Theory vs. reality

Desert Rat
Desert Rat Solar Expert Posts: 147 ✭✭✭
in Off Grid Solar & Battery Systems #1
I'm working with an Outback MX60 MPPT controller which displays PV voltage and amperage. If I multiply these two to get charging watts, what percentage of nominal total array watts should I expect to see? Assume maximum output, i.e. bulk stage & full sun @ midday.

And another question: With a series string of modules, can I measure the Isc of each individual module without disconnecting anything?
· Share on Twitter

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #2
    Re: Theory vs. reality

    This is one of those questions where the answer will be all over the place in reality. The "typical" difference between array Watts and maximum output Watts is 77%. In other words, a 1000 Watt array on an MPPT controller will have a maximum output to the batteries of 770 Watts.

    But there are so many things that will vary this that you should not expect an absolute value. The amount of power the batteries need is the first variable, as you will not see maximum output under full illumination unless the system demands it. Panel temperature is another big variable; from one day to the next they may be able to dissipate more/less heat and thus give more/less power with all other factors equal (which they rarely are). I enjoy the high elevation advantage; my array + controller runs around 82% efficient on average. :D

    No, there is no way to measure the Isc of a panel in a string. It has to be disconnected.
    · Share on Twitter
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    #3
    Re: Theory vs. reality

    that does vary and it depends on many factors. i would guess at about 77% of the stc rating, but it can be above and below that with most probably being below.
    · Share on Twitter
  • Photowhit
    Photowhit Solar Expert Posts: 6,006 ✭✭✭✭✭
    #4
    Re: Theory vs. reality

    Many/most? panels give a NOCT (Nominal Operating Cell Temperature) value. Which is a good place to start on estimating your true output. As others have said there are too many factors to give any other value. The NOCT value gives you something to compare pane to panel, I think california has some rating system based on true output as well, I don't want to hunt around to find it now, but perhaps someone knows Their system.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
    · Share on Twitter
  • Desert Rat
    Desert Rat Solar Expert Posts: 147 ✭✭✭
    #5
    Re: Theory vs. reality
    No, there is no way to measure the Isc of a panel in a string. It has to be disconnected.

    I should have been more specific with my question. I would disconnect the string from the system, then touch the ammeter probes to + and - of one panel. Would that not measure the output of that one panel only? I don't see that it would be necessary to completely isolate the panel from the string as long as the string is disconnected from the system.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #6
    Re: Theory vs. reality
    Desert Rat wrote: »
    I should have been more specific with my question. I would disconnect the string from the system, then touch the ammeter probes to + and - of one panel. Would that not measure the output of that one panel only? I don't see that it would be necessary to completely isolate the panel from the string as long as the string is disconnected from the system.

    If the panel's wires are disconnected from the others it is isolated (electrically) from the system. No need to physically remove it as there's no other electrical inter-action.
    · Share on Twitter
  • Desert Rat
    Desert Rat Solar Expert Posts: 147 ✭✭✭
    #7
    Re: Theory vs. reality

    I'm talking about disconnecting the string from the system, but leaving the series wiring in place between the panels. Can I then measure the output of each panel separately?
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #8
    Re: Theory vs. reality
    Desert Rat wrote: »
    I'm talking about disconnecting the string from the system, but leaving the series wiring in place between the panels. Can I then measure the output of each panel separately?

    Well you certain should disconnect the string as you don't want any current flowing when you disconnect the individual panels to test them for Isc.
    Since you need to measure current from one panel's (+) to its (-) with them shorted together, there really isn't any other way to do it. You have to have access to both sides in order to short them to create the current flow from the panel to be measured.

    The only thing you could measure from each individual panel with them all connected in series would be its Voc (which isn't of much value for testing) and even then you'd need some way of accessing the wiring of the panel. So again you'd be down to pulling the connectors apart.
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,613 admin
    #9
    Re: Theory vs. reality

    Yes, you can measure the series connected panels--But it does lead to some very interesting questions. It is possible to have one good panels and two bad panels in series then measure "good" Isc--Only the good panel will be supplying current, the others will be in diode bypass and not supply anything.

    For a quick check--Say you have 9 panels. 3 in series and the three strings paralleled at the Combiner box. In full sun with the charge controller in "bulk" (maximum charge rate into the battery bank--You may need to turn on some loads to draw maximum current from the array), now measure the current in each string... That should be Imp (for that amount of sunlight).

    If all panels are good, then your Vamp-array should be near specifications (or at least Vbatt-charging--if a PWM controller)--And all panels pretty much have to be good to deliver Imp per string. If one or more strings read low (compared to the others)--Start checking for poor electrical connections and, if needed, disconnect each panel in the string and do your Voc/Isc measurements to see if you can find a bad panel.

    In general, paralleled solar panel strings should all read just about identical current (in same sun/temperature/measured at about the same time). What that current will be--Imp is proportional to the amount of sun light (and loads of the charge controller). More or less, Imp measured will be around 80% to 100% of book value in full noontime sun... But in random day to day measurements, you cannot really guess that you have a bad panel unless the current is really low (say 50% or less in "full sun") due to variations in solar radiation (moisture/dust in air/on panels, sea level vs mountains, etc.). A panel that reads at 75% may be good or may be bad--difficult to tell without lab gear.

    But if you have identical parallel strings--it is is very easy to test against parallel strings--It is unlikely that all will fail exactly the same way at the same time.

    Voc/Isc measurements in full are OK to weed about 80-90% of the "problem panels" (say you are looking at buying used panels). But measurements under operational conditions (in actual array with loads) is the 100% gold standard test.

    I would use a DC current Clamp meter under full sun and with a discharged battery bank (or turn on a bunch of loads, plug in an electric heater) to check out an operational system. You only need to take them apart if you find an issue with one or more strings.

    You can do similar testing if you don't have a DC current clamp meter by turning off each strings circuit breaker (if installed)--one at a time and read the current from the charge controller/battery monitor as each string is dropped out. You could do similar if there are fuses installed--But there is a risk, fuses are not designed to be disconnected under load and it is not recommended (you can throw a blanket over the array each time you open a fuse or disconnect a wire/plug--But that is getting to be a pain).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • Desert Rat
    Desert Rat Solar Expert Posts: 147 ✭✭✭
    #10
    Re: Theory vs. reality

    Cariboocoot, I believe you're thinking of panels with MC4 connectors, and I'm thinking of panels with junction boxes. So say I have the junction box covers off and have access to the + and - terminals. I connect my ammeter to those two terminals on one panel. Yes, there are wires connected to both terminals going elsewhere, but no circuit is completed except between the panel being measured and my ammeter.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #11
    Re: Theory vs. reality
    Desert Rat wrote: »
    Cariboocoot, I believe you're thinking of panels with MC4 connectors, and I'm thinking of panels with junction boxes. So say I have the junction box covers off and have access to the + and - terminals. I connect my ammeter to those two terminals on one panel. Yes, there are wires connected to both terminals going elsewhere, but no circuit is completed except between the panel being measured and my ammeter.

    Yes, I am thinking about panels whose only electrical connection is the pig tails. Most panels these days don't have accessible junction boxes.

    With the string disconnected, run an ammeter from the (+) to the (-) and you will get an Isc on that panel only.
    · Share on Twitter
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    #12
    Re: Theory vs. reality

    just remember that it will give quite a spark if the pv is in place with full sun on it. it might be better on your meter leads to have a cheap switch inline with your meter leads and once you're connected, then flip on the switch. if you don't do this and have access to the face of each pv you could then throw a blanket or something over it until the meter is connected. once connected then pull the blanket. put the blanket back on before disconnecting and don't leave it connected with it's full isc for long as the meter wires could heat up.

    all that assumes a wired meter as the clamp on type will take away that hazard. also all of this assumes no load across the entire string and you can do it.
    · Share on Twitter
  • Desert Rat
    Desert Rat Solar Expert Posts: 147 ✭✭✭
    #13
    Re: Theory vs. reality

    Thanks for all your help, guys. I do also have a clamp meter.
    · Share on Twitter

Categories