccuracies, etc.

    It would be unreasonable to expect lab level accuracy / validation in a field test. Too many variables. However, it is reasonable to expect the customer to have some way of knowing the system is performing at 100% of expectations.

    Instead of me guessing (and causing confusion everywhere :roll:)--what is it that your customer wants measured/verified and to what level of accuracy? Does the system have some sort of monitoring built in (per inverter, per installation, to what level of accuracy, etc.)? Is there some sort of reference (solar irradiation monitor, reference panel/system, etc.) too? Is there a known/guaranteed relationship between the 2 MW system and the Reference (corrected for temperature, wind, etc.).?

    In the end, you may need to get an "expert" in solar design/qualification to setup your qualification/ongoing quality assurance for your systems to make the customer happy.,

    -Bill

    PS: The inaccuracies of the DC current clamp is probably on order of the "error bars" of solar irradiation and panel temperature (plus other variables)--if those are not taken into account too.

    Knowing one parameter accurately is not enough.

    Also, what is the specs. of each array... The 6 cells 72 volts 1.03 amps does not make sense to me.... I don't think you are doing that huge array with 75 watt panels are you?

    Bill . First thank for your help and response. This is a very large array.The six panels create one string. their are 5700 strings in this array, they feed 4 SatCon
    500kw inverters. The inverters are fed from combiner boxes with 24 strings per box. I am currently in the prcess of recording data on data sheets that include time,cell temp,wind ,weather conditions,lightmeter,total voltage ,string Open circuit voltage,total amperage,string amperage. I am just look for confirmation that the dc amperage testing method is accurate. thanks
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: S.C.E verification testing

    Are the string combiners like the "Smart sub-combiners from SatCom" which monitor string parameters (2% accuracy for current)--or something else that may or may not have current shunts/monitoring built in?

    How much accuracy are they looking for when measuring the DC current?

    A Fluke I30S AC-DC current clamp (there is both an I30 and an I30S model--the "S" may be low current-higher current accuracy, but I could not find the "S" model on the Fluke Website). Somewhere around $500 or so for ~2% overall accuracy (excluding the DVM error).

    About as accurate as I have seen--but I would suggest you get more information from Fluke on their offerings. My big problem with DC current clamps has been that they "drift" from zero pretty quickly and it makes taking accurate measurements a pain.

    -Bill

    By the way, are you inputting all of this data into spread sheets so that you can "normalize" all of the data that has been collected over many days of work (and under varying conditions) so that you can demonstrate that everything is running at 100% of specifications?
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: S.C.E verification testing

    Wow, that's some array! Too bad they didn't think of adding shunts into the combiner boxes at the install time. Add that to the Lessons Learned file
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • E.Pair
    E.Pair Registered Users Posts: 5
    Re: S.C.E verification testing
    mike90045 wrote: »
    Wow, that's some array! Too bad they didn't think of adding shunts into the combiner boxes at the install time. Add that to the Lessons Learned file

    Its all about money. The system has a Dranetz total metering modbus at the substation. However at 1000$each for the sub-combiner boxes it was outside the design budget considering their 240 sub-combiners. The issue is the helth of each string and identification for trouble shooting purposes.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: S.C.E verification testing
    E.Pair wrote: »
    Its all about money. The system has a Dranetz total metering modbus at the substation. However at 1000$each for the sub-combiner boxes it was outside the design budget considering their 240 sub-combiners. The issue is the helth of each string and identification for trouble shooting purposes.

    That's what I love about penny pinching employers, Never enough $$ to do it right, but always enough $$ to do it over.

    So to have a crew roam the array, checking amps, and logging it, and then compiling the data, to find a bad string, that costs less then an automated system, or even a manual system to just "jack in" to each combiner box and read it's amps on it's internal shunt.

    I'd think the inverter's health monitoring system, would show 20 inverters at 36.7KW, and one at 35.2KW, that would at least get you to the right ballpark of where to start looking for bird deposits on panels.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: S.C.E verification testing

    The problem is there are 4x 500 kW inverters on 5,700 strings...

    You can have 1% string failures (57 failed strings) across the entire system and not even be able to measure the loss in power output (assuming a 2% typical measurement accuracy). Throw in the variation in solar irradiation and temperature/wind--it would probably take 5-10% string failures to notice a problem using a central monitoring system.

    Finding a bad string manually with a current clamp is probably not difficult... There will be 23 strings with ~X amp output +/- 10%, and one with near zero output. But--you would have to know there is a problem (no automatic logging?) to even go out and start the measurements by popping open 60 sub-combiners and checking 24 array strings (per inverter).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • E.Pair
    E.Pair Registered Users Posts: 5
    Re: S.C.E verification testing
    mike90045 wrote: »
    That's what I love about penny pinching employers, Never enough $$ to do it right, but always enough $$ to do it over.

    So to have a crew roam the array, checking amps, and logging it, and then compiling the data, to find a bad string, that costs less then an automated system, or even a manual system to just "jack in" to each combiner box and read it's amps on it's internal shunt.

    I'd think the inverter's health monitoring system, would show 20 inverters at 36.7KW, and one at 35.2KW, that would at least get you to the right ballpark of where to start looking for bird deposits on panels.
    From E.Pair This is more about politics than money Southern Cal Edison. is the owner operator of this install. They already know the system is operating at 40% below design specs.Their are roughly 70 broken panels,one years worth of dirt build up. and maybe 20 string failures. The bottom line is this site is a prototype.
    Edison plans to build systems to produce 500megs within the next few years,two others are currently under construction in this area. I work for a third party preforming verfication we simply looking to make suggestions for improvements, in testing, monitoring, and installation. thank you for your input